Galectins: Methods and Protocols

Methods in
Molecular Biology 2442
Sean R. Stowell
Connie M. Arthur
Richard D. Cummings Editors
Galectins
Methods and Protocols
Second Edition
METHODS IN MOLECULAR BIOLOGY
Series Editor
John M. Walker
School of Life and Medical Sciences
University of Hertfordshire
Hatfield, Hertfordshire, UK
For further volumes:

http://www.springer.com/series/7651
For over 35 years, biological scientists have come to rely on the research protocols and
methodologies in the critically acclaimed Methods in Molecular Biology series. The series was
the first to introduce the step-by-step protocols approach that has become the standard in all
biomedical protocol publishing. Each protocol is provided in readily-reproducible step-by-
step fashion, opening with an introductory overview, a list of the materials and reagents
needed to complete the experiment, and followed by a detailed procedure that is supported
with a helpful notes section offering tips and tricks of the trade as well as troubleshooting
advice. These hallmark features were introduced by series editor Dr. John Walker and
constitute the key ingredient in each and every volume of the Methods in Molecular Biology
series. Tested and trusted, comprehensive and reliable, all protocols from the series are
indexed in PubMed.
Galectins
Methods and Protocols
Second Edition
Edited by
Sean R. Stowell and Connie M. Arthur

Joint Program in Transfusion Medicine Brigham and Women’s Hospital, Harvard Medical School,
Boston, MA, USA
Richard D. Cummings
Department of Surgery Beth Israel Deaconess Medical Center National Center for Functional Glycomics,
Harvard Medical School, Boston, MA, USA
Editors
Sean R. Stowell Connie M. Arthur
Joint Program in Transfusion Medicine Joint Program in Transfusion Medicine
Brigham and Women’s Hospital Brigham and Women’s Hospital
Harvard Medical School Harvard Medical School
Boston, MA, USA Boston, MA, USA
Richard D. Cummings
Department of Surgery Beth Israel
Deaconess Medical Center National
Center for Functional Glycomics
Harvard Medical School
Boston, MA, USA
ISSN 1064-3745 ISSN 1940-6029 (electronic)

Methods in Molecular Biology
ISBN 978-1-0716-2054-0 ISBN 978-1-0716-2055-7 (eBook)
https://doi.org/10.1007/978-1-0716-2055-7
© Springer Science+Business Media, LLC, part of Springer Nature 2015, 2022

This work is subject to copyright. All rights are reserved by the Publisher, whether the whole or part of the material is
concerned, specifically the rights of translation, reprinting, reuse of illustrations, recitation, broadcasting, reproduction
on microfilms or in any other physical way, and transmission or information storage and retrieval, electronic adaptation,
computer software, or by similar or dissimilar methodology now known or hereafter developed.
The use of general descriptive names, registered names, trademarks, service marks, etc. in this publication does not imply,
even in the absence of a specific statement, that such names are exempt from the relevant protective laws and regulations
and therefore free for general use.
The publisher, the authors and the editors are safe to assume that the advice and information in this book are believed to
be true and accurate at the date of publication. Neither the publisher nor the authors or the editors give a warranty,
expressed or implied, with respect to the material contained herein or for any errors or omissions that may have been
made. The publisher remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
This Humana imprint is published by the registered company Springer Science+Business Media, LLC, part of Springer
Nature.
The registered company address is: 1 New York Plaza, New York, NY 10004, U.S.A.
Preface
Galectins: Methods and Protocols, second edition, is an update of the first book solely dedicated
to methodological approaches designed to study galectin function. The galectin family
represents one of the most pleiotropic families, with individual members having been
implicated in various aspects of nearly every biological process described, from RNA splicing
to complex regulatory circuits that orchestrate adaptive immunity. Given the diverse roles of
galectins in a variety of biological systems, studying these glycan-binding proteins often
requires the assimilation of diverse technical skills to fully appreciate their biological func-
tion. Their nearly ubiquitous expression and ability to bind highly modifiable carbohydrate
ligands, in addition to a variety of other regulatory proteins, allows these glycan-binding
proteins (GBPs) to possess the capacity to regulate a wide variety of biological processes.
Individual chapters are dedicated to examining salient features of galectin functions. Written
in the successful Methods in Molecular Biology series format, chapters include introductions
to their respective topics, lists of the necessary materials and reagents, step-by-step, readily
reproducible protocols, and notes on troubleshooting and avoiding known pitfalls.
Authoritative and easily accessible, Galectins: Methods and Protocols seeks to serve both
professionals and novices with a useful framework when examining galectin function for
many years to come.
Boston, MA, USA Sean R. Stowell

Connie M. Arthur
Richard D. Cummings
v
Acknowledgments
Sean R. Stowell, Connie M. Arthur, Richard D. Cummings

We would like to thank Shang-Chuen Wu, Anu Paul, Kashyap Patel, Hans Verkerke, and
Alex D. Ho for critical reading of chapters and feedback on the entire Galectins: Methods and
Protocols, second edition.
vii
Contents
Preface . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . v
Acknowledgments. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . vii
Contributors. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . xiii
1 Galectins: An Ancient Family of Carbohydrate Binding Proteins
with Modern Functions. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1
Hans Verkerke, Marcelo Dias-Baruffi, Richard D. Cummings,
Connie M. Arthur, and Sean R. Stowell
2 Cloning, Expression, and Purification of Galectins
for In Vitro Studies . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 41
Paul A. Poland, Carol L. Kinlough, and Rebecca P. Hughey
3 Purification of Recombinant Galectins from Different Species
Using Distinct Affinity Chromatography Methods . . . . . . . . . . . . . . . . . . . . . . . . . . 55
Anu Paul, Shang-Chuen Wu, Kashyap R. Patel,
Alex D. Ho, Jerry William Lynn Allen, Hans Verkerke,
4 Alkylation of Galectin-1 with Iodoacetamide and Mass Spectrometric
Mapping of the Sites of Incorporation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 75
Shang-Chuen Wu, Anu Paul, Richard D. Cummings,
Christa L. Feasley, Connie M. Arthur, and Sean R. Stowell
5 Rapid Detection and Purification of Galectin-3 by the Capture
and Release (CaRe) Method. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 89
Priyanka D. Kadav, Jared L. Edwards, Jessica Krycia,
Purnima Bandyopadhyay, and Tarun K. Dam
6 Introducing 77Se NMR Spectroscopy to Analyzing
Galectin–Ligand Interaction. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 105
Mária Raics, István Timári, Lászlo Szilágyi,
Hans-Joachim Gabius, and Katalin E. Kövér
7 Evaluation of Galectin Binding by Surface Plasmon Resonance . . . . . . . . . . . . . . . 125
Padmaja Mehta-D’souza
8 Revealing the Identity of Human Galectin-3
as a Glycosaminoglycan-Binding Protein . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 137
Jared L. Edwards, Priyanka D. Kadav,
9 Examining Galectin Binding Specificity Using Glycan Microarrays . . . . . . . . . . . . 151
Sean R. Stowell, Lilian C. Rodrigues, Marcelo Dias-Baruffi,
Richard D. Cummings, and Connie M. Arthur
10 Mechanism of Mucin Recognition by Lectins:
A Thermodynamic Study . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 169
Tarun K. Dam, Jared L. Edwards,
Priyanka D. Kadav, and C. Fred Brewer
ix
x Contents
11 Examination of Whole-Cell Galectin Binding by Solid Phase

and Flow Cytometric Analysis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 187
Anne Lepp€ a nen, Connie M. Arthur,
Sean R. Stowell, and Richard D. Cummings
12 Dissecting Context-Specific Galectin Binding
Using Glycoengineered Cell Libraries . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 205
Mathias Ingemann Nielsen and Hans H. Wandall
13 Method for Identifying Galectin Ligands on Lymphocyte
Membrane Glycoproteins . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 215
Kashyap R. Patel, Adam W. Barb, and Sean R. Stowell
14 Transformation of Agrocybe cylindracea Galectin
into αGalNAc-Specific Lectin. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 233
Dan Hu and Jun Hirabayashi
15 What Happens If a Human Galectin Enters
the Endoplasmic Reticulum? . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 247
Tanja J. Kutzner, Alonso M. Higuero, Martina Süßmair,
Michael Hingar, Herbert Kaltner, Ingo Lindner, Jürgen Kopitz,
José Abad-Rodrı́guez, Dietmar Reusch, and Hans-Joachim Gabius
16 Investigation of Galectins in Frozen Tissue and Mammalian
Cell Culture Using Confocal Miccroscopy . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 289
Daniel Giuliano Cerri, Lilian Cataldi Rodrigues,
Marise Lopes Fermino, Marcelo Papoti, Richard D. Cummings,
Sean R. Stowell, and Marcelo Dias-Baruffi
17 Exploring the Galectin Network by Light
and Fluorescence Microscopy . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 307
Gabriel Garcı́a Caballero, Joachim C. Manning, Adele Gabba,
Donella Beckwith, Forrest G. FitzGerald, Tanja J. Kutzner,
Anna-Kristin Ludwig, Herbert Kaltner, Paul V. Murphy,
Mare Cudic, and Hans-Joachim Gabius
18 Molecular Imaging for In Vivo Tracking and Detection
of Galectin Binding Partners . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 339
Thais Canassa De Leo, Sofia Nascimento dos Santos,
Emerson Soares Bernardes, Richard D. Cummings,
19 Visualization of Cytosolic Galectin Accumulation Around
Damaged Vesicles and Organelles . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 353
Ming-Hsiang Hong, I-Chun Weng, Fang-Yen Li, and Fu-Tong Liu
20 Transcytosis of Galectin-3 in Mouse Intestine . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 367
Alena Ivashenka, Christian Wunder, Valerie Chambon,
Estelle Dransart, Ludger Johannes, and Massiullah Shafaq-Zadah
21 Evaluating the Role of Galectins in Clathrin-Independent
Endocytosis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 391
Mohit P. Mathew, Julie G. Donaldson, and John A. Hanover
22 Examination of Galectin-3 Recruitment into Multivesicular Bodies
for Exosomal Secretion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 413
Sebastian B€ a nfer, Sophie Kutscher, and Ralf Jacob
Contents xi
23 Manipulating Galectin Expression in Zebrafish (Danio rerio) . . . . . . . . . . . . . . . . . 425

Chiguang Feng, Mihai Nita-Lazar, Nuria González-Montalbán,
Jingyu Wang, Justin Mancini, Sheng Wang, Chinnarajan Ravindran,
Hafiz Ahmed, and Gerardo R. Vasta
24 Examining Galectin Gene Regulation by Reporter Assays . . . . . . . . . . . . . . . . . . . . 445
Sebastian Schmidt, Herbert Kaltner, and Hans-Joachim Gabius
25 Examining the Impact of Galectin-9 on Latent HIV Transcription . . . . . . . . . . . . 463
Opeyemi S. Adeniji, Leila B. Giron, and Mohamed Abdel-Mohsen
26 Evaluation of the Role of Galectins in Parasite Immunity . . . . . . . . . . . . . . . . . . . . 475
Jaclyn Swan, Dhanasekaran Sakthivel, Travis Beddoe,
Michael Stear, David Piedrafita, and Sarah Preston
27 Evaluation of the Bactericidal Activity of Galectins . . . . . . . . . . . . . . . . . . . . . . . . . . 517
Nourine A. Kamili, Anu Paul, Shang-Chuen Wu,
Marcelo Dias-Baruffi, Richard D. Cummings,
28 Detection of Phosphatidylserine Exposure on Leukocytes
Following Treatment with Human Galectins . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 533
Sean R. Stowell, Marcelo Dias-Baruffi,
29 Detection of Reactive Oxygen Species in Human Neutrophils
Under Various Conditions of Exposure to Galectin . . . . . . . . . . . . . . . . . . . . . . . . . 549
Lilian Cataldi Rodrigues, Daniel Giuliano Cerri,
Cleni M. Marzocchi-Machado, Richard D. Cummings,
30 Analysis of Galectin-Binding Receptors on B Cells . . . . . . . . . . . . . . . . . . . . . . . . . . 565
Asmi Chakraborty, Norhan B. B. Mohammed,
Angela E. Bernasconi, and Charles J. Dimitroff
31 Methods for Assessing the Effects of Galectins on Leukocyte
Trafficking and Clearance . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 581
Hannah L. Law and Dianne Cooper
32 Examination of the Contributions of Maternal/Placental-Derived
Galectin-1 to Pregnancy Outcome . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 603
Sophia Borowski, Nancy Freitag, Iris Urban,
Geert Michel, Gabriela Barrientos, and Sandra M. Blois
33 Method to Study the Role of Galectins in Angiogenesis
In Vivo Using the Chick Chorioallantoic Membrane Assay. . . . . . . . . . . . . . . . . . . 621
Kitty C. M. Castricum and Victor L. J. L. Thijssen
34 Untangling Galectin-Mediated Circuits that Control
Hypoxia-Driven Angiogenesis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 635
Nadia Bannoud, P. Alfredo Garcı́a, Julian Gambarte-Tudela,
Victoria Sundblad, Alejandro J. Cagnoni, Camila A. Bach,
Juan M. Pérez Saez, Ada G. Blidner, Sebastián M. Maller,
Karina V. Mariño, Mariana Salatino, Juan P. Cerliani,
Gabriel A. Rabinovich, and Diego O. Croci
xii Contents
35 Examination of the Role of Galectins and Galectin Inhibitors

in Endothelial Cell Biology . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 655
36 Evaluating Therapeutic Activity of Galectin-1 in Sarcolemma
Repair of Skeletal Muscle . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 663
Mary L. Vallecillo-Zúniga, Matthew Rathgeber, Daniel Poulson,
Braden Kartchner, Jacob Luddington, Hailie Gill, Spencer Hayes,
Matthew Teynor, Caleb S. Stowell, Connie M. Arthur,
Sean R. Stowell, and Pam M. Van Ry
37 Exploring the Role of Galectins in Cancer:
In Vitro and In Vivo Approaches . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 685
Neus Martı́nez-Bosch, Noemı́ Manero, Manero-Rupérez Moreno,
and Pilar Navarro
38 Galectin-3–U1 snRNP Complexes Initiate Splicing Activity
in U1-Depleted Nuclear Extracts . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 713
Patricia G. Voss, Kevin C. Haudek,
Ronald J. Patterson, and John L. Wang
Index . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 727
Contributors
JOSÉ ABAD-RODRÍGUEZ • Membrane Biology and Axonal Repair Laboratory, Hospital

Nacional de Parapléjicos (SESCAM), Toledo, Spain
MOHAMED ABDEL-MOHSEN • Vaccine and Immunotherapy Center, The Wistar Institute,
Philadelphia, PA, USA
OPEYEMI S. ADENIJI • The Wistar Institute, Philadelphia, PA, USA
HAFIZ AHMED • Department of Biochemistry, School of Medicine, Institute of Marine and
Environmental Technology, University of Maryland Baltimore, Baltimore, MD, USA
JERRY WILLIAM LYNN ALLEN • Joint Program in Transfusion Medicine, Department of
Pathology, Brigham and Women’s Hospital, Harvard Medical School, Boston, MA, USA
CONNIE M. ARTHUR • Joint Program in Transfusion Medicine, Department of Pathology,
Brigham and Women’s Hospital, Harvard Medical School, Boston, MA, USA; Harvard
Glycomics Center, Harvard Medical School, Boston, MA, USA
CAMILA A. BACH • Laboratorio de Inmunopatologı́a, Instituto de Biologı́a y Medicina
Experimental (IBYME), Consejo Nacional de Investigaciones Cientı́ficas y Técnicas
(CONICET), Buenos Aires, Argentina
PURNIMA BANDYOPADHYAY • Laboratory of Mechanistic Glycobiology, Department of
Chemistry, Michigan Technological University, Houghton, MI, USA
SEBASTIAN BA€ NFER • Department of Cell Biology and Cell Pathology, Philipps University of
Marburg, Marburg, Germany
NADIA BANNOUD • Instituto de Histologı́a y Embriologı́a de Mendoza (IHEM), Consejo
Nacional de Investigaciones Cientı́ficas y Técnicas (CONICET), Universidad Nacional de
Cuyo, Mendoza, Argentina; Facultad de Ciencias Médicas, Universidad Nacional de
Cuyo, Mendoza, Argentina
ADAM W. BARB • Department of Biochemistry and Molecular Biology, University of Georgia,
Athens, GA, USA; Complex Carbohydrate Research Center, University of Georgia, Athens,
GA, USA
GABRIELA BARRIENTOS • Laboratorio de Medicina Experimental, Hospital Alemán—Consejo
Nacional de Investigaciones Cientı́ficas y Técnicas, Buenos Aires, Argentina
DONELLA BECKWITH • Department of Chemistry and Biochemistry, Florida Atlantic
University, Boca Raton, FL, USA
TRAVIS BEDDOE • Department of Animal, Plant and Soil Science and Centre for Agri
Bioscience (Agri Bio), La Trobe University, Melbourne, VIC, Australia
EMERSON SOARES BERNARDES • Departamento de Radiofarmácia, Instituto de Pesquisas
Energéticas e Nucleares, IPEN/CNEN, São Paulo, SP, Brasil
ANGELA E. BERNASCONI • Department of Translational Medicine, Translational Glycobiology
Institute at FIU, Herbert Wertheim College of Medicine, Florida International University,
Miami, FL, USA
ADA G. BLIDNER • Laboratorio de Inmunopatologı́a, Instituto de Biologı́a y Medicina
SANDRA M. BLOIS • Department of Obstetrics and Fetal Medicine, University Medical Center
Hamburg-Eppendorf, Hamburg, Germany
xiii
xiv Contributors
SOPHIA BOROWSKI • Department of Obstetrics and Fetal Medicine, University Medical

Center Hamburg-Eppendorf, Hamburg, Germany; Experimental and Clinical Research
Center, a Cooperation between the Max Delbrück Center for Molecular Medicine in the
Helmholtz Association, and the Charité—Universit€ atsmedizin Berlin, Berlin, Germany
C. FRED BREWER • Departments of Molecular Pharmacology, and Microbiology and
Immunology, Albert Einstein College of Medicine, Bronx, NY, USA
ALEJANDRO J. CAGNONI • Laboratorio de Glicomica Funcional y Molecular, Instituto de
Biologı́a y Medicina Experimental (IBYME), Consejo Nacional de Investigaciones
Cientı́ficas y Técnicas (CONICET), Buenos Aires, Argentina
KITTY C. M. CASTRICUM • Amsterdam UMC location VUmc, Department of Radiation
Oncology, Cancer Center Amsterdam, Amsterdam, The Netherlands
JUAN P. CERLIANI • Laboratorio de Inmunopatologı́a, Instituto de Biologı́a y Medicina
DANIEL GIULIANO CERRI • Departamento de Análises Clı́nicas, Toxicologicas
e Bromatologicas da Faculdade de Ciências Farmacênuticas de Ribeirão Preto,
Universidade de São Paulo, Ribeirão Preto, São Paulo, Brasil
ASMI CHAKRABORTY • Department of Translational Medicine, Translational Glycobiology
Miami, FL, USA
VALERIE CHAMBON • Cellular and Chemical Biology Unit, Endocytic Trafficking and
Intracellular Delivery Team, U1143 INSERM, UMR3666 CNRS, Institut Curie, PSL
Research University, Paris Cedex, France
DIANNE COOPER • William Harvey Research Institute, Barts and The London School of
Medicine, Queen Mary University of London, London, UK
DIEGO O. CROCI • Instituto de Histologı́a y Embriologı́a de Mendoza (IHEM), Consejo
Cuyo, Mendoza, Argentina; Facultad de Ciencias Exactas y Naturales, Universidad
Nacional de Cuyo, Mendoza, Argentina
MARE CUDIC • Department of Chemistry and Biochemistry, Florida Atlantic University,
Boca Raton, FL, USA
RICHARD D. CUMMINGS • Department of Surgery, National Center for Functional
Glycomics, Beth Israel Deaconess Medical Center, Harvard Medical School, Boston, MA,
USA
TARUN K. DAM • Laboratory of Mechanistic Glycobiology, Department of Chemistry,
Michigan Technological University, Houghton, MI, USA; Health Research Institute,
Michigan Technological University, Houghton, MI, USA
THAIS CANASSA DE LEO • Departamento de Análises Clı́nicas, Toxicologicas e Bromatologicas
da Faculdade de Ciências Farmacêuticas de Ribeirão Preto, Universidade de São Paulo,
Ribeirão Preto, São Paulo, Brasil
MARCELO DIAS-BARUFFI • Department of Clinical Analysis, Toxicological and
Bromatological, School of Pharmaceutical Sciences of Ribeirão Preto, University of São
Paulo, Ribeirão Preto, Brazil
CHARLES J. DIMITROFF • Department of Translational Medicine, Translational Glycobiology
Miami, FL, USA
JULIE G. DONALDSON • National Heart, Lung and Blood Institute, Bethesda, MD, USA
Contributors xv
SOFIA NASCIMENTO DOS SANTOS • Departamento de Radiofarmácia, Instituto de Pesquisas

Energéticas e Nucleares, IPEN/CNEN, São Paulo, SP, Brasil
ESTELLE DRANSART • Cellular and Chemical Biology Unit, Endocytic Trafficking and
JARED L. EDWARDS • Laboratory of Mechanistic Glycobiology, Department of Chemistry,
CHRISTA L. FEASLEY • Cytovance Technologies, Oklahoma City, OK, USA
CHIGUANG FENG • Department of Microbiology and Immunology, Institute of Marine and
MARISE LOPES FERMINO • Departamento de Análises Clı́nicas, Toxicologicas e Bromatologicas
da Faculdade de Ciências Farmacênuticas de Ribeirão Preto, Universidade de São Paulo,
Ribeirão Preto, São Paulo, Brasil
FORREST G. FITZGERALD • Department of Chemistry and Biochemistry, Florida Atlantic
University, Boca Raton, FL, USA
NANCY FREITAG • Department of Obstetrics and Fetal Medicine, University Medical Center
Hamburg-Eppendorf, Hamburg, Germany; Experimental and Clinical Research Center, a
Cooperation between the Max Delbrück Center for Molecular Medicine in the Helmholtz
Association, and the Charité—Universit€ a tsmedizin Berlin, Berlin, Germany; Division of
General Internal and Psychosomatic Medicine, Berlin Institute of Health, Charité—
Universit€atsmedizin Berlin, Corporate Member of Freie Universit€ a t Berlin, Humboldt-
Universit€at zu Berlin, Berlin, Germany
ADELE GABBA • School of Chemistry, National University of Ireland, Galway, Ireland
HANS-JOACHIM GABIUS • Faculty of Veterinary Medicine, Institute of Physiological
Chemistry, Ludwig-Maximilians-University Munich, Munich, Germany
JULIAN GAMBARTE-TUDELA • Instituto de Histologı́a y Embriologı́a de Mendoza (IHEM),
Consejo Nacional de Investigaciones Cientı́ficas y Técnicas (CONICET), Universidad
Nacional de Cuyo, Mendoza, Argentina; Facultad de Ciencias Médicas, Universidad
Nacional de Cuyo, Mendoza, Argentina
P. ALFREDO GARCÍA • Instituto de Histologı́a y Embriologı́a de Mendoza (IHEM), Consejo
Cuyo, Mendoza, Argentina
GABRIEL GARCÍA CABALLERO • Faculty of Veterinary Medicine, Institute of Physiological
Chemistry, Ludwig-Maximilians-University Munich, Munich, Germany
HAILIE GILL • Department of Chemistry and Biochemistry, Brigham Young University,
Provo, UT, USA
LEILA B. GIRON • The Wistar Institute, Philadelphia, PA, USA
NURIA GONZÁLEZ-MONTALBÁN • Department of Microbiology and Immunology, Institute of
Marine and Environmental Technology, University of Maryland Baltimore, Baltimore,
MD, USA
JOHN A. HANOVER • National Institute of Diabetes and Digestive and Kidney Diseases,
Bethesda, MD, USA
KEVIN C. HAUDEK • Department of Biochemistry and Molecular Biology, Michigan State
University, East Lansing, MI, USA
SPENCER HAYES • Department of Chemistry and Biochemistry, Brigham Young University,
Provo, UT, USA
ALONSO M. HIGUERO • Membrane Biology and Axonal Repair Laboratory, Hospital
Nacional de Parapléjicos (SESCAM), Toledo, Spain
xvi Contributors
MICHAEL HINGAR • Pharma Biotech Development Penzberg, Roche Diagnostics GmbH,

Penzberg, Germany
JUN HIRABAYASHI • National Institute of Advanced Industrial Science and Technology,
Ibaraki, Japan; Institute for Glyco-core Research (iGCORE), Nagoya University, Furo-cho,
Chikusa-ku, Nagoya, Japan
ALEX D. HO • Joint Program in Transfusion Medicine, Department of Pathology, Brigham
and Women’s Hospital, Harvard Medical School, Boston, MA, USA
MING-HSIANG HONG • Institute of Biomedical Sciences, Academia Sinica, Taipei, Taiwan
DAN HU • Institute of Traditional Chinese Medicine and Natural Products, College of
Pharmacy, Jinan University, Guangzhou, People’s Republic of China
REBECCA P. HUGHEY • Renal-Electrolyte Division, Department of Medicine, Laboratory of
Epithelial Cell Biology, University of Pittsburgh, Pittsburgh, PA, USA; University of
Pittsburgh School of Medicine, Pittsburgh, PA, USA
ALENA IVASHENKA • Cellular and Chemical Biology Unit, Endocytic Trafficking and
RALF JACOB • Department of Cell Biology and Cell Pathology, Philipps University of
LUDGER JOHANNES • Cellular and Chemical Biology Unit, Endocytic Trafficking and
PRIYANKA D. KADAV • Laboratory of Mechanistic Glycobiology, Department of Chemistry,
HERBERT KALTNER • Faculty of Veterinary Medicine, Institute of Physiological Chemistry,
Ludwig-Maximilians-University Munich, Munich, Germany
NOURINE A. KAMILI • Joint Program in Transfusion Medicine, Department of Pathology,
Brigham and Women’s Hospital, Harvard Medical School, Boston, MA, USA
BRADEN KARTCHNER • Department of Chemistry and Biochemistry, Brigham Young
University, Provo, UT, USA
CAROL L. KINLOUGH • Renal-Electrolyte Division, Department of Medicine, Laboratory of
Epithelial Cell Biology, University of Pittsburgh, Pittsburgh, PA, USA
JÜRGEN KOPITZ • Department of Applied Tumor Biology, Institute of Pathology, Ruprecht-
Karls-University Heidelberg, Heidelberg, Germany
KATALIN E. KÖVÉR • Department of Inorganic and Analytical Chemistry, University of
Debrecen, Debrecen, Hungary; MTA-DE Molecular Recognition and Interaction Research
Group, University of Debrecen, Debrecen, Hungary
JESSICA KRYCIA • Laboratory of Mechanistic Glycobiology, Department of Chemistry,
SOPHIE KUTSCHER • Department of Cell Biology and Cell Pathology, Philipps University of
TANJA J. KUTZNER • Faculty of Veterinary Medicine, Institute of Physiological Chemistry,
HANNAH L. LAW • William Harvey Research Institute, Barts and The London School of
Medicine, Queen Mary University of London, London, UK
ANNE LEPPA€ NEN • Orion Dagnostica, Espoo, Finland
FANG-YEN LI • Institute of Biomedical Sciences, Academia Sinica, Taipei, Taiwan
INGO LINDNER • Pharma Biotech Development Penzberg, Roche Diagnostics GmbH,
Penzberg, Germany
Contributors xvii
FU-TONG LIU • Institute of Biomedical Sciences, Academia Sinica, Taipei, Taiwan

JACOB LUDDINGTON • Department of Chemistry and Biochemistry, Brigham Young
ANNA-KRISTIN LUDWIG • Faculty of Veterinary Medicine, Institute of Physiological Chemistry,
SEBASTIÁN M. MALLER • Laboratorio de Inmunopatologı́a, Instituto de Biologı́a y Medicina
JUSTIN MANCINI • Department of Microbiology and Immunology, Institute of Marine and
NOEMÍ MANERO-RUPÉREZ • Cancer Research Program, Hospital del Mar Medical Research
Institute (IMIM), Unidad Asociada IIBB-CSIC, Barcelona, Spain
JOACHIM C. MANNING • Faculty of Veterinary Medicine, Institute of Physiological Chemistry,
KARINA V. MARIÑO • Laboratorio de Glicomica Funcional y Molecular, Instituto de Biologı́a y
Medicina Experimental (IBYME), Consejo Nacional de Investigaciones Cientı́ficas y Té
cnicas (CONICET), Buenos Aires, Argentina
NEUS MARTÍNEZ-BOSCH • Cancer Research Program, Hospital del Mar Medical Research
Institute (IMIM), Unidad Asociada IIBB-CSIC, Barcelona, Spain
CLENI M. MARZOCCHI-MACHADO • Departamento de Análises Clı́nicas, Toxicologicas
e Bromatologicas da Faculdade de Ciências Farmacêuticas de Ribeirão Preto,
Universidade de São Paulo, Ribeirão Preto, SP, Brasil
MOHIT P. MATHEW • National Institute of Diabetes and Digestive and Kidney Diseases,
Bethesda, MD, USA
PADMAJA MEHTA-D’SOUZA • Oklahoma Medical Research Center, Oklahoma City, OK, USA
GEERT MICHEL • Core Facility Transgene Technology, Charité—Universit€ a tsmedizin Berlin,
Corporate Member of Freie Universit€ at Berlin, Humboldt-Universit€
a t zu Berlin, Berlin,
Germany
NORHAN B. B. MOHAMMED • Department of Translational Medicine, Translational
Glycobiology Institute at FIU, Herbert Wertheim College of Medicine, Florida
International University, Miami, FL, USA
MIREIA MORENO • Cancer Research Program, Hospital del Mar Medical Research Institute
(IMIM), Unidad Asociada IIBB-CSIC, Barcelona, Spain
PAUL V. MURPHY • School of Chemistry, National University of Ireland, Galway, Ireland
PILAR NAVARRO • Cancer Research Program, Hospital del Mar Medical Research Institute
(IMIM), Unidad Asociada IIBB-CSIC, Barcelona, Spain; Institute of Biomedical
Research of Barcelona (IIBB-CSIC), Barcelona, Spain; August Pi i Sunyer Biomedical
Research Institute (IDIBAPS), Barcelona, Spain
MATHIAS INGEMANN NIELSEN • Department of Cellular and Molecular Medicine, Faculty of
Health and Medical Sciences, Copenhagen Center for Glycomics, University of Copenhagen,
Copenhagen, Denmark
MIHAI NITA-LAZAR • Department of Microbiology and Immunology, Institute of Marine and
MARCELO PAPOTI • Escola de Educação Fı́sica e Esporte de Ribeirão Preto, Universidade de
São Paulo, Ribeirão Preto, São Paulo, Brasil
KASHYAP R. PATEL • Joint Program in Transfusion Medicine, Department of Pathology,
xviii Contributors
RONALD J. PATTERSON • Department of Microbiology and Molecular Genetics, Michigan State

ANU PAUL • Joint Program in Transfusion Medicine, Department of Pathology, Brigham
and Women’s Hospital, Harvard Medical School, Boston, MA, USA
JUAN M. PÉREZ SAEZ • Laboratorio de Inmunopatologı́a, Instituto de Biologı́a y Medicina
DAVID PIEDRAFITA • School of Science, Psychology and Sport, Federation University Australia,
Mt Helen, VIC, Australia
PAUL A. POLAND • Renal-Electrolyte Division, Department of Medicine, Laboratory of
Epithelial Cell Biology, University of Pittsburgh, Pittsburgh, PA, USA
DANIEL POULSON • Department of Chemistry and Biochemistry, Brigham Young University,
Provo, UT, USA
SARAH PRESTON • School of Science, Psychology and Sport, Federation University Australia,
Mt Helen, VIC, Australia
GABRIEL A. RABINOVICH • Laboratorio de Inmunopatologı́a, Instituto de Biologı́a y Medicina
(CONICET), Buenos Aires, Argentina; Facultad de Ciencias Exactas y Naturales,
Universidad de Buenos Aires, Buenos Aires, Argentina
MÁRIA RAICS • Department of Inorganic and Analytical Chemistry, University of Debrecen,
Debrecen, Hungary
MATTHEW RATHGEBER • Department of Chemistry and Biochemistry, Brigham Young
CHINNARAJAN RAVINDRAN • Department of Microbiology and Immunology, Institute of
Marine and Environmental Technology, University of Maryland Baltimore, Baltimore,
MD, USA; Department of Marine Biotechnology, National Institute of Oceanography
(CSIR), Dona Paula, Goa, India
DIETMAR REUSCH • Pharma Biotech Development Penzberg, Roche Diagnostics GmbH,
Penzberg, Germany
LILIAN CATALDI RODRIGUES • Departamento de Análises Clı́nicas, Toxicologicas
e Bromatologicas da Faculdade de Ciências Farmacênuticas de Ribeirão Preto,
Universidade de São Paulo, Ribeirão Preto, São Paulo, Brasil
DHANASEKARAN SAKTHIVEL • Department of Biochemistry and Molecular Biology, Monash
University, Clayton, VIC, Australia
MARIANA SALATINO • Laboratorio de Inmunopatologı́a, Instituto de Biologı́a y Medicina
SEBASTIAN SCHMIDT • Institute of Physiological Chemistry, Faculty of Veterinary Medicine,
MASSIULLAH SHAFAQ-ZADAH • Cellular and Chemical Biology Unit, Endocytic Trafficking
and Intracellular Delivery Team, U1143 INSERM, UMR3666 CNRS, Institut Curie,
PSL Research University, Paris Cedex, France
MICHAEL STEAR • Department of Animal, Plant and Soil Science and Centre for Agri
CALEB S. STOWELL • Department of Chemistry and Biochemistry, Brigham Young University,
Provo, UT, USA
SEAN R. STOWELL • Joint Program in Transfusion Medicine, Department of Pathology,
Contributors xix
VICTORIA SUNDBLAD • Laboratorio de Inmunopatologı́a, Instituto de Biologı́a y Medicina

MARTINA SÜßMAIR • Pharma Biotech Development Penzberg, Roche Diagnostics GmbH,
Penzberg, Germany
JACLYN SWAN • Department of Animal, Plant and Soil Science and Centre for Agri
LÁSZLÓ SZILÁGYI • Department of Organic Chemistry, University of Debrecen, Debrecen,
Hungary
MATTHEW TEYNOR • Department of Chemistry and Biochemistry, Brigham Young
VICTOR L. J. L. THIJSSEN • Amsterdam UMC location VUmc, Department of Radiation
Oncology, Cancer Center Amsterdam, Amsterdam, The Netherlands
ISTVÁN TIMÁRI • Department of Inorganic and Analytical Chemistry, University of
Debrecen, Debrecen, Hungary
IRIS URBAN • Core Facility Transgene Technology, Charité—Universit€ a tsmedizin Berlin,
Corporate Member of Freie Universit€ at Berlin, Humboldt-Universit€ a t zu Berlin, Berlin,
Germany
MARY L. VALLECILLO-ZÚNIGA • Department of Chemistry and Biochemistry, Brigham Young
PAM M. VAN RY • Department of Chemistry and Biochemistry, Brigham Young University,
Provo, UT, USA; Department of Biochemistry, Brigham Young University, Provo, UT,
USA
GERARDO R. VASTA • Department of Microbiology and Immunology, Institute of Marine and
HANS VERKERKE • Joint Program in Transfusion Medicine, Department of Pathology,
PATRICIA G. VOSS • Department of Biochemistry and Molecular Biology, Michigan State
HANS H. WANDALL • Department of Cellular and Molecular Medicine, Faculty of Health
and Medical Sciences, Copenhagen Center for Glycomics, University of Copenhagen,
Copenhagen, Denmark
JINGYU WANG • Department of Microbiology and Immunology, Institute of Marine and
JOHN L. WANG • Department of Biochemistry and Molecular Biology, Michigan State
SHENG WANG • Department of Microbiology and Immunology, Institute of Marine and
Environmental Technology, University of Maryland Baltimore, Baltimore, MD, USA;
State Key Laboratory for Biocontrol, School of Marine Sciences, Sun Yat-Sen University,
Guangzhou, People’s Republic of China
I-CHUN WENG • Institute of Biomedical Sciences, Academia Sinica, Taipei, Taiwan
SHANG-CHUEN WU • Joint Program in Transfusion Medicine, Department of Pathology,
Brigham and Women’s Hospital, Harvard Medical School, Boston, MA, USA
CHRISTIAN WUNDER • Cellular and Chemical Biology Unit, Endocytic Trafficking and
Chapter 1
Galectins: An Ancient Family of Carbohydrate Binding

Proteins with Modern Functions
Hans Verkerke, Marcelo Dias-Baruffi, Richard D. Cummings,
Abstract
Galectins are a large family of carbohydrate binding proteins with members in nearly every lineage of
multicellular life. Through tandem and en-mass genome duplications, over 15 known vertebrate galectins
likely evolved from a single common ancestor extant in pre-chordate lineages. While galectins have
divergently evolved numerous functions, some of which do not involve carbohydrate recognition, the
vast majority of the galectins have retained the conserved ability to bind variably modified polylactosamine
(polyLacNAc) residues on glycans that modify proteins and lipids on the surface of host cells and pathogens.
In addition to their direct role in microbial killing, many proposed galectin functions in the immune system
and cancer involve crosslinking glycosylated receptors and modifying signaling pathways or sensitivity to
antigen from the outside in. However, a large body of work has uncovered intracellular galectin functions
mediated by carbohydrate- and non-carbohydrate-dependent interactions. In the cytoplasm, galectins can
tune intracellular kinase and G-protein-coupled signaling cascades important for nutrient sensing, cell cycle
progression, and transformation. Particularly, but interconnected pathways, cytoplasmic galectins serve the
innate immune system as sensors of endolysosomal damage, recruiting and assembling the components of
autophagosomes during intracellular infection through carbohydrate-dependent and -independent activ-
ities. In the nucleus, galectins participate in pre-mRNA splicing perhaps through interactions with
non-coding RNAs required for assembly of spliceosomes. Together, studies of galectin function paint a
picture of a functionally dynamic protein family recruited during eons of evolution to regulate numerous
essential cellular processes in the context of multicellular life.
Key words Galectin, Immunology, Intracellular, Extracellular, Glycobiology, Evolution, Microbes,

Glycan binding protein
Abbreviations
CBP Carbohydrate binding protein

CRD Carbohydrate recognition domain
ECM Extracellular matrix
ER Endoplasmic reticulum
GA Golgi apparatus
GalNAc N-acetylgalactosamine
Sean R. Stowell et al. (eds.), Galectins: Methods and Protocols, Methods in Molecular Biology, vol. 2442,
https://doi.org/10.1007/978-1-0716-2055-7_1, © Springer Science+Business Media, LLC, part of Springer Nature 2022
1
2 Hans Verkerke et al.
GALNts GalNAc transferases

GBPs Glycan binding proteins
PNG Protein N glycosylation
Siglecs Sialic acid-binding immunoglobulin-type lectins
1 Introduction
Galectins are an ancient family of soluble β-galactoside carbohy-

drate binding proteins (CBPs) encoded by the LGALS gene family
in humans. Members of the galectin family have evolved numerous
intra- and extracellular functions in development and regeneration,
innate immunity and pattern recognition, adaptive immune regula-
tion [1–3], and pathogenesis of autoimmune diseases and cancer
[4]. Despite a large body of work investigating the biochemistry
and effects of recombinant and endogenous galectins in vitro,
detailed mechanistic studies have been limited for many proposed
galectin functions in vivo. Such studies have proven challenging for
several reasons including the number of galectins, their broad and
varied tissue expression patterns, localization to multiple intracellu-
lar and extracellular compartments, and the yet unproven but likely
confounding potential for galectins to complement one another by
engaging similar or identical ligands. However, many consistent
and compelling themes of galectin function have emerged since
the discovery of the historically classified sulfhydryl- or S-type
lectin, galectin-1 in 1975 among asialofetuin binding fractions
from the electric organs of an electric eel [5].
Fundamentally, mammalian galectins mediate their pleiotropic
functions through binding to both carbohydrate and
non-carbohydrate ligands. They serve as adaptors within the cell
to recruit enzymes, regulating pre-mRNA splicing in the nucleus as
well as mRNA stability, autophagy, and apoptosis in the cytoplasm.
Stimulated to non-classical secretion or released from damaged
cells, they can engage or crosslink glycoprotein receptors on
immune and neoplastic cells to modulate signaling pathways and
dictate cell fates; and free from their ECM, they serve the innate
immune system as direct antimicrobial effectors against pathogens
employing molecular mimicry [6].
Perhaps the most important challenge in our mechanistic
understanding of galectin functions is the complexity and ubiquity
of their ligands: oligosaccharide modifications of proteins and
lipids, known collectively as the glycome. The glycome is a complex
network of linear and branched carbohydrate moieties, regulated at
multiple levels by the flux of metabolites through major catabolic
pathways, in turn dictating substrate availability in the tightly con-
trolled anabolic pathways of glycan synthesis and addition, which
occur in the endoplasmic reticulum (ER) and Golgi apparatus
Galectins: An Ancient Family of Carbohydrate Binding Proteins with Modern. . . 3
(GA) of eukaryotic cells. Terminal and internal glycan and

non-glycan modifications including sulfates, phosphates, sialic
acid, fucose, mannose, and xylose further enrich the glycan code:
a complex and multifunctional set of interactions and responses
mediated by glycans and glycan binding proteins. The functional
significance of the glycan code has been particularly well studied in
the mammalian immune system, where many innate and adaptive
immune pathways have converged to utilize lectin recognition of
specific glycan signatures in order to regulate cellular trafficking,
pattern and damage recognition, signaling pathways, and differen-
tiation [7]. It is therefore essential to carefully consider the role of
known and putative glycan ligands as well as the cell types and
conditions which produce them in any rigorous study of physio-
logic galectin function.
Following the discovery of galectin-1 (Gal-1), more than
15 putative members of the galectin family (thus categorized in
1994 [8]) have been identified in various mammalian species,
including humans, wherein 12 functional galectin-encoding genes
have been found (LGALS 1, 2, 3, 4, 7, 8, 9, 10, 12, 13, 14, and 16).
Galectins 5 and 6 emerged in contemporary rodent lineages while
galectins 11 and 15 have thus far only been found in sheep and
goats. In addition, galectin orthologs with yet unknown function
have been identified in many lineages of vertebrate and invertebrate
multicellular organisms (metazoans) including representative and
model organisms from diverse evolutionary lineages: Mus musculus,
Ciona intestinalis (sea squirt), Drosophila melanogaster, Caenorhab-
ditis elegans, and even fungi and basal members of the metazoan
lineage of Poriferan sponges. The ubiquity of galectin family mem-
bers within nearly every branch of the multicellular tree of life and
the highly conserved nature of the carbohydrate recognition
domain (CRD), which defines them, supports the broad utility of
galectin activity in multiple cellular processes and offers some
explanation for experimental observations of their divergent and
pleotropic functions [9]. In this review, we aim to highlight impor-
tant observations of galectin function, focusing on aspects of
eukaryotic cell biology and vertebrate immunology set within the
broader context of galectin evolution and biochemistry. In so
doing, we hope to provide a conceptual framework for the evolving
biology of this complex family of GBPs.
1.1 The Evolution of Glycans are perhaps the most enigmatic of the four major classes of
Carbohydrate macromolecules common to all organisms. The absence of a con-
Recognition sistent genetic template and the complexity of their nonlinear
chemistry are just two intrinsic factors that have historically con-
founded the characterization of oligosaccharide modifications. By
contrast, studies of biological function that traverse the familiar
ladder from DNA (gene) to RNA (transcript) to protein (enzyme)
benefit from an enormous armamentarium of tools under
development since Gregor Mendel elaborated his theory of inheri-

tance in the mid-nineteenth century [10]. The relatively modern
field of epigenetics has built on these fundamental tools in an effort
to explain the influences of environment and development on the
heredity of gene expression patterns, a study involving the extensive
decoding of post- and co-replicative, transcriptional, and transla-
tional modification. Epigenetics has broadened the focus of life
science from what is encoded to when and why it is expressed.
Carbohydrate adducts on proteins likely evolved in the earliest
forms of life to stabilize and regulate the folding of an increasingly
large and biochemically complex repertoire of enzymes and struc-
tural proteins. As the degree and complexity of these polar mod-
ifications expanded to obscure the surfaces of their protein
substrates, carbohydrate recognition by catalytic enzymes and reg-
ulatory proteins within the evolving secretory and protein traffick-
ing networks must have become imperative [11]. According to this
theory, many of the extracellular functions for protein glycosylation
and the enzymatic pathways in the secretory system which produce
them would have evolved and diverged later. Consistent with this
“inside-out” model of glycan-GBP evolution, the major intracellu-
lar glycosyltransferases, glycosidases, and lectin chaperons within
the ER (e.g., calnexin and calreticulin) are highly conserved both
functionally and genetically in eukaryotic cells from plants, yeast,
and mammals; while enzymes that prune, extend, and terminally
modify these core structures within the Golgi apparatus are isolated
and divergent within specific metazoan lineages. In addition, there
is little evidence for ancient orthologs among GBPs with highly
specialized extracellular functions within the mammalian immune
system such as the innate and B-cell regulating sialic acid-binding
immunoglobulin-type lectins (Siglecs), which recognize specific
terminal sialic acid moieties to modulate primary signaling path-
ways, or members of the C-type lectin (CLEC) family of selectins,
which regulate leukocyte trafficking among other functions. These
specialized GBPs, now critical components of the mammalian
immune system, likely evolved to exploit and elaborate upon the
more conserved functions like protein stabilization and trafficking
common to all organisms. Importantly, all genomes including
those from unicellular protists and bacterial lineages encode pro-
teins, which have evolved convergently the ability to read or exploit
aspects of their own glycan code or that of other species. What is
known of the evolutionary history of galectins shares features of
both the ancient and modern groups of lectins.
1.2 Galectin Members of the galectin family with conserved gene and protein
Evolution and structure have been found in nearly every lineage of multicellular
Biochemistry life, suggesting the existence of a single common ancestor. How-
ever, through multiple tandem and en mass gene and genome
duplications followed by structural and functional divergence,
galectins have evolved specialized and accessory functions within

specific lineages. While some poorly characterized galectin-like
genes (e.g., GRIFIN) appear to have lost the residues necessary
for carbohydrate recognition, all formally classified galectins have
retained this functional domain. This conservation of structure and
function highlights the importance of evaluating the evolutionary
history of galectin carbohydrate recognition to appropriately con-
textualize functional observations in humans and model organisms.
1.2.1 Galectin Evolution Historically, vertebrate galectins have been categorized by the
number and linkage of their carbohydrate recognition domains
(CRDs) into mono-CRD prototypical (Gal-1, 2, 5, 7, 10, 11, 14,
15) and chimeric (Gal-3) galectins and bi-CRD tandem repeat
(Gal-4, 6, 8, 9, 12) galectins [8] (Fig. 1a). Much can be gleaned
about these ubiquitous proteins from their gene structure, chro-
mosomal location, and sequence. All carbohydrate binding pro-
teins classified as galectins share at least one ~135 amino acid
β-sandwich carbohydrate recognition domain (CRD), composed
of 6 (S1–S6) and 5 stranded (F1–F5) anti-parallel β-sheets. Strands
S4–S6 contain the conserved amino acids which mediate carbohy-
drate recognition and are all encoded by the second of three con-
secutive exons in the CRD-encoding region(s) of LGALS loci
(Fig. 1b). The first exon consistently encodes F1 and S2 strands;
while the second (middle), CRD-defining exon encodes strands S4,
S5, and S6 but can either terminate at specific positions within the
region encoding strands F3 or F4 [12] (Fig. 1c). These conserved
exon structural and genomic features are found among all known
chordate galectin genes, supporting a model in which a single
mono-CRD common ancestor to chordate galectins arose and
subsequently evolved through tandem duplication to produce the
common ancestor of vertebrate bi-CRD galectins. Because the
likelihood of a precise F3/F4 terminating gene structure arising
independently for modern galectins is vanishingly small, this par-
ticular feature is useful in tracing relationships among different
mono- and bi-CRD galectins (i.e., F3 or F4 terminating galectins
likely arose from ancestral genes sharing the same exon
structure) [12].
The specific nature of the first galectins is unknown, but the
existence of mono-CRD galectins with the F4 terminating exon
structure in both protostome and deuterostome (pre-chordate)
lineages (and lack of the F3 exon structure in protostome galectins)
implies that the original mono-CRD galectin may have also been
F4-terminating. A subsequent tandem duplication of this ancestral
galectin gene could then have given rise to the exon organization
found in all chordate bi-CRD galectins (F4-linker-F3). The F3
terminating ancestors of galectins 1, 2, and 3 would then have
arisen from further duplication of the C terminal F3 CRD of this
bi-CRD galectin, found in the common ancestor of all vertebrates.
Fig. 1 The galectin family of β-galactoside binding proteins. (a) Galectins are classified into three distinct
groups based on their quaternary structure: prototypical, chimeric, and tandem repeat. Prototypical: Gal-1,
Gal-2, Gal-7, Gal-10, Gal-13, and Gal-14. Chimeric: Gal-3. Tandem repeat: Gal-4, Gal-8, Gal-9, and Gal-12.
(b) The galectin CRD is composed of six (S1–S6) and five stranded (F1–F5) anti-parallel β-sheets oriented in a
“jelly-roll” fold. (c) Strands and amino acids required for ligand-binding (red) are encoded by LGALS loci with
conserved exon structure and orientation
Supporting the latter timeline is the unique sequence and dimeric

interface in vertebrate orthologs of Gal-1 and Gal-2. Within this
framework, all vertebrate galectins were derived from either an
ancestral F4-linker-F3 bi-CRD galectin or an F3 mono-CRD galec-
tin. The bi-CRD human galectins 4, 8, 9, and 12 are encoded by
LGALS genes in highly paralogous regions of chromosomes 1, 11,
17, and 19. And the degree of synteny (similar flanking regions)
among these members of the LGALS family, in addition to the
existence of bi-CRD orthologs in similarly flanked regions of
other vertebrate genomes, strongly supports chromosomal dupli-
cation of a single ancestral chordate bi-CRD galectin as the likely
mechanism by which they arose—perhaps in the two putative mass
genome duplications described in vertebrate evolution. More
Fig. 2 Evolution of galectins. Analysis of gene structure and organization in extant lineages of multicellular life
support the existence of a single ancestral mono-CRD galectin in the bilateral common ancestor of proto-
stomes and deuterostomes with a middle exon terminating in a gene region encoding an F4 beta strand. The
structural and functional diversity of vertebrate galectins seen today is likely to have subsequently arisen
through first tandem, then en masse gene duplication followed by additional duplications and divergent
adaptations
recently evolved mono-CRD mammalian galectins (5, 6, 7, 10,

13, 14, 15) most likely arose within more specific lineages from
duplication of single domains of these original bi-CRD galectins. In
this model, all extant prototypical galectins have emerged through
partial duplications of either the F3 or F4 CRD of these core
ancestral bi-CRD galectins while mono-CRD orthologs of galec-
tins 1, 2, and 3 evolved independently from a common F3 termi-
nating ancestor. These relationships are important to consider
when drawing inferences about galectin function in humans from
studies of galectins in model organisms such as those which may
have evolved independently for hundreds of millions of years
(Fig. 2).
1.2.2 The Galectin CRD X-ray crystallographic studies of multiple galectin family members
have revealed structural determinants of galectin activity. Despite
considerable heterogeneity in their valency and quaternary struc-
tures, all galectins possess one or more structurally conserved
CRDs with some degree of binding activity toward repeating
units of N-acetyl lactosamine (polyLacNAc). Structurally, the
galectin CRD forms a compact 11 stranded β-sandwich or jelly
roll tertiary structure and confers specificity for the 4 and 6 hydroxyl
functional groups of galactose and the 3 hydroxyl functionality of

N-acetylglucosamine (GlcNAc) through amino acid side chain
hydrogen bonding and electrostatic interactions. Ring stacking by
a highly conserved tryptophan (W) in the galectin CRD favors the
stereochemistry of galactose over mannose; while additional inter-
actions of backbone amines and side chains extend beyond the
CRD to mediate differences in specificity and affinity for length
and glycan modifications among galectin family members. In
Fig. 1b we detail structural features of the galectin CRD.
Despite overlapping specificity for LacNAc containing glycans,
studies using increasingly comprehensive glycan microarrays in
addition to more traditional biochemical approaches suggest that
each galectin exhibits unique ligand preferences [13–18]. While
glycans are the best characterized galectin ligands, it is important
to also highlight the non-carbohydrate interactions which mediate
galectin function, particularly in the intracellular compartment.
1.2.3 Galectin Prototypical galectins (e.g., Gal-1, 2, 7, 10) have evolved to form
Quaternary Structure and homodimers with outward facing CRD domains, an orientation
Function which facilitates ligand-crosslinking under certain conditions.
These oligomers exist with their monomeric counterparts in
dynamic equilibrium. The relationship between oligomeric struc-
ture and biological function has been a particular focus in biochem-
ical characterizations of Gal-1. Gal-1 dimerization promotes ligand
binding [19, 20], which in turn reduces the sensitivity of galectin-1
to oxidative inactivation [21]. Oxidation of galectin-1 leads to
stepwise formation of disulfide bonds among exposed thiols in six
conserved cysteine residues (Cys 2, 16, 42, 60, 88, 130), four of
which are solvent accessible in the predicted monomeric structure
(Cys 2, 16, 88, 130). Spectral studies have revealed that structural
changes upon oxidation of galectin-1 are reversible, suggesting the
ability of this protein to dynamically switch between a structure that
favors dimerization and ligand binding and one that does not. The
functional significance of this phenomenon has not been deter-
mined in vivo. However, the ability of galectin-1 to induce revers-
ible phosphatidyl serine exposure on and phagocytosis of cultured
neoplastic leukocytes and activated neutrophils is dependent on
dimerization and ligand binding in the reduced conformation
[22], an in vitro finding recapitulated for the C-terminus of Gal-8
[23]. These observations, combined with the finding that Lgals1
null mice exhibit increased susceptibility to tissue destruction in the
context of autoimmunity, suggest a possible mechanism by which
reduced Gal-1 is released from damaged tissue to limit leukocyte
infiltration. The temporal and spatial regulation of this process
could proceed automatically through reversible oxidative inactiva-
tion of secreted or released Gal-1 [24] (Fig. 3). Additional studies
of inflammatory processes in galectin knockout models are needed
Fig. 3 Galectin-1 is regulated by redox environment. Galectins, first called S-type lectins secondary to the
requirement of several galectins for reduced thiols to maintain carbohydrate recognition activity, can form
intra- and inter-molecular disulfide bridges that often result in significant conformational changes that
preclude carbohydrate recognition. Upon oxidation Gal-1 can for three disulfide bonds resulting in an
inactivating conformational change that dramatically inhibits ligand binding and oligomerization. At high
concentration, bridges can form between Gal-1 monomers, resulting in aggregation and the potential for
irreversible inactivation. By contrast, reducing environments activate Gal-1, promoting carbohydrate binding
and dimerization. Dimerization also enhances ligand binding affinity of Gal-1
to determine when Gal-1 is released in vivo and whether its activity

is involved in limiting immune-mediated tissue damage.
Unique among vertebrate galectins, Galectin-3 has a modular
N-terminal domain composed of a 21-amino-acid-long stretch,
bearing two sites for serine phosphorylation, and a cleavable, nine
triple-helix proline/glycine-rich repeat domain similar to collagen
domains in structure [25]. Galectin-3 is monomeric in solution,
but can form higher order structures at high concentration or in the
presence of ligands, which can seed aggregation of multimeric
complexes. The CRD of Galectin-3 also bares sites for tyrosine
phosphorylation, which modulate carbohydrate recognition and
secretion [26–28].
1.2.4 Synthesis of Analogous to promoters, transcription factor binding sites, or con-

Galectin Carbohydrate served nucleic acid structures in genetics and epigenetics, the study
Ligands of protein glycosylation (and epiproteomics in general) has been
guided by the presence of specific N- and O-linked glycosylation
motifs or sequons within the amino acid sequence of secreted
proteins. For protein N glycosylation (PNG), glycans are built
from covalent linkages to Asn-X-Ser/Thr/Cys (N-X-S/T/C)
amino acid sequences. The occupancy and complexity of glycan
Fig. 4 Glycan synthesis and lattice formation. Metabolic flux through the glycolytic pathway forms substrates
for protein N glycan synthesis (PNG), including UDP-GlcNAc, which is transferred onto dolichol phosphate on
the cytosolic face of the ER. Following co-translational transfer of this PNG stem structure onto a newly
synthesized glycoprotein, stepwise synthesis in the ER lumen builds a precursor glycan, Man5GlcNAc2, which
is then trimmed and further modified to form the common PNG backbone Glc3Man9GlcNAc2. Recognition of
this precursor in the ER licenses glycoprotein transport to the cisternae of the Golgi apparatus, where
trimming, branching, extension, and terminal modifications exponentially increase the repertoire of possible
glycan adducts on secreted and membrane associated glycoproteins. In the extracellular space, these
modifications can be recognized by non-classically secreted galectins whose multivalency can form lattice
like structures, mediating cell-cell adhesion and nano-scale localization of membrane proteins. Galectins can
also directly bind receptors baring specific glycans resulting in signal transduction and cell reprogramming.
Galectin-mediated cell-cell adhesion can also mediate paracrine signaling
modifications at PNG sequons is both encoded and conditional—

dependent on numerous genetic, structural, and environmental
factors (Fig. 4). In eukaryotes, PNG synthesis is initiated on the
cytosolic face of the endoplasmic reticulum (ER) when a precursor
glycan (Man5GlcNAc2) is transferred by dolichol lipid transporters.
Subsequently, stepwise enzymatic reactions and chaperone binding
within the ER lumen assemble, trim, and stabilize a common PNG
backbone (Glc3Man9GlcNAc2), the recognition of which serves as
a checkpoint for progression into the Golgi apparatus. Base struc-
tures for common galectin PNG ligands are formed in the medial
Golgi, first by sequential branching via N-acetylglucosamineyl-
transferase (MGAT1,2,3,4a/b, and MGAT5)-catalyzed β-1,4 addi-
tion of galactose, followed by the extension of branches in the distal
Golgi by variable addition of N-acetyllactosamine (LacNAc) units.
Gal-3, Gal-8, and Gal-9 have graded affinity for polyLacNAc
ligands, increasing with degree and length of branching; while
Gal-1 exhibits less of this effect [29]. Furthermore, Gal-1 targets
the terminal region of LacNAc structures while Gal-3 like the plant
lectin PHA-L is able to bind to internal LacNAc residues. Thus,
substrate availability and enzymatic activity of MGAT proteins
regulate extracellular galectin activity toward PNG modified
receptors.
Proteoglycan ligands for galectin binding are also formed
through O-glycan synthesis. O-linked glycosylation
(O-glycosylation) begins with the addition by GalNAc transferases
(GALNts) of N-acetylgalactosamine (GalNAc) to the side chain
oxygen atoms of hydroxyl containing amino acids, most commonly
serine or threonine. Under normal conditions, T-synthase catalyzes
the addition of galactose to the GalNAcα1-O-Ser/Thr precursor,
producing the O-glycan core-1 structure (T antigen). Branching of
the core-1 structure by the addition of GlcNAc in β6 or β3 linkages
forms the O-glycan core-2 and core-3 structures, respectively. Fur-
ther modification of the extended core structures by specific
glycosyl-, sialyl, and fucosyl transferases in the Golgi extends and
modifies core-O-linked structures, resulting in functionally signifi-
cant glycan structures including blood group antigens, Lewis and
Sialyl Lewis epitopes for selectins and other immune lectins (Lex
and sLex), as well as the best characterized ligand for galectins:
polyLacNAc. Failure of the cell to produce these structures is a
hallmark of cancer, autoimmunity, and inconsistent with life in
mouse models of embryonic development. The importance of
normal O-glycan synthesis and extension is exemplified by the
role of the human tumor-associated Thomsen–Friedenreich
(Tn) neoantigen (unmodified GalNAcα1-O-Ser/Thr) and Sialyl
Tn, which results from loss or mutation of the ER-localized chap-
erone Cosmc or its target T-synthase in malignancies
[30]. Expression and secretion of proteins baring the Tn antigen is

an important biomarker of certain cancers, particularly carcinomas
[30, 31].
For both PNG and O-glycan synthesis, terminal and internal
modifications by transferases in the medial and trans Golgi further
control the binding behaviors of lectins, including galectins, upon
secretion or surface exposure of the fully formed glycoprotein.
Exemplifying this phenomenon is Gal-8, which contains two
CRDs with different recognition of β-galactoside terminal modifi-
cations, with the N-terminal CRD strongly preferring ligands con-
taining terminal α2–3-sialic acid containing β-galactosides. Like the
N-terminal CRD of Gal-8, galectin-1 is also able to bind α2–3-sialyl
β-galactosides, but its binding is strongly inhibited by the same
modification in an α2–6 linkage while Gal-2 binding is inhibited
by terminal sialic acid regardless of linkage. These important bio-
chemical observations regarding galectin CRD specificity highlight
an essential theme in galectin biology—the ability of specific cells,
tissue types, and neoplasms to control the extracellular effect of
different extracellular galectins by branching and modification of
N- and O-linked glycan structures [32].
2 Mammalian Galectin Expression and Localization
Gal-1, Gal-3, and Gal-9 are first produced during mammalian

embryogenesis, and the patterns of their expression have been
tracked using in situ hybridization and immunohistochemistry.
Gal-1 and -3 are first expressed on the periphery of the developing
blastocyst in cells of the trophectoderm. This observation led to the
prediction that these two galectins might be necessary for the
process and patterning of embryonic implantation. However,
subsequent work found that global deletion of either or both
galectins did not prevent implantation from occurring [33], sug-
gesting accessory roles in this process. Following implantation,
Gal-1 expression marks early muscle cell precursors localized to
the myotomic pole of developing somites, a pattern which precedes
and mirrors expression in skeletal and cardiac muscle of the adult
animal. During the same embryonic timeframe, Gal-3 is expressed
primarily in cells of the notochord or primordial nervous system
[34–36]. Gal-9 has also been identified in the developing mouse
embryo, localized to the embryonic liver and thymus, where it may
play a role regulating interactions between embryonic thymocytes
and thymic epithelial cells [37]. Patterning of galectin expression
during embryogenesis is in part mediated by methylation of a small
region that encompasses Lgals transcriptional start sites [38] while
upregulation of these genes requires demethylation and activation
by certain transcription factors such as NF-κB [39]. Some galectin
expression can also be hormonally regulated during placental devel-

opment [40]. Notably, three galectin genes with conserved func-
tion in placental development (LGALS 13, 14, 15) cluster to
human Chr19 and appear to have emerged during the evolution
of semi-allogeneic fetal development in primates. Dysregulated
expression patterns of these placental galectins are associated with
pre-term preeclampsia in humans [41, 42], and they may also have
roles in directly regulating immune tolerance of the fetus [43, 44],
a function proposed for Gal-1 as well. In primate embryonic devel-
opment, Gal-1 can be hormonally regulated and is expressed most
prominently in maternal decidual stromal cells and uterine natural
killer cells, possibly playing an important role in regulating mater-
nal–fetal tolerance [40, 45, 46]. A comprehensive and inclusive
survey of the galectinome during mammalian embryonic develop-
ment would provide further insight into the evolutionary and
developmental origins of the broad galectin expression patterns
observed in adult mammals.
In adult mammals, Gal-1 is widely expressed at homeostasis in
muscle (skeletal and smooth), skin, lung, lymphoid tissue (thymo-
cytes, thymic epithelial cells, lymph nodes, and spleen), prostate,
placenta, testes, and cells of the nervous system. In noncancer
tissues, Gal-3 expression has been documented in epithelial cells
of different origins, dendritic cells, macrophages, and virus-infected
T cells [47, 48]. Gal-9 is also expressed in many cells of lymphoid
origin but also cells of the lung and myocardium. Gal-10 appears to
be solely expressed by immune cells including granulocytic neutro-
phils and eosinophils (wherein it can comprise the principal com-
ponent of Charcot–Leyden crystals) as well as regulatory T cells
[49–51]. Gal-7 expression is localized mostly to stratified squa-
mous epithelial cells of the skin, while Gal-4 and Gal-6 are mostly
found in the epithelium of the gastrointestinal tract. Rodent-
specific Gal-5 is expressed in the erythroid lineage, but its specific
function in the blood compartment of rodents has not been estab-
lished. Gal-12 is highly expressed in adipocytes and may regulate
their homeostasis, illustrated by the finding that Gal-12 null mice
exhibit decreased adiposity [52]. Further studies are needed to
delineate patterns of basal and induced galectin expression as well
as the regulation of galectin intracellular localization and secretion.
Many studies of galectin function in vitro have used cancer cells
and exogenous addition of recombinant galectins. These studies
have been critical in uncovering the scope of what galectins can do,
but may also be misleading when delineating biological function
in vivo. This challenge is confounded by overlapping and unique
tissue expression patterns. Furthermore, galectin expression and
secretion can be variably induced in stages of development, states
of inflammation, or dysregulated during oncogenesis. It is there-
fore critical to determine which galectins are present in any
experimental system and in which compartments they are localized

when evaluating putative functions in vivo.
3 Modern Galectin Functions
3.1 Nuclear and While posttranslational glycan modifications are predominantly

Cytoplasmic Galectins found in secretory compartments and the extracellular space,
nucleocytoplasmic proteins can also be modified by carbohydrate
moieties (e.g., O-GlcNAc). However, these modifications are poor
substrates for galectin binding as they have not been found to
contain LacNAc. Despite the absence of canonical galectin ligands
in the nucleocytoplasmic space, galectins have evolved intracellular
functions through non-carbohydrate interactions. These interac-
tions have been shown to mediate roles in pre-mRNA splicing
and mRNA stabilization and the modification of intracellular sig-
naling pathways. Cytoplasmic galectins have also been shown to
regulate several important signaling pathways including apoptosis.
3.2 Regulation of In multiple studies, Gal-1 and Gal-3 have been associated with
Pre-mRNA Splicing pre-mRNA splicing within the nucleus. Early experiments using
and mRNA Stability cell-free splicing assays in HeLa cell nuclear extracts found that
the addition of competitive galectin inhibitors—lactose or thiodi-
galactoside (TDG)—but not galactose alone or cellobiose potently
inhibited the formation of splicing products from pre-mRNA sub-
strates. Furthermore, after removal of nuclear lactose-binding pro-
teins from these extracts using lactose affinity chromatography,
addition of recombinant Gal-3 was sufficient to restore splicing
activity in vitro [53]. Further investigation revealed that both intra-
cellular Gal-1 and Gal-3 microscopically localize within both the
nucleus and cytoplasm of HeLa cells [54], directly interact with
splicing complexes, and are co-immunoprecipitated with
pre-mRNA [55]. For galectin-3, nuclear transport is, at least in
part, dependent on the interaction between a C-terminal poly-basic
NLS sequence (HRVKKL), which mediates interaction with trans-
locating nuclear importins [56]. Return of Gal-3 to the cytosol and
an enhancement of carbohydrate binding activity accompanies
casein kinase 1 (CK1) [57] serine phosphorylation in the nucleus.
These findings strongly support a model in which some galectin
family members can function as nucleocytoplasmic proteins in
addition to their extracellular roles [58]. Additional mechanistic
studies showed that Gal-3 specifically associates in the nucleus with
U1 snRNP, an essential initiator of splicesome assembly, to mediate
its observed pro-splicing activity [59] through both the Gal-3
CRD- and YPG-rich repeats in its N-terminus [60]. Additional
galectin–splicesome interactions have also been uncovered using
yeast two-hybrid screening, which identified survival of motor
neuron (SMN) complexes containing Gem associated protein
4 (Gemin4) and additional snRNP components as Gal-1 and Gal-3

binding partners [61]. Together, these findings strongly implicate
Gal-1 and Gal-3 involvement in the process of pre-mRNA splicing
through both protein–protein and protein–nucleic acid
interactions.
Regulation of RNA by galectins has been recently confirmed in
a study that demonstrated a role for Gal-3 in stabilizing transcripts
of membrane-associated epithelial mucin, MUC4, in cancer cells.
Mucins are important mediators of barrier function and homeosta-
sis in healthy epithelial tissue, but their over-production promotes
proliferation and motility in cancers, including pancreatic and bile
duct carcinomas, where their expression is strongly correlated with
that of LGALS3 mRNA levels. Because Gal-3 is also upregulated in
these cancers and was previously shown to regulate RNA through
direct and indirect mechanisms, the authors hypothesized that
Gal-3 might alter MUC transcript stability. They found that cyto-
plasmic Gal-3 interacts with hnRNP-L in perinuclear granules to
increase hnRNP-L association with a CA repeat element in the
30 UTR of MUC4 mRNA, thus doubling the half-life of the
MUC4 transcript [62]. Supporting this finding in vivo, they
found that LGALS3/ mice had significantly lower levels of
epithelial MUC4 mRNA and hnRNP-L protein in the
jejunum [63].
While together these findings clearly demonstrate that galectins
can be involved in pre-mRNA and mRNA processing under certain
conditions, it is important to recall that both LGALS1 and
LGALS3/ mice are viable and exhibit grossly normal develop-
ment and life span. The latter observation is key to contextualizing
the RNA regulatory functions put forward for Gal-1 and Gal-3
because null mutations in the basal components of the spliceosome
are almost universally lethal at the cellular level in both unicellular
and multicellular eukaryotes [64]. The only known human genetic
diseases resulting from mutations in essential mRNA splicing com-
ponents are sporadic autosomal dominant retinitis pigmentosa
(RP) and spinal muscular atrophy (SMA), both of which result
from only partial loss of function mutations in essential splicesome
components. It is therefore most likely that the observed roles of
Gal-1 and Gal-3 in this process are accessory, specific to certain
transcripts, or induced in neoplasm or inflammation.
3.3 Intracellular Early studies of Gal-3 revealed that it was upregulated in human
Signaling: Proliferation T-cell leukemia virus (HTLV) infected T cells and that its over-
and Apoptosis expression was sufficient to promoted proliferation and confer
resistance to apoptosis [48]. These findings led to further investi-
gation and the discovery of a functional BH1 (NWGR) domain,
common to proto-oncogenes of the Bcl-2 family, in the C-terminus
of Gal-3 [65, 66]. Consistent with a role in oncogenesis, Gal-3 has
since emerged as a biomarker of numerous cancer types, in some
cases correlating with disease progression and resistance to therapy

[67–70]. Interestingly, some have found opposing roles for cyto-
plasmic and nuclear Gal-3, with the latter localization promoting
apoptosis [71]. Additional work has demonstrated that this pro/
anti-apoptotic switch is induced by chemotherapeutic agents and
mediated by nuclear export upon Ser6 phosphorylation of Gal-3
[72]. The role of endogenous Gal-3 in homeostatic and develop-
mental apoptosis remains to be thoroughly elucidated. Several
studies have also investigated the observations that Gal-7 is down-
regulated in squamous cell carcinoma (SCC), and its overexpres-
sion in colon cancer cells can induce apoptosis [73].
3.4 Galectin Genes encoding conventional secretory proteins bare a leader or

Secretion signal sequence which targets translating ribosomes to the endo-
plasmic reticulum, where the protein is co-translationally translo-
cated to the ER lumen and trafficked to the Golgi apparatus before
secretion from the cell. However, many important secreted pro-
teins, including galectins, lack a signal sequence, instead reaching
the extracellular space by alternative mechanisms. The absence of a
galectin signal sequence initially cast doubt on whether galectins
were secreted, but metabolic labeling [65], immunohistochemical
analysis [74, 75], and detailed mechanistic studies have confirmed
the phenomenon for several family members. Release of cytoplas-
mic galectin upon cellular damage is an additional potential mech-
anism by which galectins can reach the extracellular space.
Supporting the latter mechanism are studies of muscle degenera-
tion and injury in which tissue damage is associated with increased
extracellular localization of galectin-1 which may promote myo-
genesis [76, 77]. Intriguingly, the regulated forms of galectin
secretion appear to occur through diverse mechanisms. In vitro,
export of active, surface bound Gal-1 appears to require the
engagement of cell surface β-galactoside-modified counter recep-
tors. Cells lacking these modifications are deficient for functional
Gal-1 secretion, and their presence is sufficient to drive secretion of
Gal-1 orthologs [78]. Using metabolic labeling, Cho et al. have
further demonstrated that Lec8 CHO cell mutants (incapable of
producing galactosylated glycoproteins) are indeed able to secrete
Gal-1 into the extracellular space, but it is not retained on the
surface and becomes inactivated in the absence of ligand and the
reducing conditions of the cytoplasm [79]. By contrast, Gal-3
export occurs via exosome release and requires an N-terminal tetra-
peptide P(S/T)AP motif for interaction with Tsg101 in the endo-
somal sorting complex required for transport (ESCRT) [80]
pathway. Gal-8 secretion may in part be mediated by its association
with damaged and recycling endosomes [81].
3.5 The Galectin The extracellular matrix is rich with variably modified beta-
Lattice galactoside ligands, which can be crosslinked by specific galectins
at the cell surface to form a functional lattice. In vitro binding

studies using CHO cell glycosylation mutants have shown that
the primary ligands for galectin-1, 3, 8, and 9 in particular are
complex-type N-glycans [82]. N-glycan branching and extension
is catalyzed in the medial Golgi apparatus by a family of N-acet-
ylglucosaminyltransferases encoded by MGAT1–5. The activity of
these enzymes is regulated by flux through the hexosamine path-
way, a primary metabolic route for the synthesis of glycan building
blocks, dependent on glycolysis. In addition to metabolic flux, the
number of potential N-glycan sites on transmembrane receptors
structurally encodes the graded affinity of a glycoprotein receptor
for components of the galectin lattice. Interactions of these glycans
and glycans produced by other anabolic pathways with various
galectins and other carbohydrate binding proteins can regulate
compartmentalization of the plasma membrane and control recep-
tor exposure by modulating sensitivity to endocytosis [83]. In
general, growth/activation-promoting receptors on cells encode a
higher number of N-glycan sequons per extracellular amino acid
(~70% occupied in the mammalian glycoproteome [84]), whereas
arrest and differentiation promoting receptors encode fewer such
sequons [85]. Lau et al. used computational modeling and in vitro
studies to show evidence that the number of N-glycan motifs
coevolved with metabolic glycan branching enzymes to regulate
the retention of specific receptors in the galectin lattice and control
the transition between growth and arrest/differentiation in mam-
malian cells [85].
Determining the role of galectin lattices in vivo has been more
challenging. One study by Smith et al. described a mechanism that
required galectin-3-mediated lattice formation on CD8+ T cells, by
which the regulatory cytokine interleukin-10 drives establishment
of chronic viral infection. They demonstrated that IL-10 potently
induces Mgat5 expression in antigen specific CD8+ T cells, which
in turn enhances N-glycan branching and galectin-3 lattice forma-
tion. This lattice formation reduced antigen sensitivity of the
affected cells and inhibited the ability of these antiviral effectors
to clear chronic infection in mice [86]. The latter finding suggests a
fundamental role for galectin-lattice formation in tuning the sensi-
tivity of the immune system, likely preventing unchecked activation
and tissue damage at high levels of antigen exposure. In support of
the theory that metabolic flux and environment influence N-glycan
branching and galectin lattice formation in vivo, Demetriou et al.
have found that T-cell receptor (TCR) clustering cross-linked by
galectin-3 binding (and potentially other galectins) was deficient in
T cells from Mgat5 null mice [87]. Following up on these studies,
the Demetriou group showed that genetic variants in Mgat
enzymes altered surface retention of immune cell receptors in
multiple human autoimmune diseases including multiple sclerosis
and type 1 diabetes [88–90]. A similar lattice-mediated
stabilization may also be at play on the surface of B cells during

antigen encounter. Obino et al. recently found that Gal-8 interacts
with the B-cell receptor [75].
Early studies of galectin-1 indicated its potential role in regu-
lating immune cell functions. Exogenous administration of Gal-1 in
a rabbit model of the autoimmune disease myasthenia gravis damp-
ened immune cell-mediated damage [91]. While the latter finding
demonstrated therapeutic potential of galectins, the role(s) of
Gal-1 in vivo has remained more elusive. However, endogenous
Gal-1 expression has been shown in lymphoid organs and impli-
cated in the development and function of mature lymphocytes.
Thymocyte-epithelial cells and thymocytes of the thymic cortex,
but not medulla show surface expression of human galectin-1
in vivo. The association of Gal-1 with cortical epithelial cells is
dependent on core-2 O-glycans on T-cell-associated receptors
CD43 and CD45, spatially and developmentally regulated during
thymocyte development [92]. Gal-3, Gal-8, and Gal-9 are also
notable for being broadly expressed in immune cells and lymphoid
organs, and many intra- and extracellular functions for them have
been proposed in the mammalian immune system.
3.6 Homeostasis Because many studies of cytoplasmic galectins in intracellular sig-

naling rely on tumor cell models, the proposed functions and
interactions resulting from them are often most relevant to the
dysregulated signaling environment that drives cancer. However,
galectins also appear to function in the maintenance and
re-establishment of homeostasis under normal conditions in several
tissue types including the vascular endothelium, blood, muscle, and
epithelial barriers [93].
3.7 Epithelial Barrier The process of re-epithelialization is a critical step in wound heal-
ing. Several studies support a role for galectins in the
re-establishment of epithelial homeostasis in various tissues.
Galectin-7, which is primarily expressed by stratified epithelial
cells in skin, and galectin-3 are both important factors in vitro in
models of skin and corneal wound healing. Supporting these roles,
both LGALS3 and LGALS7 null mice exhibit defects in corneal and
keratinocyte wound healing [94, 95]. A recent study by Robinson
et al. found that galectin-9 null mice are more susceptible to
intestinal epithelial damage. They further showed that galectin-9-
deficient organoids developed abnormally and showed defects in
signaling pathways associated with growth and regeneration. These
observations indicate a homeostatic role for galectin-9 in maintain-
ing and repairing healthy intestinal epithelial barriers
[96]. Together, these findings implicate multiple galectins in the
maintenance and repair of epithelial barriers.
3.8 Vascular The genesis and branching of new blood vessels involve activation
Endothelium and chemoregulation of endothelial cells to form the inner lining of
a new vessel. Galectins 1, 3, and 9 are expressed by activated and
resting endothelial cells, and extracellular addition of these galec-
tins regulates endothelial migration in vitro [97–99]. Supporting a
role in placental angiogenesis, Gal-1 null mice have vascular defects
in formation of the placental decidua and tumor associated Gal-1
has been shown to be important for tumor vascularization
[100]. These findings suggest a basal role for galectins in the
development of vascular networks through angiogenesis in normal
and neoplastic tissues.
3.9 Hemostasis Several studies have implicated galactin-1 and galectin-8 in primary
hemostasis, mediated by the activation and aggregation of platelets
in the clotting cascade. Specifically, platelets were found to contain
high levels of Gal-8, which became exposed during activation by
thrombin. This surface exposed Gal-8 amplified platelet activation
by promoting fibrinogen binding, mobilizing calcium and mem-
brane spreading mechanisms, and enhancing thromboxane and
P-selectin expression [101]. Gal-1 can also activate similar pathways
in platelets and its expression contributes to ADP-induced aggre-
gation, suggesting a potential role in clotting [102]. The latter
finding is supported by the observation of normal platelet number
but dysfunctional primary hemostasis in LGALS1/ mice
[103]. In Fig. 5, we summarize the functional roles of several
galectins in maintenance and homeostasis of different tissues.
4 Cancer
4.1 Transformation Ras genes encode a family of small GTPases, which act as binary
molecular switches to transduce intracellular signals and promote
growth, proliferation, and differentiation at the cellular level
[104]. Constitutive activation of Ras mutants results in uncon-
trolled cellular proliferation and is a molecular hallmark of many
human cancers. Gal-1 and Gal-3 have both been shown by coloca-
lization, co-immunoprecipitation, and knockdown experiments to
interact with transforming mutants of Ras proteins in cancer cells,
specifically H-Ras and K-Ras [105, 106]. Additional work in breast
carcinoma cells has demonstrated that Gal-3 is in part responsible
for the isoform switch and resulting constitutive activation of
N-Ras to K-Ras [107]. Consistent with this finding and also with
a role of extracellular Gal-3 in the tumor microenvironment, Gal-3
null mice do not support xenograft lung tumor growth [108]. In
addition, Gal-1 association enhances H-Ras localization to the
inner leaflet of the plasma membrane, which is an important step
in constitutive activation [94]. The latter finding is intriguing given
the non-conventional pathway involving membrane-associated
Fig. 5 Galectins regulate hemostasis, angiogenesis, and tissue repair. Various members of the galectin family
regulate megakaryocyte activity, hemostasis, angiogenesis, epithelial migration, and general tissue repair
following injury. Representative galectin-regulated activities are shown. Red arrows indicate an activity that
the respective galectin increases, while blue arrows signify galectin-induced decreases in the accompanying
activity. Plt platelet
counter-receptors by which Gal-1 is secreted under homeostatic

conditions [78]. Adding clinical relevance to these findings, Chung
et al. found that tumor progression is strongly associated with
Gal-1 expression in clinical isolates of lung adenocarcinoma.
Using lung cancer cell lines, they went on to confirm that effects
of Gal-1 on the cellular level were attributable to an interaction
with Ras, driving COX expression and cellular proliferation.
Knockdown of Gal-1 was sufficient to inhibit proliferative and
increase sensitivity to chemotherapeutic agents [109]. The role of
extracellular Gal-1 as a negative regulator of tumor infiltrating
lymphocytes and angiogenesis [97, 110, 111] has also been well
characterized, so studies of galectin deficiency in mouse tumor
models must be carefully designed to isolate intracellular and extra-
cellular functions.
4.2 Tumor Tumor infiltration by lymphocytes has been linked to a survival

Microenvironment benefit in multiple cancers and predicts responsiveness to immuno-
therapy [112–115]. Intriguingly, multiple galectins including
Gal-1 and Gal-3 are often overexpressed in the tumor microenvi-

ronment [116]. While many of these studies are correlative, recent
work has begun to dissect specific mechanisms by which these
galectins may modulate the tumor microenvironment to promote
cancer cell survival and progression of disease. Tumor-secreted
Gal-3 binds to glycosylated IFN-γ, inhibiting its diffusion through
the tumor stroma, which is an important chemokine gradient for
tumor-infiltrating lymphocytes [117]. Furthermore, levels of Gal-1
in head and neck tumors are inversely correlated with therapeutic
efficacy of immune checkpoint inhibitors (ICIs). Kong et al. used
mouse models of head and neck cancer to investigate this correla-
tion, finding that galectin-1 acted on endothelial cells in the tumor
microenvironment to upregulate programmed death ligand
1 (PD-L1), which induces senescence in activated tumor-
infiltrating lymphocytes bearing PD1. Blocking galectin-1 in
these models rendered the tumor microenvironment permissive
to T-cell infiltration and more responsive to anti-PD1 therapy.
Together, these findings suggest that tumors can use immune
galectins to regulate the microenvironment and suppress the host
immune response to cancer [111].
4.3 Roles in the A large body of work on galectin functions has focused on roles
Mammalian Immune within the mammalian immune system. And while many of these
System studies have compellingly implicated galectins in immune processes
involving specific cell types—from developmental programing to
activation, effector function and resolution of inflammation—
global deletion of individual or multiple galectins in mice does
not cause gross immunodeficiency. Instead of fundamentally dis-
rupting the immune system, the absence of galectins results in a
host of subtle but important alterations in susceptibility to certain
autoimmune conditions and pathogens, signaling pathways, anti-
gen sensitivity, immune cell localization, and interaction. Together,
these changes suggest that ancestral galectins may have evolved to
more generally mediate multicellular cooperative interactions in
early metazoan lineages. Further supporting this framework for
galectin evolution, expression of galectins is broad and varied
beyond cells of the immune system—suggesting conserved func-
tions in tissue organization and homeostasis more generally. In this
model, the evolving vertebrate and mammalian immune systems
would have adopted the primordial galectins—as they have other
carbohydrate binding proteins to facilitate the myriad dynamic
responses necessary for the emergent complexity of our innate
and adaptive immune systems. Consistent with this model, it has
become clear that immune and non-immune cells regulate galectin
expression for numerous functions. Rather than exhaustively detail
observations of galectin function in specific immune cell popula-
tions, we seek here to highlight the themes of galectin function
within the mammalian immune system which occur in multiple cell
Fig. 6 Galectin regulation of immune cell function. Galectins have many putative functions in development and
functioning of the mammalian immune system. Early studies suggested that galectins could influence T-cell
viability and cytokine secretion through their direct interactions with surface receptors. In parallel, researchers
uncovered the ability of galectins to regulate granulocyte activation and turnover through a non-apoptotic
mechanism termed paraparesis. Later studies utilizing galectin null animal models revealed significant roles
for galectins in the development of immature B and T lymphocytes and their polarization toward different
inflammatory states. As intracellular functions for galectins were uncovered, their relevance to innate and
adaptive immune cell differentiation has been studied—revealing roles in modulation of intracellular signaling
pathway, membrane integrity and remodeling, as well as cell cycle progression and differentiation
types. In Fig. 6, we summarize major roles for galectins in the

regulation and function of the immune system.
4.3.1 Engagement and Early efforts to determine the activity of electrolectin (EL or Gal-1)
Crosslinking of Surface led investigators to administer the recombinant protein as a thera-
Receptors peutic for autoimmune myasthenia gravis in rabbits. Surprisingly,
administration of Gal-1 in this context prevented myasthenic symp-
tom onset and resolved the autoimmune disease completely. The
researchers further noted that the therapeutic lectin had no direct
effect on the magnitude of the humoral antibody response induced
in the model or on the diseased muscular junction itself. Rather, the
galectin was seen to bind and act upon lymphocytes [91]. These
initial observation led to a host of studies investigating the thera-
peutic effect of Gal-1 in animal models of autoimmune disease,
many of which found that administration of Gal-1 consistently
reduced the pathogenesis of T-cell-mediated autoimmunity [118–
121]. Building on these observations, Perillo et al. experimentally
tested the hypothesis that Gal-1 directly killed T cells, finding that
both recombinant Gal-1 and Gal-1 expressed on the cell surface of
cultured endothelial cells was capable of inducing apoptotic cell
death in activated, but not resting human T cells. Using antibody
inhibition assays and swainsonine treatment, they further demon-
strated that this effect was dependent on an interaction between
Gal-1 and glycans decorating specific surface receptors including
the common activated leukocyte antigens CD45RO and CD43.
Antibodies against other prominent T-cell receptors CD5, CD11a,
CD28, and CD44 failed to inhibit Gal-1-induced apoptosis
[122]. While future studies by Stowell et al. demonstrated that
aspects of these in vitro observations were mediated in part by the
effect of certain reducing agents on the T cells themselves [23], the
ability of extracellular Gal-1 to alter immune cell fate and function
by engaging specific glycoprotein receptors was established as a key
mechanistic theme of galectin biology. Supporting these findings
in vivo, endogenous Gal-1 expression has been shown in lymphoid
organs and implicated in the development and function of mature T
lymphocytes. Thymocyte-epithelial cells and thymocytes of the
thymic cortex, but not medulla, show surface expression of
human galectin-1 in vivo. And the association of Gal-1 with cortical
epithelial cells is dependent on core-2 O glycans on T-cell-asso-
ciated receptors CD43 and CD45, spatially and developmentally
regulated during thymocyte development [92]. Gal-1 has also
more recently been identified as a component of CD8+ T-cell
cytotoxic granules and may regulate FAS/FasL-dependent death
of target cells by altering cell death receptor endocytosis
[123]. Later studies of galectin–receptor interactions showed that
Gal-3 could also induce cell death in T cells and bound to a host of
cell surface receptors not all bound by Gal-1 including CD29,
CD98, CD71, CD43, and CD45 [124]. Gal-3 can also bind
directly to the T-cell receptor (TCR) to decrease antigen sensitivity

under the conditions of chronic infection in a manner dependent
on metabolic flux, IL-10-mediated upregulation of MGAT branch-
ing enzymes, and nano-scale membrane compartmentalization of
CD8 and TCR [86].
A similar cell surface receptor engagement mechanism has been
proposed for the regulation by Gal-1 of dendritic cell trafficking.
Dendritic cells bridge innate and adaptive immunity through their
numerous roles in pattern recognition, cytokine production, and
potent antigen presentation. DCs develop from circulating mono-
cytes which differentiate into mature subsets in specific tissues. In
response to inflammation and pathogen recognition, activated DCs
traverse the extracellular matrix and endothelial cell layers to reach
lymphatic channels in draining lymph nodes. Several studies have
demonstrated a role of endothelial Gal-1 in regulating this process.
Gal-1 is highly expressed in lymphatic endothelial cells and vascular
endothelium of inflamed tissues as well as the inflamed tissue itself.
Thiemann et al. demonstrated that Gal-1 in this context may inhibit
egress of immunogenic dendritic cells baring different core-2 O-
glycans on CD43, while permitting the extravasation of tolerogenic
DCs [125]. In addition to controlling the entry of inflammatory
DCs to lymphoid compartments, Gal-1 exposure itself may repro-
gram DCs to a more tolerogenic phenotype. Ilarregue et al. found
that Gal-1 primes mature DCs to enhance IL-10 production of
CD4+ T cells in co-culture. While the latter experiments relied on
exogenous treatment with Gal-1, they suggest the capacity of Gal-1
to modulate a multicellular circuit in the immune system in a
specific and reproducible manner [126].
Gal-9 may also act by a similar mechanism on CD4+ T-helper
type 1 (Th1) cells through a direct interaction with glycans on
T-cell immunoglobulin and mucin protein 3 (TIM-3). Zhu et al.
found that Gal-9 can bind to TIM-3 in vitro in a carbohydrate-
dependent manner and induce calcium-dependent cell death of
ex vivo Th1 but not Th2 cells. In the same study, injection of
Gal-9 also resulted in a reduction in Th1-mediated pathology in a
mouse model of multiple sclerosis. These findings are particularly
interesting in the context of resolution of antiviral and antitumoral
CD8+ T-cell-mediated immunity and the establishment of stable
T-cell memory as recent studies have defined a long-lived pool of
CXCR5+, TIM3-CD8+ T cells, presumably refractory to the effects
of Gal-9, which remain after chronic infection and mediate the
proliferative burst after PD-1 blockade during chronic
infection [127].
4.3.2 Regulation of Neutrophils and neutrophil-like phagocytic immune cells execute

Immune Cell Turnover some of the most primordial functions of multicellular immune
systems. Neutrophils develop in the bone marrow (accounting for
up to 60% of hematopoietic derived cells at homeostasis) and travel
to sites of inflammation through blood vessels. Egress of neutro-

phils from the vasculature proceeds in a well-characterized cascade
of selectin and integrin-mediated interactions with activated endo-
thelial cells, which precede para- or transcellular migration into
infected and inflamed tissues. Recruited to these sites of inflamma-
tion, neutrophils execute effector functions to eliminate microbes
and infected cells. The latter functions include phagocytosis of
opsonized cells, production of reactive oxygen species and neutro-
phil extracellular traps, and cytotoxic and microbicidal degranula-
tion. In the mammalian immune system, neutrophils can also
bridge innate and adaptive immune responses by acting as
antigen-presenting cells and modulating inflammation through
direct interactions with antigen-specific T and B cells.
Galectins can regulate neutrophils through several well-defined
pathways. Galectin-3 was originally discovered as a high-affinity
binding partner for IgE in rat basophilic leukemia cells
[128]. The overproduction of immunoglobulin E (IgE) with reac-
tivity toward common environmental allergens and self-antigen
drives type I hypersensitivity reactions associated with asthma and
atopic dermatitis. During the late stages of asthmatic reactions,
neutrophils accumulate at sites of inflammation in a manner requir-
ing binding to IgE. However, neutrophils do not express either of
two canonical IgE (Fcε) receptors. Instead, this interaction has
been shown to be mediated via galectin-3 IgE binding activity.
Neutrophil-associated galectin-3 binds IgE and activates the respi-
ratory burst needed for ROS effector functions [129]. Galectins
may also serve as tissue factors that regulate neutrophil turnover, a
key process in the resolution of inflammation. Early studies demon-
strated that Gal-1, Gal-2, Gal-3, Gal-4, and Gal-8 are able to
induce nonapoptotic phosphatidyl serine exposure in promyelocy-
tic HL-60 cells as well as neutrophils [130, 131]. These observa-
tions combined with data from galectin null animals, which can
exhibit defects in neutrophil turnover in tissues, are consistent with
a model in which galectins can induce PS exposure on immune cells
to mark them for a non-inflammatory turnover through phagocy-
tosis. The reversible oxidative inactivation of galectin-1 may also
serve to limit the activity of this galectin to sites of tissue damage,
thereby preventing further localized damage while allowing the
immune response to proceed systemically [132]. In Fig. 4, we
illustrate the redox modulation of Gal-1 activity. While in vivo
studies have provided some evidence consistent with this mecha-
nism, studies using conditional and tissue-specific knockout ani-
mals as well as immune cell transfers are needed to parse the
contributions of immune cell intrinsic and tissue-derived galectins
in the resolution of inflammation following infection.
4.3.3 Modulation of As in cancer signaling networks, inflammatory signaling pathways

Inflammatory Intracellular have evolved to adopt intracellular galectins as accessory proteins
Signaling Pathways and adaptors in the assembly of their most essential components.
While galectins do not typically take center stage in these pathways,
they have clearly evolved important modulatory roles in the ampli-
tude and outcome of their activation. Intracellular Gal-3 in macro-
phages have been shown to mediate Nod-like receptor family, pyrin
domain containing 3 (NLRP3) inflammasome activation during
cholestatic liver injury [133]. This observation was recapitulated
in studies demonstrating that Gal-3 may act to indirectly regulate
viral pathogenesis by modulating inflammasome activity during
H5N1 infection, enhancing IL1b production and promoting
inflammatory lung pathology [134].
Mature DCs extensively remodel membranous structures to
maximize surface exposure to the environment and to mediate
phagocytosis and other uptake mechanisms. Intracellular Gal-9
has been shown to regulate these membrane dynamics by modulat-
ing Rac1-dependent actin-cytoskeleton organization. Gal-9 is con-
sequently essential for maintaining the membrane integrity of
dendritic cells in vitro. Consistent with this finding, Gal-9 null
dendritic cells exhibit impaired phagocytic capacity [135].
Galectin-3 is highly expressed by macrophages and has been
shown to antagonize pro-inflammatory polarization, favoring a
fibrotic phenotype in several model systems [136] including
mouse models of atherosclerosis [137]. Galectin-3 promotes
these fibrotic phenotypes, which can be either pathological or
adaptive depending on the disease state, by driving alternative
over classical activation of recruited macrophages in a positive
feedback loop involving extracellular crosslinking of CD98 and
PI3K signaling. The absence or inhibition of Gal-3 is sufficient to
skew macrophages away from the alternative pathway toward clas-
sical activation [138].
4.3.4 Intracellular In addition to interacting with signaling complexes, cytoplasmic

Pattern Recognition and galectins can restrict the invasion of intracellular pathogens
Damage Sensing through recognition of host and pathogen glycans exposed during
endolysosomal damage. Recognition of pathogen and danger asso-
ciated molecular patterns (PAMPs and DAMPs) by pattern recog-
nition receptors (PRRs) is a hallmark of the innate immune system.
In addition to cytoplasmic and endolysosomal membrane-
associated sensors of viral RNA and DNA (RIG-I/MDA5,
TLR3/7/8/9, and STING) immune lectins, which can recognize
pathogen-specific glycan signatures, have also been well character-
ized as innate immune effectors. However, the best-studied exam-
ples of glycan sensing by host receptors occur in the extracellular
space. Examples include mannose binding lectin (MBL), a C-type
lectin, which binds specifically to mannosylated glycans produced
in abundance by certain pathogenic bacteria and parasites, initiating
formation of bactericidal MBL attack complexes or recruitment of

phagocytes within the lectin arm of the complement cascade
[139, 140]. In addition, Dectin-1 on macrophages and dendritic
cells directly recognizes β-glucans on fungal pathogens, resulting in
downstream Syk activation of an NFκB-mediated inflammatory
program, which in turn stimulates metabolic reprogramming,
phagocytosis, and production of inflammatory cytokines: IL-6,
IL-2, IL-23, and TNF [141].
While the cytosolic localization of galectins has long made
them prime candidate sensors of intracellular pathogen glycan sig-
natures, few studies have rigorously demonstrated this phenome-
non experimentally. However, Thurston et al. [81] have elegantly
shown that intracellular galectin-8 can initiate antibacterial autop-
hagy during invasion of the cell by Salmonella typhimurium. Mech-
anistically, cytoplasmic galectin-8 is recruited to S. typhimurium
containing vesicles (SCVs) as they rupture, exposing host and
bacterial glycans to the cytosol. The 2,3-sialyl-binding
N-terminus of galectin-8 is required for recruitment to these dam-
aged vesicles. Host and bacterial glycan-bound galectin-8 then
serves as an adaptor for the autophagy initiating NDP52, the
recruitment of which is necessary for galectin-8-mediated restric-
tion of S. typhimurium invasion and replication. Consistent with its
role as a damage sensor during S. typhimurium infection, galectin-
8 also localizes to lysosomes damaged by the vacuolating cytotoxin
A (Vac-A) produced by H. pylori where it also induces autophagy.
Mechanistically, this process was found to be dependent on toxin
production by the bacteria and host O-glycan synthesis
[142]. Intriguingly, intracellular galectin-3 and galectin-9 (but
not Gal-1, 2, 4, 7, 10, 12, 13, 14, or HSPC159) were also recruited
to damaged vesicles in these studies. While they were not required
to induce autophagy or restrict infection by the specific bacterial
isolates tested, the ability of galectin-3 and galectin-9 to recognize
endolysosomal damage is likely to have alternative immune or
homeostatic functions. Indeed, galectin-3 has been shown to inter-
act with the E3 ubiquitin ligase TRIM16 to mediate selective
autophagy during Mycobacterium tuberculosis infection [143]. In
addition, earlier studies using immuno-electron microscopy
observed accumulation of galectin-3 in and around Shigella flexneri
containing vesicles in a glycan-binding dependent manner
[144]. Together these findings implicate specific cytosolic galectins
as damage associated molecular pattern sensors for the endolyso-
somal system, a common site of invasion for intracellular
pathogens.
4.3.5 Galectin–Pathogen A major strategy used to probe the endogenous roles of galectins in
Interactions the immune system has been to infect galectin-null systems (mice
or cells) with various types and subtypes of bacterial, viral, fungal,
or parasitic pathogens and monitor alterations in pathogenesis and
Fig. 7 Galectins recognize a diverse range of pathogens. While the immunoregulatory roles of galectins likely
represent some of their most well-known functions, the ability of galectins to recognize a diverse range of
pathogens may reflect some of their earliest evolutionary activities. Galectins recognition of pathogens can
result in opsonization or direct microbial killing. In contrast, pathogens may utilize galectins to facilitate
attachment and invasion. Representative galectin-regulated activities are shown. Red arrows indicate an
activity that the respective galectin increases, while blue arrows signify galectin-induced decreases in the
accompanying activity
immune response attributable to the missing galectin [145]. In

some cases, there is evidence that galectins can directly engage
glycan ligands on the pathogen causing death, blocking infection,
or altering some aspect of its life cycle. This function is exemplified
by work showing that bacterially expressed blood group antigen
mimetics are targets for bactericidal activity of Gal-4 and Gal-8
[146]. In other studies, intracellular and extracellular galectin func-
tions within or on responding immune cells or affected tissues are at
cause. In the following section, we highlight studies which suggest
that a direct interaction may occur between galectins and invading
or commensal microorganisms. In Fig. 7, we detail significant
galectin–pathogen interactions.
4.3.6 Bacteria The role of galectins in direct immunity against microorganisms

was first suggested by studies of galectin binding to glycan struc-
tures on lipopolysaccharide (LPS) from important human bacterial
pathogens including Neisseria meningitidis, Escherichia coli, Klebsi-
ella pneumonia, and Pseudomonas aeruginosa. These studies largely
focused on galectin-3 and demonstrated binding to isolated
structures on LPS. Use of glycan microarrays and genetic variants of

live bacteria demonstrated that among the innate immune lectins,
galectin-3, 4, and 8 had remarkable specificity for bacterial molecu-
lar mimics of human ABO(H) blood group antigens [147]. The
possible evolutionary reason for these observations involving the
host response to molecular mimicry was demonstrated by several
recent studies. Pathogens have evolved glycosylation enzymes that
produce terminal glycans structurally similar or identical to human
ABO(H) blood group antigens [148]. These modifications are
thought to subvert the mammalian adaptive immune system,
which has evolved a system of negative selection that deletes effec-
tor immune cells carrying receptors with autoreactive specificity
[149, 150]. As a result, individuals of each blood group are unable
to produce antibodies with binding specificity to self-blood group
antigens which may decorate an invading pathogen. Stowell et al.
demonstrated a mechanism by which galectins (in particular
galectin-4 and 8) may fill this gap in adaptive immunity by directly
killing microbes bearing ABO(H) glycan structures [146]. The
many proposed and elucidated functions of specific galectins within
immune cells poses a major challenge to demonstrating galectin
functions as direct antimicrobial effectors. Additional studies using
global and conditional knockout models are needed to dissect the
role of this activity in vivo. Despite the ability of galectin-3 to bind
both bacterial pathogens in vitro, studies demonstrating increased
susceptibility of galectin-3 null mice to Helicobacter pylori and
Neisseria meningitidis could equally be due to deficits in immune
cell function [151–153].
4.3.7 Viruses Galectins can act in the innate antiviral immune response on multi-
ple levels—either by directly interacting with viral glycoproteins or
components of viral assembly factories or by modifying innate
immune signaling pathways such as the inflammasome. Galectin-1
is upregulated in the lung during influenza A virus infection and
may directly interact with the virus to block infection or viral
budding [154]. Supporting a potential role for galectin-1 in influ-
enza virus infection, a genome-wide association study found that
variants of human LGALS1 resulting in higher mRNA expression
were associated with significant reduction in mortality during poul-
try farm derived influenza A (H7N9) infection [155]. In vitro
studies have also shown that galectin-1 can bind to Nipah virus F
glycoprotein to inhibit infection-induce syncytia formation in mul-
tiple cell types [156, 157]. On the other hand, extracellular galec-
tin-1 may also serve to mediate HIV infection or reservoir
establishment as it enhances infection in vitro and exhibits broad
and stable expression in lymphoid tissues [158]. HIV has been
shown to exploit intracellular galectin-3 activity in the ESCRT
pathway to stabilize assembly and budding of new virions
[159]. While the aforementioned role of Gal-8 in intracellular

endolysosomal damage sensing has mostly been studied in the
context of intracellular bacteria, this function may also be at play
during adenovirus infection, where gal-8 can recognize a PPxY
motif in the viral capsid protein to initiate autophagy [160].
5 Summary and Conclusions
Many of the evolving mechanisms of galectin function point to a

common theme in the evolution of ancient protein families: the
layering and elaboration of multiple functions on proteins with
single core functions. In the case of galectins, the conservation of
carbohydrate recognition and its role in stabilizing and regulating
the components of the ECM may well have been essential for the
evolution of multicellular life. The primordial extracellular role may
well have evolved within the mammalian immune system to encom-
pass a host of specific galectin-receptor interactions important for
development, activation, differentiation, and interaction of multi-
ple immune cell populations. However, just as components of the
cytoskeleton play numerous roles not just in reinforcing the cell
structurally but also in signaling, metabolite sensing, cell division,
and oncogenesis, galectin functions have diverged to include intra-
cellular danger and pattern recognition, direct microbial killing,
and even stabilization of pre-mRNA splicing components. The
ubiquity and multi-compartmentalization of these galectin locali-
zation and functions and the potential for galectins to complement
one another likely explain their roles in oncogenic transformation
and progression of disease.
References
1. Thiemann S, Baum LG (2016) Galectins and mediators of tumor progression. J Exp Med
immune responses-just how do they do those 217(2):e20182041. https://doi.org/10.
things they do? Annu Rev Immunol 34: 1084/jem.20182041
243–264. https://doi.org/10.1146/ 5. Teichberg VI, Silman I, Beitsch DD, Resheff
annurev-immunol-041015-055402 G (1975) A beta-D-galactoside binding pro-
2. Robinson BS, Arthur CM, Evavold B, tein from electric organ tissue of electropho-
Roback E, Kamili NA, Stowell CS, Vallecillo- rus electricus. Proc Natl Acad Sci U S A
Zuniga ML, Van Ry PM, Dias-Baruffi M, 72(4):1383–1387
Cummings RD, Stowell SR (2019) The 6. Arthur CM, Patel SR, Mener A, Kamili NA,
sweet-side of leukocytes: galectins as master Fasano RM, Meyer E, Winkler AM, Sola-
regulators of neutrophil function. Front Visner M, Josephson CD, Stowell SR (2015)
Immunol 10:1762. https://doi.org/10. Innate immunity against molecular mimicry:
3389/fimmu.2019.01762 examining galectin-mediated antimicrobial
3. Johannes L, Jacob R, Leffler H (2018) Galec- activity. BioEssays 37(12):1327–1337.
tins at a glance. J Cell Sci 131(9):jcs208884. https://doi.org/10.1002/bies.201500055
https://doi.org/10.1242/jcs.208884 7. Lowe JB (2001) Glycosylation, immunity,
4. Girotti MR, Salatino M, Dalotto-Moreno T, and autoimmunity. Cell 104(6):809–812.
Rabinovich GA (2020) Sweetening the hall- https://doi.org/10.1016/s0092-8674(01)
marks of cancer: galectins as multifunctional 00277-x
8. Barondes SH, Castronovo V, Cooper DN, 18. Carlsson S, Oberg CT, Carlsson MC,
Cummings RD, Drickamer K, Feizi T, Gitt Sundin A, Nilsson UJ, Smith D, Cummings
MA, Hirabayashi J, Hughes C, Kasai K et al RD, Almkvist J, Karlsson A, Leffler H (2007)
(1994) Galectins: a family of animal beta- Affinity of galectin-8 and its carbohydrate rec-
galactoside-binding lectins. Cell ognition domains for ligands in solution and
76(4):597–598. https://doi.org/10.1016/ at the cell surface. Glycobiology
0092-8674(94)90498-7 17(6):663–676. https://doi.org/10.1093/
9. Arthur CM, Baruffi MD, Cummings RD, glycob/cwm026
Stowell SR (2015) Evolving mechanistic 19. Di Lella S, Marti MA, Croci DO, Guardia
insights into galectin functions. Methods CM, Diaz-Ricci JC, Rabinovich GA, Cara-
Mol Biol 1207:1–35. https://doi.org/10. melo JJ, Estrin DA (2010) Linking the struc-
1007/978-1-4939-1396-1_1 ture and thermal stability of beta-galactoside-
10. Mendel G (1865) Experiments in plant binding protein galectin-1 to ligand binding
hybridization 41 and dimerization equilibria. Biochemistry
11. Lauc G, Kristic J, Zoldos V (2014) Glycans— 49(35):7652–7658. https://doi.org/10.
the third revolution in evolution. Front Genet 1021/bi100356g
5:145. https://doi.org/10.3389/fgene. 20. Di Lella S, Sundblad V, Cerliani JP, Guardia
2014.00145 CM, Estrin DA, Vasta GR, Rabinovich GA
12. Houzelstein D, Goncalves IR, Fadden AJ, (2011) When galectins recognize glycans:
Sidhu SS, Cooper DN, Drickamer K, from biochemistry to physiology and back
Leffler H, Poirier F (2004) Phylogenetic anal- again. Biochemistry 50(37):7842–7857.
ysis of the vertebrate galectin family. Mol Biol https://doi.org/10.1021/bi201121m
Evol 21(7):1177–1187. https://doi.org/10. 21. Stowell SR, Cho M, Feasley CL, Arthur CM,
1093/molbev/msh082 Song X, Colucci JK, Karmakar S, Mehta P,
13. Brewer CF (2002) Thermodynamic binding Dias-Baruffi M, McEver RP, Cummings RD
studies of galectin-1, -3 and -7. Glycoconj J (2009) Ligand reduces galectin-1 sensitivity
19(7–9):459–465. https://doi.org/10. to oxidative inactivation by enhancing dimer
1023/B:GLYC.0000014075.62724.d0 formation. J Biol Chem 284(8):4989–4999.
https://doi.org/10.1074/jbc.M808925200
14. Stowell SR, Arthur CM, Mehta P, Slanina KA,
Blixt O, Leffler H, Smith DF, Cummings RD 22. Dias-Baruffi M, Zhu H, Cho M, Karmakar S,
(2008) Galectin-1, -2, and -3 exhibit differen- McEver RP, Cummings RD (2003) Dimeric
tial recognition of sialylated glycans and blood galectin-1 induces surface exposure of phos-
group antigens. J Biol Chem phatidylserine and phagocytic recognition of
283(15):10109–10123. https://doi.org/10. leukocytes without inducing apoptosis. J Biol
1074/jbc.M709545200 Chem 278(42):41282–41293. https://doi.
org/10.1074/jbc.M306624200
15. Arthur CM, Rodrigues LC, Baruffi MD, Sul-
livan HC, Heimburg-Molinaro J, Smith DF, 23. Stowell SR, Arthur CM, Slanina KA, Horton
Cummings RD, Stowell SR (2015) Examin- JR, Smith DF, Cummings RD (2008)
ing galectin binding specificity using glycan Dimeric galectin-8 induces phosphatidylser-
microarrays. Methods Mol Biol 1207: ine exposure in leukocytes through polylacto-
115–131. https://doi.org/10.1007/978-1- samine recognition by the C-terminal
4939-1396-1_8 domain. J Biol Chem
283(29):20547–20559. https://doi.org/10.
16. Hirabayashi J, Hashidate T, Arata Y, Nishi N, 1074/jbc.M802495200
Nakamura T, Hirashima M, Urashima T,
Oka T, Futai M, Muller WE, Yagi F, Kasai K 24. Arthur CM, Rodrigues LC, Baruffi MD, Sul-
(2002) Oligosaccharide specificity of galec- livan HC, Cummings RD, Stowell SR (2015)
tins: a search by frontal affinity chromatogra- Detection of phosphatidylserine exposure on
phy. Biochim Biophys Acta 1572 leukocytes following treatment with human
(2–3):232–254. https://doi.org/10.1016/ galectins. Methods Mol Biol 1207:185–200.
s0304-4165(02)00311-2 https://doi.org/10.1007/978-1-4939-
1396-1_12
17. Sun Y, Cheng L, Gu Y, Xin A, Wu B, Zhou S,
Guo S, Liu Y, Diao H, Shi H, Wang G, Tao 25. Flores-Ibarra A, Vertesy S, Medrano FJ,
SC (2016) A human lectin microarray for Gabius HJ, Romero A (2018) Crystallization
sperm surface glycosylation analysis. Mol of a human galectin-3 variant with two
Cell Proteomics 15(9):2839–2851. https:// ordered segments in the shortened
doi.org/10.1074/mcp.M116.059311 N-terminal tail. Sci Rep 8(1):9835. https://
doi.org/10.1038/s41598-018-28235-x
26. Mazurek N, Conklin J, Byrd JC, Raz A, Bre- 36. Colnot C, Ripoche MA, Scaerou F, Foulis D,
salier RS (2000) Phosphorylation of the beta- Poirier F (1996) Galectins in mouse embryo-
galactoside-binding protein galectin-3 modu- genesis. Biochem Soc Trans 24(1):141–146.
lates binding to its ligands. J Biol Chem https://doi.org/10.1042/bst0240141
275(46):36311–36315. https://doi.org/10. 37. Wada J, Ota K, Kumar A, Wallner EI, Kanwar
1074/jbc.M003831200 YS (1997) Developmental regulation, expres-
27. Yoshii T, Fukumori T, Honjo Y, Inohara H, sion, and apoptotic potential of galectin-9, a
Kim HR, Raz A (2002) Galectin-3 phosphor- beta-galactoside binding lectin. J Clin Invest
ylation is required for its anti-apoptotic func- 99(10):2452–2461. https://doi.org/10.
tion and cell cycle arrest. J Biol Chem 1172/JCI119429
277(9):6852–6857. https://doi.org/10. 38. Benvenuto G, Carpentieri ML, Salvatore P,
1074/jbc.M107668200 Cindolo L, Bruni CB, Chiariotti L (1996)
28. Menon S, Kang CM, Beningo KA (2011) Cell-specific transcriptional regulation and
Galectin-3 secretion and tyrosine phosphory- reactivation of galectin-1 gene expression are
lation is dependent on the calpain small sub- controlled by DNA methylation of the pro-
unit, Calpain 4. Biochem Biophys Res moter region. Mol Cell Biol
Commun 410(1):91–96. https://doi.org/ 16(6):2736–2743. https://doi.org/10.
10.1016/j.bbrc.2011.05.112 1128/mcb.16.6.2736
29. Kamili NA, Arthur CM, Gerner-Smidt C, 39. Toscano MA, Campagna L, Molinero LL,
Tafesse E, Blenda A, Dias-Baruffi M, Stowell Cerliani JP, Croci DO, Ilarregui JM, Fuertes
SR (2016) Key regulators of galectin-glycan MB, Nojek IM, Fededa JP, Zwirner NW,
interactions. Proteomics 16(24):3111–3125. Costas MA, Rabinovich GA (2011) Nuclear
https://doi.org/10.1002/pmic.201600116 factor (NF)-kappaB controls expression of the
30. Ju T, Lanneau GS, Gautam T, Wang Y, Xia B, immunoregulatory glycan-binding protein
Stowell SR, Willard MT, Wang W, Xia JY, galectin-1. Mol Immunol 48
Zuna RE, Laszik Z, Benbrook DM, Hanigan (15–16):1940–1949. https://doi.org/10.
MH, Cummings RD (2008) Human tumor 1016/j.molimm.2011.05.021
antigens Tn and sialyl Tn arise from mutations 40. Than NG, Romero R, Erez O, Weckle A,
in Cosmc. Cancer Res 68(6):1636–1646. Tarca AL, Hotra J, Abbas A, Han YM, Kim
https://doi.org/10.1158/0008-5472.CAN- SS, Kusanovic JP, Gotsch F, Hou Z,
07-2345 Santolaya-Forgas J, Benirschke K, Papp Z,
31. Stowell SR, Ju T, Cummings RD (2015) Pro- Grossman LI, Goodman M, Wildman DE
tein glycosylation in cancer. Annu Rev Pathol (2008) Emergence of hormonal and redox
10:473–510. https://doi.org/10.1146/ regulation of galectin-1 in placental mam-
annurev-pathol-012414-040438 mals: implication in maternal-fetal immune
32. Arthur CM, Cummings RD, Stowell SR tolerance. Proc Natl Acad Sci U S A
(2014) Using glycan microarrays to under- 105(41):15819–15824. https://doi.org/10.
stand immunity. Curr Opin Chem Biol 18: 1073/pnas.0807606105
55–61. https://doi.org/10.1016/j.cbpa. 41. Than NG, Romero R, Xu Y, Erez O, Xu Z,
2013.12.017 Bhatti G, Leavitt R, Chung TH,
33. Colnot C, Fowlis D, Ripoche MA, El-Azzamy H, LaJeunesse C, Wang B,
Bouchaert I, Poirier F (1998) Embryonic Balogh A, Szalai G, Land S, Dong Z, Hassan
implantation in galectin 1/galectin 3 double SS, Chaiworapongsa T, Krispin M, Kim CJ,
mutant mice. Dev Dyn 211(4):306–313. Tarca AL, Papp Z, Bohn H (2014) Evolution-
https://doi.org/10.1002/(SICI)1097-0177 ary origins of the placental expression of chro-
(199804)211:4<306::AID-AJA2>3.0. mosome 19 cluster galectins and their
CO;2-L complex dysregulation in preeclampsia. Pla-
centa 35(11):855–865. https://doi.org/10.
34. Poirier F, Robertson EJ (1993) Normal 1016/j.placenta.2014.07.015
development of mice carrying a null mutation
in the gene encoding the L14 S-type lectin. 42. Ely ZA, Moon JM, Sliwoski GR, Sangha AK,
Development 119(4):1229–1236 Shen XX, Labella AL, Meiler J, Capra JA,
Rokas A (2019) The impact of natural selec-
35. Poirier F, Timmons PM, Chan CT, Guenet tion on the evolution and function of placen-
JL, Rigby PW (1992) Expression of the L14 tally expressed galectins. Genome Biol Evol
lectin during mouse embryogenesis suggests 11(9):2574–2592. https://doi.org/10.
multiple roles during pre- and post- 1093/gbe/evz183
implantation development. Development
115(1):143–155 43. Than NG, Romero R, Goodman M,
Weckle A, Xing J, Dong Z, Xu Y, Tarquini F,
Szilagyi A, Gal P, Hou Z, Tarca AL, Kim CJ, eosinophil Charcot-Leyden crystal protein
Kim JS, Haidarian S, Uddin M, Bohn H, (galectin-10): a crystallographic study at 1.8
Benirschke K, Santolaya-Forgas J, Grossman A resolution. Biochemistry
LI, Erez O, Hassan SS, Zavodszky P, Papp Z, 38(42):13837–13843. https://doi.org/10.
Wildman DE (2009) A primate subfamily of 1021/bi990756e
galectins expressed at the maternal-fetal inter- 52. Yang RY, Yu L, Graham JL, Hsu DK, Lloyd
face that promote immune cell death. Proc KC, Havel PJ, Liu FT (2011) Ablation of a
Natl Acad Sci U S A 106(24):9731–9736. galectin preferentially expressed in adipocytes
h t t p s : // d o i . o r g / 1 0 . 1 0 7 3 / p n a s . increases lipolysis, reduces adiposity, and
0903568106 improves insulin sensitivity in mice. Proc
44. Than NG, Romero R, Kim CJ, McGowen Natl Acad Sci U S A 108(46):18696–18701.
MR, Papp Z, Wildman DE (2012) Galectins: h t t p s : // d o i . o r g / 1 0 . 1 0 7 3 / p n a s .
guardians of eutherian pregnancy at the 1109065108
maternal-fetal interface. Trends Endocrinol 53. Dagher SF, Wang JL, Patterson RJ (1995)
Metab 23(1):23–31. https://doi.org/10. Identification of galectin-3 as a factor in
1016/j.tem.2011.09.003 pre-mRNA splicing. Proc Natl Acad Sci U S
45. Koopman LA, Kopcow HD, Rybalov B, Boy- A 92(4):1213–1217. https://doi.org/10.
son JE, Orange JS, Schatz F, Masch R, Lock- 1073/pnas.92.4.1213
wood CJ, Schachter AD, Park PJ, Strominger 54. Vyakarnam A, Lenneman AJ, Lakkides KM,
JL (2003) Human decidual natural killer cells Patterson RJ, Wang JL (1998) A comparative
are a unique NK cell subset with immuno- nuclear localization study of galectin-1 with
modulatory potential. J Exp Med other splicing components. Exp Cell Res
198(8):1201–1212. https://doi.org/10. 242(2):419–428. https://doi.org/10.1006/
1084/jem.20030305 excr.1998.4111
46. Garin MI, Chu CC, Golshayan D, Cernuda- 55. Wang W, Park JW, Wang JL, Patterson RJ
Morollon E, Wait R, Lechler RI (2007) (2006) Immunoprecipitation of spliceosomal
Galectin-1: a key effector of regulation RNAs by antisera to galectin-1 and galectin-3.
mediated by CD4+CD25+ T cells. Blood Nucleic Acids Res 34(18):5166–5174.
109(5):2058–2065. https://doi.org/10. https://doi.org/10.1093/nar/gkl673
1182/blood-2006-04-016451 56. Nakahara S, Hogan V, Inohara H, Raz A
47. Liu FT (2000) Galectins: a new family of (2006) Importin-mediated nuclear transloca-
regulators of inflammation. Clin Immunol tion of galectin-3. J Biol Chem
97(2):79–88. https://doi.org/10.1006/ 281(51):39649–39659. https://doi.org/10.
clim.2000.4912 1074/jbc.M608069200
48. Hsu DK, Hammes SR, Kuwabara I, Greene 57. Huflejt ME, Turck CW, Lindstedt R, Bar-
WC, Liu FT (1996) Human T lymphotropic ondes SH, Leffler H (1993) L-29, a soluble
virus-I infection of human T lymphocytes lactose-binding lectin, is phosphorylated on
induces expression of the beta-galactoside- serine 6 and serine 12 in vivo and by casein
binding lectin, galectin-3. Am J Pathol kinase I. J Biol Chem 268(35):26712–26718
148(5):1661–1670 58. Wang JL, Gray RM, Haudek KC, Patterson
49. Ackerman SJ, Corrette SE, Rosenberg HF, RJ (2004) Nucleocytoplasmic lectins. Bio-
Bennett JC, Mastrianni DM, Nicholson- chim Biophys Acta 1673(1–2):75–93.
Weller A, Weller PF, Chin DT, Tenen DG https://doi.org/10.1016/j.bbagen.2004.
(1993) Molecular cloning and characteriza- 03.013
tion of human eosinophil Charcot-Leyden 59. Haudek KC, Voss PG, Locascio LE, Wang JL,
crystal protein (lysophospholipase). Similari- Patterson RJ (2009) A mechanism for incor-
ties to IgE binding proteins and the S-type poration of galectin-3 into the spliceosome
animal lectin superfamily. J Immunol through its association with U1 snRNP. Bio-
150(2):456–468 chemistry 48(32):7705–7712. https://doi.
50. Dor PJ, Ackerman SJ, Gleich GJ (1984) org/10.1021/bi900071b
Charcot-Leyden crystal protein and eosino- 60. Gray RM, Davis MJ, Ruby KM, Voss PG,
phil granule major basic protein in sputum of Patterson RJ, Wang JL (2008) Distinct effects
patients with respiratory diseases. Am Rev on splicing of two monoclonal antibodies
Respir Dis 130(6):1072–1077. https://doi. directed against the amino-terminal domain
org/10.1164/arrd.1984.130.6.1072 of galectin-3. Arch Biochem Biophys
51. Swaminathan GJ, Leonidas DD, Savage MP, 475(2):100–108. https://doi.org/10.1016/
Ackerman SJ, Acharya KR (1999) Selective j.abb.2008.04.010
recognition of mannose by the human
61. Park JW, Voss PG, Grabski S, Wang JL, Pat- of galectin-3 expression promotes both
terson RJ (2001) Association of galectin-1 in vitro and in vivo drug-induced apoptosis
and galectin-3 with Gemin4 in complexes of human pancreatic carcinoma cells. Clin Exp
containing the SMN protein. Nucleic Acids Metastasis 28(4):367–376. https://doi.org/
Res 29(17):3595–3602. https://doi.org/10. 10.1007/s10585-011-9376-x
1093/nar/29.17.3595 71. Califice S, Castronovo V, Bracke M, van den
62. Coppin L, Vincent A, Frenois F, Duchene B, Brule F (2004) Dual activities of galectin-3 in
Lahdaoui F, Stechly L, Renaud F, Villenet C, human prostate cancer: tumor suppression of
Van Seuningen I, Leteurtre E, Dion J, nuclear galectin-3 vs tumor promotion of
Grandjean C, Poirier F, Figeac M, cytoplasmic galectin-3. Oncogene
Delacour D, Porchet N, Pigny P (2017) 23(45):7527–7536. https://doi.org/10.
Galectin-3 is a non-classic RNA binding pro- 1038/sj.onc.1207997
tein that stabilizes the mucin MUC4 mRNA 72. Takenaka Y, Fukumori T, Yoshii T, Oka N,
in the cytoplasm of cancer cells. Sci Rep 7: Inohara H, Kim HR, Bresalier RS, Raz A
4 3 9 2 7 . h t t p s : // d o i . o r g / 1 0 . 1 0 3 8 / (2004) Nuclear export of phosphorylated
srep43927 galectin-3 regulates its antiapoptotic activity
63. Bafna S, Kaur S, Batra SK (2010) Membrane- in response to chemotherapeutic drugs. Mol
bound mucins: the mechanistic basis for Cell Biol 24(10):4395–4406. https://doi.
alterations in the growth and survival of can- org/10.1128/mcb.24.10.4395-4406.2004
cer cells. Oncogene 29(20):2893–2904. 73. Bernerd F, Sarasin A, Magnaldo T (1999)
https://doi.org/10.1038/onc.2010.87 Galectin-7 overexpression is associated with
64. Golling G, Amsterdam A, Sun Z, the apoptotic process in UVB-induced sun-
Antonelli M, Maldonado E, Chen W, burn keratinocytes. Proc Natl Acad Sci U S
Burgess S, Haldi M, Artzt K, Farrington S, A 96(20):11329–11334. https://doi.org/
Lin SY, Nissen RM, Hopkins N (2002) Inser- 10.1073/pnas.96.20.11329
tional mutagenesis in zebrafish rapidly identi- 74. Cooper DN, Barondes SH (1990) Evidence
fies genes essential for early vertebrate for export of a muscle lectin from cytosol to
development. Nat Genet 31(2):135–140. extracellular matrix and for a novel secretory
https://doi.org/10.1038/ng896 mechanism. J Cell Biol 110(5):1681–1691.
65. Yang RY, Hsu DK, Liu FT (1996) Expression https://doi.org/10.1083/jcb.110.5.1681
of galectin-3 modulates T-cell growth and 75. Obino D, Fetler L, Soza A, Malbec O, Saez JJ,
apoptosis. Proc Natl Acad Sci U S A Labarca M, Oyanadel C, Del Valle BF,
93(13):6737–6742. https://doi.org/10. Goles N, Chikina A, Lankar D, Segovia-
1073/pnas.93.13.6737 Miranda F, Garcia C, Leger T, Gonzalez A,
66. Akahani S, Nangia-Makker P, Inohara H, Kim Espeli M, Lennon-Dumenil AM, Yuseff MI
HR, Raz A (1997) Galectin-3: a novel anti- (2018) Galectin-8 favors the presentation of
apoptotic molecule with a functional BH1 surface-tethered antigens by stabilizing the B
(NWGR) domain of Bcl-2 family. Cancer Res cell immune synapse. Cell Rep
57(23):5272–5276 25(11):3110–3122 e3116. https://doi.org/
67. Wang Y, Balan V, Gao X, Reddy PG, Kho D, 10.1016/j.celrep.2018.11.052
Tait L, Raz A (2013) The significance of 76. Cerri DG, Rodrigues LC, Stowell SR, Araujo
galectin-3 as a new basal cell marker in pros- DD, Coelho MC, Oliveira SR, Bizario JC,
tate cancer. Cell Death Dis 4:e753. https:// Cummings RD, Dias-Baruffi M, Costa MC
doi.org/10.1038/cddis.2013.277 (2008) Degeneration of dystrophic or injured
68. Cheng D, Liang B, Li Y (2015) Serum skeletal muscles induces high expression of
galectin-3 as a potential marker for gastric galectin-1. Glycobiology 18(11):842–850.
cancer. Med Sci Monit 21:755–760. https:// https://doi.org/10.1093/glycob/cwn079
doi.org/10.12659/MSM.892386 77. Vallecillo-Zuniga ML, Rathgeber MF, Poul-
69. Mazurek N, Byrd JC, Sun Y, Hafley M, son PD, Hayes S, Luddington JS, Gill HN,
Ramirez K, Burks J, Bresalier RS (2012) Teynor M, Kartchner BC, Valdoz J,
Cell-surface galectin-3 confers resistance to Stowell C, Markham AR, Arthur C,
TRAIL by impeding trafficking of death Stowell S, Van Ry PM (2020) Treatment
receptors in metastatic colon adenocarcinoma with galectin-1 improves myogenic potential
cells. Cell Death Differ 19(3):523–533. and membrane repair in dysferlin-deficient
https://doi.org/10.1038/cdd.2011.123 models. PLoS One 15(9):e0238441.
70. Kobayashi T, Shimura T, Yajima T, Kubo N, https://doi.org/10.1371/journal.pone.
Araki K, Wada W, Tsutsumi S, Suzuki H, 0238441
Kuwano H, Raz A (2011) Transient silencing
78. Seelenmeyer C, Wegehingel S, Tews I, 87. Demetriou M, Granovsky M, Quaggin S,

Kunzler M, Aebi M, Nickel W (2005) Cell Dennis JW (2001) Negative regulation of
surface counter receptors are essential compo- T-cell activation and autoimmunity by
nents of the unconventional export machin- Mgat5 N-glycosylation. Nature
ery of galectin-1. J Cell Biol 171(2):373–381. 409(6821):733–739. https://doi.org/10.
https://doi.org/10.1083/jcb.200506026 1038/35055582
79. Cho M, Cummings RD (1995) Galectin-1, a 88. Mkhikian H, Grigorian A, Li CF, Chen HL,
beta-galactoside-binding lectin in Chinese Newton B, Zhou RW, Beeton C, Torossian S,
hamster ovary cells. II. Localization and bio- Tatarian GG, Lee SU, Lau K, Walker E, Simi-
synthesis. J Biol Chem 270(10):5207–5212. novitch KA, Chandy KG, Yu Z, Dennis JW,
https://doi.org/10.1074/jbc.270.10.5207 Demetriou M (2011) Genetics and the envi-
80. Banfer S, Schneider D, Dewes J, Strauss MT, ronment converge to dysregulate
Freibert SA, Heimerl T, Maier UG, Elsasser N-glycosylation in multiple sclerosis. Nat
HP, Jungmann R, Jacob R (2018) Molecular Commun 2:334. https://doi.org/10.1038/
mechanism to recruit galectin-3 into multi- ncomms1333
vesicular bodies for polarized exosomal secre- 89. Li CF, Zhou RW, Mkhikian H, Newton BL,
tion. Proc Natl Acad Sci U S A 115(19): Yu Z, Demetriou M (2013) Hypomorphic
E4396–E4405. https://doi.org/10.1073/ MGAT5 polymorphisms promote multiple
pnas.1718921115 sclerosis cooperatively with MGAT1 and
81. Thurston TL, Wandel MP, von Muhlinen N, interleukin-2 and 7 receptor variants. J Neu-
Foeglein A, Randow F (2012) Galectin 8 tar- roimmunol 256(1–2):71–76. https://doi.
gets damaged vesicles for autophagy to defend org/10.1016/j.jneuroim.2012.12.008
cells against bacterial invasion. Nature 90. Yu Z, Li CF, Mkhikian H, Zhou RW, Newton
482(7385):414–418. https://doi.org/10. BL, Demetriou M (2014) Family studies of
1038/nature10744 type 1 diabetes reveal additive and epistatic
82. Patnaik SK, Potvin B, Carlsson S, Sturm D, effects between MGAT1 and three other poly-
Leffler H, Stanley P (2006) Complex morphisms. Genes Immun 15(4):218–223.
N-glycans are the major ligands for galectin- https://doi.org/10.1038/gene.2014.7
1, -3, and -8 on Chinese hamster ovary cells. 91. Levi G, Tarrab-Hazdai R, Teichberg VI
Glycobiology 16(4):305–317. https://doi. (1983) Prevention and therapy with electro-
org/10.1093/glycob/cwj063 lectin of experimental autoimmune myasthe-
83. Nabi IR, Shankar J, Dennis JW (2015) The nia gravis in rabbits. Eur J Immunol
galectin lattice at a glance. J Cell Sci 13(6):500–507. https://doi.org/10.1002/
128(13):2213–2219. https://doi.org/10. eji.1830130613
1242/jcs.151159 92. Baum LG, Pang M, Perillo NL, Wu T,
84. Apweiler R, Hermjakob H, Sharon N (1999) Delegeane A, Uittenbogaart CH, Fukuda M,
On the frequency of protein glycosylation, as Seilhamer JJ (1995) Human thymic epithelial
deduced from analysis of the SWISS-PROT cells express an endogenous lectin, galectin-1,
database. Biochim Biophys Acta which binds to core 2 O-glycans on thymo-
1473(1):4–8. https://doi.org/10.1016/ cytes and T lymphoblastoid cells. J Exp Med
s0304-4165(99)00165-8 181(3):877–887. https://doi.org/10.1084/
85. Lau KS, Partridge EA, Grigorian A, Silvescu jem.181.3.877
CI, Reinhold VN, Demetriou M, Dennis JW 93. Lee-Sundlov MM, Stowell SR, Hoffmeister
(2007) Complex N-glycan number and KM (2020) Multifaceted role of glycosylation
degree of branching cooperate to regulate in transfusion medicine, platelets, and red
cell proliferation and differentiation. Cell blood cells. J Thromb Haemost
j.cell.2007.01.049 1111/jth.14874
86. Smith LK, Boukhaled GM, Condotta SA, 94. Gendronneau G, Sidhu SS, Delacour D,
Mazouz S, Guthmiller JJ, Vijay R, Butler Dang T, Calonne C, Houzelstein D,
NS, Bruneau J, Shoukry NH, Krawczyk CM, Magnaldo T, Poirier F (2008) Galectin-7 in
Richer MJ (2018) Interleukin-10 directly the control of epidermal homeostasis after
inhibits CD8(+) T cell function by enhancing injury. Mol Biol Cell 19(12):5541–5549.
N-glycan branching to decrease antigen sensi- h t t p s : // d o i . o r g / 1 0 . 1 0 9 1 / m b c . E 0 8 -
tivity. Immunity 48(2):299–312 e295. 02-0166
https://doi.org/10.1016/j.immuni.2018. 95. Panjwani N (2014) Role of galectins in
01.006 re-epithelialization of wounds. Ann Transl
Med 2(9):89. https://doi.org/10.3978/j. stimulates platelet activation, and controls

issn.2305-5839.2014.09.09 primary hemostasis. FASEB J
96. Robinson BS, Saeedi B, Arthur CM, Owens J, 26(7):2788–2798. https://doi.org/10.
Naudin C, Ahmed N, Luo L, Jones R, 1096/fj.11-197541
Neish A, Stowell SR (2020) Galectin-9 is a 104. Pylayeva-Gupta Y, Grabocka E, Bar-Sagi D
novel regulator of epithelial restitution. Am J (2011) RAS oncogenes: weaving a tumori-
Pathol 190(8):1657–1666. https://doi.org/ genic web. Nat Rev Cancer
10.1016/j.ajpath.2020.04.010 11(11):761–774. https://doi.org/10.1038/
97. Thijssen VL, Postel R, Brandwijk RJ, Dings nrc3106
RP, Nesmelova I, Satijn S, Verhofstad N, 105. Paz A, Haklai R, Elad-Sfadia G, Ballan E,
Nakabeppu Y, Baum LG, Bakkers J, Mayo Kloog Y (2001) Galectin-1 binds oncogenic
KH, Poirier F, Griffioen AW (2006) H-Ras to mediate Ras membrane anchorage
Galectin-1 is essential in tumor angiogenesis and cell transformation. Oncogene
and is a target for antiangiogenesis therapy. 20(51):7486–7493. https://doi.org/10.
Proc Natl Acad Sci U S A 1038/sj.onc.1204950
103(43):15975–15980. https://doi.org/10. 106. Elad-Sfadia G, Haklai R, Balan E, Kloog Y
1073/pnas.0603883103 (2004) Galectin-3 augments K-Ras activation
98. D’Haene N, Sauvage S, Maris C, Adanja I, Le and triggers a Ras signal that attenuates ERK
Mercier M, Decaestecker C, Baum L, Salmon but not phosphoinositide 3-kinase activity. J
I (2013) VEGFR1 and VEGFR2 involvement Biol Chem 279(33):34922–34930. https://
in extracellular galectin-1- and galectin-3- doi.org/10.1074/jbc.M312697200
induced angiogenesis. PLoS One 8(6): 107. Shalom-Feuerstein R, Cooks T, Raz A, Kloog
e67029. https://doi.org/10.1371/journal. Y (2005) Galectin-3 regulates a molecular
pone.0067029 switch from N-Ras to K-Ras usage in human
99. Nangia-Makker P, Honjo Y, Sarvis R, breast carcinoma cells. Cancer Res
Akahani S, Hogan V, Pienta KJ, Raz A 65(16):7292–7300. https://doi.org/10.
(2000) Galectin-3 induces endothelial cell 1158/0008-5472.CAN-05-0775
morphogenesis and angiogenesis. Am J 108. Vuong L, Kouverianou E, Rooney CM,
Pathol 156(3):899–909. https://doi.org/ McHugh BJ, Howie SEM, Gregory CD, For-
10.1016/S0002-9440(10)64959-0 bes SJ, Henderson NC, Zetterberg FR, Nils-
100. Freitag N, Tirado-Gonzalez I, Barrientos G, son UJ, Leffler H, Ford P, Pedersen A,
Herse F, Thijssen VL, Weedon-Fekjaer SM, Gravelle L, Tantawi S, Schambye H, Sethi T,
Schulz H, Wallukat G, Klapp BF, Nevers T, MacKinnon AC (2019) An orally active
Sharma S, Staff AC, Dechend R, Blois SM galectin-3 antagonist inhibits lung adenocar-
(2013) Interfering with Gal-1-mediated cinoma growth and augments response to
angiogenesis contributes to the pathogenesis PD-L1 blockade. Cancer Res
of preeclampsia. Proc Natl Acad Sci U S A 79(7):1480–1492. https://doi.org/10.
110(28):11451–11456. https://doi.org/10. 1158/0008-5472.CAN-18-2244
1073/pnas.1303707110 109. Chung LY, Tang SJ, Sun GH, Chou TY, Yeh
101. Romaniuk MA, Tribulatti MV, Cattaneo V, TS, Yu SL, Sun KH (2012) Galectin-1 pro-
Lapponi MJ, Molinas FC, Campetella O, motes lung cancer progression and chemore-
Schattner M (2010) Human platelets express sistance by upregulating p38 MAPK, ERK,
and are activated by galectin-8. Biochem J and cyclooxygenase-2. Clin Cancer Res
BJ20100538 1158/1078-0432.CCR-11-3348
102. Pacienza N, Pozner RG, Bianco GA, D’Atri 110. Banh A, Zhang J, Cao H, Bouley DM,
LP, Croci DO, Negrotto S, Malaver E, Kwok S, Kong C, Giaccia AJ, Koong AC, Le
Gomez RM, Rabinovich GA, Schattner M QT (2011) Tumor galectin-1 mediates tumor
(2008) The immunoregulatory glycan- growth and metastasis through regulation of
binding protein galectin-1 triggers human T-cell apoptosis. Cancer Res
platelet activation. FASEB J 71(13):4423–4431. https://doi.org/10.
22(4):1113–1123. https://doi.org/10. 1158/0008-5472.CAN-10-4157
1096/fj.07-9524com 111. Nambiar DK, Aguilera T, Cao H, Kwok S,
103. Romaniuk MA, Croci DO, Lapponi MJ, Tri- Kong C, Bloomstein J, Wang Z, Rangan VS,
bulatti MV, Negrotto S, Poirier F, Jiang D, von Eyben R, Liang R, Agarwal S,
Campetella O, Rabinovich GA, Schattner M Colevas AD, Korman A, Allen CT,
(2012) Binding of galectin-1 to alphaIIbbeta Uppaluri R, Koong AC, Giaccia A, Le QT
(3) integrin triggers “outside-in” signals, (2019) Galectin-1-driven T cell exclusion in
the tumor endothelium promotes immuno- 8(1):793. https://doi.org/10.1038/

therapy resistance. J Clin Invest s41467-017-00925-6
129(12):5553–5567. https://doi.org/10. 118. Santucci L, Fiorucci S, Cammilleri F,
1172/JCI129025 Servillo G, Federici B, Morelli A (2000)
112. Galon J, Costes A, Sanchez-Cabo F, Galectin-1 exerts immunomodulatory and
Kirilovsky A, Mlecnik B, Lagorce-Pages C, protective effects on concanavalin A-induced
Tosolini M, Camus M, Berger A, Wind P, hepatitis in mice. Hepatology
Zinzindohoue F, Bruneval P, Cugnenc PH, 31(2):399–406. https://doi.org/10.1002/
Trajanoski Z, Fridman WH, Pages F (2006) hep.510310220
Type, density, and location of immune cells 119. Santucci L, Fiorucci S, Rubinstein N,
within human colorectal tumors predict clini- Mencarelli A, Palazzetti B, Federici B, Rabi-
cal outcome. Science 313(5795):1960–1964. novich GA, Morelli A (2003) Galectin-1 sup-
https://doi.org/10.1126/science.1129139 presses experimental colitis in mice.
113. Herbst RS, Soria JC, Kowanetz M, Fine GD, Gastroenterology 124(5):1381–1394.
Hamid O, Gordon MS, Sosman JA, McDer- https://doi.org/10.1016/s0016-5085(03)
mott DF, Powderly JD, Gettinger SN, Kohrt 00267-1
HE, Horn L, Lawrence DP, Rost S, 120. Toscano MA, Commodaro AG, Ilarregui JM,
Leabman M, Xiao Y, Mokatrin A, Bianco GA, Liberman A, Serra HM,
Koeppen H, Hegde PS, Mellman I, Chen Hirabayashi J, Rizzo LV, Rabinovich GA
DS, Hodi FS (2014) Predictive correlates of (2006) Galectin-1 suppresses autoimmune
response to the anti-PD-L1 antibody retinal disease by promoting concomitant
MPDL3280A in cancer patients. Nature Th2- and T regulatory-mediated anti-inflam-
515(7528):563–567. https://doi.org/10. matory responses. J Immunol
1038/nature14011 176(10):6323–6332. https://doi.org/10.
114. Peranzoni E, Lemoine J, Vimeux L, 4049/jimmunol.176.10.6323
Feuillet V, Barrin S, Kantari-Mimoun C, 121. Baum LG, Blackall DP, Arias-Magallano S,
Bercovici N, Guerin M, Biton J, Ouakrim H, Nanigian D, Uh SY, Browne JM,
Regnier F, Lupo A, Alifano M, Damotte D, Hoffmann D, Emmanouilides CE, Territo
Donnadieu E (2018) Macrophages impede MC, Baldwin GC (2003) Amelioration of
CD8 T cells from reaching tumor cells and graft versus host disease by galectin-1. Clin
limit the efficacy of anti-PD-1 treatment. Proc Immunol 109(3):295–307. https://doi.org/
Natl Acad Sci U S A 115(17):E4041–E4050. 10.1016/j.clim.2003.08.003
h t t p s : // d o i . o r g / 1 0 . 1 0 7 3 / p n a s . 122. Perillo NL, Pace KE, Seilhamer JJ, Baum LG
1720948115 (1995) Apoptosis of T cells mediated by
115. Savas P, Virassamy B, Ye C, Salim A, Mintoff galectin-1. Nature 378(6558):736–739.
CP, Caramia F, Salgado R, Byrne DJ, Teo ZL, https://doi.org/10.1038/378736a0
Dushyanthen S, Byrne A, Wein L, Luen SJ, 123. Clemente T, Vieira NJ, Cerliani JP, Adrain C,
Poliness C, Nightingale SS, Skandarajah AS, Luthi A, Dominguez MR, Yon M, Barrence
Gyorki DE, Thornton CM, Beavis PA, Fox FC, Riul TB, Cummings RD, Zorn TM,
SB, Kathleen Cuningham Foundation Con- Amigorena S, Dias-Baruffi M, Rodrigues
sortium for Research into Familial Breast C, MM, Martin SJ, Rabinovich GA, Amarante-
Darcy PK, Speed TP, Mackay LK, Neeson PJ, Mendes GP (2017) Proteomic and functional
Loi S (2018) Single-cell profiling of breast analysis identifies galectin-1 as a novel regu-
cancer T cells reveals a tissue-resident memory latory component of the cytotoxic granule
subset associated with improved prognosis. machinery. Cell Death Dis 8(12):e3176.
Nat Med 24(7):986–993. https://doi.org/ https://doi.org/10.1038/cddis.2017.506
10.1038/s41591-018-0078-7
124. Stillman BN, Hsu DK, Pang M, Brewer CF,
116. Thijssen VL, Heusschen R, Caers J, Griffioen Johnson P, Liu FT, Baum LG (2006)
AW (2015) Galectin expression in cancer Galectin-3 and galectin-1 bind distinct cell
diagnosis and prognosis: a systematic review. surface glycoprotein receptors to induce T
Biochim Biophys Acta 1855(2):235–247. cell death. J Immunol 176(2):778–789.
https://doi.org/10.1016/j.bbcan.2015. https://doi.org/10.4049/jimmunol.176.
03.003 2.778
117. Gordon-Alonso M, Hirsch T, Wildmann C, 125. Thiemann S, Man JH, Chang MH, Lee B,
van der Bruggen P (2017) Galectin-3 cap- Baum LG (2015) Galectin-1 regulates tissue
tures interferon-gamma in the tumor matrix exit of specific dendritic cell populations. J
reducing chemokine gradient production and Biol Chem 290(37):22662–22677. https://
T-cell tumor infiltration. Nat Commun doi.org/10.1074/jbc.M115.644799
126. Ilarregui JM, Croci DO, Bianco GA, Toscano 134. Chen YJ, Wang SF, Weng IC, Hong MH, Lo
MA, Salatino M, Vermeulen ME, Geffner JR, TH, Jan JT, Hsu LC, Chen HY, Liu FT
Rabinovich GA (2009) Tolerogenic signals (2018) Galectin-3 enhances avian H5N1
delivered by dendritic cells to T cells through influenza a virus-induced pulmonary inflam-
a galectin-1-driven immunoregulatory circuit mation by promoting NLRP3 inflammasome
involving interleukin 27 and interleukin 10. activation. Am J Pathol 188(4):1031–1042.
Nat Immunol 10(9):981–991. https://doi. https://doi.org/10.1016/j.ajpath.2017.
org/10.1038/ni.1772 12.014
127. Im SJ, Hashimoto M, Gerner MY, Lee J, Kis- 135. Querol Cano L, Tagit O, Dolen Y, van
sick HT, Burger MC, Shan Q, Hale JS, Lee J, Duffelen A, Dieltjes S, Buschow SI, Niki T,
Nasti TH, Sharpe AH, Freeman GJ, Germain Hirashima M, Joosten B, van den Dries K,
RN, Nakaya HI, Xue HH, Ahmed R (2016) Cambi A, Figdor CG, van Spriel AB (2019)
Defining CD8+ T cells that provide the pro- Intracellular galectin-9 controls dendritic cell
liferative burst after PD-1 therapy. Nature function by maintaining plasma membrane
537(7620):417–421. https://doi.org/10. rigidity. iScience 22:240–255. https://doi.
1038/nature19330 org/10.1016/j.isci.2019.11.019
128. Robertson MW, Albrandt K, Keller D, Liu FT 136. Henderson NC, Mackinnon AC, Farnworth
(1990) Human IgE-binding protein: a solu- SL, Kipari T, Haslett C, Iredale JP, Liu FT,
ble lectin exhibiting a highly conserved inter- Hughes J, Sethi T (2008) Galectin-3 expres-
species sequence and differential recognition sion and secretion links macrophages to the
of IgE glycoforms. Biochemistry promotion of renal fibrosis. Am J Pathol
29(35):8093–8100. https://doi.org/10. 172(2):288–298. https://doi.org/10.2353/
1021/bi00487a015 ajpath.2008.070726
129. Truong MJ, Gruart V, Kusnierz JP, Papin JP, 137. Di Gregoli K, Somerville M, Bianco R,
Loiseau S, Capron A, Capron M (1993) Thomas AC, Frankow A, Newby AC, George
Human neutrophils express immunoglobulin SJ, Jackson CL, Johnson JL (2020) Galectin-
E (IgE)-binding proteins (Mac-2/epsilon 3 identifies a subset of macrophages with a
BP) of the S-type lectin family: role in potential beneficial role in atherosclerosis.
IgE-dependent activation. J Exp Med Arterioscler Thromb Vasc Biol
177(1):243–248 40(6):1491–1509. https://doi.org/10.
130. Stowell SR, Karmakar S, Stowell CJ, Dias- 1161/ATVBAHA.120.314252
Baruffi M, McEver RP, Cummings RD 138. MacKinnon AC, Farnworth SL, Hodkinson
(2007) Human galectin-1, -2, and -4 induce PS, Henderson NC, Atkinson KM,
surface exposure of phosphatidylserine in acti- Leffler H, Nilsson UJ, Haslett C, Forbes SJ,
vated human neutrophils but not in activated Sethi T (2008) Regulation of alternative mac-
T cells. Blood 109(1):219–227. https://doi. rophage activation by galectin-3. J Immunol
org/10.1182/blood-2006-03-007153 180(4):2650–2658. https://doi.org/10.
131. Stowell SR, Qian Y, Karmakar S, Koyama NS, 4049/jimmunol.180.4.2650
Dias-Baruffi M, Leffler H, McEver RP, Cum- 139. Takahashi K, Ezekowitz RA (2005) The role
mings RD (2008) Differential roles of of the mannose-binding lectin in innate
galectin-1 and galectin-3 in regulating leuko- immunity. Clin Infect Dis 41(Suppl 7):
cyte viability and cytokine secretion. J Immu- S440–S444. https://doi.org/10.1086/
nol 180(5):3091–3102 431987
132. Stowell SR, Karmakar S, Arthur CM, Ju T, 140. Garred P, Larsen F, Seyfarth J, Fujita R, Mad-
Rodrigues LC, Riul TB, Dias-Baruffi M, sen HO (2006) Mannose-binding lectin and
Miner J, McEver RP, Cummings RD (2009) its genetic variants. Genes Immun
Galectin-1 induces reversible phosphatidyl- 7(2):85–94. https://doi.org/10.1038/sj.
serine exposure at the plasma membrane. gene.6364283
Mol Biol Cell 20(5):1408–1418. https:// 141. Brown GD (2006) Dectin-1: a signalling
doi.org/10.1091/mbc.E08-07-0786 non-TLR pattern-recognition receptor. Nat
133. Tian J, Yang G, Chen HY, Hsu DK, Rev Immunol 6(1):33–43. https://doi.org/
Tomilov A, Olson KA, Dehnad A, Fish SR, 10.1038/nri1745
Cortopassi G, Zhao B, Liu FT, Gershwin ME, 142. Li FY, Weng IC, Lin CH, Kao MC, Wu MS,
Torok NJ, Jiang JX (2016) Galectin-3 regu- Chen HY, Liu FT (2019) Helicobacter pylori
lates inflammasome activation in cholestatic induces intracellular galectin-8 aggregation
liver injury. FASEB J 30(12):4202–4213. around damaged lysosomes within gastric epi-
https://doi.org/10.1096/fj.201600392RR thelial cells in a host O-glycan-dependent
manner. Glycobiology 29(2):151–162. 152. Park AM, Hagiwara S, Hsu DK, Liu FT,
https://doi.org/10.1093/glycob/cwy095 Yoshie O (2016) Galectin-3 plays an impor-
143. Chauhan S, Kumar S, Jain A, Ponpuak M, tant role in innate immunity to gastric infec-
Mudd MH, Kimura T, Choi SW, Peters R, tion by Helicobacter pylori. Infect Immun
Mandell M, Bruun JA, Johansen T, Deretic 84(4):1184–1193. https://doi.org/10.
V (2016) TRIMs and galectins globally coop- 1128/IAI.01299-15
erate and TRIM16 and galectin-3 co-direct 153. Fowler M, Thomas RJ, Atherton J, Roberts
autophagy in endomembrane damage IS, High NJ (2006) Galectin-3 binds to Heli-
homeostasis. Dev Cell 39(1):13–27. https:// cobacter pylori O-antigen: it is upregulated
doi.org/10.1016/j.devcel.2016.08.003 and rapidly secreted by gastric epithelial cells
144. Paz I, Sachse M, Dupont N, Mounier J, in response to H. pylori adhesion. Cell Micro-
Cederfur C, Enninga J, Leffler H, Poirier F, biol 8(1):44–54. https://doi.org/10.1111/
Prevost MC, Lafont F, Sansonetti P (2010) j.1462-5822.2005.00599.x
Galectin-3, a marker for vacuole lysis by inva- 154. Yang ML, Chen YH, Wang SW, Huang YJ,
sive pathogens. Cell Microbiol Leu CH, Yeh NC, Chu CY, Lin CC, Shieh
12(4):530–544. https://doi.org/10.1111/j. GS, Chen YL, Wang JR, Wang CH, Wu CL,
1462-5822.2009.01415.x Shiau AL (2011) Galectin-1 binds to influ-
145. Vasta GR (2009) Roles of galectins in infec- enza virus and ameliorates influenza virus
tion. Nat Rev Microbiol 7(6):424–438. pathogenesis. J Virol 85(19):10010–10020.
https://doi.org/10.1038/nrmicro2146 https://doi.org/10.1128/JVI.00301-11
146. Stowell SR, Arthur CM, Dias-Baruffi M, 155. Chen Y, Zhou J, Cheng Z, Yang S, Chu H,
Rodrigues LC, Gourdine JP, Heimburg- Fan Y, Li C, Wong BH, Zheng S, Zhu Y, Yu F,
Molinaro J, Ju T, Molinaro RJ, Rivera- Wang Y, Liu X, Gao H, Yu L, Tang L, Cui D,
Marrero C, Xia B, Smith DF, Cummings RD Hao K, Bosse Y, Obeidat M, Brandsma CA,
(2010) Innate immune lectins kill bacteria Song YQ, To KK, Sham PC, Yuen KY, Li L
expressing blood group antigen. Nat Med (2015) Functional variants regulating
16(3):295–301. https://doi.org/10.1038/ LGALS1 (galectin 1) expression affect
nm.2103 human susceptibility to influenza A(H7N9).
147. Stowell SR, Arthur CM, McBride R, Sci Rep 5:8517. https://doi.org/10.1038/
Berger O, Razi N, Heimburg-Molinaro J, srep08517
Rodrigues LC, Gourdine JP, Noll AJ, von 156. Levroney EL, Aguilar HC, Fulcher JA,
Gunten S, Smith DF, Knirel YA, Paulson JC, Kohatsu L, Pace KE, Pang M, Gurney KB,
Cummings RD (2014) Microbial glycan Baum LG, Lee B (2005) Novel innate
microarrays define key features of host- immune functions for galectin-1: galectin-1
microbial interactions. Nat Chem Biol inhibits cell fusion by Nipah virus envelope
10(6):470–476. https://doi.org/10.1038/ glycoproteins and augments dendritic cell
nchembio.1525 secretion of proinflammatory cytokines. J
148. Springer GF, Williamson P, Brandes WC Immunol 175(1):413–420. https://doi.org/
(1961) Blood group activity of gram-negative 10.4049/jimmunol.175.1.413
bacteria. J Exp Med 113(6):1077–1093 157. Garner OB, Aguilar HC, Fulcher JA, Levro-
149. Stowell CP, Stowell SR (2019) Biologic roles ney EL, Harrison R, Wright L, Robinson LR,
of the ABH and Lewis histo-blood group Aspericueta V, Panico M, Haslam SM, Morris
antigens part I: infection and immunity. Vox HR, Dell A, Lee B, Baum LG (2010) Endo-
Sang 114(5):426–442. https://doi.org/10. thelial galectin-1 binds to specific glycans on
1111/vox.12787 nipah virus fusion protein and inhibits matu-
ration, mobility, and function to block syncy-
150. Stowell SR, Stowell CP (2019) Biologic roles tia formation. PLoS Pathog 6(7):e1000993.
of the ABH and Lewis histo-blood group https://doi.org/10.1371/journal.ppat.
antigens part II: thrombosis, cardiovascular 1000993
disease and metabolism. Vox Sang
114(6):535–552. https://doi.org/10.1111/ 158. Ouellet M, Mercier S, Pelletier I, Bounou S,
vox.12786 Roy J, Hirabayashi J, Sato S, Tremblay MJ
(2005) Galectin-1 acts as a soluble host factor
151. Quattroni P, Li Y, Lucchesi D, Lucas S, Hood that promotes HIV-1 infectivity through sta-
DW, Herrmann M, Gabius HJ, Tang CM, bilization of virus attachment to host cells. J
Exley RM (2012) Galectin-3 binds Neisseria Immunol 174(7):4120–4126. https://doi.
meningitidis and increases interaction with org/10.4049/jimmunol.174.7.4120
phagocytic cells. Cell Microbiol
14(11):1657–1675. https://doi.org/10. 159. Wang SF, Tsao CH, Lin YT, Hsu DK, Chiang
1111/j.1462-5822.2012.01838.x ML, Lo CH, Chien FC, Chen P, Arthur Chen
YM, Chen HY, Liu FT (2014) Galectin-3 EM, Schneider C, Reimer R, Grunewald K,
promotes HIV-1 budding via association Wiethoff CM, Wodrich H (2017) Multi-
with Alix and Gag p6. Glycobiology layered control of Galectin-8 mediated autop-
24(11):1022–1035. https://doi.org/10. hagy during adenovirus cell entry through a
1093/glycob/cwu064 conserved PPxY motif in the viral capsid.
160. Montespan C, Marvin SA, Austin S, Burrage PLoS Pathog 13(2):e1006217. https://doi.
AM, Roger B, Rayne F, Faure M, Campell org/10.1371/journal.ppat.1006217
Chapter 2
Cloning, Expression, and Purification of Galectins

for In Vitro Studies
Paul A. Poland, Carol L. Kinlough, and Rebecca P. Hughey
Abstract
Galectins are best known for their ability to bind glycoconjugates containing β-galactose, but classification
of these small proteins within the galectin family is also defined by amino acid homology within structural
domains and exon/intron junctions within genes. As galectins are expressed by organisms as diverse as
some fungi, C. elegans, fish, birds and mammals, and biological activities attributed to galectins are equally
diverse, it becomes essential to identify, clone, and characterize galectins from many sources. Glutathione S-
transferase (GST) fused to the amino-terminus of galectin cDNAs has proven to be especially useful for the
preparation of recombinant galectins in bacteria for use on glycan arrays, in experiments with cultured or
isolated cells, and in pull-down assays with immunopurified glycoproteins. Many galectins are stabilized by
reducing reagents, such that binding and elution of GST-galectins from glutathione-conjugated Sepharose
with excess glutathione is both efficient and innocuous. The ability to bind and elute GST-galectins from
lactose-conjugated Sepharose with excess lactose provides a relatively easy means to insure that galectins are
competent for glycoconjugate binding prior to experimentation. This chapter focuses primarily on the
varied approaches to use GST-galectin binding to glutathione- and lactose-conjugated Sepharose to purify
recombinant galectins and then develop effective experimental protocols to characterize the specificity,
interactions and function of galectins cloned from any source. We provide one example where a pull-down
assay with all the GST-tagged canine galectins reveals that the C-terminal carbohydrate recognition domain
of galectin-9 (Gal-9C) specifically recognizes the glycan-dependent apical targeting signal from the
glycoprotein MUC1.
Key words Galectin, GST-galectin, Recombinant galectin, Pull-down assay
1 Introduction
Our interest in the role of galectins (Gal) in glycan-dependent

apical targeting of transmembrane MUC1, in the well-
characterized model system of polarized Madin Darby canine kid-
ney (MDCK) epithelial cells, led us down the unforeseen path of
cloning and characterizing all of the canine galectins [1, 2].
At the start of our odyssey, Gal-3 and Gal-4 were implicated in
glycan-dependent and lipid raft-dependent sorting of apical pro-
teins, respectively, in polarized epithelial cells [3–7]. Gal-3 was
41
42 Paul A. Poland et al.
purported to crosslink glycoproteins while Gal-4 was thought to

crosslink glycolipids, thereby producing sorting platforms for clus-
tering of proteins. As our preliminary data were most consistent
with glycan-dependent targeting of MUC1 [2], we focused on
knockdown of Gal-3 with siRNAs in MDCK cells but found no
effect on MUC1 apical targeting [8]. In order to consider a role for
other galectins in apical targeting in MDCK cells, we had to deter-
mine both which galectins were actually expressed by dogs and
which were present in MDCK cells [9].
Using nucleotide BLAST and TBLASTIN programs with both
the dog genome and the NCBI Nucleotide Database, we identified
sequences that aligned with nucleotide sequences and exon profiles,
or amino acid sequences, for mammalian Gal-1, -2, -3, -4, -7, -8,
-9, and -12, as well as the galectin-related proteins GRIFIN and
HSPC159. GRIFIN and HSPC159 lack essential residues in the
carbohydrate recognition domains for sugar binding. We found no
canine genes homologous to rat Gal-5, mouse Gal-6, human
Gal-10, human Gal-13, ovine Gal-14, ovine Gal-15, human
PP13, or human PPL13.
We estimated the levels of transcripts for the canine galectins in
MDCK cells using RT-PCR and primers designed to yield ~200 bp
sequences. RNA from canine jejunum was used as a positive con-
trol. Amplification of RNA for Gal-1, -3, -8 and -9 was readily
achieved by this approach while amplification of RNA for Gal-4
and -7 required nested primers to produce a band on an agarose gel
with ethidium bromide staining. Based on these data, we concluded
that transcript levels for Gal-3 > Gal-9 > Gal-8 > Gal-1 > Gal-
4 > Gal-7. While we did not obtain evidence for Gal-2 or Gal-12
transcripts in MDCK cells, a subsequent study by Friedrichs et al.
[10] using real-time PCR reported that transcripts for Gal-3 were
100 times more abundant than transcripts for Gal-1, -8 and -9, and
1000 times more abundant than for Gal-12, essentially consistent
with our findings [9].
We first attempted to clone the canine galectins by RT-PCR
from RNA isolated from MDCK cells using primers designed to
match nucleotide sequences overlapping the predicted translational
start and stop codons. Full-length cDNAs for canine Gal-1, -3 and
-8 were obtained by this approach, while cloning of Gal-4 and -7
required amplification with nested primers where external primers
matched nucleotide sequences within the 30 or 50 UTR, and inter-
nal primes overlapped the start and stop codons. The cDNAs for
Gal-2 and -9 required amplification from a commercial source of
canine jejunum RNA, and Gal-12 cDNA was eventually obtained
by amplification from canine heart RNA. The cDNAs were cloned
and sequenced for alignment with the canine genome and determi-
nation of the exon/intron structure of the genes for comparison to
other mammalian galectin genes. We found 73–86% identity
between canine and human amino acid sequences of the galectin
Cloning, Expression, and Purification of Galectins for In Vitro Studies 43
with missing or extra exons representing previously described short

(Gal-8 and Gal-12) and long (Gal-9) isoforms, respectively. We
found five residues within exon 1 of Gal-7 as described for cow,
but unique from that of mouse (two) and human (six), and six
9-mer amino acid repeats in Gal-3 as compared to cow (four),
mouse (three), pig (three), human (two), rat (two), and rabbit
(one). Although these differences are small, they could affect galec-
tin biological activities, and we did proceed to isolate the
GST-tagged recombinant canine galectins for characterization on
glycan arrays of Core H of the NIH Consortium for Functional
Glycomics. This involved expression of GST-galectins in bacteria
and the realization that GST-Gal-9 and -12 formed aggregates in
bacteria and were not useful constructs. We subsequently sub-
cloned the two different carbohydrate recognition domains from
Gal-9 as it was prevalent in MDCK, while Gal-12 was not, produc-
ing GST-Gal-9 N and GST-Gal-9C. GST-Gal-3, -4, -7, -8, -9N and
-9C were purified from bacterial extracts by affinity binding to
glutathione-conjugated and then lactose-conjugated Sepharose.
Binding of GST-galectins on glycan arrays was then followed with
an Alexa 488-conjugated anti-GST antibody [9].
Finally, to assess which endogenous galectins are likely inter-
acting with MUC1 expressed in polarized MDCK cells, we carried
out preliminary experiments where human MUC1 immunopuri-
fied from stably transfected MDCK cell extracts was incubated with
recombinant GST-tagged Gal-1, -3, -4, -7, -8, -9N, and -9C in
pull-down assays. Interestingly, we found that MUC1, whether
metabolically labeled with [35S]Met/Cys or followed by immuno-
blotting, interacted best with Gal-3 and Gal-9, the two most abun-
dant galectins expressed in MDCK cells [9]. In subsequent
experiments, we found that core-glycosylated mucin-like tandem
repeats (0TR) from MUC1 are an apical targeting signal in polar-
ized MDCK cells when appended to a non-polarized protein (Tac)
[8]. We now present data from galectin pull-down assays compar-
ing interactions with Tac and 0TR-Tac that clearly shows that
Gal-9C specifically binds to the apical targeting motif of MUC1.
2 Materials
2.1 Cloning of 1. RNA from specified organisms and/or tissues (Zyagen) (see
Galectins Note 1).
2. RNA Superscript II reverse transcriptase and Taq DNA poly-
merase (Invitrogen).
3. PCR primers (Integrated DNA Technologies).
4. pCR2.1 TOPO vector for DNA cloning (Invitrogen).
5. pGEX-6P-1 for the production of glutathione S-transferase

(GST) fusion galectins (GE Health Care Life Sciences) (see
Note 2).
6. Bacteria XL-1 Blue (Agilent Technologies).
2.2 Expression and 1. E. coli strain BL21 (EMD Millipore).

Purification of GST- 2. Isopropyl β-D-1-thiogalactopyranoside (IPTG) (Sigma).
Galectins
3. Luria Bertani (LB) broth.
4. Ampicillin (Sigma).
5. Lysis-Sonication buffer: 300 mM NaCl, 5 mM MgCl2, 1 mM
dithiothreitol, 0.2% Triton X-100 and 50 mM Tris–HCl,
0.1 mM phenylmethanesulfonylfluoride (PMSF), pH 8.0.
6. Glutathione-conjugated to Sepharose (GE Healthcare).
7. PD-10 gel filtration columns (GE Healthcare).
8. Sepharose CL-6B (Sigma).
9. Lactose-conjugated Sepharose (Sigma) (see Note 3).
10. Elution buffer (with glutathione): 100 mM Tris–HCl, pH 8.0,
15 mM glutathione, 1 mM dithiothreitol (see Note 4).
11. Wash buffer: 0.1 M Tris–HCl, pH 8, 150 mM NaCl, 14 mM
beta-mercaptoethanol (βME).
12. Elution buffer (with lactose): 0.1 M Tris–HCl, pH 8, 150 mM
NaCl, 14 mM beta-mercaptoethanol (βME), 0.1 mM α-lactose
(Sigma) (see Note 4).
13. 1-cm diameter column (8 cm tall) containing a sintered glass
filter.
14. 1.5-mL conical plastic tubes.
2.3 Preparation of 1. Lactose-conjugated Sepharose (Sigma).

GST-Galectins for Pull- 2. Glutathione-conjugated to Sepharose (GE Healthcare).
Down Assays
3. Sepharose CL-6B (Sigma).
4. Hamilton syringe, 100 μL size with a 22S-gauge needle
(Gastight).
5. PreScission protease (GE Healthcare).
6. 15-mL conical tubes.
2.4 Preparation of 1. Glutathione-conjugated to Sepharose (GE Healthcare).

Immunopurified 2. Protein G or Protein A conjugated to Sepharose (Invitrogen).
Proteins
3. [35S]Met/Cys.
4. Phosphate-buffered saline, 14 mM beta-mercaptoethanol
(βME), 1% Triton X-100.
5. Phosphate-buffered saline, 0.1% SDS.
6. Phosphate-buffered saline, 14 mM beta-mercaptoethanol,

15 mM glutathione.
7. Hamilton syringe, 100 μL size with a 22S-gauge needle
(Gastight).
8. Criterion Precast Gels (4–15% Tris–HCl, 1 mm).
9. BioRad Mini Protean Gel system.
10. SDS-PAGE sample buffer (BioRad).
11. Nitrocellulose (EMD Millipore, 0.45 μm).
12. Bio-safe Coomassie (BioRad).
13. 1.5-mL tubes with snap cap.
2.5 Special 1. Thermocycler.

Equipment 2. BioRad Phosphorimager.
3. Quantity One Software.
3 Methods
3.1 Cloning 1. First-strand cDNA is synthesized from 1 μg RNA using RNA

Superscript II reverse transcriptase and amplified using Taq
DNA polymerase as described by the manufacturer.
2. Primers for PCR to amplify the full-length transcripts are
designed to overlap the predicted start and stop codons using
nucleotide sequence information from database queries (see
Note 5).
3. Amplify DNA using standard PCR technique.
4. Amplified DNA is cloned into pCR2.1 TOPO vector as
described by the supplier.
5. Transform bacteria XL-1 Blue by the heat shock protocol.
6. Culture bacteria and isolate colonies.
7. Plasmid preps from multiple colonies should be generated.
8. Sequence isolated cDNAs from individual colonies in order to
verify the predicted amino acid sequence of new galectins.
3.2 Expression and 1. Galectin cDNAs are subcloned into pGEX-6P-1 for expression
Purification of GST- in bacteria as N-terminal GST-fusion proteins based on a pre-
Galectins vious protocol [11]. Details are found below.
2. LB medium (20 mL) containing ampicillin is inoculated with
200 μL of glycerol stock of E. coli strain BL21 transformed with
pGEX-6P-1 encoding GST-galectin and grown overnight with
shaking at 37 C.
3. This overnight 20-mL starter culture is used to inoculate 2 L of
culture medium with ampicillin (divided between two 2-L
Fig. 1 Affinity purification of GST-galectins on Lac-Sepharose. After batch elution of recombinant

GST-galectins from glutathione conjugated to Sepharose, the GST-galectins can be stored at 80 C. Before
experimental use, GST-galectins are affinity purified by incubation with lactose conjugated to Sepharose
(Lac-Sepharose) and eluted after incubation in a column equilibrated with α-lactose (a, c). Lactose is removed
from the GST-galectins using a PD-10 gel filtration column (b, d). See text for protocol details. Example
profiles for GST-Gal-9N (a, b) and GST-Gal-9C (c, d) are shown here using calculated extinction coefficients of
55,010 and 56,620 M1, respectively (A280 on left axis and concentration on right axis)
flasks) and grown with shaking for 2–3 h at 37 C until it

reaches an OD600 of 0.8.
4. Isopropyl β-D-1-thiogalactopyranoside is then added to
0.1 mM to induce expression of the fusion protein, and bacte-
ria are grown with shaking overnight at room temperature.
5. Bacteria are collected by centrifugation at 25,000 g for
12 min at 4 C in 250-mL centrifuge bottles. Pellets of bacteria
can be frozen at 20 C at this step for later use.
6. Pellets of bacteria are subsequently resuspended in 10 mL of
Lysis-Sonication buffer (with PMSF) using a glass rod and
transfer pipettes.
7. Bacteria are further lysed by sonication on ice with a Fisher
Sonic Dismembrator Model 100 (5 times, at setting
5 (8–12 W) for 30 s with 30-s intervals).
8. The lysate is centrifuged for 30 min at 10,000 g at 4 C to
pellet cell debris.
9. The supernatant from the lysed bacteria are incubated at 4 C
overnight with 0.75 mL of a 50% slurry of glutathione-
conjugated Sepharose in a 15-mL conical tube with end-over-
end mixing.
10. Beads are subsequently pelleted after centrifugation for 2 min
at 500 g (Eppendorf model 5702 swinging bucket centri-
fuge) and the supernatant removed with a transfer pipette.
11. Beads are washed three times by the addition of 5 mL of Lysis-
Sonication buffer, mixing and centrifugation, as already
described.
12. GST-galectins are released from the washed glutathione-
conjugated Sepharose beads by adding 5 mL of Elution buffer,
mixing, and incubation at 4 C for 15 min.
13. After centrifugation for 2 min at 500 g, supernatant (eluate)
is removed with a transfer pipette and saved (see Note 6).
Alternative protocol for purifying galectin from glutathione
eluate:
14. The entire eluate from the glutathione-conjugated Sepharose
beads is incubated with 1 mL of a 50% slurry of lactose-
conjugated Sepharose in a 15-mL conical tube at 4 C for
30 min with end-over-end mixing.
15. The slurry is then poured into a 1-cm diameter column con-
taining a sintered glass filter and the packed column bed is
overlayed with 1 mL of a 50% slurry of Sepharose 6B. The
column bed is washed with 5 mL buffered saline containing
βME.
16. To elute the GST-galectins from the column, 2 mL of buffered

saline containing both βME and 0.1 M α-lactose is added to the
column (Elution buffer).
17. 0.5 mL is collected from the bottom, and the column closed
and capped for 30 min at 4 C.
18. Fractions of 0.25 mL are subsequently collected after adding
buffer to the top of the column.
19. The absorbance at 280 nm is determined with a spectropho-
tometer for each fraction.
20. The peak fractions are pooled and lactose is removed on a
PD-10 gel filtration column as described by the manufacturer
(see Fig. 1 for representative elution profiles) (see Notes 7
and 8).
3.3 Preparation of 1. Aliquots of thawed GST-galectins (see Note 7) for pull-down

GST-Galectins for Pull- assays are first affinity purified on lactose-conjugated Sepharose
Down Assays before binding to glutathione beads and incubation with
immunopurified proteins (see Subheading 3.4). Buffered saline
containing βME is used throughout the protocol (see Note 4).
2. An aliquot of GST-galectin (1 μg) is added to 30 μL of a mixed
slurry of lactose-conjugated Sepharose and carrier Sepharose
CL-6B in a conical 1.5-mL tube with snap cap and incubated
for 30 min at 4 C with end-over-end mixing (see Note 9).
3. After centrifugation at 500 g in a table-top microcentrifuge
for 2 min, the supernatant is removed and discarded, and the
pellet is washed 2 times with 150 μL buffered saline to remove
any unbound GST-galectin that is potentially “inactive.” Liq-
uid is always removed from the beads with a Hamilton syringe
(100 μL size with a 22S-gauge needle) (see Note 10).
4. GST-galectin is eluted from the lactose-conjugated Sepharose
by incubation with 30 μL of buffered saline containing
100 mM α-lactose for 15 min at 4 C with end-over-end
mixing.
for 2 min, the supernatant containing the “active” GST-galec-
tin is transferred to a mixed slurry of glutathione-conjugated
Sepharose and carrier Sepharose CL-6B in a 1.5-mL conical
tube with snap cap, and incubated for 15 min at 4 C with end-
over-end mixing (see Note 11).
for 2 min, the beads are washed twice with 200 μL of buffered
saline and all liquid removed from the pellet with a Hamilton
syringe.
7. The final pellet of GST-galectin bound to glutathione-
conjugated beads is subsequently mixed with immunopurified
protein already prepared from cell extracts (see Subheading

3.4). Alternatively, the beads can be resuspended into 100 μL
of buffered saline and 30–45 μL can be transferred to different
tubes of immunopurified proteins to produce duplicate or
triplicate samples.
8. The GST tag on GST-galectin prepared in the pGEX-6P-1
vector can also be removed by treatment of GST-galectin
pre-bound to glutathione-conjugated beads using PreScission
protease to cut at the single engineered consensus sequence
LEVLFQ/GP between the GST and galectin as directed by the
manufacturer (where / is the cleavage site and letters are
abbreviations for amino acids).
3.4 Pull-Down 1. Pull-down assays can be carried out either by following protein
Assays with GST- binding by immunoblotting or by following binding of [35S]
Galectins proteins recovered by metabolic labeling of cell cultures
(described here). The exact protocol for the preparation of
each immunopurified protein should be optimized by the
researcher.
2. Multiple wells (or permeable supports) of cultured cells expres-
sing the protein of interest are metabolically labeled with [35S]
Met/Cys for 15–60 min and chased in medium containing Met
and Cys for 60–120 min (see Note 12). Cells are extracted from
each well in a mild detergent such as octyl-glucoside or Triton
X-100, centrifuged to remove cell debris and nuclei, and the
supernatants combined.
3. [35S]Proteins are recovered by immunoprecipitation from the
combined supernatants in 1.5-mL conical plastic tube with
snap cap using specific antibodies and either Protein G or
Protein A conjugated to Sepharose beads. Proteins are eluted
from the beads by heating at 90 C for 2 min in 200 μL
buffered saline containing 0.1% SDS (heating is optional).
The samples are centrifuged at 9300 g in a table-top micro-
centrifuge twice for 30 s to pellet the beads.
4. Equal aliquots of eluant (20 μL) are transferred to clean snap
cap tubes. One of the aliquots is retained as the input sample
and mixed with 10 μL SDS gel sample buffer. The remainder of
the aliquots are diluted tenfold with buffered saline containing
βME and 1% Triton X-100 to neutralize the 0.1% SDS prior to
addition of freshly prepared GST-galectins pre-bound to
glutathione-conjugated Sepharose beads. One tube receives
Sepharose CL-6B beads as a negative control (i.e., there should
be no binding to the beads alone). See Fig. 2 for a representa-
tive experiment with eight different GST-galectins, beads alone
and input sample (total).
Fig. 2 Pull-down experiments with GST-galectins reveals specific binding of GST-Gal-9C to the apical
targeting signal from MUC1. Core-glycosylated mucin-like tandem repeats (0TR) from MUC1 are an apical
targeting signal in polarized MDCK cells when appended to a non-polarized protein (Tac) [8]. To identify
specific galectins that bind to 0TR, we compared Tac and 0TR-Tac binding to GST-galectins in pull-down
assays. (a) Equal aliquots of immunopurified [35S]Tac or [35S]0TR-Tac were incubated overnight with fresh
GST-galectins bound to GSH-beads (or beads alone) and eluted with sucrose (S), then lactose (L), and
analyzed after SDS-PAGE with a BioRad imager. The percent [35S]Tac or [35S]0TR-Tac eluted with lactose was
calculated from the input total (TOT) aliquot and is presented as the mean and SEM for three experiments.
[35S]0TR-Tac binding to only GST-Gal-9C was consistently greater than [35S]Tac binding ( p < 0.05). Note that
mature (M), but not precursor (P) forms were bound. One representative experiment is presented in (b)
showing a side-by-side comparison of [35S]Tac or [35S]0TR-Tac binding with MW markers to the right of the
gels. The GST-galectins from each sample were eluted with glutathione for SDS-PAGE and analyzed by
scanning the Coomassie-stained SDS gel. Mobility of MW markers in kDa are noted between the two gels (c).
See text for protocol details
5. [35S]Proteins are incubated with the GST-galectins pre-bound

to glutathione beads overnight at 4 C with end-over-end
mixing. After centrifugation at 500 g in a table-top
microcentrifuge for 2 min, the supernatant is removed and the

pellet washed 2 times with 200 μL buffered saline containing
βME to remove any unbound [35S]proteins.
6. Beads are incubated for 30 min at 4 C with 10 μL buffered
saline containing βME and 100 mM sucrose to elute proteins
bound non-specifically. After centrifugation at 500 g in a
table-top microcentrifuge for 2 min, the eluant is removed with
a Hamilton syringe and saved.
saline containing βME and 100 mM lactose to elute bound
proteins. After centrifugation at 500 g in a table-top micro-
centrifuge for 2 min, the eluant is removed with a Hamilton
syringe and saved.
saline containing βME and 15 mM glutathione to elute
GST-galectins. After centrifugation at 500 g in a table-top
microcentrifuge for 2 min, the eluant is removed with a Hamil-
ton syringe and saved.
9. Eluants from sucrose, lactose, and glutathione incubations are
subjected to SDS-PAGE by mixing 10 μL of sample with 10 μL
SDS sample buffer and heating at 90 C for 2 min.
10. After electrophoresis, the proteins eluted by sucrose and lactose
can be electrophoretically transferred to nitrocellulose. Nitro-
cellulose is dried for 1 h at room temperature, stored
0.5–3 days with a Kodak TR screen and bands analyzed with
a BioRad Phosphorimager and Quantity One Software. The
SDS-gel containing proteins eluted with glutathione
(GST-galectins) is stained with Bio-safe Coomassie. The
stained gel is scanned and bands analyzed with Quantity One
Software. See Fig. 2 for a representative experiment with
immunopurified [35S]proteins and pull-down with eight dif-
ferent GST-galectins.
11. The dry nitrocellulose can be subsequently hydrated and used
for immunoblotting after blocking using standard protocols.
Bands are visualized with a BioRad Versadoc or by scanning
bands on exposed film, and quantified using Quantity One
Software.
12. The fraction of immunopurified protein bound and specifically
eluted from each GST-galectin is calculated using the input
aliquot as 100%. These values are normalized with the amount
of each GST-galectin bound to the glutathione-conjugated
beads (based on the Coomassie stained gel and divided by the
individual formula weight of the GST-galectin).
4 Notes
1. Alternatively, RNA can be isolated from tissues or cell lines

using commercial kits (Ambion RNAqueous 4PCR Kit) fol-
lowing the directions of the manufacturer.
2. It may be necessary to express carbohydrate binding domains
from tandem repeat galectins separately to assess binding spe-
cificities individually. On the other hand, we found that canine
GST-Gal-9 repeatedly aggregated in the bacteria, so we sub-
cloned the N-terminal and C-terminal CRDs separately as
GST-Gal-9N (residues 1–148) and GST-Gal-9C (residues
225–355), respectively, for further study.
3. Sepharose CL-6B is used because it is cross-linked and will not
be altered by heating in 1% SDS.
4. Beta-mercaptoethanol or dithiothreitol is included in buffers to
stabilize the galectins as many have reactive Cys residues.
5. When either the start or stop codon cannot be located in the
database, the full-length cDNA should be obtained using stan-
dard approaches such as RACE-PCR [12]. Amplification of
rare transcripts may also require the use of nested primers,
whereby an aliquot (1%) of the amplified DNA obtained by
PCR with external primers after 28 cycles is amplified a second
time for 28 cycles with internal primers.
6. At this stage, the GST-galectin can be divided into aliquots and
frozen at 80 C after addition of βME to 14 mM. Before any
further use, GST-galectins should be affinity purified on
lactose-conjugated Sepharose to be sure that the galectin is
active (i.e., competent for binding glycans). Protein concentra-
tion of GST-galectin is estimated from Coomassie staining after
SDS-PAGE as glutathione interferes with spectrophotometer
readings at A280.
7. GST-galectins can be batch eluted from the lactose-conjugated
Sepharose. Elution from the column allows retrieval of the
most concentrated fractions as needed for protocols such as
screening glycan arrays. Sufficient GST-galectin for pull-down
assays can be obtained from a 50-mL culture of bacteria (~5 μg
is eluted from the glutathione-conjugated Sepharose).
8. Protein concentrations of GST-galectins are calculated from
A280 using individual extinction coefficients determined from
the amino acid content of each GST-Gal using the Peptide
Property Calculator found online (http://www.basic.north
western.edu/biotools/proteincalc.html). Be sure to include
the amino acid content of GST in the calculation. Pooled
peak samples (1–1.5 mL) routinely contain ~0.25–1 mg GST-
galectin. This active concentrated GST-galectin is optimal for

use on glycan arrays.
9. For slurry mixture, a 50% slurry of lactose-conjugated Sephar-
ose (25 μL) is mixed with a 50% slurry of carrier Sepharose
CL-6B (225 μL), and 30 μL of the mixed slurry is transferred
to a 1.5-mL conical tube with snap cap for subsequent addition
of 1 μg of GST-galectin.
10. This size of Hamilton syringe allows removal of all the liquid
while excluding the Sepharose beads when the tip is placed at
the bottom of the tube in the middle of the pellet.
11. A 50% slurry of glutathione conjugated Sepharose (90 μL) is
mixed with a 50% slurry of Sepharose CL-6B (270 μL). A
40 μL aliquot of the mixed slurry is used for each aliquot of
GST-galectin eluted from the mixed slurry of lactose-
conjugated Sepharose.
12. Our investigation of MUC1 binding to canine GST-galectins
were carried out by metabolic labeling of polarized cultures of
MDCK epithelial cells with [35S]Met/Cys for 30 min and a
chase in normal culture medium for 90 min, a time profile
previously determined for efficient surface expression of
MUC1. [35S]MUC1 eluted from GST-galectins (or the input
control) was subjected to SDS-PAGE and transferred to nitro-
cellulose such that we were able to analyze both newly synthe-
sized radiolabeled MUC1 and then steady state MUC1 by
immunoblotting. Data can also be obtained from analysis of a
dried SDS-gel.
Acknowledgments
This work was funded by grants to RPH by National Institutes of

Health (DK054787), Genzyme Renal Innovations Program, NIH
DK051391 and DK038470 and Dialysis Clinics, Inc.
References
1. Potter BA, Hughey RP, Weisz OA (2006) Role 3. Delacour D, Gouyer V, Zanetta JP,
of N- and O-glycans in polarized biosynthetic Drobecq H, Leteurtre E, Grard G, Moreau-
sorting. Am J Physiol Cell Physiol 290(1): Hannedouche O, Maes E, Pons A, Andre S,
C1–C10. https://doi.org/10.1152/ajpcell. Le Bivic A, Gabius HJ, Manninen A, Simons K,
00333.2005 Huet G (2005) Galectin-4 and sulfatides in
2. Mattila PE, Kinlough CL, Bruns JR, Weisz apical membrane trafficking in enterocyte-like
OA, Hughey RP (2009) MUC1 traverses api- cells. J Cell Biol 169(3):491–501. https://doi.
cal recycling endosomes along the biosynthetic org/10.1083/jcb.200407073
pathway in polarized MDCK cells. Biol Chem 4. Stechly L, Morelle W, Dessein AF, Andre S,
390(7):551–556. https://doi.org/10.1515/ Grard G, Trinel D, Dejonghe MJ,
BC.2009.088 Leteurtre E, Drobecq H, Trugnan G, Gabius
HJ, Huet G (2009) Galectin-4-regulated deliv- Chem 286(45):39072–39081. https://doi.

ery of glycoproteins to the brush border mem- org/10.1074/jbc.M111.289504
brane of enterocyte-like cells. Traffic 9. Poland PA, Rondanino C, Kinlough CL,
10(4):438–450. https://doi.org/10.1111/j. Heimburg-Molinaro J, Arthur CM, Stowell
1600-0854.2009.00882.x SR, Smith DF, Hughey RP (2011) Identifica-
5. Delacour D, Cramm-Behrens CI, Drobecq H, tion and characterization of endogenous galec-
Le Bivic A, Naim HY, Jacob R (2006) Require- tins expressed in Madin Darby canine kidney
ment for galectin-3 in apical protein sorting. cells. J Biol Chem 286(8):6780–6790.
Curr Biol 16(4):408–414. https://doi.org/ https://doi.org/10.1074/jbc.M110.179002
10.1016/j.cub.2005.12.046 10. Friedrichs J, Torkko JM, Helenius J, Teravai-
6. Delacour D, Greb C, Koch A, Salomonsson E, nen TP, Fullekrug J, Muller DJ, Simons K,
Leffler H, Le Bivic A, Jacob R (2007) Apical Manninen A (2007) Contributions of
sorting by galectin-3-dependent glycoprotein galectin-3 and -9 to epithelial cell adhesion
clustering. Traffic 8(4):379–388. https://doi. analyzed by single cell force spectroscopy. J
org/10.1111/j.1600-0854.2007.00539.x Biol Chem 282(40):29375–29383. https://
7. Delacour D, Koch A, Ackermann W, Eude-Le doi.org/10.1074/jbc.M701867200
Parco I, Elsasser HP, Poirier F, Jacob R (2008) 11. Self AJ, Hall A (1995) Purification of recombi-
Loss of galectin-3 impairs membrane polarisa- nant Rho/Rac/G25K from Escherichia coli.
tion of mouse enterocytes in vivo. J Cell Sci Methods Enzymol 256:3–10
121(Pt 4):458–465. https://doi.org/10. 12. Frohman MA, Dush MK, Martin GR (1988)
1242/jcs.020800 Rapid production of full-length cDNAs from
8. Kinlough CL, Poland PA, Gendler SJ, Mattila rare transcripts: amplification using a single
PE, Mo D, Weisz OA, Hughey RP (2011) gene-specific oligonucleotide primer. Proc
Core-glycosylated mucin-like repeats from Natl Acad Sci U S A 85(23):8998–9002
MUC1 are an apical targeting signal. J Biol
Chapter 3
Purification of Recombinant Galectins from Different

Species Using Distinct Affinity Chromatography Methods
Anu Paul, Shang-Chuen Wu, Kashyap R. Patel, Alex D. Ho,
Jerry William Lynn Allen, Hans Verkerke, Connie M. Arthur,
and Sean R. Stowell
Abstract
Galectins are lectins having the capacity to recognize β-galactose-containing glycan structures and are
widely distributed among various taxa. However, the exact physiological and biochemical functions
mediated by galectins that necessitate their wide occurrence among diverse species have not yet been
delineated in a precise manner. Purification of recombinant galectins in active form is a fundamental
requirement to elucidate their biological function. In this chapter, we are describing methods to recombi-
nantly express and purify galectins using three different methods of affinity purification, i.e., lactosyl-
Sepharose chromatography for fungal galectin Coprinopsis cinerea galectin 2 (CGL2), nickel-
chromatography for histidine-tagged human galectin-7, and glutathione-Sepharose chromatography for
Glutathione S-transferase-tagged (GST-tagged) human galectin-7. Step-by-step instructions are provided
for obtaining the above-mentioned recombinant galectins that retain carbohydrate-binding activity and are
suitable for conducting biochemical experiments.
Key words Recombinant galectin purification, Coprinopsis cinerea galectin-2 (CGL2), Divinyl
sulfone-coupled lactosyl-Sepharose, Lactosyl-Sepharose affinity chromatography, Nickel
NTA-affinity purification of histidine-tagged human galectin-7, Glutathione-Sepharose affinity purifi-
cation of GST-tagged human galectin-7
1 Introduction
Galectins belong to a family of lectins that recognize β-galactosyl

structures by means of carbohydrate recognition domain (CRD)
that has many evolutionarily conserved residues among multiple
eukaryotic species [1]. Galectins are ubiquitously expressed in
mammalian tissues and are characterized by wide taxonomic distri-
bution that spans across multiple phyla [2]. Galectins are known to
be involved in varied roles in biological processes, ranging from
Anu Paul and Shang-Chuen Wu contributed equally with all other contributors.
55
56 Anu Paul et al.
regulating neutrophil function to neurodevelopment [3, 4]. In

particular, galectins appear to influence many aspects of both innate
and adaptive immune system in mammals [5, 6]. To better under-
stand the structural and functional relationship of galectins in vari-
ous organisms, purification of galectins in active form is a
fundamental requirement.
The history of galectin discovery is attributed to the efficient
screening of various tissues using lactose or asialofetuin immobi-
lized on Sepharose matrix [7–9]. Briefly, the tissue homogenates
were subjected to interaction with the ligand-immobilized affinity
matrix and the proteins that were sugar-specifically bound to the
matrix were subsequently eluted using high molar concentration of
ligand. This method has obvious limitations to meet the require-
ments of present-day researchers who study galectins from various
organisms. Resourcefully combining the immense array of data
available in genomic databases to obtain information on galectin
sequences as well as following standard recombinant protein purifi-
cation methods equip researchers to purify galectins from any
organism of interest. The purpose of this chapter is to give a
detailed description on the purification of bacterially expressed
recombinant galectins using multiple purification strategies. Differ-
ent strategies may be necessary for optimal recovery of the protein.
In this chapter, we are describing three purification strategies of
recombinant galectins, i.e., lactosyl-Sepharose chromatography,
Nickel NTA-chromatography, and glutathione-Sepharose chroma-
tography for purifying fungal galectin CGL2, 6 His-tagged
human galectin-7 and GST-tagged human galectin-7, respectively.
This chapter also includes PD-10 desalting protocol to remove
lactose from galectin solution that is purified using lactosyl-
Sepharose chromatography. This is crucial to obtain ligand-free
galectins to be used for biochemical experiments.
In this chapter, we describe purification of the fungal galectin
Coprinopsis cinerea galectin-2 (CGL2) using lactosyl-Sepharose
affinity chromatography (Fig. 1). Coprinopsis cinerea gray shag is a
species of fungus that expresses two lectins, i.e., CGL1 and CGL2
that are related in sequence and carbohydrate-binding specificity to
other galectins [10, 11]. Lactosyl-Sepharose affinity chromatogra-
phy has been applied for purifying active galectins from various
sources of organisms, i.e., CchG from Cinachyrella sp. (ball
sponge) [12], LEC-6 from nematode Caenorhabditis elegans
[13], CvGal-1 from eastern oyster Crassostrea virginica [14], vari-
ous human galectins, etc. Given the possible role of galectins in
providing innate immunity against microbes [15], defining the
specificity of recombinant galectins using array technology holds
promise in defining key aspects of their biological activity [16, 17].
Many researchers use Nickel-NTA affinity chromatography
method to purify recombinant proteins in general, including galec-
tins using a 6 His-tag [18–22]. Here, we use human galectin-7 as
Purification of Recombinant Galectins from Different Species Using. . . 57
Fig. 1 SDS-PAGE analysis of a fungal galectin CGL2 purified by lactosyl-Sepharose column. SDS-PAGE stained
with Coomassie Brilliant Blue. Whole-cell lysate (Lysate), supernatant obtained after high-speed centrifuga-
tion, flow through (FT) obtained after passing through the lactose-Sepharose, final wash and 100 mM lactose
eluate fractions are shown
an example for Ni-NTA affinity chromatography purification

(Fig. 2). Human galectin-7 is a prototypic galectin with a single
CRD which forms divalent homodimers and is mainly found in
epithelial tissues [23]. The gene sequence for human galectin-7
(NCBI Reference Sequence: NP_002298.1) was cloned into a
pET-28a vector with the addition of an N-terminal histidine tag.
We confirmed that human galectin-7 purified using Nickel-NTA
affinity chromatography retains activity by verifying its binding
capacity to lactosyl-Sepharose column, and galectin-7 purified by
this method is suitable to be used for biochemical studies. How-
ever, in our experience, not all galectins can be purified with this
method. For example, full-length human galectin-9 purified using
Ni-NTA method does not retain its carbohydrate-binding activity
as evident by its lack of binding to lactosyl-Sepharose column [24].
As a third method, purification of human galectin-7 using
glutathione-agarose affinity chromatography is also described in
this chapter (Fig. 3). Glutathione S-transferase (GST) tag is a
211 amino acid protein (26 kDa) added to the N-terminus of the
58 Anu Paul et al.
Fig. 2 SDS-PAGE analysis of a His-tagged human galectin-7 purified by Nickel-NTA column. SDS-PAGE
stained with Coomassie Brilliant Blue. Whole-cell lysate (Lysate), supernatant obtained after high-speed
centrifugation, flow through (FT) obtained after passing through the Ni-NTA, washes with 20 mM imidazole
and 40 mM imidazole and eluate fractions with 300 mM Imidazole are shown
target protein by cloning it in pGEX expression vector [25]. With

the GST tag, protein folds rapidly into a stable and highly soluble
form upon translation, leading to greater expression and solubility
of recombinant proteins. Therefore, many researchers use
GST-fusion proteins to perform glycan microarrays [26] and pull-
down assays with immunopurified glycoproteins [27]. Besides,
many galectins are commonly stabilized with reducing agents like
β-mercaptoethanol. During the process of GST protein purifica-
tion, presence of reducing agents like DTT and reduced glutathi-
one can help galectin maintain stability. However, it should be
noted that DTT or similar reducing agents should be removed as
these reagents can significantly influence the outcome of galectin
incubation with cells [28–31].
In conclusion, galectins can be purified using multiple affinity
chromatography methods but the most important factor to con-
sider while working with galectins is to ensure that the purified
protein retains its carbohydrate binding activity. Otherwise, it is
possible that binding data generated may not be reflective of the
biological activity of galectins, especially in the cases where galectin
binding is not observed. The ideal way to ensure lectin activity of
galectin is to look for its binding to lactosyl-Sepharose column.
Fig. 3 SDS-PAGE analysis of a GST-tagged human galectin-7 purified by Glutathione-Sepharose column.

SDS-PAGE stained with Coomassie Brilliant Blue. Whole-cell lysate (Lysate), supernatant obtained after high-
speed centrifugation, flow through (FT) obtained after passing through the glutathione-Sepharose, first wash,
final wash, eluate fractions with 50 mM reduced glutathione and eluate from lactose-Sepharose column are
shown
However, while working with small amounts of galectins that are

commercially purchased, it may be difficult to analyze using this
approach. In this case, hemagglutination is a simple way to check
galectin activity in a preparation of protein, even though this is not
a quantitative approach [31]. To ensure that biological readouts
actually reflect carbohydrate-dependent binding, inclusion of
carbohydrate-based inhibitors, such as lactose or thiodigalactoside
(TDG) is critical.
2 Materials
The first step is to clone the gene of fungal galectin CGL2 (NCBI
Reference Sequence: XP_001836012.2 with amino acid 1–150 and
codon optimized for E. coli expression) into pET-21a vector and
transform into BL-21(DE3) E. coli.
60 Anu Paul et al.
2.1 Expression and

Purification of Fungal
Galectin CGL2 Using
Lactosyl-Sepharose
Affinity
Chromatography
2.2 Preparation of Use Milli-Q-purified water or its equivalent in all solutions and
Luria-Bertani Broth for protocol steps wherever H2O is required.
Bacterial Cultivation
1. LB broth powder (RPI Research Products International
L24060).
2. 250-mL conical flask (1 flask).
3. 2000-mL Erlenmeyer conical flasks (6 flasks).
4. Double-distilled water.
5. Laboratory autoclave.
6. Fisherbrand™ Aluminum Foil 24 25 ft (01-213-100).
7. Indicator Tape, Autoclave Steam Sterilization, 3/400 1650 :
VWR 10153-150.
8. Ampicillin-sodium salt (Sigma Aldrich A9518).
Prepare a stock solution of 100 mg/mL. Pass the ampicil-
lin solution through a 0.22-μm filter and aliquot into 1.5-mL
vials and store at 20 C.
9. LB-Ampicillin media: Dissolve ampicillin solution in LB broth
to a final concentration of 100 μg/mL.
2.3 Purification of 1. Glycerol stocks of BL-21 transformed with CGL2-expressing

Recombinant Fungal plasmid stored at 80 C.
Galectin 2. 100 mL LB-Ampicillin media.
2.3.1 Day 1: Starter 3. Fisherbrand Disposable Inoculating Loops and Needles,
Culture Length: 8 in. (20 cm) (Fisher Scientific 22-363-605).
2.3.2 Day 2: Secondary 1. 6 mL of Ampicillin-sodium solution (100 mg/mL).

Culture 2. Six 2000-mL flasks of autoclaved LB broth (1 L LB broth in
each flask).
3. Isopropyl β-D-1-thiogalactopyranoside (IPTG) (Fisher Bio
Reagents BP1755100). Prepare 1 M IPTG in PBS and sterilize
using 0.22-μm filter.
4. Phosphate-buffered saline (PBS), 1, 0.0067 M PO4, without
calcium, magnesium (Fisher scientific SH30256.02).
5. Shaking incubator for bacterial growth (should be able to set at
temperatures range of 16–37 C).
2.3.3 Day 3: Pelleting of 1. Centrifuge (Sorvall RC5B Plus).

Bacterial Cultures 2. Sorvall plastic centrifuge bottles with a capacity of 1 L.
2.4 Processing of 1. Bacterial lysis buffer components.

Bacterial Pellets (a) PBS.
(b) 14 mM β-mercaptoethanol (Sigma Aldrich M3148) (see
Notes 1 and 2).
(c) Pierce™ protease inhibitor tablets, EDTA-free (Thermo
Fisher Scientific A32965).
(d) Lysozyme from chicken egg white, lyophilized powder
(Sigma Aldrich L6876). Dissolve in sterile double distilled
water to a stock concentration of 100 mg/mL and store at
20 C.
(e) Ribonuclease A from bovine pancreas (Sigma Aldrich
R4875). Dissolve in sterile double distilled water to a
stock of concentration 10 mg/mL and store at 20 C.
(f) DNase I, lyophilized powder from bovine pancreas
(Sigma Aldrich 11284932001). Dissolve in sterile
double-distilled water to a stock concentration 10 mg/
mL and store at 20 C.
2. Sonicator (Fisher Scientific Model FB120).
3. Beckman Centrifuge.
4. JA-25.50 fixed angle rotor and Nalgene™ Oak Ridge High-
speed PPCO Centrifuge tubes with Sealing Cap.
2.5 Coupling of 1. Divinyl sulfone, 98% purity, (VWR-77-77-0) (see Note 3).
Lactose to Sepharose 2. Sepharose™ Gel Filtration Base Matrix (GE Healthcare
4B Using Divinyl 17-0120-01, Sepharose™ 4B).
Sulfone
3. Sodium carbonate (Na2CO3): (Sigma-Aldrich 330361). Pre-
pare 2L of 0.5 M Na2CO3 (pH 11.0).
4. Glass beakers (2 L).
5. Buchner funnel.
6. pH meter.
7. Fume hood.
8. PBS.
9. Double-distilled water.
10. Whatman qualitative filter paper (GE Health care Life Sciences,
150 mm, CAT No. 1540-150).
11. Magnetic stirrer.
12. Scienceware F371220050 Polygon Spinbar, Magnetic Stir Bar
w/Ring, 50 8 mm; 1/Pk.
62 Anu Paul et al.
13. Sodium azide, ReagentPlus®, 99.5%, CAS 26628-22-

8 (Sigma-Aldrich S2002).
14. D-Lactose monohydrate, ACS Certified Grade (Fisher Chemi-
cal L5-500).
15. Spatula to transfer the gel.
16. Buffers:
(a) Buffer A: 0.5 M Na2CO3.
(b) Buffer B: 10% lactose in 100 mL Buffer A.
(c) Column storage buffer: PBS with 14 mM
β-mercaptoethanol. Add 0.02% (w/v) sodium azide and
mix well.
2.6 Packing of 1. Coupled lactosyl-Sepharose (50% [v/v] in PBS).

Lactosyl-Sepharose 2. Econo-Column® chromatography columns, 1.5 50 cm
Chromatography (LSC) (Bio-Rad: 7370717).
Column
3. Iron stand for attaching the Econo-column.
4. Tubing for passing solutions to the chromatography column.
5. 3-Prong dual adjustable clamps to connect column to the
stand.
6. Sand, white quartz, 50–70 mesh particle size.
7. LSC Wash buffer: PBS with 14 mM β-mercaptoethanol.
2.7 Galectin (This procedure must be done inside the cold room at 4 C)
Purification
1. Lactosyl-Sepharose column that is washed and equilibrated in
LSC wash buffer.
2. Connectors and adaptors for Econo-Column® chromatogra-
phy columns.
3. Pipette-aid.
4. Measuring pipettes.
5. Fraction collector and tubes.
6. 1.5-mL vials.
7. LSC wash buffer: 1 L PBS with 14 mM β-mercaptoethanol.
8. LSC elution buffer: 100 mL PBS with 14 mM
β-mercaptoethanol and 100 mM lactose.
9. Amicon Ultra centrifugal filters Ultracel-10k, Sigma Aldrich,
UFC901024.
10. Spectrophotometer for measuring OD at 280 nm.
11. Column Storage Buffer: PBS with 0.02% (w/v) sodium azide
and 14 mM β-mercaptoethanol.
2.8 Removal of 1. Disposable PD 10 desalting columns (Cytiva 17085101).

Lactose from Purified 2. 1.5-mL vials.
Galectin Solution
3. Spectrophotometer for measuring OD at 280 nm.
Using PD10-Desalting
Column 4. Iron Stand to connect the PD10 desalting column.
5. PBS.
6. Galectin eluted with 100 mM lactose (freshly purified or stored
at 80 C and thawed immediately before use).
7. Amicon Ultra-15 Centrifugal Filter Unit with Ultracel-10
membrane.
2.9 Expression and Additional items and reagents required for purifying galectin-7
Purification of Human using Nickel-NTA purification are mentioned here.
Galectin-7 Using 1. Human galectin-7 plasmid: NCBI Reference Sequence:
Nickel-NTA NP_002298.1 cloned into pET-28a vector.
Chromatography
2. Kanamycin sulfate (Sigma Aldrich K1377).
Prepare a stock solution of 50 mg/mL. Pass the kanamycin
sulfate solution through a 0.22-μm filter and aliquot into 1.5-
mL vials and store at 20 C.
3. LB-kanamycin broth: Add kanamycin stock solution to LB
broth to make a final concentration of 50 μg/mL.
4. Ni Sepharose™ excel (GE Healthcare 17-3712-01).
5. Imidazole (Sigma-Aldrich 1202).
6. Trizma HCl (Sigma-Aldrich RDD009).
7. Sodium chloride (Sigma-Aldrich S7653).
8. Hydrochloric acid 37% (Sigma-Aldrich 1003170510).
9. Sodium hydroxide pellets 97% (Sigma-Aldrich V000101).
2.9.1 Reagents and 1 M Tris–HCl, pH 8.0.

Solutions 2 M NaCl.
2 M Imidazole (avoid exposure of the solution to light).
Ni-NTA lysis buffer: 50 mM Tris–HCl, pH 8.0, 250 mM NaCl,
10 mM imidazole, Lysozyme (100 μg/mL), DNase (10 μg/
mL), RNase (10 μg/mL), protease inhibitor tablets (1 per
50 mL lysis buffer).
Ni-NTA binding buffer: 50 mM Tris–HCl, pH 8.0, 250 mM NaCl,
10 mM imidazole.
Ni-NTA Wash buffer 1: 50 mM Tris–HCl, pH 8.0, 250 mM NaCl,
20 mM imidazole.
Ni-NTA Wash buffer 2: 50 mM Tris–HCl, pH 8.0, 250 mM NaCl,
40 mM imidazole.
Ni-NTA elution buffer: 50 mM Tris–HCl, pH 8.0, 250 mM NaCl,
300 mM imidazole.
Ni-NTA dialysis buffer: 50 mM Tris–HCl, pH 8.0, 250 mM NaCl.
64 Anu Paul et al.
2.10 Purification of 1. Human galectin-7 plasmid: NCBI Reference Sequence:

Human Galectin-7 by NP_002298.1 cloned into pGEX-2TK vector.
Glutathione S- 2. Trizma HCl (Sigma Aldrich RDD009).
Transferase Affinity
3. NaCl (Sigma Aldrich 746398).
Chromatography
4. Triton-X-100 (EM Science 9410).
5. Dithiothreitol (DTT) (GE Healthcare 17-1318-01).
6. Glutathione Sepharose 4B (GE17-0756-01).
7. L-Glutathione reduced (Sigma-Aldrich G4251).
8. Benchmark Roto-Mini™ Rotator Series.
2.10.1 Reagents and 1 M Tris–HCl, pH 7.5.

Solutions 1 M Tris–HCl, pH 8.0.
2 M NaCl.
500 mM DTT.
10% Triton-X-100.
GST Lysis buffer: 50 mM Tris–HCl, pH 7.5, 100 mM NaCl,
Lysozyme (100 μg/mL), DNase (10 μg/mL), RNase
(10 μg/mL), 1 mM DTT, Protease inhibitor tablets, 1% Tri-
ton-X-100.
GST Binding buffer: 50 mM Tris-HCl, pH 7.5, 100 mM NaCl,
1 mM DTT.
GST Wash buffer 1: 50 mM Tris-HCl, pH 7.5, 100 mM NaCl,
1 mM DTT.
GST Wash buffer 2: 50 mM Tris-HCl, pH 8.0, 100 mM NaCl,
1 mM DTT.
GST Elution buffer: 50 mM Tris-HCl, pH 8.0 with 50 mM
reduced glutathione, 1 mM DTT, 100 mM NaCl.
Check the pH after adding reduced glutathione. Adjust pH to 8.0
using NaOH.
3 Methods
3.1 Expression and The purification of galectin using affinity chromatography is a 3-day
Purification of procedure. So, the experiment should be planned accordingly.
Galectins Using 1. Dissolve 20 g of LB broth powder in 1 L of water into each 2 L
Lactosyl-Sepharose flask, preparing total 6 flasks.
Chromatography (LSC)
2. Dissolve 2 g of LB broth powder in 100 mL of water in one
Column
250-mL flask.
3. Autoclave at 121 C for 20 min.
4. Let the bacterial culture media cool down to room temperature

before use.
5. Add ampicillin (final concentration 100 μg/mL) to LB-broth
before use.
3.1.1 Protocol for 1. Add 100 μL of ampicillin solution to 100 mL of autoclaved LB

Bacterial Culture broth to make a final ampicillin concentration of 100 μg/mL.
2. Take the glycerol stock from 80 C freezer and place on
3.1.1.1 Day 1: Starter dry ice.
Culture
3. Take a loopful of bacterial glycerol stock and inoculate into
100 mL LB-ampicillin media.
4. Incubate in the bacterial shaker overnight at 37 C at 250 rpm.
3.1.1.2 Day 2: Secondary 1. Place six flasks containing 1 L of LB media in the

Culture bacterial hood.
2. Add 1 mL of ampicillin (stock concentration 100 mg/mL) to
each flask.
3. Add 10 mL of the bacterial starter culture or primary inoculum
(1% of the total) to each of the flasks.
4. Incubate all flasks at 37 C at 250 rpm in a bacterial shaker
incubator until the OD600 reaches 0.4–0.6. Blank with LB
Media. Usually, this OD is attained in 2–3 h.
5. Prepare 1 M solution of IPTG in PBS. Add 1 mL of 1 M IPTG
to each flask to attain a final concentration of 1 mM IPTG.
6. After adding IPTG, continue incubation at 16 C for 20 h at
250 rpm.
3.1.1.3 Day 3: Pelleting 1. After incubation, pour the bacterial culture into six clean 1-L
of Bacterial Cultures plastic centrifuge bottles.
2. Collect bacterial pellets by centrifuging the bacterial culture at
6000 g for 30 min at 4 C in a Sorvall RC5B plus centrifuge.
3. Carefully pour off the supernatant without disturbing the
pellets.
4. The bacterial pellets can be stored at 80 C until further use.
3.2 Lysis of Bacterial 1. Add 60 μL each of Lysozyme, Ribonuclease A and DNase

Pellets 1 from the stock solution into the 60 mL PBS with 14 mM
β-mercaptoethanol, pH 7.4 and mix well (see Note 4).
3.2.1 Preparation of LSC
Lysis Buffer 2. Add two protease inhibitor tablets and let them dissolve in the
LSC lysis buffer.
3.2.2 Method of 1. Take the bacterial pellets from 80 C and thaw on ice.
Bacterial Lysis 2. Add 10 mL of LSC Lysis Buffer to each bottle and resuspend
pellets gently until the pellets get homogenized in an even
manner.
66 Anu Paul et al.
3. Let the resuspension mix sit on ice for 30 min.

4. Keeping on ice, sonicate the pellets at 60% amplitude (sonica-
tion intensity setting) for 5 min with 10 s ON and 10 s OFF.
Collect 20 μL of lysate for SDS-PAGE analysis (see Note 5).
5. Transfer the lysate to Nalgene™ Oak Ridge High-speed PPCO
Centrifuge tubes with Sealing Cap and spin at 20,000 g at
4 C for 30 min in a Beckman centrifuge using a JA-25.50 fixed
angle rotor.
6. Collect the supernatant to a clean Oak Ridge tube without
disrupting the pellets.
7. Repeat the centrifugation at 20,000 g at 4 C for 30 min.
8. Collect 20 μL of supernatant for SDS-PAGE analysis.
3.3 Coupling of For the purification of the galectin from the bacterial culture super-
Lactose to Sepharose natant, affinity chromatography using lactose-Sepharose matrix is
4B Using Divinyl necessary to isolate only active protein fraction. Lactose can be
Sulfone coupled to Sepharose using divinyl sulfone (VWR-77-77-0) as
previously described but a succinct method, adapted from this
protocol, is illustrated below [7]. All steps are done at room tem-
perature. The procedure can be scaled up by multiplying all
volumes by 3.
1. Place a Whatman qualitative filter paper on the Buchner funnel
and pour double-distilled water gently over it to make the
paper adhere firmly to the funnel. This ensures that there will
be no leakage of Sepharose gel through the funnel during the
coupling procedure.
2. Add 100 mL of Sepharose 4B in the Buchner funnel and wash
with buffer A (0.5 M Na2CO3) as required until pH 11 is
obtained.
3. Drain the gel in a beaker using a spatula and resuspend in
100 mL buffer A.
4. Place the beaker on a magnetic stirrer plate and place a stir bar
inside.
5. Add 10 mL divinyl sulfone slowly (2 mL aliquots over
10–15 min) to the gel with gentle stirring.
6. The mixture will become grayish brown. Some drops of the
divinyl sulfone might appear undissolved but they will dissolve
eventually.
7. Keep the gel mixture for slow stirring on the magnetic stirrer
plate for up to 70 min at 23 C.
8. Transfer the gel mixture to Buchner funnel and wash with
300 mL of buffer A (0.5 M Na2CO3).
9. Drain the gel mixture into a beaker.

10. Resuspend the gel mixture in a 100 mL of buffer B (10%
lactose in 100 mL Buffer A).
11. Place the beaker on a magnetic stirrer plate and put a stir bar
in it.
12. Let the mixture stir gently to facilitate the coupling reaction.
13. Allow the reaction to proceed for 15 h (or overnight).
14. Finally, transfer the gel to a Buchner funnel with a fresh What-
man filter paper as done previously.
15. Transfer the gel to the Buchner funnel.
16. Wash the gel with 300 mL of Buffer A (0.5 M Na2CO3).
17. Wash the gel with 300 mL of distilled water.
18. Wash the gel with 300 mL of PBS.
19. If not used immediately, store the gel in column storage buffer
at 4 C.
3.3.1 Packing and 1. Pour sand into the column to cover the length of 6–8 cm at the
Preparation of Lactosyl- bottom of the column (see Note 6).
Sepharose Column 2. Pour the 60 mL coupled lactosyl-sepharose (50% v/v in PBS)
into the Econo-column.
3. Connect the adaptor to the top of the column.
4. Pass 10 times column volumes of PBS with 14 mM
β-mercaptoethanol through the column to remove the column
storage buffer.
3.4 Galectin 1. Set the Lactosyl-Sepharose Chromatography (LSC) column

Purification ready and connected to the flask containing wash buffer (see
Note 7).
2. Apply the bacterial culture supernatant to the column.
3. Collect 20 μL of sample flow-through for SDS-PAGE analysis.
4. Wash the LSC with ten times column volumes of LSC Wash
Buffer. Check OD280 of wash fraction using a spectrophotom-
eter intermittently. Continue with additional washing till the
OD reaches zero.
5. Collect a fraction of wash to be analyzed by SDS-PAGE.
6. Elute the galectin protein with LSC Elution Buffer. Collect
1 mL fractions using a fraction collector in 1.5 mL vials.
7. Read the OD280 using a spectrophotometer. Blank with LSC
Elution Buffer. For calculation of the protein concentration, see
Note 8.
68 Anu Paul et al.
8. Determine the fractions showing highest OD values and com-

bine them. If the concentration of the protein is less than
2–3 mg/mL, concentrate the solution using a 10 kDa
MWCO Amicon centrifugal filter at 3361 g at 4 C.
9. Aliquot into 0.5 mL with a final concentration of 2–3 mg/mL
in each tube and store at 80 C.
3.5 Re-equilibration 1. Wash LSC with ten column volumes of column storage buffer.
and Storage of 2. Close the column and store at 4 C until further use.
Lactose-Sepharose
Column
3.6 Removal of Galectin that is eluted using affinity chromatography method con-
Lactose from Galectin tains lactose. For biochemical binding experiments with the pro-
Solution Using tein, lactose should be removed. This can be done with PD-10
Desalting PD10 desalting column.
Column 1. Wash the column with five times column volumes of 1 PBS.
2. Thaw the vials containing purified galectin on ice.
3. After washing the column with PBS, load 2 mL of protein
through the column.
4. Collect the flow through in 1.5-mL Eppendorf vials with each
fraction of 500 μL volume.
5. Continue washing with 1 PBS.
6. Up to 10 fractions can be collected (see Note 9).
7. Later, check the OD at 280 nm using a spectrophotometer.
8. Pool the fractions with positive OD reading.
9. Concentrate the protein at 4 C using Amicon 10K centrifugal
filter unit by following manufacturer’s guidelines. De-salted
galectin can be used for biochemical studies.
3.7 Purification of 1. Clone human galectin-7 gene in pET-28a vector and transform
Human Galectin-7 with E. coli BL21 (DE3) and store the clones in glycerol at
Using Nickel-NTA 80 C.
Affinity
Chromatography
3.7.1 Day 1: Starter 2. Aseptically inoculate 100 mL of LB broth containing 50 μg/

Culture mL kanamycin with a loopful of bacterial glycerol stock.
3. Incubate bacterial culture in a shaker incubator with agitation
(250 rpm) at 37 C overnight.
3.7.2 Day 2: Main Culture 4. Inoculate six flasks containing 1 L LB-kanamycin broth each
with 10 mL of the starter culture (1% of total volume).
5. Incubate the flasks at 37 C with shaking at 250 rpm for

approximately 2–3 h. Check the OD600 (blank with LB
broth) using a spectrophotometer.
6. When bacteria are grown to the mid-log phase (OD600
reaches 0.4–0.6), induce protein expression by adding 1 mL
of 1 M IPTG solution to each flask to attain a final concentra-
tion of 1 mM IPTG. Incubate the culture at 16 C for 20 h
with 250 rpm shaking.
3.7.3 Day 3: Pelleting of 7. Harvest bacteria pellets by centrifugation at 6000 g at 4 C

Bacterial Cultures and for 30 min and resuspend in Ni-NTA lysis buffer (10 mL lysis
Lysis of Bacterial Pellets buffer per liter of bacterial culture).
8. After resuspension in Ni-NTA lysis buffer, keep the suspension
on ice and sonicate at 60% amplitude (sonication intensity
setting) for 5 min with 10 s ON and 10 s OFF for cell disrup-
tion. Collect 20 μL of lysate for SDS-PAGE analysis.
9. Centrifuge the lysate in an Oak Ridge tube at 20,000 g at
4 C for 30 min.
10. Transfer the supernatant to a fresh Oak Ridge tube and
centrifuge at 20,000 g at 4 C for 30 min. Collect 20 μL
of the supernatant for SDS-PAGE analysis.
3.7.4 Nickel-NTA Column 11. Apply the supernatant after the second centrifugation step to
Purification of Human 1 mL resin-bed GE-Ni-Sepharose preequilibrated with
Galectin-7 Ni-NTA binding buffer.
12. Collect 20 μL of the sample flow-through for SDS-PAGE
analysis.
13. Wash resin with five resin-bed volumes of Ni-NTA binding
buffer.
14. Collect a fraction of the wash for analysis using SDS-PAGE
Gel as “10 mM imidazole.”
15. Wash resin with five resin-bed volumes of Ni-NTA Wash
buffer 1.
16. Collect a portion of Wash 1 to analyze using SDS-PAGE Gel
as “20 mM imidazole.”
17. Wash resin with Ni-NTA wash buffer 2 until the OD280
reaches zero.
18. Collect a portion of Wash 2 to analyze using SDS-PAGE Gel
as “40 mM imidazole.”
19. Pass 6 mL of Ni-NTA elution buffer through the column and
collect 2 mL eluate each in 15-mL conical tubes.
20. Check the OD280 in spectrophotometer and collect the frac-
tions with higher values.
70 Anu Paul et al.
21. Extensively dialyze the eluate and buffer exchange at 4 C to

Ni-NTA dialysis buffer with an Amicon centrifugal filter 10K at
3361 g.
22. Concentrate the protein to 2–3 mg/mL.
23. Aliquot into 0.5 mL each in 1.5 mL tube and store at 80 C.
3.8 Purification of 1. Clone human galectin-7 plasmid: NCBI Reference Sequence:

Human Galectin-7 NP_002298.1 into pGEX-2TK vector.
Using Glutathione- 2. Bacterial culture was done in LB broth containing 100 μg/mL
Affinity ampicillin with agitation (250 rpm) at 37 C.
Chromatography 3. When bacteria are grown to the mid-log phase (0.4–0.6 OD),
induce protein expression by the addition of 1 mL of 1 M IPTG
solution to each flask to attain a final concentration of
1 mM IPTG.
4. After 20 h induction in 16 C, harvest bacteria pellets by
centrifugation at 6000 g and then resuspend in lysis buffer
(10 mL per liter of bacterial culture).
5. Keeping the sample on ice, sonicate at 60% amplitude for 5 min
with 10 s ON and 10 s OFF.
6. Collect 20 μL of lysate for SDS-PAGE analysis.
7. Add Triton X-100 at a final concentration of 1% to the bacteria
lysate (see Note 10).
8. Add a magnetic stir bar to the bacteria lysate and stir for 30 min
on a stir plate at 150 rpm in cold room at 4 C.
9. Collect the sample in an Oak Ridge tube and spin at 20,000 g
in a JA-25.50 rotor for 30 min at 4 C.
10. Transfer the supernatant to a fresh Oak Ridge tube and repeat
the spin at 20,000 g for 30 min at 4 C.
11. Collect the supernatant in 50-mL conical tubes and keep
on ice.
12. Collect 20 μL of supernatant for SDS-PAGE.
13. Take 2 mL of Glutathione 4B Sepharose in a 50-mL conical
tube and remove the alcohol-containing storage solution by
spinning at 700 g for 5 min at 4 C and removing superna-
tant (see Note 11).
14. Add 10 times column volumes of PBS to glutathione resin and
keep in end-over-end mixing in a rotating mixer for 3 min. Spin
at 700 g for 5 min at 4 C and remove supernatant.
15. Pre-equilibrate glutathione-Sepharose 4B with 10 times col-
umn volumes of GST binding buffer. Spin at 700 g for 5 min
at 4 C and remove supernatant.
16. Mix the protein supernatant with glutathione-Sepharose pre--

equilibrated with binding buffer and keep in gentle rotation at
4 C for 3–16 h (see Note 12).
17. Transfer the sample to an empty gravity column.
18. Collect 20 μL of sample flow-through for SDS-PAGE.
19. Wash the resin with ten resin-bed volumes of GST Wash
buffer I.
20. Collect a small fraction of wash for SDS-PAGE.
21. Wash the resin with ten resin-bed volumes GST Wash buffer II.
22. Collect a small fraction of wash for SDS-PAGE.
23. Check the OD280 in spectrophotometer. If necessary, con-
tinue washing and stop washing once the OD reaches zero.
24. Add 2 mL GST elution buffer pH 8.0 to the column, close the
column, and leave at room temperature for 20 min. Using a
1-mL micropipette tip, gently mix the GST-bead suspension
with GST elution buffer 2–3 times.
25. Collect the 0.5 mL eluate volumes into 1.5-mL vials.
26. Check the OD at 280 nm in spectrophotometer.
27. Repeat the elution step twice.
28. Extensively dialyze the eluate and buffer exchange against
50 mM Tris–HCl, pH 7.5 at 4 C in an Amicon 10K centrifu-
gal filter.
29. Concentrate the protein to 2–3 mg/mL (see Note 13).
30. Aliquot into 0.5 mL to final concentration of 2–3 mg/mL
protein in each tube and store at 80 C.
4 Notes
1. It is known that not all members of the galectin family neces-

sarily require the addition of β-mercaptoethanol in buffers to
maintain the protein activity. Even though CGL2 could remain
active in the absence of β-mercaptoethanol, we have given the
protocol with the addition of β-mercaptoethanol as a standard
protocol for galectin purification.
2. Always use a fume hood while working with
β-mercaptoethanol.
3. Divinyl sulfone is a toxic reagent. Store and use in fume hood
and avoid contact with skin.
4. Add lysozyme/DNase/RNase/protease inhibitor tablets to
lysis buffer immediately before use.
72 Anu Paul et al.
5. For analyzing lysate/supernatant/flow-through in an

SDS-PAGE gel, mix 5 μL of the sample after sonication,
10 μL of distilled water, and 5 μL of 4 SDS loading dye.
Load 5 μL of the mix in the SDS-PAGE well for analysis. For
analyzing samples from washes, prepare the sample mix in the
same manner and load the whole mix. We have used Tru-
PAGE™ LDS Sample Buffer 4 (PCG 3009: Sigma Aldrich)
for SDS loading.
6. The purpose of packing sand in the column is to make further
pipetting and handling of the column easy.
7. This procedure must be done inside the cold room.
8.
OD at 280 nm
Concentration ¼
Extinction coefficient Pathlength
Molecular weight Dilution factor
[32].
9. OD at 280 nm starts showing higher values approximately
from the fifth–sixth fraction onwards. Pool those fractions for
further use.
10. Preparation of 10% Triton-X-100: Warm the Triton-X-100 at
30–40 C in a water bath.
Using a cut tip P1000, add 1 mL of Triton-X-100 to 9 mL
distilled water in a conical tube. Keep the conical tube in a
water bath until all Triton-X-100 is dissolved. Cool and then
add 10% Triton-X-100 to the lysate to a final concentration
of 1%.
11. While pipetting GST beads, use a cut tip P1000.
12. Binding of GST to glutathione is flow-rate dependent, and
binding capacity increases with lower flow rates. This should
be considered during sample loading and elution.
13. Checking the binding of purified galectin to lactosyl-Sepharose
column is essential to confirm the activity of galectin.
Acknowledgments
This work was supported in part by National Institutes of Health

Grants R01 HL154034 to CMA and U01 CA242109 to SRS.
Conflict of Interest No conflict of interest declared.
References
1. Barondes SH, Castronovo V, Cooper DN, sequence and specificity of two isolectins from
Cummings RD, Drickamer K, Feizi T, Gitt Coprinus cinereus. J Biol Chem
MA, Hirabayashi J, Hughes C, Kasai K et al 272(3):1514–1521. https://doi.org/10.
(1994) Galectins: a family of animal beta- 1074/jbc.272.3.1514
galactoside-binding lectins. Cell 11. Walser PJ, Haebel PW, Kunzler M, Sargent D,
76(4):597–598. https://doi.org/10.1016/ Kues U, Aebi M, Ban N (2004) Structure and
0092-8674(94)90498-7 functional analysis of the fungal galectin
2. Cooper DN, Barondes SH (1999) God must CGL2. Structure 12(4):689–702. https://doi.
love galectins; he made so many of them. Gly- org/10.1016/j.str.2004.03.002
cobiology 9(10):979–984. https://doi.org/ 12. Freymann DM, Nakamura Y, Focia PJ, Sakai R,
10.1093/glycob/9.10.979 Swanson GT (2012) Structure of a tetrameric
3. Robinson BS, Arthur CM, Evavold B, galectin from Cinachyrella sp. (ball sponge).
Roback E, Kamili NA, Stowell CS, Vallecillo- Acta Crystallogr D Biol Crystallogr 68
Zuniga ML, Van Ry PM, Dias-Baruffi M, (Pt 9):1163–1174. https://doi.org/10.
Cummings RD, Stowell SR (2019) The 1107/S0907444912022834
sweet-side of leukocytes: galectins as master 13. Hirabayashi J, Ubukata T, Kasai K (1996) Puri-
regulators of neutrophil function. Front fication and molecular characterization of a
Immunol 10:1762. https://doi.org/10. novel 16-kDa galectin from the nematode Cae-
3389/fimmu.2019.01762 norhabditis elegans. J Biol Chem
4. Srejovic I, Selakovic D, Jovicic N, Jakovljevic V, 271(5):2497–2505. https://doi.org/10.
Lukic ML, Rosic G (2020) Galectin-3: roles in 1074/jbc.271.5.2497
neurodevelopment, neuroinflammation, and 14. Feng C, Ghosh A, Amin MN, Giomarelli B,
behavior. Biomolecules 10(5):798. https:// Shridhar S, Banerjee A, Fernandez-Robledo
doi.org/10.3390/biom10050798 JA, Bianchet MA, Wang LX, Wilson IB, Vasta
5. Stowell SR, Arthur CM, Dias-Baruffi M, GR (2013) The galectin CvGal1 from the east-
Rodrigues LC, Gourdine JP, Heimburg- ern oyster (Crassostrea virginica) binds to
Molinaro J, Ju T, Molinaro RJ, Rivera- blood group A oligosaccharides on the hemo-
Marrero C, Xia B, Smith DF, Cummings RD cyte surface. J Biol Chem
(2010) Innate immune lectins kill bacteria 288(34):24394–24409. https://doi.org/10.
expressing blood group antigen. Nat Med 1074/jbc.M113.476531
16(3):295–301. https://doi.org/10.1038/ 15. Stowell SR, Arthur CM, McBride R, Berger O,
nm.2103 Razi N, Heimburg-Molinaro J, Rodrigues LC,
6. Rabinovich GA, Liu FT, Hirashima M, Ander- Gourdine JP, Noll AJ, von Gunten S, Smith
son A (2007) An emerging role for galectins in DF, Knirel YA, Paulson JC, Cummings RD
tuning the immune response: lessons from (2014) Microbial glycan microarrays define
experimental models of inflammatory disease, key features of host-microbial interactions.
autoimmunity and cancer. Scand J Immunol Nat Chem Biol 10(6):470–476. https://doi.
66(2–3):143–158. https://doi.org/10.1111/ org/10.1038/nchembio.1525
j.1365-3083.2007.01986.x 16. Arthur CM, Cummings RD, Stowell SR
7. Levi G, Teichberg VI (1981) Isolation and (2014) Using glycan microarrays to under-
physicochemical characterization of electrolec- stand immunity. Curr Opin Chem Biol 18:
tin, a beta-D-galactoside binding lectin from 55–61. https://doi.org/10.1016/j.cbpa.
the electric organ of electrophorus electricus. 2013.12.017
J Biol Chem 256(11):5735–5740 17. Kamili NA, Arthur CM, Gerner-Smidt C,
8. Nowak TP, Kobiler D, Roel LE, Barondes SH Tafesse E, Blenda A, Dias-Baruffi M, Stowell
(1977) Developmentally regulated lectin from SR (2016) Key regulators of galectin-glycan
embryonic chick pectoral muscle. Purification interactions. Proteomics 16(24):3111–3125.
by affinity chromatography. J Biol Chem https://doi.org/10.1002/pmic.201600116
252(17):6026–6030 18. Hsieh TJ, Lin HY, Tu Z, Lin TC, Wu SC,
9. de Waard A, Hickman S, Kornfeld S (1976) Tseng YY, Liu FT, Hsu ST, Lin CH (2016)
Isolation and properties of beta-galactoside Dual thio-digalactoside-binding modes of
binding lectins of calf heart and lung. J Biol human galectins as the structural basis for the
Chem 251(23):7581–7587 design of potent and selective inhibitors. Sci
10. Cooper DN, Boulianne RP, Charlton S, Farrell Rep 6:29457. https://doi.org/10.1038/
EM, Sucher A, Lu BC (1997) Fungal galectins, srep29457
74 Anu Paul et al.
19. Hsieh TJ, Lin HY, Tu Z, Huang BS, Wu SC, 26. Poland PA, Rondanino C, Kinlough CL,
Lin CH (2015) Structural basis underlying the Heimburg-Molinaro J, Arthur CM, Stowell
binding preference of human Galectins-1, -3 SR, Smith DF, Hughey RP (2011) Identifica-
and -7 for Galbeta1-3/4GlcNAc. PLoS One tion and characterization of endogenous galec-
10(5):e0125946. https://doi.org/10.1371/ tins expressed in Madin Darby canine kidney
journal.pone.0125946 cells. J Biol Chem 286(8):6780–6790. https://
20. Verkerke H, Horwath M, Saeedi B, Boyer D, doi.org/10.1074/jbc.M110.179002
Allen JW, Owens J, Arthur CM, Nakahara H, 27. Poland PA, Kinlough CL, Hughey RP (2015)
Rha J, Patel K, Wu SC, Paul A, Yasin N, Cloning, expression, and purification of galec-
Wang J, Shin S, Brown D, Normile K, Cole L, tins for in vitro studies. Methods Mol Biol
Meyers M, Lin H, Woods E, Isaac J, Broder K, 1207:37–49. https://doi.org/10.1007/978-
Wade J, Kauffman RC, Patel R, Josephson CD, 1-4939-1396-1_2
Reynolds S, Sherman M, Wrammert J, Alter D, 28. Stowell SR, Karmakar S, Arthur CM, Ju T,
Guarner J, Roback JD, Neish A, Stowell SR Rodrigues LC, Riul TB, Dias-Baruffi M,
(2021) Comparison of antibody class specific Miner J, McEver RP, Cummings RD (2009)
SARS-CoV-2 serology for the diagnosis of Galectin-1 induces reversible phosphatidylser-
acute COVID-19. J Clin Microbiol 59(4): ine exposure at the plasma membrane. Mol
e02026-20. https://doi.org/10.1128/JCM. Biol Cell 20(5):1408–1418. https://doi.org/
02026-20 10.1091/mbc.E08-07-0786
21. Verkerke H, Saeedi B, Boyer D, Allen JW, 29. Stowell SR, Karmakar S, Stowell CJ, Dias-
Owens J, Shin S, Horwath M, Patel K, Baruffi M, McEver RP, Cummings RD
Paul A, Wu S, Wang J, Ho A, Arthur CM, (2007) Human galectin-1, -2, and -4 induce
Roback JD, Neish AS, Lough C, Josephson surface exposure of phosphatidylserine in acti-
CD, Stowell SR (2021) Are we forgetting vated human neutrophils but not in activated T
about IgA? A re-examination of COVID-19 cells. Blood 109(1):219–227. https://doi.
convalescent plasma. Transfusion org/10.1182/blood-2006-03-007153
61(6):1740–1748 30. Stowell SR, Cho M, Feasley CL, Arthur CM,
22. Allen JWL, Verkerke H, Owens J, Saeedi B, Song X, Colucci JK, Karmakar S, Mehta P,
Boyer D, Shin S, Roback JD, Neish AS, Stowell Dias-Baruffi M, McEver RP, Cummings RD
SR (2021) Serum pooling for rapid expansion (2009) Ligand reduces galectin-1 sensitivity
of anti-SARS-CoV-2 antibody testing capacity. to oxidative inactivation by enhancing dimer
Transfus Clin Biol 28(1):51–54. https://doi. formation. J Biol Chem 284(8):4989–4999.
org/10.1016/j.tracli.2020.10.008 https://doi.org/10.1074/jbc.M808925200
23. Magnaldo T, Fowlis D, Darmon M (1998) 31. Stowell SR, Arthur CM, Mehta P, Slanina KA,
Galectin-7, a marker of all types of stratified Blixt O, Leffler H, Smith DF, Cummings RD
epithelia. Differentiation 63(3):159–168. (2008) Galectin-1, -2, and -3 exhibit differen-
https://doi.org/10.1046/j.1432-0436.1998. tial recognition of sialylated glycans and blood
6330159.x group antigens. J Biol Chem
24. Wu SC, Paul A, Ho A, Patel KR, Allen JWL, 283(15):10109–10123. https://doi.org/10.
Verkerke H, Arthur CM, Stowell SR (2021) 1074/jbc.M709545200
Generation and use of recombinant galectins. 32. Arthur CM, Rodrigues LC, Baruffi MD, Sulli-
Curr Protoc 1(3):e63. https://doi.org/10. van HC, Heimburg-Molinaro J, Smith DF,
1002/cpz1.63 Cummings RD, Stowell SR (2015) Examining
25. Harper S, Speicher DW (2011) Purification of galectin binding specificity using glycan micro-
proteins fused to glutathione S-transferase. arrays. Methods Mol Biol 1207:115–131.
Methods Mol Biol 681:259–280. https://doi. https://doi.org/10.1007/978-1-4939-
org/10.1007/978-1-60761-913-0_14 1396-1_8
Chapter 4
Alkylation of Galectin-1 with Iodoacetamide and Mass

Spectrometric Mapping of the Sites of Incorporation
Shang-Chuen Wu, Anu Paul, Richard D. Cummings, Christa L. Feasley,
Abstract
Galectins can display unique sensitivity to oxidative changes that result in significant conformational
alterations that prevent carbohydrate recognition. While a variety of approaches can be utilized to prevent
galectin oxidation, several of these require inclusion of reducing agents that not only prevent galectins from
undergoing oxidative inactivation but can also interfere with normal redox potentials required for funda-
mental cellular processes. To overcome the limitations associated with placing cells in an artificial reducing
environment, cysteine residues on galectins can be directly alkylated with iodoacetamide to form a stable
thioether adduct that is resistant to further modification. Iodoacetamide alkylated galectin remains stable
over prolonged periods of time and retains the carbohydrate binding and biological activities of the protein.
As a result, this approach allows examination of the biological roles of a stabilized form of galectin-1
without introducing the confounding variables that can occur when typical soluble reducing agents are
employed.
Key words Alkylation, Galectin-1, Mass spectrometry, Oxidation, Reducing agents
1 Introduction
Unlike most proteins secreted from cells, galectins primarily reside

within the cytosol where they can be secreted into the extracellular
space through an endoplasmic reticulum and Golgi apparatus-
independent pathway [1, 2]. Synthesized on free ribosomes in the
cytosol of cells, galectins are maintained in a reduced state prior to
secretion from the cell secondary to the reducing environment
within the cell [3]. Following secretion, galectins can engage car-
bohydrate ligands outside the cells, which are critical ligands
responsible for mediating many galectin functions [4, 5]. Ligand
engagement appears to reduce galectin sensitivity to oxidative inac-
tivation in the extracellular environment and allows galectins to
Shang-Chuen Wu and Anu Paul contributed equally with all other contributors.
75
76 Shang-Chuen Wu et al.
participate in a wide range of biological activities, from regulating a

variety of cellular activities to directly providing innate immunity
[6–11]. However, in the absence of carbohydrate ligand, some
members of the galectin family, in particular galectin-1, can
undergo intramolecular disulfide oxidation, which results in signif-
icant alteration in protein conformation that precludes dimeriza-
tion and carbohydrate recognition [6, 12–15]. However, in the
presence of ligand, galectin-1 experiences enhanced dimerization,
which limits the formation of monomers that appear to be an
intermediate for effective galectin-1 oxidation [6].
While galectin-1 oxidation likely reflects a normal pathway that
regulates protein function [16, 17], oxidation tends to complicate
the assessment of galectin-1 activity in the context of other
biological systems. As a result, many studies include reducing
agents in treatment buffers to reduce the impact of oxidation
when assessing the carbohydrate recognition-dependent activity
of galectin-1 [18, 19]. While this approach facilitates biochemical
assessment of galectin-1 activity, such as carbohydrate binding
specificity, when added to biological systems reducing agents can
unfortunately complicate the interpretation of results. For example,
while inclusion of reducing agents such as dithiothreiotol (DTT)
and beta-mercaptoethanol (βME) prevent galectin oxidation, these
reagents also possess the ability to cross cell membranes where they
can induce the unfolded protein response within the endoplasmic
reticulum [20–22], likely secondary to preventing appropriate oxi-
dation during protein folding [21]. While utilization of membrane
impermeable reducing agents, such as reduced glutathione, can
reduce the deleterious intracellular consequences associated with
DTT and βME inclusion, these reagents can also reduce proteins
normally oxidized on the cell surface, which can likewise alter their
biological activity (Fig. 1). As a result, a method to prevent galectin
oxidation in the absence of residual soluble reducing agents is
needed to maintain galectin-1 activity while avoiding the con-
founding influence of reducing agent inclusion.
Cysteine oxidation typically reflects disulfide bond formation,
as occurs during galectin-1 oxidation, which can readily reverse
upon inclusion of free thiol reducing agents, such as βME, DTT,
or reduced glutathione [23]. As a result, these reducing agents
form distinct molecular adducts that do not typically reflect the
type of stable interactions commonly observed among other cova-
lent linkages in biological systems [23] (Fig. 2). Thus, reduction of
galectin-1 with βME, DTT, or reduced glutathione, followed by
the elimination of excess reducing agent results in spontaneous
adduct loss as reciprocal intramolecular disulfides form. In contrast,
cysteine modification driven by reaction with iodoacetamide or
maleimide results in the formation of a thioether, which represents
a relatively irreversible adduct that can prevent disulfide bond for-
mation [24] (Fig. 2). Thus, following iodoacetamide alkylation,
Alkylation of Galectin-1 with Iodoacetamide and Mass Spectrometric Mapping. . . 77
Fig. 1 Graphical abstract of galectin-1 inside and outside the cell. Galectin-1 located in the cytosol is
maintained in a reduced state prior to secretion through endoplasmic reticulum and Golgi apparatus—
independent pathway. When galectin-1 is secreted outside the cell, it can engage with carbohydrate ligands,
making the protein stable and mediate normal biological functions. Nevertheless, in the absence of ligands,
galectin-1 will undergo intramolecular oxidation and loss of dimerization and carbohydrate recognition ability.
IAM-mediated alkylation of galectin-1 is better than using other reducing agents such as 2-ME, DTT, and
reduced glutathione. By doing so, impairment of normal cell functions caused by reducing agents can be
avoided. When we remove the IAM using PD-10 column, the protein remains stable with normal carbohydrate
binding and biological activities. In conclusion, galectin-1 alkylation by IAM is a better way to study the protein
function in biological systems
excess iodoacetamide can be removed from the alkylated protein,

while galectin-1 retains a stable adduct that prevents intramolecular
disulfide bond formation and inactivation of the protein [24].
In this study, we will describe the process of galectin-1 alkyl-
ation with iodoacetamide, including a description of methods
required for mapping iodoacetamide incorporation (Fig. 3). In
addition, we will discuss approaches for examining the effect of
iodoacetamide alkylation on carbohydrate binding activity in addi-
tion to methods used to assess the sensitivity of the protein to
oxidation (Fig. 4). These methods should be of significant value
when seeking to understand the potential role of galectin-1 in a
variety of contexts while controlling for protein oxidation.
Fig. 2 Cysteine residues can be modified with various adducts to reduce intramolecular disulfide bond
formation. Dithiothreitol (DTT, a), β-mercaptoethanol (βME, b) and reduced glutathione (c) behave in a similar
way, forming reversible disulfides with free thiols that can reduce intramolecular disulfide bond formation. In
contrast, iodoacetamide (IAM, d) forms a stable thioether with cysteine that does not readily reverse upon
removal of excess iodoacetamide
2 Materials
2.1 Iodoacetamide 1. Recombinant galectin-1 (see Note 1).

Alkylation of Free 2. PD10 gel filtration column (GE Healthcare) (see Note 2).
Sulfhydryls on
3. Phosphate Buffered Saline (PBS), standard pH 7.4 (Hyclone).
Galectin-1
4. α-D-(+) Lactose, ACS grade (Fisher).
5. Iodoacetamide (IAM): 1 M stock freshly made in water
(Sigma-Aldrich, I1149).
6. Purified water (dH2O): The use of Barnstead/Millipore water
(deionized, UV, and carbon-filtered H2O) for all buffer pre-
parations is recommended.
Fig. 3 Incubation of Gal-1 with iodoacetamide results in acetamide

incorporation. (a, b) Mass spectrometry of galectin-1 (Gal-1) (a) or
iodoacetamide treated galectin-1 (iGal-1) (b). (c, d) Mass spectrometry of
tryptic fragments of Gal-1 or iGal-1. Peak 2002.47 corresponds to the tryptic
peptide fragment from Gal-1, while peak 2116.25 corresponds to same peptide
after reaction with iodoacetamide. This research was originally published in the
Journal of Biological Chemistry. Stowell SR, Cho M, Feasley CL, Arthur CM,
Song X, Colucci JK, Karmakar S, Mehta P, Dias-Baruffi M, McEver RP, Cummings
RD. Ligand reduces galectin-1 sensitivity to oxidative inactivation by enhancing
dimer formation. 2009 Feb 20;284(8):4989–99 © the American Society for
Biochemistry and Molecular Biology
2.2 Galectin 1. Lactosyl-Agarose column (Sigma Aldrich) (see Note 3).

Activity Assay 2. 2-Mercaptoethanol (2-ME) (Fisher) (see Note 4).
3. Affinity column loading buffer: Phosphate buffered saline with
2-mercaptoethanol (MEPBS), 0.01 M Na2HPO4, 0.85%
NaCl, pH 7.4 with 14 mM 2-ME.
4. Affinity column elution buffer: PBS with 100 mM lactose.
5. PD-10 gel filtration column (GE Healthcare) (see Note 2).
2.3 Proteolytic 1. Sequencing grade modified trypsin, 1 mg/mL in 50 mM acetic

Digestion of Alkylated acid, store at 80 C (Promega).
Galectin-1 2. C18 Sep-Pak: 100 mg cartridge (Waters).
Fig. 4 Alkylation protects Gal-1 from oxidative inactivation. Galectin-1 (Gal-1) or

iodoacetamide treated Gal-1 (iGal-1) were incubated for 24 h in PBS at 37 C
followed by subjection to affinity chromatography over lactosyl-Sepharose. This
research was originally published in the Journal of Biological Chemistry. Stowell
SR, Cho M, Feasley CL, Arthur CM, Song X, Colucci JK, Karmakar S, Mehta P,
Dias-Baruffi M, McEver RP, Cummings RD. Ligand reduces galectin-1 sensitivity
to oxidative inactivation by enhancing dimer formation. 2009 Feb 20;284
(8):4989–99 © the American Society for Biochemistry and Molecular Biology
3. C18 Sep-Pak pre-equilibration solution: HPLC grade metha-

nol, water, 50% ACN, 0.1% TFA in water.
2.4 HPLC Separation 1. Vydac analytical C18 column.

and Fraction Collection 2. HPLC and MS solvents: Purified, deionized H2O, HPLC
grade or better is recommended (see Note 5).
3. Acetonitrile (ACN), LC-MS grade.
4. Trifluoroacetic acid (TFA), sequencing grade 99.5% (Thermo
Scientific).
5. Formic acid, LC-MS grade, 99% (Thermo Scientific).
6. HPLC buffers (see Note 5):
HPLC buffer A: 94.9% dH2O, 5% ACN, 0.1% TFA. Add
50 mL ACN to 1 mL ampule TFA and dilute to 1 L with
dH2O.
HPLC buffer B: 94.9% ACN, 5% dH2O, 0.1% TFA.
2.5 Mass 1. α-Cyano-4-hydroxycinnamic acid (CHCA) matrix solution.

Spectrometry 2. MALDI target plate (Bruker).
Reagents and
3. Methanol: MS grade (Fisher, Optima grade or better).
Software
4. Acetic acid (Fisher).
5. Flow splitter (Upchurch).
6. Peptide standard mixture for external mass calibration: brady-

kinin, angiotensin I, angiotensin II, ACTH clip 18–39 (Ana-
spec; Fremont, CA) at 1 nmol/μL in 50% ACN, 0.1% formic
acid in dH2O.
7. α-Cyano-4-hydroxycinnaminic acid (CHCA) matrix solution,
recrystallized (see Note 6): Make fresh daily, 10 mg/mL in 50%
ACN, 0.1% TFA in dH2O (Sigma-Aldrich).
8. C18 column (Vydac).
9. LC-MS buffers (see Note 5):
LC-MS buffer A: 94.9% dH2O, 5% ACN, 0.1% formic acid.
LC-MS buffer B: 94.9% ACN, 5% dH2O, 0.1% formic acid.
2.5.1 Mass Spectrometry 10. GRAMS: PerSeptive (see Note 7).

Analysis Software 11. FlexAnalysis 2.4, BioTools 3.0: Bruker Daltonics.
12. ChemStation (Agilent Technology).
13. BioTools 3.0.
2.6 Special 1. Centrifugal vacuum concentrator, centrivap (Labconco) (see

Equipment Note 8).
2. Fast Protein Liquid Chromatography (FPLC) (see Note 9).
3. High-performance liquid chromatography (HPLC) (see
Note 10).
4. MALDI-TOF(/TOF)-MS: Voyager DE-STR (PerSeptive Bio-
systems) or Ultraflex II (Bruker Daltonics).
5. ESI-MS: Agilent 1100 MSD ion trap with quaternary pump,
diode array detector, and flow splitter into the ion trap (see
Note 11).
6. Syringe pump (Harvard Apparatus).
7. Fraction collector, such as BioRad 2110.
3 Methods
3.1 Iodoacetamide 1. Frozen galectin stocks should be stored in PBS with 100 mM
Alkylation of Free lactose and 14 mM 2-ME (or other appropriate reducing
Sulfhydryls on agent) (see Note 4). Remove stock from storage at 80 C
Galectin-1 and thaw on ice.
2. Equilibrate a PD-10 column by washing the column with
10 column volumes of PBS. Carefully load 2 mL of Gal-1
(2–5 mg/mL) over the PBS-equilibrated PD-10 column for
buffer exchange.
3. Collect 0.5 mL fractions from PD-10 column and determine
protein concentration using OD280 measurements. Pool
fractions with protein concentrations greater than 1 mg/mL

and add PBS such that the final galectin-1 concentration is
~2–5 mg/mL in PBS.
4. To the purified galectin (2–5 mg/mL) sample, add 10% v/v of
1 M IAM stock and incubate at 4 C overnight (see Note 12).
5. Remove excess iodoacetamide by applying the protein solution
to a PD-10 column pre-equilibrated with PBS. Collect frac-
tions containing Gal-1 from PD-10 column into PBS.
6. Verify Gal-1 activity after iodoacetamide treatment.
3.2 Galectin 1. Dilute recombinant galectin, both alkylated and non-alkylated

Activity Assay in PBS to the desired concentration (typically 1–20 μM) (see
Note 13).
2. Add 250 μL of galectin solution to a 48-well sterile tissue
culture plate.
3. Incubate this plate at the 37 C in a humidified chamber for
pre-established time intervals (see Note 14).
4. To determine residual activity, remove recombinant galectin
solution and apply galectin to 1 mL packed column of
lactosyl-agarose that has been equilibrated with PBS (see
Note 15).
5. Once the galectin solution has penetrated the column, add
5 column volumes of PBS to the column and collect 0.5 mL
fractions.
6. Add 100 mM lactose in PBS as an elution buffer and continue
to collect 0.5 mL fractions.
7. Determine the approximate protein concentration of each frac-
tion by measuring the OD280 nm. Evaluate the percent of
galectin in total lactose eluted fractions as a percent of the
total starting material (see Note 16).
3.3 Proteolytic 1. To iodoacetamide-treated Galectin-1 samples, add sequencing

Digestion of Alkylated grade trypsin (1 mg/mL stock) at a 1:50 enzyme:substrate
Galectin-1 (w/w) ratio and incubate digest at 37 C for 14 h.
2. To desalt peptides by C18 Sep-Pak, apply digested Galectin-1
peptide sample to a Sep-Pak cartridge pre-equilibrated with
3 mL sequential washes of C18 Sep-Pak pre-equilibration
solution.
3. Then elute unbound species from the Sep-Pak by washing with
3 1 mL of C18 Sep-Pak pre-equilibration solution.
4. Finally, elute peptides with 3 1 mL of 50% ACN, 0.1% TFA in
water.
5. Dry samples by rotary evaporation to 10–20 μL. Tryptic pep-
tides from this step can be analyzed directly (go to Subheading
3.6) or after further purification (continue to Subheading 3.4)

by LC-MS/MS.
3.4 HPLC Separation 1. Add an equal volume (10–20 μL) of HPLC buffer A to Galec-
and Fraction Collection tin-1 tryptic peptides.
2. Inject tryptic peptides onto the Vydac analytical C18 column
pre-equilibrated with HPLC buffer A.
3. Elute buffer salts and unbound peptides with a two column
volume of 98% buffer A/2% buffer B wash at 1 mL/min
flow rate.
4. Separate peptides using a 1 mL/min linear gradient from 2%
buffer B to 100% HPLC buffer B over 80 min, monitoring UV
absorption at 215 nm. Collect 1 min fractions.
5. Concentrate fractions by vacuum centrifugation to a uniform
volume of 100 μL.
3.5 MALDI-TOF and 1. To analyze HPLC separated peptide fractions in positive or

TOF/TOF Analysis negative ion mode, spot 1 μL of each peptide fraction on a
ground steel MALDI target plate (Bruker), add 1 μL of
α-cyano-4-hydroxycinnamic acid (CHCA) matrix solution,
and dry under vacuum.
2. In parallel, prepare a peptide standard spot with 1 μL of the
peptide calibration mixture with 1 μL of CHCA matrix
solution.
3. Analyze galectin peptides with a Ultraflex II MALDI-TOF-
TOF MS in reflectron positive ion mode with an accelerating
voltage of 23 kV. Acquire spectra with 66 Hz laser frequency
and sum 1000–2000 individual spectra for each sample.
4. Externally calibrate the TOF with the peptide calibration mix-
ture using the flexControl 2.4 software.
5. Galectin-1 peptides observed in the MS are accelerated by
8.5 kV and selected by a timed ion gate for MS/MS
fragmentation.
6. TOF/TOF fragmentation is achieved by accelerating the frag-
ment ions to 19 kV using the LIFT apparatus.
3.6 ESI-MS, Direct 1. Combine peptide fractions with an equal volume of methanol
Infusion with 1% acetic acid (100 μL) and directly infuse (5 μL/min)
into the Agilent 1100 LC/MSD instrument equipped with an
ion trap mass analyzer by a syringe pump.
2. Trap parameters are set as follows: Dry nitrogen was intro-
duced into the capillary region at a flow rate of 5 L/min and
heated to 220 C. Compound stability set to 35% and trap
drive level parameters was 70%. Nebulizer gas pressure was set
to 15 psi, and the spray voltage was adjusted to 3 kV.
3. Use helium for collision-induced dissociation.

4. MS and tandem MS (MS/MS) data were acquired and pro-
cessed using Bruker Trap software 4.1 in our lab.
3.7 LC-MS Analysis 1. Inject galectin-1 tryptic peptides (10–20 μL) on an Agilent
(See Note 17) 1100 HPLC-MSD-Trap with a 4.6 250 mm C18 column
pre-equilibrated with LC-MS buffer A.
2. Use a flow rate of 0.5 mL/min and a linear gradient 2% LC-MS
buffer B to 95% LC-MS buffer B over 95 min and split the flow
1:10 using flow splitter and direct 1/10 total column elution
into the MSD trap.
3. Trap parameters should be set same as in Subheading 3.6 with
the following exceptions: Dry nitrogen is introduced into the
capillary region at a flow rate of 8 L/min and heated to 300 C.
Compound stability set to 30%. Nebulizer gas pressure set to
25 psi.
4. MS/MS fragmentation should be data driven to select and
fragment the most abundant three ions.
3.8 Data Processing 1. Use flexAnalysis 2.1 software for MALDI-MS spectra proces-
and Analysis sing and peak picking.
2. Use ChemStation software to process ESI-MSD trap data.
3. Use BioTools 3.0 software for peptide MS/MS database
searching and de novo sequencing.
4 Notes
1. Purification of recombinant galectins is reported

elsewhere [25].
2. Alternatively, gel filtration or desalting columns can be utilized
(spin columns), but this desalting step should be rapid and
remove excess lactose used to affinity purify galectin.
3. Buffer exchange protein into low lactose containing buffers
before the application of cellular extracts or purified proteins.
4. 2-Mercaptoethanol or DTT can be utilized as a reducing agent;
however, both should be removed from galectin before use in
biological assays as both these thiols also possess the ability to
cross cell membranes where they can induce the unfolded
protein response with the endoplasmic reticulum [20–22]
likely secondary to preventing appropriate oxidation of pro-
teins during protein folding [21]. Thus, each may cause cell
damage if not removed prior to biological assays.
5. All FPLC and HPLC buffers should be vacuum filtered with

nylon filter (0.22 μm, 47 mm) and degassed with a helium
sparge or vacuum degassed with sonication.
6. Recrystallize CHCA matrix as follows: Mass 5 g CHCA into a
clean, dry beaker. Add 10 mL hot methanol and heat the
matrix/methanol solution. Add enough methanol (the mini-
mum amount required) to dissolve all of the CHCA. Add
100 mg of activated charcoal and filter the CHCA/methanol
solution. Cool the filtrate to room temperature, then cool to
4 C on ice. Vacuum filter CHCA crystals and wash with ice
cold water. Store at 20 C in a desiccated container.
7. Data processing and handling on mass spectrometers are
instrument and vendor dependent. Equivalent (or newer) soft-
ware package that are available may be utilized. Software tools
that de novo sequence peptides from MS/MS fragmentation
data can also be used for determining sites of iodoacetamide
incorporation.
8. We used Labconco centrivap containing a glass-lined cold trap
chilled to 50 C with an Edwards vacuum pump capable of
reaching 50 mTorr or equivalent.
9. Our lab used AKTA FPLC for protein purification and gel
filtration.
10. HPLC system capable of high-pressure pumping, flow rates
0.1–10 mL/min, with detector capable of monitoring at
215 nm or other UV-Vis wavelengths. We used a Beckman
System Gold with a UV-Vis detector system equipped with an
analytical C18 column (Vydac 4.6 250 mm) was used for
protein separations.
11. There are many additional systems available including nanoLC-
MS versions capable of mass spectrometric peptide mapping.
MSn studies are performed (if needed) by direct infusion of
peptide fractions collected from the HPLC.
12. Alkylation of free sulfhydryls can be achieved in shorter incu-
bation times at elevated temperatures. This may be determined
empirically.
13. This is typically done by examining galectin at a similar con-
centration as typically employed in biological assays, which can
range from 1 to 20 μM. As a control, wells containing 14 mM
βME or 3 mM DTT should also be included in separate plates.
14. This can range from examining galectin hours to days follow-
ing the initiation of the incubation period. There are many
methodological approaches to inducing protein oxidation.
We prefer to measure the potential impact of protein oxidation
under different conditions and following protein modification
by incubating galectins in a similar environment to which they
will be exposed during a typical incubation with cells as this will

be the commonly employed situation where galectin may sig-
nificantly impact the biological outcome of galectin–ligand
interactions, but where inclusion of reducing agents can obfus-
cate results.
15. It is possible, depending on the protein concentration
employed, that oxidative inactivation will result in protein
aggregate formation. If this is to occur, the final concentration
of soluble galectin should be noted and may be used to calcu-
late the overall loss of protein activity secondary to galectin
oxidation.
16. If determining galectin protein concentration by measurement
of OD, then be sure to correct concentration by extinction
coefficient. For short discussion of alternative methods of
assaying activity, see Note 4. Ensure that column is not
saturated such that active galectin is found in the flow through
simply because the column limit has been reached. This can be
achieved by examining for potential flow through in the reduc-
ing agent including control and/or by examining the satura-
tion limit prior to experimental setup.
17. Peptides analyzed by this method are described in an offline
LC-MS manner. There are numerous other high-resolution
LC-MS systems that can be substituted for the ones
described here.
Acknowledgments

Grants R01 HL154034 to CMA and U01 CA242109 to SRS.
References
1. Poland PA, Rondanino C, Kinlough CL, extracellular matrix and for a novel secretory
Heimburg-Molinaro J, Arthur CM, Stowell mechanism. J Cell Biol 110(5):1681–1691.
SR, Smith DF, Hughey RP (2011) Identifica- https://doi.org/10.1083/jcb.110.5.1681
tion and characterization of endogenous galec- 4. Arthur CM, Cummings RD, Stowell SR
tins expressed in Madin Darby canine kidney (2014) Using glycan microarrays to under-
cells. J Biol Chem 286(8):6780–6790. stand immunity. Curr Opin Chem Biol 18:
https://doi.org/10.1074/jbc.M110.179002 55–61. https://doi.org/10.1016/j.cbpa.
2. Dias-Baruffi M, Stowell SR, Song SC, Arthur 2013.12.017
CM, Cho M, Rodrigues LC, Montes MA, 5. Kamili NA, Arthur CM, Gerner-Smidt C,
Rossi MA, James JA, McEver RP, Cummings Tafesse E, Blenda A, Dias-Baruffi M, Stowell
RD (2009) Differential expression of immuno- SR (2016) Key regulators of galectin-glycan
modulatory galectin-1 in peripheral leukocytes interactions. Proteomics 16(24):3111–3125.
and adult tissues and its cytosolic organization https://doi.org/10.1002/pmic.201600116
in striated muscle. Glycobiology 6. Stowell SR, Cho M, Feasley CL, Arthur CM,
20(5):507–520 Song X, Colucci JK, Karmakar S, Mehta P,
3. Cooper DN, Barondes SH (1990) Evidence Dias-Baruffi M, McEver RP, Cummings RD
for export of a muscle lectin from cytosol to
(2009) Ligand reduces galectin-1 sensitivity to 16. Inagaki Y, Sohma Y, Horie H, Nozawa R,
oxidative inactivation by enhancing dimer for- Kadoya T (2000) Oxidized galectin-1 pro-
mation. J Biol Chem 284(8):4989–4999. motes axonal regeneration in peripheral nerves
https://doi.org/10.1074/jbc.M808925200 but does not possess lectin properties. Eur J
7. Cerri DG, Rodrigues LC, Stowell SR, Araujo Biochem 267(10):2955–2964
DD, Coelho MC, Oliveira SR, Bizario JC, 17. Cerliani JP, Stowell SR, Mascanfroni ID,
Cummings RD, Dias-Baruffi M, Costa MC Arthur CM, Cummings RD, Rabinovich GA
(2008) Degeneration of dystrophic or injured (2011) Expanding the universe of cytokines
skeletal muscles induces high expression of and pattern recognition receptors: galectins
Galectin-1. Glycobiology 18(11):842–850 and glycans in innate immunity. J Clin Immu-
8. Cho M, Cummings RD (1995) Galectin-1, a nol 31(1):10–21. https://doi.org/10.1007/
beta-galactoside-binding lectin in Chinese s10875-010-9494-2
hamster ovary cells. II. Localization and bio- 18. Toscano MA, Bianco GA, Ilarregui JM, Croci
synthesis. J Biol Chem 270(10):5207–5212 DO, Correale J, Hernandez JD, Zwirner NW,
9. Stowell SR, Arthur CM, McBride R, Berger O, Poirier F, Riley EM, Baum LG, Rabinovich GA
Razi N, Heimburg-Molinaro J, Rodrigues JP, (2007) Differential glycosylation of TH1, TH2
Noll AJ, von Gunten S, Smith DF, Knirel YA, and TH-17 effector cells selectively regulates
Paulson JC, Cummings RD (2014) Microbial susceptibility to cell death. Nat Immunol
glycan microarrays define key features of host- 8(8):825–834. https://doi.org/10.1038/
microbial interactions. Nat Chem Biol ni1482
10(6):470–476 19. Perillo NL, Pace KE, Seilhamer JJ, Baum LG
10. Stowell SR, Arthur CM, Dias-Baruffi M, (1995) Apoptosis of T cells mediated by
Rodrigues LC, Gourdine JP, Heimburg- galectin-1. Nature 378(6558):736–739.
Molinaro J, Ju T, Molinaro RJ, Rivera- https://doi.org/10.1038/378736a0
Marrero C, Xia B, Smith DF, Cummings RD 20. Tartier L, McCarey YL, Biaglow JE, Kochevar
(2010) Innate immune lectins kill bacteria IE, Held KD (2000) Apoptosis induced by
expressing blood group antigen. Nat Med dithiothreitol in HL-60 cells shows early acti-
16(3):295–301. https://doi.org/10.1038/ vation of caspase 3 and is independent of mito-
nm.2103 chondria. Cell Death Differ 7(10):1002–1010.
11. van Kooyk Y, Rabinovich GA (2008) Protein- https://doi.org/10.1038/sj.cdd.4400726
glycan interactions in the control of innate and 21. Braakman I, Helenius J, Helenius A (1992)
adaptive immune responses. Nat Immunol Manipulating disulfide bond formation and
9(6):593–601. https://doi.org/10.1038/ni. protein folding in the endoplasmic reticulum.
f.203 EMBO J 11(5):1717–1722
12. Tracey BM, Feizi T, Abbott WM, Carruthers 22. Stowell SR, Karmakar S, Stowell CJ, Dias-
RA, Green BN, Lawson AM (1992) Subunit Baruffi M, McEver RP, Cummings RD
molecular mass assignment of 14,654 Da to (2007) Human galectin-1, -2, and -4 induce
the soluble beta-galactoside-binding lectin surface exposure of phosphatidylserine in acti-
from bovine heart muscle and demonstration vated human neutrophils but not in activated T
of intramolecular disulfide bonding associated cells. Blood 109(1):219–227. https://doi.
with oxidative inactivation. J Biol Chem org/10.1182/blood-2006-03-007153
267(15):10342–10347 23. Go YM, Jones DP (2013) Thiol/disulfide
13. Hirabayashi J, Kasai K (1991) Effect of amino redox states in signaling and sensing. Crit
acid substitution by sited-directed mutagenesis Revi Biochem Mol Biol 48(2):173–181.
on the carbohydrate recognition and stability https://doi.org/10.3109/10409238.2013.
of human 14-kDa beta-galactoside-binding 764840
lectin. J Biol Chem 266(35):23648–23653 24. Clerch LB, Whitney P, Hass M, Brew K,
14. Cho M, Cummings RD (1995) Galectin-1, a Miller T, Werner R, Massaro D (1988)
beta-galactoside-binding lectin in Chinese Sequence of a full-length cDNA for rat lung
hamster ovary cells. I. Physical and chemical beta-galactoside-binding protein: primary and
characterization. J Biol Chem secondary structure of the lectin. Biochemistry
270(10):5198–5206 27(2):692–699. https://doi.org/10.1021/
15. Stowell SR, Qian Y, Karmakar S, Koyama NS, bi00402a030
Dias-Baruffi M, Leffler H, McEver RP, Cum- 25. Wu SC, Paul A, Ho A, Patel KR, Allen JWL,
mings RD (2008) Differential roles of galectin- Verkerke H, Arthur CM, Stowell SR (2021)
1 and galectin-3 in regulating leukocyte viabil- Generation and use of recombinant galectins.
ity and cytokine secretion. J Immunol Curr Protoc 1(3):e63. https://doi.org/10.
180(5):3091–3102 1002/cpz1.63
Chapter 5
Rapid Detection and Purification of Galectin-3

by the Capture and Release (CaRe) Method
Priyanka D. Kadav, Jared L. Edwards, Jessica Krycia,
Abstract
Specific interactions between lectins and glycoproteins determine the outcomes of numerous biological
processes. To elucidate the roles of lectins and glycoproteins in those processes, it is essential to detect these
proteins in biological samples and purify them to homogeneity. Conventional protein detection and
purification techniques are multi-step, time-intensive, and expensive. They often require rigorous trial
and error experimentations and fairly larger volumes of crude extracts. To minimize some of these
challenges, we recently formulated a new method named Capture and Release (CaRe). This method is
rapid, facile, precise, and inexpensive, and it works even when the sample volume is smaller. We developed
this method to detect and purify recombinant human Galectin-3 and subsequently validated this method by
purifying several other lectins. Besides lectins, CaRe is capable of detecting/purifying glycoproteins. In this
method, targets (lectins and glycoproteins) are captured by multivalent ligands called target capturing
agents (TCAs). The captured targets are then released and separated from their TCAs to obtain purified
targets. CaRe can potentially be used as a tool to discover new lectins and glycoconjugates and elucidate
their functions.
Key words Galectin-3, Lectin and glycoprotein purification, Affinity chromatography, Non-covalent
cross-linking, Method development
1 Introduction
Interactions of glycoconjugates [namely glycoproteins (GPs) and

proteoglycans (PGs)] with glycan-binding proteins (lectins and
glycosaminoglycan-binding proteins, GAGBPs) are crucial for a
variety of biological processes [1–5]. Lectins interact with the
glycan chains of GPs, while glycosaminoglycan chains of PGs
serve as the binding epitopes of GAGBPs [6]. To understand the
functional consequences of GP-lectin and PG-GAGBP interac-
tions, these proteins must be identified in biological samples and
then those proteins, in native forms or their recombinant versions,
should be purified to homogeneity. Thus, protein identification and
89
90 Priyanka D. Kadav et al.
purification techniques play indispensable roles in protein discovery

and the elucidation of their structure–function relationship.
Affinity column chromatography [7], size exclusion chroma-
tography [8], and ion exchange chromatography [9] are the most
widely used techniques for protein (including lectins and glycopro-
teins) purification. Purification methods based on column chroma-
tography have their own limitations. They are multi-step, time
intensive, and expensive. They often need extensive trial and error
experimentations, elaborate infrastructure and the requirement of
larger sample volume. Other problems associated with these tech-
niques include over-dilution of the purified proteins and nonspe-
cific binding.
Immunoprecipitation (IP) [10–12] is an extensively used
method that is employed to detect/purify proteins from limited
sample volumes. The use of IP for lectin and glycoprotein detec-
tion/purification remains limited. IP heavily depends on the prior
knowledge of the target protein(s). Therefore, it is not exactly
suitable for unknown lectins and glycoproteins. In addition, IP
requires expensive antibodies. Nonspecific binding, dissociation
under extreme conditions, and antibody contamination often cre-
ate problems in IP. In IP, antibodies are attached directly to agarose
beads or superparamagnetic microbeads or to protein A/G
attached to the beads [13]. Therefore, the beads and Protein
A/G can nonspecifically interact with other entities present in the
crude lysates. The types of beads and the amount of antibodies
covalently linked to the beads significantly control the outcome
of IP.
We recently developed a purification/detection method to
overcome some of the challenges described above. The method,
named as “Capture and Release” (CaRe), was originally designed
to detect/purify recombinant human galectin-3 (Gal-3) and was
subsequently validated by purifying other lectins [13]. The method
was also found capable of purifying/detecting glycoproteins
[13]. In CaRe (Fig. 1), a target (a lectin or a glycoprotein) in a
crude mixture (tissue or cell extracts) is captured by a multivalent
target-capturing agent (TCA) in solution phase. The TCA specifi-
cally binds to the target and spontaneously forms insoluble cross-
linked complexes. The target-TCA complex is then isolated from
other components of the crude mixture. The complex is then
dissolved by a monovalent ligand to release the captured target.
Once released, the target is separated from the TCA by membrane
filtration or gel filtration. CaRe is a relatively fast and facile method
that does not need solid affinity matrices, antibodies, beads, col-
umns, elaborate arrangements, harsh experimental conditions,
specialized detectors, controlled environments, or functionaliza-
tion of the reagents. This method can detect/purify lectins and
glycoproteins from a limited volume of crude extracts by using
minimal space and infrastructural support. CaRe can be done in a
Rapid Detection and Purification of Galectiny the Capture. . . 91
Fig. 1 Different steps of the Capture and Release (CaRe) method. 1, Crude extract containing the target (lectin
or glycoprotein). 2, Extracts mixed with a target capturing agent (TCA). 3, TCA binds and cross-links the target
and forms insoluble complexes (precipitates). 4, Isolation of the precipitates by centrifugation. 5, Addition of
monovalent ligand (ML) to the complex. 6, ML dissolves the complex and releases the target. 7, Separation of
the released target from the TCA by filtration
few microfuge tubes. Detection, purification, and sequence identi-

fication of an unknown lectin/glycoprotein could be done in a
single day. This method provides a better and faster option to
detect and purify Gal-3 and has the potential to accelerate the
discovery of other lectins and glycoconjugates.
This chapter contains CaRe protocols for purifying a lectin
(Gal-3) and a glycoprotein (chicken ovalbumin).
2 Materials
2.1 Extraction of 1. E. coli (BL21 DE3) transformed with a plasmid expressing

Recombinant Gal-3 human Gal-3.
from E. coli Cells 2. Luria Bertani media (Fisher Scientific, Millipore Sigma).
3. Protease Inhibitor cocktail mixture (Millipore Sigma).
4. Microfuge tubes.
5. 20 mM Phosphate-buffered saline (PBS), pH 7.4.
6. 22 mM Tris–HCl buffer containing 5 mM EDTA, pH 7.5.
7. 50- and 15-mL centrifuge tubes.
8. Magnetic stir bars.
9. Magnetic stir plate.
10. Standard analytical balance.

11. Sonicator (Fisher Scientific).
12. Blender (Oster Duralast Classic).
13. Centrifuge (Thermo Scientific Sorvall Legend X1R).
2.2 Preparation of 1. Blender (Oster Duralast Classic).

Crude Extracts from 2. Centrifuge (Thermo Scientific Sorvall Legend X1R).
Chicken Egg Whites
3. Chicken eggs.
(Source of Ovalbumin)
4. 100 mM HEPES buffer (pH 7.4). pH of 100 mM HEPES was
adjusted with NaOH solution.
5. 100 mM Sodium acetate (pH 5.2) pH of 100 mM sodium
acetate was adjusted with acetic acid.
2.3 Identification of 1. Multivalent Ligands [chondroitin sulfate A (CSA), chondroitin

Multivalent Inhibitors sulfate C (CSC), bovine thyroglobulin (Tg), ConA, all from
as Target Capturing Millipore Sigma].
Agents (TCAs) by 2. 2% v/v rabbit RBC suspension. Rabbit blood was collected
Hemagglutination from Michigan Tech Animal Care Facility.
Inhibition 3. PBS (20 mM sodium phosphate dibasic and sodium phosphate
monobasic with 150 mM NaCl, pH 7.4).
4. 100 mM HEPES, 100 mM HEPES with 5 mM CaCl2, 5 mM
MnCl2, and 150 mM NaCl. pH adjusted with 2 M NaOH
solution.
5. 96-well U-bottom Nunc brand microtiter plates.
6. Micropipette.
7. Micropipette tips.
2.4 Testing the 1. Multivalent ligands that showed micromolar to nanomolar

Capturing Abilities of affinity in the assay described above.
TCAs by Precipitation 2. Purified Concanavalin A (ConA) (Millipore-Sigma, Cat #
Assays L7647).
3. Crude extract of E. coli containing recombinant human Gal-3
(described under Subheading 3.1).
4. Crude extract of egg white.
5. 20 mM PBS buffer [(20 mM sodium phosphate dibasic and
sodium phosphate monobasic with 150 mM NaCl, pH 7.4) or
100 mM HEPES buffer [100 mM HEPES with 5 mM CaCl2,
5 mM MnCl2, and 150 mM NaCl, pH adjusted with 2 M
NaOH].
6. Micropipettes.
8. Quartz cuvettes.
9. Shimadzu UV-Vis spectrophotometer.

10. 1.5-mL microcentrifuge tubes.
2.5 Capturing of 1. Crude extracts of E. coli expressing recombinant Gal-3 and egg
Targets (Lectins and whites (stored at 4 C).
GPs) in the Crude 2. TCA solutions (solutions of chondroitin sulfate A (CSA),
Extracts by TCAs and chondroitin sulfate C (CSC), bovine thyroglobulin (Tg), and
Separation of the ConA, all from Millipore Sigma).
Target-TCA Complex 3. 20 mM PBS buffer (20 mM sodium phosphate dibasic and
from Other sodium phosphate monobasic with 150 mM NaCl, pH 7.4)
Components of the or 100 mM HEPES buffer (100 mM HEPES with 5 mM
Crude Extracts CaCl2, 5 mM MnCl2, and 150 mM NaCl, pH adjusted with
2 M NaOH). If any lectin requires metal ions, HEPES is used.
Otherwise, PBS is used (see Note 1).
4. Micropipettes.
2.6 Releasing the 1. Precipitated complexes from previous section.

Captured Targets 2. Monovalent saccharides (Lactose for Gal-3, mannose for
(Lectins and GPs) by ConA) (see Note 2).
Dissolving the
3. 20 mM PBS buffer (20 mM Sodium phosphate dibasic and
Complex sodium phosphate monobasic with 150 mM NaCl, pH 7.4) or
100 mM HEPES buffer (100 mM HEPES, 100 mM HEPES
with 5 mM CaCl2, 5 mM MnCl2, and 150 mM NaCl. pH 7.4
adjusted with 2 M NaOH solution).
4. 100 mM sodium acetate buffer without ions (100 mM sodium
acetate with 5 mM CaCl2, 5 mM MnCl2, and 150 mM NaCl.
pH 5.2).
5. Micropipettes.
2.7 Separation of the 1. Dissociated target-TCA complexes.

Targets (Lectins and 2. Amicon membrane filtration tubes MWCO: 100, 50, 30,
Glycoproteins) from 10 kDa.
Their Respective TCAs
3. 20 mM PBS buffer.
4. 100 mM sodium acetate buffer (pH 5.2) (100 mM sodium
acetate with 5 mM CaCl2, 5 mM MnCl2, and 150 mM NaCl.
pH 5.2).
5. Glass test tubes from Fisher Scientific.
6. Micropipette.
9. Quartz cuvettes.
10. Shimadzu UV-Vis spectrophotometer.
2.8 Verification of 1. 10 SDS-PAGE running buffer (Tris–base, glycine, and SDS,
the Purity of the pH 8.3).
Isolated Targets 2. SDS-PAGE sample buffer (Tris–HCl, pH 6.8 with 50% glyc-
(Lectins or GPs) erol, 1.0% bromophenol blue and 10% SDS).
3. SDS-PAGE staining solution (to 175 mL ultrapure water add
50 mL methanol and 25 mL acetic acid. Dissolve 0.625 g
Coomassie brilliant blue in that solvent).
4. SDS-PAGE de-staining solution (to 375 mL ultrapure water
add 75 mL methanol then and 50 mL acetic acid).
5. TGX Stain-free gels 12% (Bio-Rad, Catalog #4568043).
6. Broad range prestained protein standard (New England Bio-
labs, Catalog #P7719S).
7. 0.6-mL microcentrifuge tubes.
8. Micropipettes.
10. 100-mL beaker.
11. Erlenmeyer flask.
12. Glass pipettes.
13. Gel loading tips.
14. Mini Protein Tetra cell (Bio-Rad).
15. PowerPac Power Supply (Bio-Rad).
16. Microcentrifuge compatible heater (Themo-Fisher).
3 Methods
3.1 Extraction of 1. Centrifuge E. coli cells that contain expressed recombinant

Recombinant Gal-3 Gal-3 [13] at 1957 g for 30 min at 4 C.
from E. coli Cells 2. Dislodge cell pellet (8–10 mL) into a 50-mL Falcon tube with
22 mM Tris–HCl buffer containing 5 mM EDTA (30 mL of
buffer) and 300 μL of Protease cocktail inhibitor. Limit the
volume of buffer to prevent dilution of Gal-3.
3. Sonicate the cell suspension in the buffer. Split the cell suspen-
sion in multiple tubes to limit the volume in each tube so that
sample is not lost due to splashing during sonication. The
sonication process is 50% amplitude, 30 s under sonication,
60 s on ice after each round. Eight cycles of sonications should
take place to ensure complete release of all Gal-3.
4. Centrifuge the sonicated sample at 1957 g for 30 min at

4 C. Test the Gal-3 activity from supernatant by hemaggluti-
nation assay.
5. After confirmation, centrifuge the supernatant again at
9469 g for 30 min at 4 C.
6. Use this supernatant to purify Gal-3 by CaRe.
3.2 Preparation of 1. Place a 500-mL beaker on the lab bench.

Crude Extracts from 2. Carefully separate the chicken egg yolks from six egg whites
Chicken Egg Whites while placing the egg whites into the beaker.
(Source of Ovalbumin)
3. Decant the egg whites into a blender and add 50 mL of HEPES
buffer.
4. Blend the mixture for 30 s three times (90 s total).
5. Decant the mixture into a beaker and add a stir bar.
6. Stir the mixture on a stir plate for 2 h at 4 C.
7. After 2 h, decant the mixture into 50-mL centrifuge tubes.
8. Spin the mixture at 1957 g for 30 min at 4 C. This speed
gives best results.
9. Spin the supernatant again at 9469 g for 30 min at 4 C. This
speed gives best results.
10. Pool the supernatant together. Discard the precipitate.
11. Use the supernatant for CaRe.
3.3 Identification of 1. Adjust the titer of the target lectin you are trying to test to 2–3
Multivalent Inhibitors agglutination titers (meaning the target lectin sample should
as Target Capturing agglutinate up to 2–3 wells. Titer is adjusted by diluting or
Agents (TCAs) by concentrating the initial test samples. This is important because
Hemagglutination a nonoptimal titer gives misleading results).
Inhibition 2. Pipette 25 μL of PBS or HEPES buffer into 8 wells of the
96-well round bottom plate using a micropipette.
3. Pipette 25 μL of PBS or HEPES buffer into 2 additional wells
of the 96-well round bottom plate using a micropipette.
4. Serial dilute (twofold serial dilution) 25 μL of the ligand into
seven of the eight wells.
5. Pipette 25 μL of the ligand into the eighth well.
6. Pipette 25 μL of the lectin crude into the seven wells that
contains the serial diluted ligand. Do NOT pipette your lectin
into the eighth well that contains your ligand control.
7. Pipette 25 μL of the lectin into the one of the two additional
wells that contain PBS or HEPES buffer. This will act as the
lectin control.
8. Gently shake the plate and let it sit at room temperature for
30–60 min.
9. Pipette 25 μL of the 2% v/v RBC suspension into all 10 wells.
10. Pipette 25 μL of the 2% v/v RBC suspension in a well contain-
ing only 25 μL of buffer (this well is the RBC and buffer
control).
30–45 min.
12. Visualize and note the results. The multivalent ligands that
show inhibitory activities are selected for the precipitation
experiments. Because those ligands show measurable precipita-
tion activities. Non-inhibitory ligands do not form precipitates
with the targets.
3.4 Testing the 1. Change the absorbance on the spectrophotometer to 420 nm.
Capturing Abilities of 2. Pipette 1 mL of buffer into the reference and sample cuvettes.
TCAs by Precipitation
3. Autozero the spectrophotometer.
Assays
3.4.1 Preparing the
Spectrophotometer
3.4.2 Precipitation of 1. Perform “Preparing the Spectrophotometer” before proceed-

Target Lectin (Gal-3) from ing with step 2.
the Crude Solution 2. Remove the sample cuvette from the spectrophotometer and
decant the buffer.
3. Add 1 mL of the lectin crude with known hemagglutination
titer to the sample cuvette.
4. Place the cuvette into the spectrophotometer.
5. Record the absorbance.
7. Remove the sample cuvette from the spectrophotometer.
8. Pipette 5 μL of the multivalent ligand (stock solution with
known concentration, 100 μM to 5 mM) into the cuvette once.
9. Invert the cuvette once by capping the cuvette with a cover.
10. Use a Kimwipe to wipe the cuvette and place it back into the
spectrophotometer.
12. Continue steps 5–8 until the curve plateaus and slightly
decreases. Make a note of the amount of ligand added where
it plateaus before it decreases.
13. Decant the mixture in the sample cuvette into a 1.5-mL
microcentrifuge tube.
14. Incubate the mixture at room temperature for 1–12 h (see

Note 3).
15. Determine the stoichiometric ratio for the target lectin and
multivalent ligand. The concentration value of a ligand that
produces the highest OD point with a given crude solution on
a precipitation curve is noted. That concentration value of the
ligand with regard to the given crude solution is the stoichio-
metric ratio.
3.4.3 Precipitation of 1. Perform “Preparing the Spectrophotometer” before

Target Glycoprotein proceeding.
(Chicken Ovalbumin) from 2. Remove the sample cuvette from the spectrophotometer and
the Crude Solution decant the buffer.
3. Add 1 mL of 100 μM of purified ConA to the sample cuvette.
4. Place the cuvette into the spectrophotometer.
7. Remove the sample cuvette from the spectrophotometer.
8. Pipette 5 μL of the glycoprotein crude into the cuvette once.
9. Invert the cuvette once, after putting a lid on the cuvette.
10. Use a Kimwipe to wipe the cuvette and place it back into the
spectrophotometer.
12. Continue steps 8–11 until the curve plateaus and slightly
decreases. Make a note of the amount of glycoprotein crude
added where it plateaus before it decreases.
13. Decant the mixture in the sample cuvette into a 1.5-mL
14. Incubate the mixture at room temperature for 1–12 h (incuba-
tion time varies based on the samples. It is decided by checking
the OD at different time points).
3.5 Capturing of 1. Capture the target lectin with the proper capturing agent
Targets (Lectins and (TCA). For capturing the targeted GP from the crude, use
GPs) in the Crude purified ConA as the TCA (see Note 2).
Extracts by TCAs and 2. Incubate the mixture for 1–12 h at 4 C (incubation time varies
Separation of the based on the samples. It is decided by checking the OD at
Target-TCA Complex different time points).
from Other 3. Centrifuge the mixture at 9469 g for 30 min at 4 C. This
Components of the g force is optimal for this purpose.
Crude Extracts
4. Discard the supernatant.
3.5.1 Capturing the
Targets (Lectins or GPs)
(Fig. 2)
Fig. 2 Capturing of targets by TCAs and visualization of insoluble target-TCA

complex. An example is shown with Gal-3 as a target and chondroitin sulfate A
(CSA) as the TCA. CSA binds and cross-link Gal-3 and forms insoluble complex.
The solution becomes turbid due to the formation of target-TCA complex
3.5.2 Washing the 1. Pipette 1 mL of cold buffer (the same buffer, the crude sample,
Insoluble Precipitate and the TCA were in) along the wall of the microcentrifuge
tube with the precipitate.
2. Centrifuge the precipitate with the buffer at 6339 g for 5 min
at 4 C. This g force is optimal for this purpose.
3. Discard the buffer (without disturbing the cell pellet).
4. Repeat steps 1–3 two more times.
5. Store at 4 C until the next step.
3.6 Releasing the 1. Make a high concentration (200–400 mM) of monovalent

Captured Targets saccharides in PBS.
(Lectins and GPs) by 2. Remove the recently precipitated complexes from 4 C (from
Dissolving the see Subheading 3.5 above).
Complex 3. Pipette 1 mL of the monovalent saccharide solution along the
wall of the microcentrifuge tube containing the complex.
4. Incubate the mixture at 4 C for 4–18 h until the complex
dissolves and the opaque solution becomes gradually clear
(checked visually, can also be checked by measuring OD at
420 nm, turbidity formed by precipitation best detected at
420 nm). For most protein tested, one can vortex the mixture
to dissolve the precipitate quickly (see Note 4).
5. If the complex does not dissolve, perform the following steps:
(a) Centrifuge the mixture at 9469 g for 15 min at 4 C.
This g force was found out to be optimum for this
purpose.
(b) Pipette 1 mL of cold buffer (PBS, HEPES or sodium
acetate, based on the target protein) along the wall of
the microcentrifuge tube holding the precipitate.
(c) Centrifuge the precipitate with the buffer at 6339 g for

5 min at 4 C.
(d) Discard the supernatant.
(e) Pipette 1 mL of 100 mM sodium acetate buffer (pH 4.9)
into the microcentrifuge tube.
(f) Incubate the solution for 4–18 h. For some proteins,
precipitation happens relatively quickly, for others, the
process is much slower. The exact incubation time for
each protein is determined by checking precipitation of
each protein at different time intervals.
(g) Vortex the solution if some precipitate is still observed.
3.7 Separation of the 1. Wash the membrane filtration tubes per the manufacturer’s
Targets (Lectins and instructions.
Glycoproteins) from 2. Pipette the dissociated complex onto the 100 or 50 kDa molec-
Their Respective TCAs ular weight cutoff membrane filtration tube (see Note 5).
3.7.1 Separating the 3. Centrifuge the mixtures at 5000 g for 10 min at 4 C.
Galectin-3-TCA Complexes 4. Remove the filter from the centrifuge.
5. Pipette the bottom fraction onto a 10 kDa membrane
filtration tube.
6. Centrifuge the fraction at 5000 g for 10 min at 4 C.
7. Remove the filter from the centrifuge.
8. Pipette 5 mL of PBS buffer onto the top of the filter.
9. Centrifuge the membrane filtration tube until the final volume
on the top of the filter is approximately 1 mL.
10. Use this fraction to check the purity of the lectin by SDS-PAGE
(see Subheading 3.8).
11. The sample at the bottom of the filter is also checked by
SDS-PAGE. Gal-3 will stay at the top of the 10 kDa filter.
Sample at the bottom of that filter is also checked by
SDS-PAGE to see the presence of any proteins or leakage of
Gal-3 through the filter.
3.7.2 Separating the 1. Wash the membrane filtration tubes per the manufacturer’s
Ovalbumin-ConA instructions.
Complexes 2. Pipette the dissociated complex onto the 100 kDa membrane
filtration tube.
3. Centrifuge the mixture at 5000 g for 10 min at 4 C.
filtration tube.

filtration tube.
11. Pipette 5 mL of PBS buffer onto the top of the filter.
12. Centrifuge the membrane filtration tube until the final volume
is approximately 1 mL.
13. Use this fraction to check the purity of the lectin using
SDS-PAGE (see Subheading 3.8).
3.8 Verification of 1. Remove SDS-PAGE running buffer from 4 C.

the Purity of the 2. Turn the heater on and set the temperature to 95 C.
Isolated Targets
3. Make a 1 solution of the SDS-PAGE running buffer.
(Lectins or GPs)
4. Decant 50 mL of the 1 running buffer into a 100-mL beaker.
5. Remove the TGX stain-free gels from 4 C and wash them with
the 100 mL of 1 running buffer according to the manufac-
turer’s instructions.
6. Set up the Mini-Protein Tetra cell according to the manufac-
7. Remove the protein standard from 20 C and bring it to
room temperature.
8. Pipette 950 μL of sample buffer into a 1.5-mL
9. Pipette 50 μL of β-mercaptoethanol into 950 μL sample buffer.
Mix the solution thoroughly by pipetting up and down or
vortex.
10. Pipette 20 μL of each sample into the 0.6 mL microcentrifuge
tubes. Include commercially available purified samples as refer-
ences (see Note 6).
11. Pipette 40 μL of the sample buffer into each sample tube. Mix
thoroughly by pipetting up and down.
12. Place the tube in the heater set for 95 C.
13. Let the tubes incubate at 95 C for 5 min.
14. Remove the tubes from the heater.
15. Pipette 10 μL of the protein standard into one of the wells on
the gel.
16. Pipette 20 μL of each sample into the wells of the gel.
17. Run the gel at 150 V until finished (the dye reaches the bottom
of the gel).
Fig. 3 Confirmation of the purity of the separated target by SDS-PAGE. An

example of purification by CaRe is shown by using a Gal-3 mutant (a single
amino acid substitution at the binding site of Gal-3). Lane 1, molecular eight
markers, lane 2, E. coli lysate containing the recombinant Gal-3 mutant, lane
3, Gal-3 mutant purified by CaRe
18. Stain the gel for 2 h using enough SDS-PAGE staining solution
to cover the gel.
19. Discard the staining solution after staining.
20. Add the de-staining solution to cover the gel and let the gel
de-stain until the desired background is achieved.
21. Note the results (Fig. 3) (see Note 7).
4 Notes
1. Activity of certain lectins (such as ConA) is dependent on

calcium and manganese ions, while some other lectins includ-
ing Galectin-3 do not require metal ions to be active. There-
fore, the buffer must contain those ions while working with
ConA. HEPES buffer with required metal ions is used for this
purpose. For Galectin-3 and other metal ion-independent lec-
tins, PBS is suitable.
2. To release the targets from the capturing agents, specific mono-
valent saccharides are used: mannose for mannose-binding
lectins and lactose for galactose/GalNAc/lacose/LacNAc-

binding lectins.
3. While capturing targets (lectins or glycoproteins), incubation
time with the target capturing agents (TCAs) varies based on
the samples. Incubation time is decided by checking the OD at
different time points in preliminary experiments. In certain
cases, placing the TCAs in cuvette and adding crudes (contain-
ing the targets) to them yields more precipitates.
4. The ability of monovalent ligands to dissolve target-TCA com-
plexes depends on the constituents of the complexes. Some
complexes get dissolved quickly (in minutes) when treated with
the monovalent ligands. Others need a longer incubation time.
To ensure proper dissociation of targets and TCAs, overnight
incubation is often performed. Dissolution can be checked
visually or by measuring OD at 420 nm. Formation and disso-
lution of target-TCA complexes are best detected at 420 nm.
For most protein/glycoprotein tested, vortexing the mixture
helped dissolve the precipitates quickly. If Galectin-3 is a con-
stituent of a target-TCA complex, it should not be vortexed.
Galectin-3 tends to come out of solution when stirred
vigorously.
5. Molecular weight cutoff of the filters is selected based on the
masses of the multivalent ligands (TCAs) and their relative
differences from those of the targets. For example, the molec-
ular mass of bovine thyroglobulin (Tg), the TCA of Gal-3, is
670 kDa, whereas the molecular weight of the target (Gal-3) is
29 kDa (monomeric) and 150 kDa (oligomeric).
6. In the current study, commercially available purified samples of
ovalbumin was used as a reference to confirm the purification of
ovalbumin. When the targets are lectins and glycoproteins,
their purified forms can be purchased from vendors and be
used as references.
7. Known targets are identified by their molecular masses. The
identity of unknown targets is established by mass
spectrometry.
Acknowledgments
This work was supported by National Science Foundation Grant

1608537 (to T.K.D.). A part of this work was supported by a
startup fund (to P.B.) and by the Research Excellence Fund (to T.
K.D.) provided by Michigan Technological University.
References
1. Stanley P, Taniguchi N, Aebi M (2017) 7. Arora S, Saxena V, Ayyar BV (2017) Affinity
N-glycans. In: Varki A et al (eds) Essentials of chromatography: a versatile technique for anti-
glycobiology, 3rd edn. Cold Spring Harbor, body purification. Methods 116:84–94
New York 8. Burgess RR (2018) A brief practical review of
2. Brockhausen I, Stanley P (2017) O-GalNAc size exclusion chromatography: rules of
glycans. In: Varki A et al (eds) Essentials of thumb, limitations, and troubleshooting. Pro-
glycobiology, 3rd edn. Cold Spring Harbor, tein Expr Purif 150:81–85
New York 9. Cummins PM, Rochfort KD, O’Connor BF
3. Lindahl U, Couchman J, Kimata K et al (2017) (2017) Ion-exchange chromatography: basic
Proteoglycans and sulfated principles and application. Methods Mol Biol
glycosaminoglycans. In: Varki A et al (eds) 1485:209–223
Essentials of glycobiology, 3rd edn. Cold 10. Speth C, Toledo-Filho LA, Laubinger S (2014)
Spring Harbor, New York Immunoprecipitation-based analysis of
4. Varki A (2017) Biological roles of glycans. Gly- protein-protein interactions. Methods Mol
cobiology 27:3–49 Biol 1158:175–185
5. Taylor ME, Drickamer K, Schnaar RL et al 11. Lin JS, Lai EM (2017) Protein-protein inter-
(2017) Discovery and classification of glycan- actions: co-immunoprecipitation. Methods
binding proteins. In: Varki A et al (eds) Essen- Mol Biol 1615:211–219
tials of glycobiology, 3rd edn. Cold Spring 12. Kaboord B, Perr M (2008) Isolation of pro-
Harbor, New York teins and protein complexes by immunoprecip-
6. Esko JD, Prestegard JH, Linhardt RJ (2017) itation. Methods Mol Biol 424:349–364
Proteins that bind sulfated 13. Welch CJ, Talaga ML, Kadav PD et al (2020) A
glycosaminoglycans. In: Varki A et al (eds) capture and release method based on noncova-
Essentials of glycobiology, 3rd edn. Cold lent ligand cross-linking and facile filtration for
Spring Harbor, New York purification of lectins and glycoproteins. J Biol
Chem 295:223–236
Chapter 6
Introducing 77Se NMR Spectroscopy to Analyzing

Galectin–Ligand Interaction
Mária Raics, István Timári, László Szilágyi, Hans-Joachim Gabius,
and Katalin E. Kövér
Abstract
Their emerging nature as multifunctional effectors explains the large interest to monitor glycan binding to
galectins and to define bound-state conformer(s) of their ligands in solution. Basically, NMR spectroscopy
facilitates respective experiments. Towards developing new and even better approaches for these purposes,
extending the range of exploitable isotopes beyond 1H, 13C, and 15N offers promising perspectives. Having
therefore prepared selenodigalactoside and revealed its bioactivity as galectin ligand, monitoring of its
binding by 77Se NMR spectroscopy at a practical level becomes possible by setting up a 2D 1H, 77Se
CPMG-HSQBMC experiment including CPMG-INEPT long-range transfer. This first step into applying
77
Se as sensor for galectin binding substantiates its potential for screening relative to inhibitory potencies in
compound mixtures and for achieving sophisticated epitope mapping. The documented strategic combi-
nation of synthetic carbohydrate chemistry and NMR spectroscopy prompts to envision to work with
isotopically pure 77Se-containing β-galactosides and to build on the gained experience with 77Se by adding
19
F as second sensor in doubly labeled glycosides.
Key words Galectin, HSQMBC, Selenoglycoside, 77Se NMR, Thiodigalactoside
1 Introduction
A prerequisite for phylogenesis to reach the level of cellular com-

plexity of mammals is the establishment of sophisticated means of
molecular communication. Fittingly, respective analysis has discov-
ered the occurrence of several biochemical languages, each one
apparently serving distinct purposes. Nucleic acids, to turn to
heredity, are ideal to copy genetic information in the germline
with minimal error rates. Building blocks like nucleotides or
amino acids, the equivalents of “letters,” are connected to form
oligo- and polymers, the molecular messages. Compared to nucleic
acids and proteins, the products of the first and second alphabets of
life, glycans of the ubiquitous glycoconjugates stand apart: they
“are much more complex, variegated, and difficult to study”
105
106 Mária Raics et al.
[1]. However, this obvious problem for analytical work has its
inherent advantage for cellular communication: their structural
complexity (oligo- and polysaccharides have therefore formerly
been called “complex carbohydrates”) equips cells with a large
diversity of glycan-encoded signals in a minimum of space, e.g.,
on the cell surface. Clearly, the occurrence of a highly organized
enzymatic machinery for glycosylation will be required toward this
end, and this assumption has been confirmed entirely [2–4]. The
growing realization of “how deeply glycan functions pervade all
aspects of organismic biology” [5] attests that the exceptional
talents of carbohydrates to generate molecular “words” of unsur-
passed coding capacity are indeed put to use physiologically, and
widely so [6–8]. As in the genetic code, the flow of information
from code word to bio-effect involves intermolecular recognition.
Translating these glycan-based signals into cellular activities are
sugar receptors with antibody-like capacity for specificity. This
property is the origin of the term “lectin” (“from Latin lectus, the
past principle of legere meaning to pick, choose or select” [9]) [10–
12]. It also accounts for the term “galectin”: these homologous
proteins share a sequence signature and bind to a galactose moiety
by involving a central Trp residue, resulting in the common C–H/π
interaction. This led to the abbreviation of ga(lactose-binding)
lectin as name for the members of the respective family, now
known to be broad-range effectors in cellular homeostasis and
pathophysiology [13–18]. Analysis by crystallography [19, 20],
molecular modeling [21], and binding assays, for example, using
glycan arrays or frontal affinity chromatography as test platforms
[22–24], revealed that there is an intimate cross-talk between the
cellular display of β-galactosides and the productive cell binding by
galectins. As a consequence, their activity can readily be fine-tuned
by modulating the glycome profile and counterreceptor density as
well as galectin expression dynamically, enabling context-
dependent multifunctionality.
In solution, the physical contact of a galectin with its ligand was
first monitored by fluorescence spectroscopy to detect the typical
blue shift in the Trp emission spectrum, still used effectively to
determine the binding strength [25, 26]. The scope of structural
analysis was then extended by different techniques such as, as
summarized recently [27], nuclear magnetic resonance (NMR)
spectroscopy among them. It made mapping of contact patterns
and identification of switches in protein structure possible when
complete 1H, 13C, and 15N assignments are available [28, 29]. In
this field, the inevitably low degree of resolution of proton spectra
(with their inevitably considerable overlap) has encouraged efforts
to find ways to much more favorable resolution. We here introduce
readers to a novel application of the 77Se isotope by strategically
combining synthetic carbohydrate chemistry and NMR spectros-
copy with the aim to characterize the glycan-galectin recognition.
Introducing 77Se NMR Spectroscopy to Analyzing Galectin–Ligand Interaction 107
Central to its implementation is the pioneering observation

that a thioglycoside, i.e., thiodigalactoside (TDG), is a potent
inhibitor of galectin-dependent hemagglutination (at an IC50-
value of 32 μM for electrolectin) [25]. If this activity were main-
tained for the respective selenoglycoside, i.e., selenodigalactoside
(SeDG), then its 77Se isotope could serve as reporter isotope on
binding to a galectin. The individual C-O/C-S/C-Se bond angles
(115 /95 /95 ) and lengths (1.4 Å/1.8 Å/1.9 Å) already indicate
a close similarity between these two thio- and seleno-glycosides.
Taken to the level of measuring intermolecular flexibility of
thio- and seleno-oligosaccharides, i.e., histo-blood group ABH
determinants, and docking them into binding sites of lectins by
computational simulations [30, 31], the hypothesis was shaped for
SeDG to be a potent ligand for human galectins. Experimental
testing fully confirmed this expectation [32, 33], hereby setting
the stage for work to establish interaction analysis for galectins by
77
Se NMR spectroscopy. However, the low-level natural occur-
rence (7.63% abundance) and sensitivity (0.7% of 1H) require signal
enhancements, that is to turn from direct to indirect 77Se detection
through protons via an initial polarization transfer to selenium.
Owing to our favorable experience with the Carr-Purcell-Mei-
boom-Gill (CPMG) heteronuclear single quantum multiple-bond
correlation (HSQMBC) approach [34–37], we calculated a sensi-
tivity enhancement up to a theoretical factor of 63 when bringing
this method together with achieving minimal intensity loss and
suppression of line broadening by using CPMG-INEPT
[38, 39]. As shown previously on the proof-of-principle level [40]
and described herein, 2D 1H, 77Se HSQMBC spectra for a human
galectin and SeDG that were obtained by the given protocol detect
the interaction. This experimental evidence illustrates the validity of
our concept, thus giving direction to take the next step along this
route. Preparing and applying isotopically pure 77SeDG as ligand is
such a project line. Equally important, combining the 77Se sensor
with monitoring 19F, this substitution in a β-galactoside such as
lactose is shown to be tolerated at certain sites such as the 30 -OH
group by galectins [41], let improvements for contact mapping
with a doubly labeled glycan come within reach. In fact, a recent
model study on a plant lectin with a trifluorotrimannoside illu-
strated how 2D STD-TOCSYreF NMR spectroscopy refined the
analysis of bound-state conformers up to measuring the ratio of the
two binding modes that coexist in solution [42]. Teaming up 1H,
19
F, and 77Se in NMR spectroscopy is therefore likely to gain
information on the ligand when bound to its cognate galectin at
an unprecedented level of precision in solution.
2 Materials
2.1 Chemical 1. 2,3,4,6-Tetra-O-acetyl-α-D-galactopyranosyl bromide (CAS

Synthesis no. 3068-34-4).
2. 2,3,4,6-Tetra-O-acetyl-α-D-glucopyranosyl bromide (CAS
no. 572-09-8).
3. 2,3,4,6-Tetra-O-acetyl-β-D-glucopyranosyl isoselenuronium
bromide [43, 44].
4. Selenourea 98% (CAS no. 630-10-4).
5. Acetone (see Note 1).
6. Dichloromethane (see Note 1).
7. Dry methanol (see Note 1).
8. Sodium methoxide in methanol (see Note 1).
9. Solution of KOH (0.22 g) in water (1 mL) (see Note 1).
10. Amberlist® 15H ion exchange (see Note 1).
11. CaCl2 solid (see Note 1).
12. Argon gas cylinder.
13. Heated magnetic stirrer.
14. Round bottomed and Erlenmeyer flasks 50 and 100 mL,
respectively.
15. Reflux condenser.
16. Separating funnel.
17. Rotary evaporator.
18. Glass filter.
2.2 Equipment 1. NMR spectrometer equipped with a broadband inverse (1H)

(Hardware, Software) detected liquid state probe suitable for excitation and decou-
and Materials Used for pling of the 77Se nucleus. Bruker Avance II 500 spectrometer
NMR Measurements equipped with BBI probehead operating at 500 MHz 1H and
95.38 MHz 77Se frequency (Bruker BioSpin GmbH, Rhein-
stetten, Germany) was used.
2. Pulse sequence scheme of the refocused 2D 1H,77Se CPMG-
HSQMBC experiment with 77Se decoupling (see Fig. 1 and
Note 7 for Bruker Pulse Sequence code).
3. Data processing software: TopSpin 2.1 or 3.5 (Bruker Biospin
GmbH, Karlsruhe, Germany).
Fig. 1 Pulse sequence scheme of the 2D 1H,77Se CPMG-HSQMBC experiment. Narrow and broad bars indicate
90 and 180 pulses, respectively, applied with phase x, unless indicated otherwise. ϕ1 is incremented
according to the XY16 scheme [45, 46], used in the CPMG-INEPT module. Further phases: ϕ2 ¼ y; ϕ3 ¼ x, x;
ϕ4 ¼ x, x, x, x; ϕ5 ¼ x, x, x, x, x, x, x, x; ϕrec ¼ x, x, x, x, x, x, x, x. The last 90 pulse on
77
Se (gray) is an optional CLIP (CLean In-Phase) purge pulse to destroy residual 2Hx,ySez antiphase
components. The CPMG echo delay τ was set to 150 μs while the number n of simultaneous composite
inversion pulses was adjusted to set the desired 2,3JH,77Se coupling evolution time, Δ 0.5/2,3JH,77Se. Clean,
phase-sensitive 77Se coherence selection is achieved by setting gradient since bell-shaped pulse strengths
G2: G4 ¼ 80: 15.257 with alternating sign in successive FIDs. z-Spoil gradients G1 and G3 have arbitrary but
distinct strengths. Sine bell-shaped gradient pulses of 1 ms duration were followed by a recovery delay of
200 μs. For 77Se CPD decoupling during FID acquisition, the WALTZ16 (or 64) scheme with a 90 pulse length
of 0.5/BW(77Se) was used, where BW is the bandwidth (in Hz) of the 77Se spectrum (from [40]; with
permission)
4. Selenodiglycosides (SeDG, MW 405), diselenodigalactoside

(DSeDG, MW 484), and selenodiglucoside (SeDGlc, MW
405, used as inert reference compound). See synthesis Sub-
headings 3.1 and 3.2 for their preparation.
(a) Buffer: 10 mM phosphate in D2O (pH 7.4), 0.5 M NaCl.
(b) Ligand stock solutions: 250 mM SeDG, 250 mM
DSeDG, and 250 mM SeDGlc each in D2O buffer.
(c) Final ligand concentration in NMR samples of the volume
of 500 μL: 2.5 mM of each ligand in all NMR measure-
ments including screening and competition experiments.
5. Galectin used as lyophilized powder.
3 Methods
3.1 Synthesis of 1. Plan the strategy for the steps of preparing SeDG (4b; Scheme
SeDG 1) [32], also applicable for the glucose derivative (4a).
2. Prepare the tetra-O-acetyl-β-D-galactopyranosyl-isoselenuro-
nium bromide (2b) by dissolving 2,3,4,6-tetra-O-acetyl-α-D-
galactopyranosyl bromide (7.0 g, 17 mmol) in acetone
(30 mL) and adding selenourea (2.1 g, 17 mmol) with stirring.
Heat the solution under reflux for 30 min. Allow the reaction
mixture to cool to room temperature (RT) and filter off the
precipitated pale yellow solid (5.9 g). Concentrate the filtrate in
vacuo to gain further 1.0 g of the product. Total yield: 78%.
Pale yellow solid, 1H NMR (D2O, 500 MHz): δ 5.60 (br.d,
1H, H-4, J3.4 3.1 Hz); 5.53–5.54 (m, 2H, H-1, H-2); 5.32 m,
(1H, H-3); 4.40 (m, 1H, H-5); 4.29 (dd, 1H, H-6a, J6a,6b
11.8 Hz, J5,6a 6.8 Hz); 4.26 (dd, 1H, H-6b, J5,6b 5.2 Hz);
2.03; 2.10; 2.15; 2.22 (12H, 4CH3(CO)); 13C NMR
(CDCl3, 125 MHz): δ 173.5; 172.9; 172.8; 172.5; (4
(CH3)CO); 165.1 (C-Se); 79.8 (C-1); 76.2 (C-5); 71.5
(C-3); 68.0 (C-4); 67.7 (C-2); 62.1 (C-6); 20.0, 20.1
(4CH3(CO)); HRMS: Calculated for C15H23N2O9Se:
455.057 [M Br]+, Found: 455.054.
3. Prepare the tetra-O-acetylated selenodigalactoside SeDG
(SeDGal, 3b) by dissolving 2,3,4,6-tetra-O-acetyl-β-D-galacto-
pyranosyl isoselenuronium bromide (from step 2; 1 g,
1.9 mmol) and 2,3,4,6-tetra-O-acetyl-α-D-galactopyranosyl
bromide (0.82 g, 2.0 mmol) in acetone (4 mL) and adding a
solution of KOH (0.22 g) in water (1 mL). Stir for 30 min
under argon atmosphere at RT. Remove the acetone under
reduced pressure and extract the residue with dichloromethane
(20 mL). Shake the CH2Cl2 solution twice with 10 mL water,
separate the organic phase using a separating funnel, add CaCl2
for drying, filter, and evaporate the filtrate to dryness under
reduced pressure. White powder recrystallize from methanol to
furnish 0.65 g of pure 3b (63%, see Note 2) [α]D22 – 5.2 (c 0.2
CHCl3); 1H NMR (CDCl3, 500 MHz): δ 5.46 (br.d, 1H, H-4,
J3.4 3.0 Hz); 5.30 (t, 1H, H-2, J1,2 ¼ J2,3 10.3 Hz); 5.06 (d,
1H, H-1); 5.05 (br.d, 1H, H-3); 4.13; 4.17(m, 2H, H-6a,
H-6b); 3.90 (m, 1H, H-5); 1.99; 2.05; 2.06; 2.17 (12H,
4CH3(CO)); 13C NMR (CDCl3, 125 MHz): δ 169.7;
169.9; 170.1; 170.3 (4(CH3)CO); 77.0 (C-1); 75.8 (C-5);
71.6 (C-3); 68.1 (C-2); 67.2 (C-4); 61.6 (C-6); 20.74, 20.65
(2), 20.74 (4CH3(CO)); HRMS: Calculated for
C28H38O18: 765.122 [M + Na]+, Found: 765.111.
Apply the same procedure to synthesize SeDGlc (3a) by
replacing 2,3,4,6-tetra-O-acetyl-β-D-galactopyranosyl
Scheme 1 Strategy of SeDG (and SeDGlc) synthesis in three steps via an isoselenuronium bromide derivative
isoselenuronium bromide and 2,3,4,6-tetra-O-acetyl-α-D-

galactopyranosyl bromide with analogs of gluco configuration
(see Subheading 2). Pure 3a (74%, see Note 2) is characterized
by [α]D22 – 35.2. (c 0.2 CHCl3); 1H NMR (CDCl3,
500 MHz): δ 5.08–5.24 (m, 3H, H-2, H-3, H-4); 5.03 (d,
1H, H-1, J1,2 9.5 Hz); 4.26 (dd, 1H, H-6a, J6a,6b 12.5 Hz,
J5,6a 4.7 Hz); 4.17 (dd, 1H, H-6b, J5,6b 1.5 Hz); 3.69 (m, 1H,
H-5); 2.12, 2.04, 2.03, 2.00 (12H, 4CH3(CO)); 13C NMR
(CDCl3, 125 MHz): δ 170.6; 170.1; 169.4; 169.3 (4(CH3)
CO); 77.2 (C-5); 76.4 (C-1); 73.7; 70.9; 68.1 (C-2, C-3,
C-4); 62.2 (C-6); 20.7; 20.6; 20.5 (4CH3(CO)); HRMS:
Calculated for C28H38O18Se [M + Na]+: 765.112, Found:
765.112.
Obtain the products SeDG (SeDGal, 4b) or SeDGlc (4a)
by deprotection via Zemplén deacetylation by dissolving bis
(2,3,4,6-tetra-O-acetyl-β-D-galactopyranosyl) selenide (3b) or
(2,3,4,6-tetra-O-acetyl-β-D-glucopyranosyl) selenide (3a) in
dry methanol (80 mL), then adding a solution of sodium
methoxide in methanol (30%) and keeping the mixture at RT
for 15 min. Add 1 g of Amberlist® 15H ion exchange resin
thereafter, shake the suspension well and filter off the resin.
Evaporate the filtrate to dryness under reduced pressure. Crys-
tallize from methanol to furnish 0.3 g of white solid (4b, 86%),
or 0.21 g (4a, 81%). For 4b: [α]D22–30.9 (c 0.2 CHCl3); 1H
NMR (D2O, 500 MHz): δ 4.98 (d, 1H, H-1, J1,2 10.0 Hz);
3.94 (dd, 1H, H-4, J3,4 3.3 Hz J4,5 ~ 1 Hz); 3.64–3.74 over-
lapping signals (4H, H-2, H-5, H-6a, H-6b); 3.60 (dd, 1H,
H-3, J2,3 9.1 Hz); 13C NMR (D2O, 125 MHz): δ 80.6 (C-1);
80.4 (C-5); 73.9 (C-3); 70.6 (C-2); 69.1 (C-4); 61.4 (C-6);
77
Se NMR (D2O, 95.4 MHz): δ 394; HRMS: C12H22O10Se
[M + Na]+: 429.030, Found: 429.032.
For 4a: [α]D22 – 75.1 (c 0.17 methanol-H2O 5:1); 1H
NMR (D2O, 500 MHz): δ 4.93 (br m, 1H, H-1); 3.76 (dd,
1H, H-6a, J6a,6b 12.3 Hz, J5,6a 1.8 Hz); 3.56 (dd, 1H, H-6b,
J5,6b 5.2 Hz); 3.27–3.38 overlapping signals (m, 4H, H-2,
H-3, H-4, H-5); 13C NMR (D2O, 125 MHz): δ 80.7 (C-1);
79.3 (C-5); 76.7; 72.6; 69.1 (C-2, C-3, C-4); 60.5 (C-6); 77Se
NMR (D2O, 95.4 MHz): δ 393.9; HRMS: Calcd. for
C12H22O10Se [M + Na]+: 429.030, Found: 429.027.
3.2 Synthesis of 1. Plan the strategy for the steps of preparing DSeDG (6b;
Diselenodigalactoside Scheme 2), conveniently using the tetra-O-acetylated isosele-
(DSeDG, DSeDGal, 6b) nuronium bromide derivative 2b prepared as described in Sub-
heading 3.1, step 2 (see also [32]).
2. Prepare the bis(2,3,4,6-tetra-O-acetyl-β-D-galactopyranosyl)
diselenide (5b) by dissolving 2,3,4,6-tetra-O-acetyl-β-D-galac-
topyranosyl isoselenuronium bromide (2b, 2.65 g) in acetone
(10 mL), then adding the KOH solution and stirring at RT for
15 min. Remove the acetone via evaporation under reduced
pressure and extract the residual aqueous phase with dichlor-
omethane (2 30 mL). Wash the extract with water
(2 20 mL), dry the organic phase with CaCl2 and evaporate
to dryness. Pale yellow solid, 1.48 g (73%). [α]D22–10.7 (c 0.2
CHCl3); 1H NMR (CDCl3, 500 MHz): δ 5.45 (dd, 1H, H-4,
J4,5 ~ 1 Hz); 5.36 (t, 1H, H-2, J2,3 9.8 Hz); 5.07 (dd, 1H,
H-3, J3,4 3.5 Hz); 4.91 (d, 1H, H-1, J1,2 10.0 Hz); 4.20 (dd,
Scheme 2 Strategy of DSeDG synthesis in two steps
1H, H-6a, J6a,6b 10.7 Hz, J5,6a 6.2 Hz); 4.10 (dd, 1H, H-6b,
J5,6b 7.2 Hz); 2.16, 2.07, 2.03, 1.98 (12H, 4CH3(CO)); 13C
NMR (CDCl3, 125 MHz): δ 170.2; 170.1; 170.0 (2); 169.5
(4(CH3)CO); 81.3 (C-1); 75.5 (C-5); 71.6 (C-3); 69.4

(C-2); 67.1 (C-4); 60.8 (C-6); 20.9; 20.6; 20.5
(4CH3(CO)); HRMS: C28H38O18Se2 [M + Na]+: 845.029,
Found: 845.027.
3. Obtain the product DSeDG (DSeDGal, 6b) via Zemplén dea-
cetylation form bis(2,3,4,6-tetra-O-acetyl-β-D-galactopyrano-
syl)diselenide 5b (1.26 g, 1.5 mmol) in dry methanol
(50 mL), add a solution of sodium methoxide in methanol
(0.28 mL), and keep at RT for 10 min, then add 1 g of Amber-
list® 15H ion exchange resin, shake the suspension well and
then filter off the resin. Evaporate the filtrate to dryness under
reduced pressure. White powder, 0.68 g (91%). [α]D22 – 7.7
(c 0.17 methanol-H2O 5:1); 1H NMR (D2O, 500 MHz): δ
4.94 (d, 1H, H-1, J1,2 9.8 Hz); 4.04 (br.d, 1H, H-4, J3,4
3.3 Hz J4,5 < 1 Hz); 3.90 (t, 1H, H-2, J2,3 9.8 Hz);
3.75–3.84 overlapping signals (3H, H-5, H-6a, H-6b); 3.74
(dd, 1H, H-3); 13C NMR (D2O, 125 MHz): δ 83.8 (C-1);
80.7 (C-5); 73.8 (C-3); 70.6 (C-2); 69.1 (C-4); 61.2 (C-6);
77
Se NMR (D2O, 95.4 MHz): δ 381; HRMS: Calc. for:
C12H22O10Se2 [M + Na]+: 508.943, Found: 508.948.
3.3 77Se NMR- 1. 2D 1H,77Se Correlation spectra were recorded at natural 77Se
Spectroscopical abundance by the 1H, 77Se CPMG-HSQMBC sequence shown
Monitoring and in Fig. 1 at 303 K. Simultaneous composite π pulses on the 1H
Competition Assay and 77Se channels were applied with equal 90 pulse lengths of
15 μs (60 μs for the 90 x-180 y-90 x composite pulse) after
careful adjustment of both power levels. Spectra were recorded
with 2048 total data points in the 1H (t2) dimension and
36 total points in the 77Se (t1) dimension, using spectral win-
dows of 5 ppm (2500 Hz) for 1H and 6 ppm (570 Hz) for 77Se.
Three hundred twenty scans per t1 increment were accumu-
lated, and the polarization recovery (relaxation) delay between
consecutive scans (d1 in Bruker code, see Note 7) was set to
1.7 s. The CPMG-INEPT delay Δ for long-range heteronuc-
lear coupling evolution was set to 54.5 ms. For 77Se CPD
decoupling during FID acquisition (210 ms), the WALTZ16
scheme with a 90 pulse length of 400 μs was used (see Note 3).
2. For processing of the raw NMR data, multiplication with
squared cosine bell function and zero filling was applied in
both dimensions prior to 2D Fourier processing in accordance
with the echo–antiecho protocol resulting in 4096 512 total
data points.
3. For screening of mixtures using 77Se as sensor, prepare 500 μL
of ligand mixture sample containing DSeDG, SeDG, and
SeDGlc at 2.5 mM concentration each and perform the 2D
1
H,77Se CPMG-HSQMBC experiment with the experimental
Fig. 2 Selenoglycoside binding to Gal-3 monitored by 2D 1H,77Se CPMG-HSQMBC. The 2D spectra of the
ligand mixture containing DSeDG (red), SeDG (blue), and SeDGlc (black) were recorded before (left) and after
(right) adding Gal-3 to a protein:ligand ratio of 1:20 (top) and 1:60 (bottom). 1H (F2) traces extracted at the
three 77Se signals are shown next to the contour plots. The indicated attenuated signal intensities from binding
are relative to the free ligand intensity. The observed signal attenuation is a valid indicator of ligand binding.
The intensity drop is due to the enhanced 77Se and 1H T2 relaxation of ligands in complex with the protein.
Moreover, the relative signal attenuations detected upon binding allow affinity ranking of the three ligands.
The 2D 1H,77Se CPMG-HSQMBC spectra were plotted with identical parameters (from [40]; with permission)
setup described above (see Note 4). Then add the appropriate
amount of lectin to the sample in protein:ligand ratio of 1:20 or
1:60, i.e., protein concentration of 125 or 42 μM, respectively
(see Note 5), let it completely dissolve and start the next NMR
data acquisition using the same experimental setup as in the
first run. The resulting 2D 1H,77Se CPMG-HSQMBC corre-
lation maps together with respective 1H (F2) traces extracted at
the 77Se signals are exemplarily shown in Fig. 2 for the case of
human galectin (Gal-3).
4. For detecting competitive inhibition, first record the 2D
1
H,77Se CPMG-HSQMBC spectra of 2.5 mM ligand (SeDG
and DSeDG)-containing solutions (500 μL), as detailed above.
These correlation maps serve as reference spectra in the
subsequent competition experiment. Add the lyophilized pro-
tein to the DSeDG-containing solution (for Gal-3, use a pro-
tein:ligand ratio of ~0.75:60, to avoid aggregation of protein)
and record the 2D 1H,77Se spectrum as above. Finally, add
5 μL of 250 mM stock solution of the SeDG to the sample
Fig. 3 Competitive displacement of DSeDG bound to Gal-3 by SeDG. (a) 1H (F2) traces of 2D 1H,77Se CPMG-
HSQMBC spectra of DSeDG (red, top) and SeDG (blue, bottom) in the absence of Gal-3 provide the reference
signal intensities I0 ¼ 100%. (b) The 1H (F2) trace of DSeDG (2.5 mM) after adding Gal-3 (30 μM, i.e., molar
ratio ¼ 0.75:60) indicates binding induced signal attenuation to I/I0 ¼ 67%. (c) Equimolar addition of SeDG
(2.5 mM) causes a rebound of the attenuated DSeDGal signal to 86% (red, top), thus indicating its competitive
displacement. The SeDG spectrum is conversely attenuated to 57% (blue, bottom), confirming its preferred
binding by Gal-3 (from [40]; with permission)
(for Gal-3, use a protein:ligand1:ligand2 ratio of ~0.75:60:60)

and start the next NMR data acquisition. After processing the
raw NMR data, monitor and compare the signal intensity
changes in the resulting spectra, as illustrated in Fig. 3 (see
Note 6).
4 Notes
1. Amberlist® 15H is a trademark of Sigma-Aldrich. Analytical

grade solvents and reagents from Sigma-Aldrich have been
used in all preparations.
2. The mother liquor contains diselenide 5b as side product
formed in concurrent reaction common in all types of syntheses
based on glycosyl-isoselenuronium salts [47].
3. It is recommended to perform quantitative PULCON [48] 1H
NMR measurements before and after the 2D (1H-77Se) runs to
check the protein concentration and to detect aggregation
(if any). Experimental parameters for PULCON: 1H spectral
width: 15 ppm, recovery delay between transients: 30 s, num-
ber of scans: 16. Protein concentration is estimated by the
integration of aliphatic region up to the water signal at
4.7 ppm, dividing by the number of contributing protons
(calculated for each protein sequence) and referencing the
integral intensity to that of a sucrose solution of known con-
centration (in our case: 10.08 mM, in D2O). The ratio of
integral intensities (protein vs. sucrose) is corrected according
to the principle of reciprocity, taking into account the respec-
tive ratio of 90 pulse lengths determined/calibrated at the
same power level in both samples.
4. Typically, ~30–40 min after inserting the tube into the magnet,
thermal equilibrium is reached to start the NMR measurement.
5. Routinely, lyophilized protein stored dry at 20 C (or colder)
is dissolved just prior to the experiment. An appropriate quan-
tity of protein lyophilized from salt-free water can be used or a
freshly prepared solution.
6. A calibration with a ligand of known affinity (or titrations with
SeDG) is helpful to estimate inhibitory capacity of test
substances.
7. Bruker Pulse Sequence code of 2D 1H,77Se CPMG-HSQMBC
experiment for AVANCE II and TopSpin 2.x.
;ek_ti_cpmg_hsqmbc_dc
;refocused and X-decoupled CPMG-HSQMBC pulse sequence
;2D 1H-X correlation via double inept transfer
;phase sensitive using Echo/Antiecho gradient selection
;using CPMG for polarization transfer to avoid evolution of J
(HH)
;using composite 1H and X 180 pulses in CPMG
;with gradient z,z filter purge pulse between last 90 degree
pulse pair
[40]
;with X-decoupling during acquisition (please see for
details)
;
;Bruker Avance II version
#include <Avance.incl>
#include <Grad.incl>
"p2=p1*2"
"p4=p3*2"
"d0=3u"
"d11=30m"
"d13=3u"
"d7=d13+p16+d16+4u"
"d20=p16+d16+p2+d0*2"
"l3=(td1/2)"
"in0=inf1/2"
1 ze
d11 pl12:f2
2 d1 do:f2
d11
3 d11 pl1:f1
4 (p1 ph1)
;---- CPMG-INEPT using XY16 phase cycling ----
5 d15 pl2:f2
(p1 ph20) (p3 ph20):f2
3u
(p2 ph21) (p4 ph21^):f2
3u
(p1 ph20) (p3 ph20^):f2
d15 ;d15=140-150us
lo to 5 times l1 ;p1 should be calibrated to p1=p3 at pl1 !
;l1=multiple of 16
;long-range coupling evolution = (2*d15+2*p4+6)*l1
(p1 ph2)
;---- end of INEPT, z-spoil on 2HzSez ----
d13 UNBLKGRAD
p16:gp3 ;gpz3=19 purging
d16
;---- t1 on 77Se ----
(p3 ph3):f2
d0
(p2 ph5)
d0
p16:gp1 ;gpz1=80 for coherence selection
d16
(p3 ph14):f2 ;comp. X 180 pulse
(p4 ph4):f2
(p3 ph14):f2
d20
(p3 ph4):f2
;---- end of t1, z-spoil on 2HzSez ----
d13
p16:gp4 ;gpz4=10 purging
d16
;---- refocusing CPMG-INEPT ----
(p1 ph1)
;---- CPMG-reINEPT using XY16 phase cycling ----
6 d15
(p1 ph20) (p3 ph20):f2
3u
(p2 ph21) (p4 ph21^):f2
3u
(p1 ph20) (p3 ph20^):f2
d15 ;d15=140-150us
lo to 6 times l1 ;p1 should be calibrated to p1=p3 at pl1!!!
;l1=multiple of 16
;long-range coupling evolution = (2*d15+2*p4+6)*l1
d7
(p2 ph1)
d13
p16:gp2*EA ;gpz2=15.26 for echo-antiecho 77Se coherence
selection
d16 pl12:f2
4u BLKGRAD
go=2 ph31 cpd2:f2

d1 do:f2 wr #0 if #0 zd
d11 igrad EA
lo to 3 times 2
d11 id0
lo to 4 times l3
exit
ph1=0
ph2=1
ph20=1 2 1 2 2 1 2 1 3 0 3 0 0 3 0 3
ph21=0 1 0 1 1 0 1 0 2 3 2 3 3 2 3 2
ph3=0 2
ph4=0 0 0 0 2 2 2 2
ph14=1 1 1 1 3 3 3 3
ph5=0 0 2 2
ph31=0 2 0 2 2 0 2 0
;pl1 : f1 channel - power level for pulse (default)

;pl2 : f2 channel - power level for pulse (default)
;pl12: f2 channel - power level for CPD/BB decoupling
;p1 : f1 channel - 90 degree high power pulse
;p16: homospoil/gradient pulse
;d0 : incremented delay [3 usec]
;d1 : relaxation delay; 1-5 * T1
;d7: =d13+p16+d16+4u
;d11: delay for disk I/O [30 msec]
;d13: short delay [3 usec]
;d15: interpulse delay [140-150 usec]
;d16: delay for homospoil/gradient recovery
;d20: =p16+d16+p2+d0*2
;l1: loop for CPMG
;NS: number of scans
;DS: number of dummy scans, >= 128!
;FnMODE1: Echo-Antiecho
;cpd2: decoupling according to sequence defined by cpdprg2
;pcpd2: f2 channel - 90 degree pulse for decoupling sequence
;gpz1: 80%
;gpz2: 20.1% for C-13, 15.26% for Se-77
;gpz3: purging gradient (~19%)
;gpz4: purging gradient (~10%)
;
;use gradient files:
;gpnam1: SINE.100
;gpnam2: SINE.100
;gpnam3: SINE.100
;gpnam4: SINE.100
Acknowledgments
This research was supported by the National Research, Develop-

ment and Innovation Office of Hungary (grant numbers: NKFI/
OTKA NN 128368) (to M.R., L.Sz., and K.E.K.) and NKFI/
OTKA PD 135034 (to I.T.)) and co-financed by the European
Regional Development Fund (project GINOP-2.3.2-15-2016-
00044). I.T. acknowledges the support of the János Bolyai
Research Scholarship of the Hungarian Academy of Sciences
(BO/00372/20/7). L.Sz. is indebted to Sára Balla for excellent
help in organic syntheses, to Lajos Nagy for MS spectra and to
Attila Kiss for optical rotation data (all in the Department of

Chemistry, Faculty of Science & Technology, University of
Debrecen).
References
1. Roseman S (2001) Reflections on glycobiol- 199–222. https://doi.org/10.1007/s00418-

ogy. J Biol Chem 276(45):41527–41542. 016-1524-6
https://doi.org/10.1074/jbc.R100053200 12. Kaltner H, Garcı́a Caballero G, Ludwig A-K,
2. Roth J (1987) Subcellular organization of gly- Manning JC, Gabius H-J (2018) From glyco-
cosylation in mammalian cells. Biochim Bio- phenotyping by (plant) lectin histochemistry to
phys Acta 906(3):405–436. https://doi.org/ defining functionality of glycans by pairing
10.1016/0304-4157(87)90018-9 with endogenous lectins. Histochem Cell Biol
3. Corfield AP (2017) Eukaryotic protein glyco- 149(6):547–568. https://doi.org/10.1007/
sylation: a primer for histochemists and cell s00418-018-1676-7
biologists. Histochem Cell Biol 147(2): 13. Barondes SH (1984) Soluble lectins: a new
119–147. https://doi.org/10.1007/s00418- class of extracellular proteins. Science
016-1526-4 223(4642):1259–1264. https://doi.org/10.
4. Kopitz J (2017) Lipid glycosylation: a primer 1126/science.6367039
for histochemists and cell biologists. Histo- 14. Cooper DNW (2002) Galectinomics: finding
chem Cell Biol 147(2):175–198. https://doi. themes in complexity. Biochim Biophys Acta
org/10.1007/s00418-016-1518-4 1572(2–3):209–231. https://doi.org/10.
5. Hart GW (2013) Thematic minireview series 1016/s0304-4165(02)00310-0
on glycobiology and extracellular matrices: gly- 15. Arthur CM, Baruffi MD, Cummings RD,
can functions pervade biology at all levels. J Stowell SR (2015) Evolving mechanistic
Biol Chem 288(10):6903. https://doi.org/ insights into galectin functions. Methods Mol
10.1074/jbc.R113.453977 Biol 1207:1–35. https://doi.org/10.1007/
6. Gabius H-J, Roth J (2017) An introduction to 978-1-4939-1396-1_1
the sugar code. Histochem Cell Biol 147(2): 16. Kaltner H, Toegel S, Garcı́a Caballero G, Man-
111–117. https://doi.org/10.1007/s00418- ning JC, Ledeen RW, Gabius H-J (2017)
016-1521-9 Galectins: their network and roles in immu-
7. Cummings RD (2019) Stuck on sugars: how nity/tumor growth control. Histochem Cell
carbohydrates regulate cell adhesion, recogni- Biol 147(2):239–256. https://doi.org/10.
tion, and signaling. Glycoconj J 36(4): 1007/s00418-016-1522-8
241–257. https://doi.org/10.1007/s10719- 17. Kasai K-i (2018) Galectins: quadruple-faced
019-09876-0 proteins. Trends Glycosci Glycotechnol
8. Kaltner H, Abad-Rodrı́guez J, Corfield AP, 30(172):SE221–SE223. https://doi.org/10.
Kopitz J, Gabius H-J (2019) The sugar code: 4052/tigg.1745.7SE
letters and vocabulary, writers, editors and 18. Garcı́a Caballero G, Kaltner H, Kutzner TJ,
readers and biosignificance of functional Ludwig A-K, Manning JC, Schmidt S,
glycan-lectin pairing. Biochem J 476(18): Sinowatz F, Gabius H-J (2020) How galectins
2623–2655. https://doi.org/10.1042/ have become multifunctional proteins. Histol
BCJ20170853 Histopathol 35(6):509–539. https://doi.org/
9. Boyd WC, Shapleigh E (1954) Specific preci- 10.14670/HH-18-199
pitating activity of plant agglutinins (lectins). 19. Kamitori S (2018) Three-dimensional struc-
Science 119(3091):419. https://doi.org/10. tures of galectins. Trends Glycosci Glycotech-
1126/science.119.3091.419 nol 30(172):SE41–SE50. https://doi.org/10.
10. Lis H, Sharon N (1998) Lectins: carbohydrate- 4052/tigg.1731.1SE
specific proteins that mediate cellular recogni- 20. Romero A, Gabius H-J (2019) Galectin-3: is
tion. Chem Rev 98(2):637–674. https://doi. this member of a large family of multifunctional
org/10.1021/cr940413g lectins (already) a therapeutic target? Expert
11. Manning JC, Romero A, Habermann FA, Gar- Opin Ther Targets 23(10):819–828. https://
cı́a Caballero G, Kaltner H, Gabius H-J (2017) doi.org/10.1080/14728222.2019.1675638
Lectins: a primer for histochemists and cell 21. Ford MG, Weimar T, Köhli T, Woods RJ
biologists. Histochem Cell Biol 147(2): (2003) Molecular dynamics simulations of
galectin-1-oligosaccharide complexes reveal
the molecular basis for ligand diversity. Pro- 30. Strino F, Lii JH, Gabius H-J, Nyholm PG
teins 53(2):229–240. https://doi.org/10. (2009) Conformational analysis of thioglyco-
1002/prot.10428 side derivatives of histo-blood group ABH
22. Arthur CM, Cataldi Rodrigues L, Dias antigens using an ab initio-derived reparame-
Baruffi M, Sullivan HC, Heimburg-Molinaro J, terization of MM4: implications for design of
Smith DF, Cummings RD, Stowell SR (2015) non-hydrolysable mimetics. J Comput Aided
Examining galectin binding specificity using Mol Des 23(12):845–852. https://doi.org/
glycan microarrays. Methods Mol Biol 1207: 10.1007/s10822-009-9301-4
115–131. https://doi.org/10.1007/978-1- 31. Strino F, Lii JH, Koppisetty CA, Nyholm PG,
4939-1396-1_8 Gabius H-J (2010) Selenoglycosides in silico:
23. Kamili NA, Arthur CM, Gerner-Smidt C, ab initio-derived reparameterization of MM4,
Tafesse E, Blenda A, Dias-Baruffi M, Stowell conformational analysis using histo-blood
SR (2016) Key regulators of galectin-glycan group ABH antigens and lectin docking as
interactions. Proteomics 16(24):3111–3125. indication for potential of bioactivity. J Com-
https://doi.org/10.1002/pmic.201600116 put Aided Mol Des 24(12):1009–1021.
24. Iwaki J, Hirabayashi J (2018) Carbohydrate- https://doi.org/10.1007/s10822-010-
binding specificity of human galectins: an over- 9392-y
view by frontal affinity chromatography. 32. André S, Kövér KE, Gabius H-J, Szilágyi L
Trends Glycosci Glycotechnol 30(172): (2015) Thio- and selenoglycosides as ligands
SE137–SE153. https://doi.org/10.4052/ for biomedically relevant lectins: valency-
tigg.1728.1SE activity correlations for benzene-based dithio-
25. Teichberg VI, Silman I, Beitsch DD, Resheff G galactoside clusters and first assessment for
(1975) A β-D-galactoside-binding protein from (di)selenodigalactosides. Bioorg Med Chem
electric organ tissue of Electrophorus electricus. Lett 25(4):931–935. https://doi.org/10.
Proc Natl Acad Sci U S A 72(w4):1383–1387. 1016/j.bmcl.2014.12.049
https://doi.org/10.1073/pnas.72.4.1383 33. Kaltner H, Szabo T, Fehér K, André S, Balla S,
26. Sindrewicz P, Li X, Yates EA, Turnbull JE, Lian Manning JC, Szilágyi L, Gabius H-J (2017)
LY, Yu LG (2019) Intrinsic tryptophan fluo- Bivalent O-glycoside mimetics with S/disul-
rescence spectroscopy reliably determines fide/Se substitutions and aromatic core: syn-
galectin-ligand interactions. Sci Rep 9(1): thesis, molecular modeling and inhibitory
11851. https://doi.org/10.1038/s41598- activity on biomedically relevant lectins in
019-47658-8 assays of increasing physiological relevance.
Bioorg Med Chem 25(12):3158–3170.
27. Solı́s D, Bovin NV, Davis AP, Jiménez- https://doi.org/10.1016/j.bmc.2017.04.011
Barbero J, Romero A, Roy R, Smetana K Jr,
Gabius H-J (2015) A guide into glycosciences: 34. Williamsson RT, Márquez BL, Gerwick WH,
how chemistry, biochemistry and biology Kövér KE (2000) One-and two-dimensional
cooperate to crack the sugar code. Biochim gradient-selected HSQMBC NMR experi-
Biophys Acta 1850(1):186–235. https://doi. ments for the efficient analysis of long-range
org/10.1016/j.bbagen.2014.03.016 heteronuclear coupling constants. Magn
Reson Chem 28(4):265–273. https://doi.
28. Eckardt V, Miller MC, Blanchet X, Duan R, org/10.1002/(SICI)1097-458x
Leberzammer J, Duchene J, Soehnlein O,
Megens RT, Ludwig A-K, Dregni A, 35. Boros S, Kövér KE (2011) Low-power com-
Faussner A, Wichapong K, Ippel H, posite CPMG HSQMBC experiment for accu-
Dijkgraaf I, Kaltner H, Doring Y, rate measurement of long-range heteronuclear
Bidzhekov K, Hackeng TM, Weber C, Gabius coupling constants. Magn Reson Chem 49(3):
H-J, von Hundelshausen P, Mayo KH (2020) 106–110. https://doi.org/10.1002/mrc.
Chemokines and galectins form heterodimers 2717
to modulate inflammation. EMBO Rep 21(4): 36. Timári I, Illyes TZ, Adams RW, Nilsson M,
e47852. https://doi.org/10.15252/embr. Szilágyi L, Morris GA, Kövér KE (2015) Pre-
201947852 cise measurement of long-range heteronuclear
29. Miller MC, Nesmelova IV, Daragan VA, coupling constants by a novel broadband
Ippel H, Michalak M, Dregni A, Kaltner H, proton-proton-decoupled CPMG-HSQMBC
Kopitz J, Gabius H-J, Mayo KH (2020) Pro4 method. Chem Eur J 21(8):3472–3479.
prolyl peptide bond isomerization in human https://doi.org/10.1002/chem.201405535
galectin-7 modulates the monomer-dimer 37. Timári I, Kövér KE (2018) Broadband homo-
equilibrum to affect function. Biochem J nuclear decoupled HSQMBC methods. Magn
477(17):3147–3165. https://doi.org/10. Reson Chem 56(10):910–917. https://doi.
1042/BCJ20200499 org/10.1002/mrc.4700
38. Koskela H, Kolpelainen I, Halkkinen S (2003) 43. Wagner G, Nuhn P (1964) Synthese von Sele-
LR-CAHSQC: an application of a Carr-Pur- noglykosiden mit acetyl-glykosyl-isoselenuro-
cell-Meiboom-Gill-type sequence to hetero- nium-bromiden. 4. Mitt. über
nuclear multiple bond correlation “Selenoglykoside”. Arch Pharm 297(8):
spectroscopy. J Magn Reson 164(2):228–232. 461–473. https://doi.org/10.1002/ardp.
https://doi.org/10.1016/s1090-7807(03) 19642970804
00250-7 44. Schneider W, Wrede F (1917) Synthese eines
39. Kövér KE, Batta G, Fehér K (2006) Accurate schwefelhaltigen und eines selenhaltigen disac-
measurement of long-range heteronuclear cou- charides. Ber Dtsch Chem Ges 50(1):
pling constants from undistorted multiplets of 793–804. https://doi.org/10.1002/cber.
an enhanced CPMG-HSQMBC experiment. J 191705001131
Magn Reson 181(1):89–97. https://doi.org/ 45. Pervushin K, Gallius V, Ritter C (2001)
10.1016/j.jmr.2006.03.015 Improved TROSY-HNCA experiment with
40. Raics M, Timári I, Diercks T, Szilágyi L, Gabius suppression of conformational exchange
H-J, Kövér KE (2019) Selenoglycosides as lec- induced relaxation. J Biomol NMR 21(2):
tin ligands: 77Se-edited CPMG-HSQMBC 1 6 1 – 1 6 6 . h t t p s : // d o i . o r g / 1 0 . 1 0 2 3 /
NMR to monitor biomedically relevant inter- a:1012484027195
actions. ChemBioChem 20(13):1688–1692. 46. Zhuravleva A, Orekhov VY (2008) Divided
https://doi.org/10.1002/cbic.201900088 evolution: a scheme for suppression of line
41. Solı́s D, Romero A, Kaltner H, Gabius H-J, broadening induced by conformational
Dı́az-Mauriño T (1996) Different architecture exchange. J Am Chem Soc 130(11):
of the combining sites of two chicken galectins 3260–3261. https://doi.org/10.1021/
revealed by chemical-mapping studies with ja710056t
synthetic ligand derivatives. J Biol Chem 47. Illyés T-Z, Balla S, Bényei A, Kumar AA,
271(22):12744–12748. https://doi.org/10. Timári I, Kövér KE, Szilágyi L (2016) Explor-
1074/jbc.271.22.12744 ing the synthesis of novel glycomimetics. Car-
42. Diercks T, Infantino AS, Unione L, Jiménez- bohydrate derivatives with Se-S- or Se-Se-
Barbero J, Oscarson S, Gabius H-J (2018) glycosidic linkages. ChemistrySelect 1(10):
Fluorinated carbohydrates as lectin ligands: 2383–2388. https://doi.org/10.1002/slct.
synthesis of OH/F-substituted N-glycan core 201600628
trimannoside and epitope mapping by 2D 48. Wider G, Dreier L (2006) Measuring protein
STD-TOCSYreF NMR spectroscopy. Chem concentrations by NMR spectroscopy. J Am
Eur J 24(59):15761–15765. https://doi.org/ Chem Soc 128(8):2571–2576. https://doi.
10.1002/chem.201803217 org/10.1021/ja055336t
Chapter 7
Evaluation of Galectin Binding by Surface Plasmon

Resonance
Padmaja Mehta-D’souza
Abstract
Surface plasmon resonance (SPR) instruments, like the BIAcore 3000, are useful for studying the binding
between macromolecules in real time. The high sensitivity and low sample consumption in the Biacore
enables the measurement of rapid kinetics and low affinities characteristics of many biological interactions.
This chapter describes the affinity measurement of Galectins-1, -2 and -3 and their glycoside ligands using
this approach.
Key words Surface plasmon resonance (SPR), Streptavidin (SA) sensor chip, Affinity, Ligand, Analyte,
Flow cell (fc), Galectin, Glycosides
1 Introduction
The BIAcore 3000 measures the interaction between two macro-

molecules using the principle of surface plasmon resonance (SPR).
One molecule is coupled to a sensor surface and is referred to as the
ligand, the other is in free flow and is referred to as the analyte [1–
3]. Each sensor chip creates four reaction chambers, referred to as
flow cells (fc). The integrated microfluidic cartridge (IFC) is
designed such that samples can be run over each individual flow
cell, over selected pairs of flow cells, or serially over all four flow
cells. It is always recommended that one flow cell be used as a
control surface to minimize nonspecific binding to the carboxy-
methylated dextran surface of the sensor chip. An ideal control
surface is created by immobilizing a non-binding ligand control
and should always be on fc1 so that it is upstream of the test ligands.
Ligands can be covalently coupled to the sensor surface or
captured with an affinity tag such that all immobilized molecules
are oriented the same way. However, if it is possible to label a
protein with biotin or synthesize a ligand with a biotin tag, that
should be the method of choice, as the ligand can then be conve-
niently captured on a sensor chip pre-coupled with streptavidin
125
126 Padmaja Mehta-D’souza
(SA). Interactions on the BIAcore require small amounts of each

macromolecule; the analyte molecule does not need to be labeled,
and data is obtained in a relatively short time. Choice of ligand and
analyte depends on several factors, but especially on relative molec-
ular size, availability, and relative purity.
Given the potential roles of individual galectin family members
in a wide variety of biological processes, many studies have sought
to identify the carbohydrate ligands through which galectins medi-
ate their regulatory activities [4–10]. Many studies have evaluated
galectin ligand interactions using a variety of approaches, including
frontal affinity chromatography, isothermal calorimetry, fluores-
cence polarization, and more recently glycan microarrays [8, 9,
11–16]. While these each of these approaches can enable unique
insight into the specificity and biological activities of galectin family
members [15–26], BIAcore also provides a distinct approach with
unique advantages when evaluating galectin–glycan interactions
[2, 14].
When using BIAcore to evaluate galectin–glycan interactions,
samples can be injected onto the sensor chip surface using several
different flow rates (from 1 to 100 μL/min). During analyte injec-
tion, a constant concentration of analyte is delivered to the ligand
surface. The flow rate selected and the volume of the sample
injected determine the contact time of the analyte with the ligand.
Binding is measured in arbitrary resonance units (RU) and
recorded in real time in a sensorgram, which is the raw data output
from a BIAcore experiment. A sensorgram always provides infor-
mation about the association and dissociation phases of an interac-
tion, and often contains information about steady state binding as
well (depending on the affinity of the interaction, the analyte
concentration used, and the length of the injection). Antigen–
antibody interactions are usually depicted by a typical profile
shown in panel a of Fig. 1. For some low-affinity interactions with
rapid kinetics, a representative sensorgram often looks like that
depicted in panel b of Fig. 1. In binding studies for these interac-
tions, a control surface is very essential, so that the nonspecific
binding response arising from changes in the bulk refractive index
can be subtracted [27].
One of the advantages of using the BIAcore to look at binding
of macromolecules is that the same surface can be used repeatedly.
After each analyte injection, the bound analyte is removed with a
regeneration solution that disrupts the ligand–analyte interaction
but does not affect the functionality of the ligand. When the
interaction between analyte and ligand is fairly weak, as in most
lectin–carbohydrate interactions, the bound analyte is often washed
away with continuous buffer flow alone without the need for
regeneration. In this study [14], the BIAcore 3000 was used to
compare the binding affinities of Galectin-1, -2 and -3 for lactose,
(LacNAc)2 and (LacNAc)3.
Evaluation of Galectin Binding by Surface Plasmon Resonance 127
A steady-state regeneration
dissociation
2000
Response (RU)
association
0
baseline baseline
-2000
200 400 600

Time (s)
B Total Binding
600 steady-state
Response (RU)
400
association dissociation
200
600
Response (RU)
400
Non-specific Binding
200
0
80 100 120
Time (s)
Fig. 1 Sensorgrams showing different phases of the interaction. (a) Typical

profile with measurable association and dissociation kinetics. (b) Rapid associ-
ation and dissociation kinetics showing total binding (top) on ligand surface and
nonspecific binding (bottom) on control surface
2 Materials (See Notes 1 and 2)
2.1 Starting the 1. BIAcore 3000 instrument (GE Healthcare Life Sciences): a
BIAcore 3000 high performance research system that enables label-free mea-
surements of biomolecular interactions.
2.2 Preparing the 1. Sensor Chip SA: sensor chip that is precoated with streptavidin.
Sensor Surface 2. Biotinylated proteins, glycosides or peptides (National Insti-
tutes for Health/NIGMS-funded Consortium for Functional
Glycomics) (see Note 3).
3. Plastic sample vials: 7 mm plastic vials with caps (see Note 4).
4. Preconditioning solution: 1 M NaCl in 50 mM NaOH.
5. 40% glycerol for the Normalize routine.
2.3 Galectin Binding 1. Recombinant human Galectin-1, Galectin-2, and Galectin-3

were purified by affinity chromatography on lactosyl-
Sepharose [28].
2. BIAcore Running Buffer: Phosphate-buffered saline (see Notes
5 and 6).
2.4 Data Analysis 1. BIAevaluation software.
3 Methods
3.1 Starting Up 1. Stop the Standby routine.

BIAcore 3000 (See 2. Undock, then remove the maintenance chip, and replace it with
Notes 7 and 8) a fresh, unused research grade SA chip (see Note 9).
3. Remove the D/W bottle used for Standby and replace it with
room temperature filtered and degassed running buffer (see
Note 10).
4. Empty the waste bottle.
5. Prime the system twice using running buffer (see Note 11).
6. Normalize the signal in all flow cells according to the manu-
facturer’s instructions.
7. Prime the system once more with running buffer.
8. The instrument and the docked sensor chip are now ready
to use.
3.2 Preparing the 1. Start a new sensorgram at a flow rate of 5 μL/min.

Sensor Surface 2. Set flow path to fc1, 2, 3, 4, over all four flow cells (see
Note 12).
3. Inject 5 μL of preconditioning solution, 1 M NaCl in 50 mM

NaOH, 3 times (see Note 13) (Fig. 1).
4. Dilute each biotinylated glycoside to a concentration of
10 fmol/mL in running buffer.
5. Change the flow rate to 2 μL/min.
6. Change the flow path to fc1.
7. Inject 12 μL of the control glycoside arabinose on fc1.
8. Change the flow path first to fc2, then to fc3 and then to fc4
(see Note 14).
9. Stop the sensorgram and run Standby for 3–4 h or overnight to
stabilize the baseline (see Note 15).
3.3 Galectin Binding 1. To test the ligand surface, run Prime before injecting the
analyte.
2. First run a few pilot experiments in manual mode to test the
prepared surfaces.
3. Start a new sensorgram at a flow rate of 30 μL/min.
4. Set the flow path to fc1, 2, 3, 4, with inline reference subtrac-
tion of 2–1, 3–1, 4–1 (see Note 16).
5. Make three serial dilutions of Galectin-1 (or one of the analytes
being tested; a wild-type protein if testing binding mutants).
6. Inject 30 μL of the lowest concentration over the sensor chip
(see Notes 17 and 18).
7. Inject the remaining concentrations in turn, waiting for the
protein from each prior injection to be completely washed away
before the next injection.
8. Examine the subtracted sensorgrams showing the 2–1, 3–1,
and 4–1 profiles. Dose-dependent specific binding on each
surface indicates that the biotinylated glycosides are functional,
and the surfaces are ready for the kinetic experiments.
9. For the affinity measurements, make a set of serial dilutions
from 100 to 0.1 μM; for Galectin-1, Galectin-2, and Galectin-
3.
10. Write out a customized wizard setting the flow rate to 60 μL/
min, a flow path of fc1, 2, 3, 4 and a detection readout for
fc2–1, 3–1, 4–1 (see Note 19).
11. Specify sample injections for 60 μL using the Kinject com-
mand, allowing adequate dissociation time for the response
to come back to baseline after each injection.
12. Check the boxes to Prime the instrument before starting the
wizard and to run the Standby routine once all the injections
have been completed.
13. Start the wizard and enter a file name for saving the results.
14. Each Galectin concentration will be recorded as a new cycle

with a new sensorgram.
3.4 Data Analysis 1. Using the BIAevaluation software, open the results file from
the kinetic experiment.
2. Import the 2–1, 3–1, and 4–1 subtracted sensorgrams for all
three Galectins.
3. Overlay all sensorgrams for Galectin-1 (fc2–1) and determine
the apparent Kd as described below (Fig. 2).
4. Select a 15-s section of the curves just before the end of the
injection, representing the steady-state binding.
5. Get the average value for each concentration of Galectin-1,
using Fit: Average.
6. Using this average response for each concentration, in the
menu bar, select Plot Parameters and generate a plot of
Response Units vs. Concentration.
7. Select all the points on the plot. In the menu bar, select Fit:
Steady State Analysis (see Note 20) (Fig. 3).
8. The table below the curve fitting gives the apparent KD value
and other statistics associated with the curve fitting (Fig. 4).
9. Repeat analysis for Galectins-2 and -3.
4 Notes
1. All aqueous reagents were made with MilliQ water. Common

reagents were all of analytical grade. All buffers and other
solutions meant for the BIAcore were filtered using 0.45-μm
filter units. After filtration was complete, the buffers were sub-
jected to vacuum for an additional 20 min to effectively
degas them.
2. Buffers should be thoroughly degassed prior to using them on
the BIAcore. Even though air bubbles are not likely to cause
damage due to pressure changes, the small diameter of the
channels in the IFC can be easily blocked by a large air bubble,
causing a spike in the sensorgram, thereby interfering with the
signal.
3. It is advantageous to couple or capture proteins on the sensor
chip such that all molecules are uniformly oriented and hence
present a homogenous ligand surface. Attaching a biotin label
to one end of a glycoside helps to orient all molecules in the
same uniform presentation. A pre-coated SA chip can be pur-
chased or a streptavidin chip can be made by covalently cou-
pling streptavidin to a CM5 sensor chip.
36000
Response (RU)
34000
32000
2
30000
0 1000 2000 3000 4000
Time (s)
Fig. 2 Ligand capture on an SA chip. Sensorgram shows a typical profile for

capturing ligand on an SA chip. The first three injections (1) represented by a
large increase in response are for the 1 M NaCl/50 mM NaOH preconditioning
solution. Subsequent injection of the glycoside ligand (2) produces a very small
response
400
300
Response (RU)
200
100
0
60 80 100 120 140
Time (s)
Fig. 3 Specific binding of Galectin-1 to (LacNAc)2. An overlay of binding

sensorgrams, obtained after subtracting nonspecific binding on the control
glycoside, for each concentration of Galectin-1
4. All 7-mm plastic vials are supplied with caps and should be
capped to prevent evaporation of sample. Sample evaporation
changes the concentration and hence the refractive index of the
sample solution, thereby altering the magnitude of the signal
obtained.
5. Although several pre-made buffers are available from GE for
using as running buffers on the BIAcore, we preferred to use
phosphate-buffered saline (PBS) in which the galectins were
known to be stable. The PBS was filtered using a 0.45-μm filter
unit and P20 (GE Healthcare), which is a 10% filtered solution
400
300
Specific Binding (RU)
200
100
0
0.0 20.0 40.0 60.0 80.0 100.0
Galectin-1 (µM)
Fig. 4 Affinity of binding of Galectin-1 to (LacNAc)2. Average specific binding responses were plotted against
each Galectin-1 concentration. The equilibrium affinity of binding was determined from nonlinear curve fitting
of this data
of Tween 20, was added to a final concentration of 0.05% to

generate the running buffer (see Note 4).
6. The addition of a small amount of Tween 20 is highly recom-
mended to prevent nonspecific adsorption to the IFC that
delivers sample to the sensor surface.
7. The B3000 instrument should be routinely maintained accord-
ing to the manufacturer’s instructions so that no additional
cleaning is required prior to the start of the experiment.
8. Desorb and Sanitize routines should be performed on a weekly
basis to ensure that there is no build-up in the IFC. For multi-
user instruments, a more rigorous maintenance schedule may
be required depending on the frequency of usage and the
nature of samples used.
9. Prior to docking a brand new sensor chip, always ensure that
the chip is pushed all the way into the cassette assembly, as even
a slight protrusion of the chip can interfere with the docking
process.
10. Take 10 mL of running buffer in a separate tube before insert-
ing the tubing into the container. This tube of buffer will be
useful for diluting analyte samples.
11. The Prime routine should be used whenever the signal quality
is erratic or seen to deteriorate suddenly. Prime should always
be used each morning and whenever the buffer is changed. It is
always recommended to Prime the system twice before starting
up a new experiment aimed at collecting data for kinetic

analysis.
12. There are several different injection options available for use on
the BIAcore 3000. Typically, injections that reduce dispersion
from the buffer require a larger volume of sample. For pilot
tests or for samples that are in limiting amounts, use the
Quickinject command. For kinetic experiments and when the
dissociation phase is to be monitored, use the Kinject
command.
13. The SA chip is convenient because it comes pre-coupled with
streptavidin. However, before a biotinylated ligand is injected,
it is necessary to precondition the sensor surface in order to
remove any loosely bound streptavidin. This ensures that
subsequent injection of the biotinylated ligand will deliver it
only to the immobilized streptavidin.
14. Each time, inject a different glycoside on the respective flow
cell. Inject 12 μL first and then if the ligand density is lower
than the control flow cell, inject some more ligand to get
approximately the same molar amount as the control glycoside.
The manual inject command may be used to control ligand
density.
15. Especially when a surface is prepared for running kinetic
experiments, or when the magnitude of the signal obtained is
likely to be low, it is advisable to stabilize the baseline before
injecting analyte.
16. When immobilizing several ligands, it is always recommended
that the control ligand be immobilized on fc1, upstream of the
remaining ligands to be tested. If ligands are to be tested in
pairs, the control ligand can be immobilized on fc1 and fc3; in
which case, when collecting data, detection should be set to
fc2–1 and fc4–3.
17. Galectin-1 dissociates rapidly from the sensor chip, hence
regenerating the surface is not necessary for this analyte.
18. In this instance, when evaluating galectin binding to glycoside
ligands, the bound galectin dissociated fairly quickly, and it was
not necessary to regenerate the surface. However, when regen-
eration is required, it should be carried out with the mildest
solution and for the shortest time necessary. This will ensure
that the bound analyte is completely dissociated and the ligand
functionality has been minimally affected.
19. Analyte injections can be performed at flow rates ranging from
1 to 100 μL/min. For experiments where the data is to be
analyzed for determining kinetic parameters, however, the
highest flow-rate possible should be selected. The flow rate
should be at least 30 μL/min and will depend on the amount

of analyte that is available.
20. Kinetic analysis for a 1:1 binding model using the BIAevalua-
tion software uses the Levenberg–Marquardt algorithm for
nonlinear curve fitting.
Acknowledgments
The author thanks Drs. Sean Stowell, Richard Cummings, and

Richard Alvarez for helpful discussions during the course of this
study. The BIAcore 3000 instrument used in this study was pur-
chased with an NIH shared instrumentation grant awarded to
Rodger P. McEver. The study was carried out in the BIAcore
Core Facility at the Oklahoma Medical Research Foundation.
References
1. Malmqvist M, Karlsson R (1997) Biomolecular insights into galectin functions. Methods Mol
interaction analysis: affinity biosensor technol- Biol 1207:1–35. https://doi.org/10.1007/
ogies for functional analysis of proteins. Curr 978-1-4939-1396-1_1
Opin Chem Biol 1(3):378–383 8. Arthur CM, Rodrigues LC, Baruffi MD, Sulli-
2. Morton TA, Myszka DG (1998) Kinetic analy- van HC, Heimburg-Molinaro J, Smith DF,
sis of macromolecular interactions using sur- Cummings RD, Stowell SR (2015) Examining
face plasmon resonance biosensors. Methods galectin binding specificity using glycan micro-
Enzymol 295:268–294 arrays. Methods Mol Biol 1207:115–131.
3. O’Shannessy DJ, Brigham-Burke M, Soneson https://doi.org/10.1007/978-1-4939-
KK, Hensley P, Brooks I (1993) Determina- 1396-1_8
tion of rate and equilibrium binding constants 9. Kamili NA, Arthur CM, Gerner-Smidt C,
for macromolecular interactions using surface Tafesse E, Blenda A, Dias-Baruffi M, Stowell
plasmon resonance: use of nonlinear least SR (2016) Key regulators of galectin-glycan
squares analysis methods. Anal Biochem interactions. Proteomics 16(24):3111–3125.
212(2):457–468 https://doi.org/10.1002/pmic.201600116
4. Cerliani JP, Stowell SR, Mascanfroni ID, 10. Stowell SR, Qian Y, Karmakar S, Koyama NS,
Arthur CM, Cummings RD, Rabinovich GA Dias-Baruffi M, Leffler H, McEver RP, Cum-
(2011) Expanding the universe of cytokines mings RD (2008) Differential roles of galectin-
and pattern recognition receptors: galectins 1 and galectin-3 in regulating leukocyte viabil-
and glycans in innate immunity. J Clin Immu- ity and cytokine secretion. J Immunol
nol 31(1):10–21. https://doi.org/10.1007/ 180(5):3091–3102
s10875-010-9494-2 11. Sorme P, Kahl-Knutsson B, Huflejt M, Nilsson
5. Leffler H, Carlsson S, Hedlund M, Qian Y, UJ, Leffler H (2004) Fluorescence polarization
Poirier F (2004) Introduction to galectins. as an analytical tool to evaluate galectin-ligand
Glycoconj J 19(7–9):433–440. https://doi. interactions. Anal Biochem 334(1):36–47.
org/10.1023/B:GLYC.0000014072. https://doi.org/10.1016/j.ab.2004.06.042
34840.04 12. Hirabayashi J, Hashidate T, Arata Y, Nishi N,
6. Leppanen A, Arthur CM, Stowell SR, Cum- Nakamura T, Hirashima M, Urashima T,
mings RD (2015) Examination of whole cell Oka T, Futai M, Muller WE, Yagi F, Kasai K
galectin binding by solid phase and flow cyto- (2002) Oligosaccharide specificity of galectins:
metric analysis. Methods Mol Biol 1207: a search by frontal affinity chromatography.
91–104. https://doi.org/10.1007/978-1- Biochim Biophys Acta 1572(2–3):232–254.
4939-1396-1_6 https://doi.org/10.1016/s0304-4165(02)
7. Arthur CM, Baruffi MD, Cummings RD, 00311-2
Stowell SR (2015) Evolving mechanistic
13. Dam TK, Gabius HJ, Andre S, Kaltner H, leukocytes requires expression of complex-
Lensch M, Brewer CF (2005) Galectins bind type N-glycans. Glycobiology 18(10):770–778
to the multivalent glycoprotein asialofetuin 21. Stowell SR, Cho M, Feasley CL, Arthur CM,
with enhanced affinities and a gradient of Song X, Colucci JK, Karmakar S, Mehta P,
decreasing binding constants. Biochemistry Dias-Baruffi M, McEver RP, Cummings RD
44(37):12564–12571. https://doi.org/10. (2009) Ligand reduces galectin-1 sensitivity
1021/bi051144z to oxidative inactivation by enhancing dimer
14. Stowell SR, Arthur CM, Mehta P, Slanina KA, formation. J Biol Chem 284(8):4989–4999.
Blixt O, Leffler H, Smith DF, Cummings RD https://doi.org/10.1074/jbc.M808925200
(2008) Galectin-1, -2, and -3 exhibit differen- 22. Ahmad N, Gabius HJ, Sabesan S, Oscarson S,
tial recognition of sialylated glycans and blood Brewer CF (2004) Thermodynamic binding
group antigens. J Biol Chem studies of bivalent oligosaccharides to
283(15):10109–10123. https://doi.org/10. galectin-1, galectin-3, and the carbohydrate
1074/jbc.M709545200 recognition domain of galectin-3. Glycobiol-
15. Carlsson S, Oberg CT, Carlsson MC, ogy 14(9):817–825. https://doi.org/10.
Sundin A, Nilsson UJ, Smith D, Cummings 1093/glycob/cwh095
RD, Almkvist J, Karlsson A, Leffler H (2007) 23. Brewer CF, Miceli MC, Baum LG (2002)
Affinity of galectin-8 and its carbohydrate rec- Clusters, bundles, arrays and lattices: novel
ognition domains for ligands in solution and at mechanisms for lectin-saccharide-mediated cel-
the cell surface. Glycobiology 17(6):663–676. lular interactions. Curr Opin Struct Biol
https://doi.org/10.1093/glycob/cwm026 12(5):616–623
16. Brewer CF (2004) Thermodynamic binding 24. Arthur CM, Cummings RD, Stowell SR
studies of galectin-1, -3 and -7. Glycoconj J (2014) Using glycan microarrays to under-
19(7–9):459–465. https://doi.org/10.1023/ stand immunity. Curr Opin Chem Biol
B:GLYC.0000014075.62724.d0 18C:55–61. https://doi.org/10.1016/j.cbpa.
17. Earl LA, Bi S, Baum LG (2011) Galectin multi- 2013.12.017
merization and lattice formation are regulated 25. Stowell SR, Arthur CM, McBride R, Berger O,
by linker region structure. Glycobiology Razi N, Heimburg-Molinaro J, Rodrigues JP,
21(1):6–12. https://doi.org/10.1093/ Noll AJ, von Gunten S, Smith DF, Knirel YA,
glycob/cwq144 Paulson JC, Cummings RD (2014) Microbial
18. Stowell SR, Arthur CM, Dias-Baruffi M, glycan microarrays define key features of host-
Rodrigues LC, Gourdine JP, Heimburg- microbial interactions. Nat Chem Biol
Molinaro J, Ju T, Molinaro RJ, Rivera- 10(6):470–476
Marrero C, Xia B, Smith DF, Cummings RD 26. Robinson BS, Arthur CM, Kamili NA, Stowell
(2010) Innate immune lectins kill bacteria SR (2018) Galectin regulation of host micro-
expressing blood group antigen. Nat Med bial interactions. Trends Glycosci Glycotechnol
16(3):295–301. https://doi.org/10.1038/ 30(172):SE185–SE198
nm.2103 27. Mehta P, Cummings RD, McEver RP (1998)
19. Poland PA, Rondanino C, Kinlough CL, Affinity and kinetic analysis of P-selectin bind-
Heimburg-Molinaro J, Arthur CM, Stowell ing to P-selectin glycoprotein ligand-1. J Biol
SR, Smith DF, Hughey RP (2011) Identifica- Chem 273(49):32506–32513
tion and characterization of endogenous galec- 28. Stowell SR, Dias-Baruffi M, Penttila L,
tins expressed in Madin Darby canine kidney Renkonen O, Nyame AK, Cummings RD
cells. J Biol Chem 286(8):6780–6790. https:// (2004) Human galectin-1 recognition of
doi.org/10.1074/jbc.M110.179002 poly-N-acetyllactosamine and chimeric poly-
20. Karmakar S, Stowell SR, Cummings RD, saccharides. Glycobiology 14(2):157–167.
McEver RP (2008) Galectin-1 signaling in https://doi.org/10.1093/glycob/cwh018
Chapter 8
Revealing the Identity of Human Galectin-3 as a

Glycosaminoglycan-Binding Protein
Jared L. Edwards, Priyanka D. Kadav, Purnima Bandyopadhyay,
and Tarun K. Dam
Abstract
Human galectin-3 (Gal-3) is a β-galactoside-binding lectin. This multitasking protein preferentially inter-
acts with N-acetyllactosamine moieties on glycoconjugates. Specific hydroxyl groups (4-OH, 6-OH of
galactose, and 3-OH of glucose/N-acetylglucosamine) of lactose/LacNAc are essential for their binding to
Gal-3. Through hemagglutination inhibition, microcalorimetry, and spectroscopy, we have shown that
despite being a lectin, Gal-3 possesses the characteristics of a glycosaminoglycan (GAG)-binding protein
(GAGBP). Gal-3 interacts with sulfated GAGs [heparin, chondroitin sulfate-A (CSA), -B (CSB), and -C
(CSC)] and chondroitin sulfate proteoglycans (CSPGs). Heparin, CSA, and CSC showed micromolar
affinity for Gal-3, while the affinity of CSPGs for Gal-3 was much higher (nanomolar). Interestingly,
CSA, CSC, and a bovine CSPG, not heparin and CSB, were multivalent ligands for Gal-3, and they formed
reversible noncovalent cross-linked complexes with the lectin. Binding of sulfated GAGs to Gal-3 was
completely inhibited when Gal-3 was preincubated with β-lactose. Cross-linking of Gal-3 by CSA, CSC,
and the bovine CSPG was also reversed by β-lactose. These findings strongly suggest that GAGs primarily
occupy the lactose/LacNAc binding site of Gal-3. Identification of Gal-3 as a GAGBP should help to reveal
new functions of Gal-3 mediated by GAGs and proteoglycans. The GAG- and CSPG-binding properties of
Gal-3 make the lectin a potential competitor/collaborator of other GAGBPs such as growth factors,
cytokines, morphogens, and extracellular matrix proteins.
Key words Galectin-3, Glycoproteins, Glycosaminoglycans, Proteoglycans, Isothermal titration calo-

rimetry, Cross-linking
1 Introduction
Glycan-binding proteins (GBPs) belong to two distinct groups,

namely lectins and glycosaminoglycan (GAG) binding proteins
(GAGBPs) [1, 2]. From human lectins to viral hemagglutinins,
lectins comprise a large group of evolutionarily conserved ubiqui-
tous proteins. On the other hand, GAGBPs represent a variety of
structurally unrelated proteins including growth factors, cytokines,
morphogens, many extracellular matrix (ECM) proteins, nuclear
proteins, and pathogen surface proteins [3–6]. Lectins interact with
137
138 Jared L. Edwards et al.
Fig. 1 Structures of glycosaminoglycans and a proteoglycan. (a) Heparin, (b) smaller heparin fragment, (c)
chondroitin sulfate, (d) chondroitin sulfate disaccharide, (e) chondroitin sulfate proteoglycan (CSPG)
N- and O-linked glycans of glycoproteins and the glycan moieties of

glycolipids, but GAGBPs recognize free and proteoglycan-linked
GAGs including heparin/heparan sulfate (HS) and chondroitin/
dermatan sulfates including chondroitin sulfate A (CSA), chondroi-
tin sulfate B (CSB) or dermatan sulfate, and chondroitin sulfate C
(CSC). Most of the GAGs are negatively charged linear polysac-
charides that possess repeating disaccharide units (Fig. 1)
[GlcNAc6Sα1–4GlcA/IdoA2S (for heparin/HS);
GlcAβ1–3GalNAc4S (for CSA); IdoAα1–3GalNAc4S (for CSB);
GlcAβ1–3GalNAc6S (for CSC)] [1, 3]. The majority of GAGs are
sulfated. Compared to GAGBPs, the binding sites of lectins show
strict stereospecificity, but, unlike the binding sites of GAGBPs,
they do not possess a patch of positively charged amino acids. While
GAGBPs bind to GAGs with micromolar to nanomolar affinity,
lectins, in general, show weaker affinity (in the millimolar range) for
their monovalent ligands [2]. Such low affinity is enhanced by
multivalency. Multivalent lectin–ligand interactions not only
amplify the binding affinity but also lead to cross-linked complex
formation, which is essential for most lectin-mediated biological
functions including cell signaling. Although GAGBPs often
undergo ligand-mediated oligomerization, they are unlikely to
form extensive cross-linked complexes with their ligand (GAGs
and proteoglycans). Thus, lectins and GAGBPs represent two fun-
damentally different groups of GBPs. In general, members of one
Revealing the Identity of Human Galectin-3 as a Glycosaminoglycan-Binding. . . 139
group do not show the characteristics of the other. However, the

human β-galactoside-specific lectin galectin-3 (Gal-3) is a notable
exception to this rule [7].
Gal-3 is a bona fide lectin with no known characteristics of a
GAGBP. It recognizes neutral galactosyl moieties, mostly N-acet-
yllactosamine (LacNAc) residues of glycan chains, on glycopro-
teins. Most biological interactions of Gal-3 have been shown to
occur through its binding to N- and O-linked glycans of glycopro-
teins. Three specific hydroxyl groups (4-OH, 6-OH of galactose,
and 3-OH of glucose/N-acetylglucosamine) of lactose/LacNAc
are essential for binding to Gal-3 [8]. Therefore, it was unexpected
that Gal-3 could bind to sulfated GAGs and proteoglycans through
its glycan binding site. However, we have recently shown that Gal-3
indeed possesses the binding properties of GAGBPs [7]. Gal-3
recognizes sulfated GAGs (heparin, CSA, CSB, and CSC) as well
as chondroitin sulfate proteoglycans (CSPGs) at physiological pH
and ionic concentration. In addition, interactions of Gal-3 with
CSA, CSC, and a CSPG lead to the formation of non-covalent
cross-linked complexes (Fig. 2). Such complex formation could
be reversed by lactose.
We performed hemagglutination inhibition assays to screen the
affinities of proteoglycans, sulfated GAGs [heparin, HS, CSA, CSC,
and CSB (dermatan sulfate)], and some of their constituent
mono- and disaccharides for recombinant human Gal-3 [7]. We
then elucidated the binding thermodynamics by isothermal titra-
tion calorimetry (ITC). Formation of noncovalent cross-linking of
Gal-3 by GAGs and proteoglycans was monitored by quantitative
precipitation assays and by precipitation kinetics studies [7]. Pro-
teoglycans are found as transmembrane proteins, in secretory gran-
ules, and as extracellular matrix components. They and free GAG
play important roles in many cellular events. The discovery of
GAG-binding properties of Gal-3 significantly changes our percep-
tion of Gal-3.
2 Materials
2.1 Preparation of E. 1. E. coli (BL21 DE3) transformed with a plasmid (commercially

coli Cell Lysates obtained from ATUM, Newark, CA, USA) expressing human
Containing Human Gal-3.
Galectin-3 (Gal-3) 2. Luria Bertani media (Fisher Scientific, Millipore Sigma).
3. Tris–HCl (pH 7.5) with 5 mM EDTA.
4. 20 mM phosphate-buffered saline (PBS), pH 7.4.
5. Beakers of various sizes.
6. Erlenmeyer flasks.
7. Graduate cylinders of various sizes.
Fig. 2 Schematic representation of Gal-3-mediated cross-linking of CSPG. A CSPG contains multiple

chondroitin sulfate chains covalently linked to a core protein. Gal-3 engages in multivalent binding with
those chains. The binding is both intermolecular and intramolecular. Intermolecular multivalent binding by
Gal-3 cross-links CSPG molecules and forms complexes
8. Spatula.
9. 50- and 15-mL centrifuge tubes.
10. Magnetic stir bars.
11. Magnetic stir plate.
12. Standard analytical balance.
13. Sonicator (Fisher Scientific).

15. Protease inhibitor cocktail (Millipore Sigma).
2.2 Preparation of 1. Unwashed rabbit blood from rabbits maintained at Michigan

2% v/v Red Blood Cell Tech Animal Care Facility.
(RBC) Suspension 2. Glass pipette.
3. 15-mL centrifuge tubes.
5. Micropipette.
2.3 Detection of 1. 2% v/v RBC suspension.

Lectin Activity of Gal-3 2. Micropipette.
in the Crude Cell
Extracts
4. 20 mM PBS buffer (20 mM phosphate buffer with 150 mM
NaCl, pH 7.4).
5. 96-well U-bottom microtiter plates (Thermo Fisher Scientific,
Cat #262162).
6. Crude extract of human Galectin-3 (from section described
above).
2.4 Purification of 1. Lactose-agarose matrix (Millipore Sigma).

Gal-3 by Affinity 2. 50-mL plastic syringe (to make the column).
Chromatography
3. Tubing.
4. Stands to attach the column.
5. Glass wool.
6. PBS.
7. Crude cell extract (from Subheading 3.1).
8. 200 mM lactose (from Millipore Sigma) in PBS.
2.5 Dialysis of Gal-3 1. Dialysis tubing, flat width 1.0 in., length 6–8 in., MWCO
for Removal of Lactose 8000 Da.
2. PBS.
3. Magnetic stirrer and magnetic bars.
4. Beakers.
5. 2-L conical flasks.
2.6 Determination of 1. Shimadzu UV Spectrophotometer.

the Concentration of 2. Quartz cuvettes.
Purified Gal-3
3. PBS.
4. Micropipette and micropipette tips.

5. Calculator.
6. Wash bottle with de-ionized water.
2.7 Hemagg- 1. Ligands (CSA, CSB, CSC, Proteoglycans (MilliporeSigma)).

lutination Inhibition 2. 2% v/v RBC suspension.
Assays
4. 96-well U-bottom Nunc brand microtiter plates.
5. Micropipette.
2.8 Isothermal 1. GE/Microcal VP-ITC unit.

Microtitration 2. Hamilton syringe.
Calorimetry (ITC) of
3. Purified Gal-3.
Gal-3 with
Glycosaminoglycans 4. Ligands (CSA, CSC, proteoglycan (MilliporeSigma)).
(GAGs) and 5. Origin software(GE/Microcal).
Proteoglycan 6. Vortex.
8. Shimadzu UV Spectrophotometer.
2.9 UV-Vis 1. Purified Gal-3.

Spectroscopic 2. Ligands (CSA, CSC, proteoglycan, lactose from
Analysis of Gal-3 and MilliporeSigma).
GAGs Precipitation
3. Vortex.
(Complex Formation
Through Noncovalent
4. Centrifuge.
Cross-linking) and 5. Shimadzu UV Spectrophotometer.
Testing the Reversible 6. Micropipette and microtips.
Nature of This Cross- 7. 1.5-mL microcentrifuge tubes.
linking
8. Kimwipes.
9. Lactose solution (200–400 mM in 20 mM PBS, pH 7.4).
3 Methods
3.1 Preparation of E. 1. Centrifuge E. coli (BL21 DE3) transformed with a plasmid

coli Cell Lysates expressing human Gal-3 [7] at 1957 g for 30 min at 4 C
Containing Human in order to obtain the cell pellets from 1 to 3 L of culture. This
Galectin-3 (Gal-3) specific g force was found to be most efficient for pelleting the
cells.
2. Sonicate the cell pellets in 20–60 mL of sonication buffer
[22 mM Tris–HCl (pH 7.5) with 5 mM EDTA] containing
protease inhibitor cocktail (10 μL/mL of sonication buffer).

The sonication should be done on ice and at 50% amplitude for
8 cycles of 30 s each with 1-min interval in between 2 cycles.
3. Centrifuge the sonicated cell lysate at 1957 g for 30 min at
4 C. Collect the supernatant. This g value was used to remove
larger cell debris formed after sonication.
4. Centrifuge the supernatant again at 9469 g for 30 min at
4 C. Collect the supernatant. This g value was used to remove
the residual finer particles.
5. Use this supernatant for affinity chromatography using a
lactose-agarose column.
3.2 Methods 1. Fill six 15-mL centrifuge tubes with 10 mL of PBS buffer.
3.2.1 Washing the Red 2. Using a glass pipette, remove 500 μL of unwashed blood cells
Blood Cells from the blood collection tube.
3. Pipette the 500 μL of blood into each of the six tubes contain-
ing PBS (see Note 1).
4. Gently invert every tube several times to make sure the blood is
thoroughly mixed into the PBS.
5. Centrifuge the tubes on 350 g for 10 min at 4 C.
6. Carefully remove the tubes from the centrifuge without dis-
turbing the packed red blood cells (RBCs).
7. Carefully decant the PBS without pouring out the RBCs.
8. Pour PBS back into the tube to the 10 mL line.
9. Gently invert the tubes several times to mix the RBC pellet in
the buffer and place them back in the centrifuge.
10. Repeat steps 5–9 four more times.
11. On the final wash, remove the PBS from the tubes using the
glass pipette. Do NOT disturb the packed RBCs.
12. Store the washed RBCs at 4 C.
3.2.2 Making a 2% v/v 1. Fill a 15-mL centrifuge tube with 9.8 mL of PBS buffer.
RBC Suspension 2. Remove one tube of the packed RBCs from 4 C.
3. Pipette 200 μL of packed RBCs into the 15-mL centrifuge tube
using a micropipette.
(a) Pipette up and down several times until there is no longer
any RBCs adhering to the wall of the pipette tip.
4. Invert the centrifuge tube several times and store at 4 C.
3.3 Methods 1. Pipette 25 μL of PBS buffer into 13 wells of the 96-well round
bottom plate using a micropipette.
3.3.1 Activity Assay
2. Serial dilute (twofold serial dilution) 25 μL of the crude extract

up to 12 wells.
3. Add 25 μL of 2% v/v RBC suspension to all 13 wells.
4. Gently shake the plate a few times.
5. Let the plate sit at room temperature undisturbed for
45–60 min.
6. Visualize the absence or presence of lectin activity (agglutina-
tion) and note the results. Formation of a button at the bottom
of the titer plate wells indicates an absence of lectin activity
(agglutination), whereas a smear-like profile is the signature of
lectin activity (agglutination), as shown in Fig. 4. The total
numbers of wells that show the sign of lectin activity is called
the titer value.
3.4 Methods 1. Pack a lactose-agarose column (25 mL of matrix, average yield

could be 170 mg of Gal-3 per purification cycle).
3.4.1 Affinity purification
of Gal-3 2. Wash the prepared column with 500 mL of PBS. Washing is
done at room temperature to ensure proper removal of impu-
rities that are nonspecifically bound to the column matrix.
Then store the column at 4 C.
3. Load the column with Gal-3 containing crude cell extract
(from Subheading 3.1 described above) at 4 C. While loading
the crude on the column, collect and discard the first few
milliliters (approximately one third of the bed volume) of the
runoff crude. Then start collecting 1.5-mL fractions in glass
tubes. Check the activity of these fractions by hemagglutina-
tion to make sure active sample fractions are not lost. Less or no
activity of these runoff fractions indicate proper binding of
Gal-3 to the lactose-agarose matrix in the column.
4. After loading, leave the column at 4 C overnight. This ensures
effective binding of Gal-3 to the column.
5. Next day, wash the column with ice-cold buffer (PBS) at 4
C. Wash the column with ~500 mL of PBS buffer. Then collect
a 2 mL fraction of the runoff and check the absorbance at
280 nm. An absorbance lower than 0.010 indicates that the
column is free of all the impurities, only Gal-3 is bound to the
matrix and the column is ready for elution.
6. Load the column with ice-cold 200 mM lactose solution
(in PBS). Collect the eluent during loading and check OD at
280 nm to see if Gal-3 is in the eluent.
7. Keep the column loaded with 200 mM lactose solution at 4 C
overnight.
8. Before elution, place the column at room temperature for ~3 h

to warm up. This will aide in elution, because temperature will
affect the binding interactions of the protein to the column.
9. Elute the column with 200 mM lactose solution at room
temperature. Collect 20–40 fractions (based on the column
volume). Properly label the fractions for easy identification.
10. Check the OD of the collected fractions at 280 nm. Continue
collecting till the absorbance goes down to 0.05.
11. After elution, store the Gal-3 containing fractions at 4 C (or at
80 C for long term preservation of the protein structure).
3.5 Methods 1. Gal-3 was eluted from affinity column with lactose. The eluted
samples must be dialyzed to separate Gal-3 from lactose.
3.5.1 Dialysis of Purified
Gal-3 2. Soak and wash appropriate dialysis tubing with de-ionized
ultrapure water. This removes contaminants or impurities
from the tubing.
3. Once washed, tie the dialysis tubing at one end. Pour purified
Gal-3 into the tubing using a glass pipette. When the tube is
75% filled, tie it off as close to the Gal-3 eluent level as possible.
This limits the dilution of Gal-3 during dialysis.
4. Place the Gal-3 containing dialysis bag in a 2 L flask filled with
PBS and place that on a magnetic stirrer. Dialyze the secured
Gal-3 filled dialysis bag at 4 C. Change the PBS every 4 h until
the lactose concentration reaches sub-nano-molar level (after
every buffer change, the concentration of residual lactose in the
tubing was calculated from the following relationship: initial
lactose concentration of the Gal-3 sample in the tubing
volume of the Gal-3 sample in the tubing ¼ C volume of
buffer in which the tubing is being dialyzed. C is the final
lactose concentration the Gal-3 sample in the tubing). The
number of changes necessary will vary depending upon the
volume of Gal-3 in the dialysis tubing and the volume of the
buffer in which the tubing is placed. Roughly 6–8 changes will
be needed.
5. When dialysis is complete, remove the dialysis tubes from the
flask and release the contents into appropriate containers.
When all Gal-3 is obtained and pooled, determine the protein
concentration at 280 nm.
3.6 Methods 1. Take 100 μL of purified Gal-3 and add 900 μL of PBS to it.
3.6.1 Determination of 2. Check the O.D. at 280 nm.
Gal-3 Concentration 3. Perform the following equation:

OD per mL
0:61 mg1
100
3 ¼ μM Gal3 (extinction coefficient of
Gal-3 is 6.1).
4. Depending upon the experiment, dilute to desired concentra-

tion and reassess the concentration as described above. The
final concentration of Gal-3 is adjusted to 10–100 μM, based
on the type of experiments.
5. If any experiment needs a higher concentration of Gal-3, it can
be done by using Amicon Filters. Wash a 10 kDa filter with
PBS. Place Gal-3 samples into the filter and centrifuge at
1957 g. Continue the process until the desired concentration
is reached.
3.7 Methods 1. Adjust the titer value of the target lectin you are trying to test
to 2–3 agglutination titers.
3.7.1 Inhibition Assay
2. Pipette 25 μL of PBS or HEPES buffer into 8 wells of the
96-well round bottom plate using a micropipette.
3. Pipette 25 μL of PBS or HEPES buffer into 2 additional wells
of the 96-well round bottom plate using a micropipette.
4. Serial dilute (twofold serial dilution) 25 μL of the ligands into
7 of the 8 wells.
5. Pipette 25 μL of the ligand into the eighth well. This will act as
the ligand control. Do NOT serial dilute.
6. Pipette 25 μL of the lectin crude into the seven wells that
contain the serial diluted ligand. Do NOT pipette your lectin
into the eighth well that contains your ligand control.
7. Pipette 25 μL of the lectin into the one of the two additional
wells that contain PBS or HEPES buffer.
30–60 min.
9. Pipette 25 μL of the 2% v/v RBC suspension into all 10 wells.
10. Pipette 25 μL of the 2% v/v RBC suspension in a well contain-
ing only 25 μL of buffer (this well is the RBC and buffer
control).
30–45 min.
12. Visualize and record the results (Fig. 4).
3.8 Methods 1. Wash ITC syringe and the sample cell extensively with deio-
nized ultrapure H2O. Completely flush and remove water after
3.8.1 ITC Studies
washing. This ensures a clean and dry apparatus.
2. Rinse the syringe and the cell with PBS. Make sure that there is
no residual PBS in the cell or syringe.
3. Adjust (based on the experiments) the concentration of pur-
ified Gal-3 with PBS. Fill the cell with required volume of Gal-3
after degassing. Make sure that no air bubbles are trapped in
the cell.
Fig. 3 Inhibition of hemagglutination of rabbit red blood cells (RBCs) by Gal-3

with bovine proteoglycan (A), and CSC (B). The buffer used was PBS, pH 7.4,
with 0.15 M NaCl. The ligands were serially diluted from well 1 to well 8. The red
buttons (accumulated RBCs) at the bottom of the wells indicate inhibition (A1–A5
and B1–B4)
4. Prepare ligands in desired concentrations in PBS. Fill ITC

syringe very carefully. There should not be any empty space in
the syringe.
5. Place the long needle of the syringe properly inside the cell.
Open up the program to run the experiment.
6. Enter proper experimental parameters (Temperature: 27 C,
injection volume: 4–10 μL, time between 2 injections: 4 min).
7. Allow equilibration to occur in chamber.
8. Monitor the quality of the ITC profile and the baseline. Run
should be stopped if the quality of the profiles is not good.
9. Continue injections until saturation.
10. Once the run is complete, analyze data using Origin software
by entering correct concentrations of the ligands and lectins,
and by choosing appropriate number of binding sites per sub-
unit (Fig. 3).
3.9 Methods 1. Take two quartz cuvettes and fill with buffer (PBS) and place
them in the reference and sample chambers. Auto zero at
3.9.1 Cross-linking
420 nm.
Studies
2. After that, discard buffer from the sample cuvette alone.
3. Place between 900 and 1000 μL of Gal-3 in the cuvette.
4. Put the cuvette in the chamber and auto zero at 420 nm again.
5. Remove the cuvette containing Gal-3 from the chamber. Take
the ligand to be tested and quickly pipette 5 μL of it into the
cuvette containing Gal-3. Rapidly cover and invert once the
cuvette, place back into sample chamber of spectroscope.
Record absorbance after each incremental addition of 5 μL
ligand.
6. Repeat this process of addition of ligand, inversion, and record-
ing absorbance until a plateau is reached.
7. Once plateau is reached, stop adding ligands (Fig. 5).
Fig. 4 ITC profile of intact human Gal-3 with CSA (a) at 27 C and pH 7.4. The
data obtained are for automatic injections (4 μL each) of CSA and are shown at
the top panel of (a). The integrated curve showing experimental points and the
best fit are shown in the bottom panel of (a). CSA did not show binding when
injected into a solution of intact Gal-3 preincubated with 200 mM β-lactose in
20 mM PBS, pH 7.4 (b). The binding affinity (Ka) was 37.8 0.9 M1 104,
ΔH was 21.7 0.8 kcal/mol, binding stoichiometry (n) was 0.28 0.01, ΔG
was 7.6 0.02, and TΔS was 14.1 0.8
a
1.4
1.2
OD at 420 nm
0.8
0.6
0.4
0.2
0
0 10 20 30 40 50 60 70 80
μM)
CSA concentration (μ
b
1.2
1
OD at 420 nm
0.8
0.6
0.4
0.2
0
0 5 10 15 20 25 30 35
Lactose concentration (mM)
Fig. 5 CSA cross-links Gal-3 and forms complexes as measured at 420 nm (a).
Turbidity, produced by CSA-Gal-3 cross-linking, is efficiently measured at
420 nm Addition of lactose dissolves the cross-linked complexes, which was
also monitored at 420 nm (b)
8. To test if lactose can reverse the complex formation, add 5 μL

of 200 mM solution incrementally to the precipitates and
record the decreasing OD at 420 nm (Fig. 5).
Acknowledgments
This work was supported by National Science Foundation Grant

1608537 (to T.K.D.). A part of this work was supported by a
startup fund (to P.B.) and by the Research Excellence Fund (to T.
K.D.) provided by Michigan Technological University.
References
1. Esko JD, Kimata K, Lindahl U (2009) Proteo- 5. Xu D, Esko JD (2014) Demystifying heparan
glycans and sulfated glycosaminoglycans. In: sulfateprotein interactions. Annu Rev Biochem
Varki A et al (eds) Essentials of glycobiology, 83:129–157
2nd edn. Cold Spring Harbor Laboratory 6. Lindahl U, Hook M (1978) Glycosaminogly-
Press, New York cans and their binding to biological macromole-
2. Esko JD, Linhardt RJ (2009) Proteins that bind cules. Annu Rev Biochem 47:385–417
sulfated glycosaminoglycans. In: Varki A et al 7. Talaga M, Fan N, Fueri A et al (2016) Multitask-
(eds) Essentials of glycobiology, 2nd edn. Cold ing human lectin galectin-3 interacts with sul-
Spring Harbor Laboratory Press, New York fated glycosaminoglycans and chondroitin
3. Vallen MJ, van der Steen SC, van Tilborg AA sulfate proteoglycans. Biochemistry 55:
et al (2014) Sulfated sugars in the extracellular 4541–4551
matrix orchestrate ovarian cancer development: 8. Cummings RD, Liu FT (2009) Galectins. In:
‘when sweet turns sour’. Gynecol Oncol 135: Varki A et al (eds) Essentials of glycobiology,
371–381 2nd edn. Cold Spring Harbor Laboratory
4. Kitagawa H (2014) Using sugar remodeling to Press, New York
study chondroitin sulfate function. Biol Pharm
Bull 37:1705–1712
Chapter 9
Examining Galectin Binding Specificity Using Glycan

Microarrays
Sean R. Stowell, Lilian C. Rodrigues, Marcelo Dias-Baruffi,
Abstract
Glycan binding proteins (GBPs) possess the unique ability to regulate a wide variety of biological processes
through interactions with highly modifiable cell surface glycans. While many studies demonstrate the
impact of glycan modification on GBP recognition and activity, the relative contribution of subtle changes
in glycan structure on GBP binding can be difficult to define. To overcome limitations in the analysis of
GBP–glycan interactions, recent studies utilized glycan microarray platforms containing hundreds of
structurally defined glycans. These studies not only provided important information regarding GBP–glycan
interactions in general but have also resulted in significant insight into binding specificity and biological
activity of the galectin family. We will describe the methods used when employing glycan microarray
platforms to examine galectin–glycan binding specificity and function.
Key words Glycan binding protein (GBP), Galectin, Glycan microarray, GBP–glycan interactions
1 Introduction
Members of the galectin family of carbohydrate binding proteins

have been implicated in a wide variety of biological processes
including roles in cell signaling, development, and immunity [1–
5]. While galectins initially reside within the cytosol where they can
regulate various intracellular processes [6–8], most documented
activities of galectin family members reflect modulation of cellular
behavior following release from cells, where they recognize and
cross-link highly modifiable cell surface carbohydrates [2, 9,
10]. Although all galectins share a common affinity for
β-galactose terminating carbohydrate structures, variations within
the carbohydrate-binding domains of individual galectin family
members can lead to differential binding activity following
β-galactose modification [11–14]. Thus, knowledge of the specific
binding preferences of individual galectins can provide significant
151
152 Sean R. Stowell et al.
insight into the function and regulatory activity of this protein

family.
Various methods have been employed to determine binding
preferences of galectins. Early studies utilized simple monosacchar-
ides, typically in the setting of hemagglutination inhibition assays,
to understand general binding preferences of glycan binding pro-
teins (GBPs), including individual members of the galectin family
[15–17]. These early studies suggested that galectins typically rec-
ognize terminal β-galactoside containing glycans, although the fine
specificity for these ligands and the potential impact of β-galactose
modification appeared to differ between individual family members
[16, 18]. Subsequent studies utilized a variety of biochemical
assays, including isothermal calorimetry, fluorescence polarization,
surface plasmon resonance, and frontal affinity chromatography
[11, 13, 19–21]. These approaches often used expanded libraries
of glycans, which not only contributed to a greater understanding
of galectin carbohydrate binding properties but also provided sig-
nificant insight into the thermodynamics of galectin–glycan inter-
actions [19, 20].
In an effort to expand on previous findings [15–17, 22], several
groups began to generate larger libraries of test glycans [23–
25]. These libraries typically contained combinations of naturally
occurring and uniquely modified glycans designed to specifically
elucidate the binding specificity of GBPs [21, 23, 24, 26–
29]. Although these libraries can be utilized in a variety of formats,
glycan microarrays allow interrogation of hundreds of structurally
distinct glycans while employing small amounts of target glycan and
test GBPs [11, 20, 30–32]. As synthetic limitations may limit the
overall breadth of a particular glycan library, recent strategies also
generated glycan microarrays from glycans directly harvested from
natural sources [28, 29, 33–39]. Using a combined approach of
different glycan microarray strategies, significant insight into galec-
tin–glycan binding specificity and overall biological function can be
obtained [13, 21, 27, 40–45].
In this work, we will discuss the methods used for derivatizing
and examining galectin–glycan binding specificity using a glycan
microarray format. The potential limitations and impact of galectin
concentration on glycan microarray results and analysis will also be
explored. Given the utility of this tool in elucidating the binding
specificity and discovery of previously unrecognized biological
activities of individual members of the galectin family [13, 27,
43], additional studies using various glycan microarray approaches
will likely continue to provide significant insight into the biochemi-
cal and biological roles of galectins and other GBPs.
Examining Galectin Binding Specificity Using Glycan Microarrays 153
2 Materials
2.1 Galectin 1. Lactosyl-Sepharose (Sigma-Aldrich).

Preparation 2. IPTG (Thermo Scientific).
3. PBS (Hyclone).
4. Lactose (Fisher).
5. β-Mercaptoethanol (Fisher).
6. CelLytic B-II Bacterial Cell Lysis/Extraction reagent (Sigma).
7. Complete Mini EDTA-free Protease Inhibitor cocktail tablets
(Roche).
8. Lysozyme (Sigma Aldrich).
9. RNase (Thermo Scientific).
10. DNase (Thermo Scientific).
11. Sodium azide (Sigma Aldrich).
12. Tris-HCl base.
13. Glycine.
14. Sodium dodecyl sulfate.
15. Loading buffer (Thermo Scientific).
16. Broad Range STD Molecular Weight Markers (Santa Cruz).
17. 4–20% or 16% Tris–Glycine Gel (10 Well) (Life Technologies).
18. Coomassie staining solution (Life Technologies).
19. Methanol.
20. Glacial acetic acid.
21. 1.7 mL snap cap microcentrifuge tube (Sigma-Aldrich).
22. 2-L Erlenmeyer flask.
23. 50-mL Erlenmeyer flask.
24. 1-L centrifuge bottles.
2.1.1 Special Equipment 1. Centrifuge.

2. Sonicator.
3. Fraction collector (BioRad).
4. Gel apparatus including voltage source.
2.1.2 Buffers 1. Lysis buffer (for 6 L of culture). Use 60 mL of CelLytic B-II

Bacterial Cell Lysis/Extraction reagent, 14 mM
β-mercaptoethanol (βME) (67.2 μL), 6 Complete Mini
EDTA-free Protease Inhibitor cocktail tablets, Lysozyme
(6 mg), RNase with a stock of 10 mg/mL (60 μL), and
DNase with a stock of 10 mg/mL (60 μL).
2. Wash buffer (1 L): 1 PBS without Azide with 14 mM βME

(1.12 mL).
3. Elution buffer (1 L): 1 PBS without Azide, 14 mM βME
(1.12 mL) and 100 mM Lactose (36 g).
4. Column storage buffer (total volume 500 mL): 1 PBS with
0.02% Azide, 14 mM βME.
5. 1 Electrophoresis buffer: 3 g Tris base, 14.4 g glycine, and 1 g
sodium dodecyl sulfate in total volume of 1 L dH2O.
6. Destain solution: 40% methanol and 10% glacial acetic acid in
dH2O.
2.2 Galectin 1. PD10 column (GE Healthcare).

Derivitiztion 2. PBS (Hyclone).
3. NHS-LC-Biotin (Thermo-Scientific).
5. β-Mercaptoethanol (Fisher).
6. Lactosyl-Sepharose (Fisher).
2.3 Glycan Array 1. Glycan printed slides (Core D), printed on the side of the slide
Screening with the white etched bar code and black marks (DO NOT
TOUCH THIS AREA).
2. Cover slips, 24 50 (Fisher scientific).
3. Humidified Slide processing chambers (Fisher), or homemade
system made with petri dish, with wet paper towels in the
bottom of the chamber.
4. 100 mL Coplin jars for slide washing.
5. Tris–HCl.
6. NaCl.
7. CaCl2.
8. MgCl2.
9. Potassium phosphate monobasic.
10. dH2O.
11. BSA (Fisher scientific).
12. Alexa Fluor-488-Streptavidin (Invitrogen).
13. Tween-20 (EMD Biosciences).
2.3.1 Special Equipment 1. ProScanArray Scanner (Perkin Elmer).
2.3.2 Buffers 1. TSM ¼ 20 mM Tris–HCl, pH 7.4150 mM NaCl, 2 mM CaCl2,

2 mM MgCl2.
2. TSM Wash Buffer (TSMW) ¼ TSM buffer + 0.05% Tween-20.

3. TSM Binding Buffer (TSMBB) ¼ TSM buffer + 0.05% Tween
20 + 1% BSA.
3 Methods
3.1 Galectin 1. In a laminar flow hood, pour 10 mL of autoclaved LB media

Preparation into a sterile 50-mL flask or other appropriate container. Add
10 μL of ampicillin (stock 50 μg/mL) to media (see Note 1).
2. Remove the galectin glycerol stock from the –80 C freezer and
place on ice. Place a loop full of bacteria into the LB/Amp
media to inoculate starter culture.
3. Place cover on flask loosely and incubate overnight at 37 C,
shaking at 250 revolutions per minute (rpm).
4. Prepare a 2-L Erlenmeyer flasks containing 1 L of LB media by
adding 20 g of LB broth powder in 1 L of dH2O in the flask.
Mix well then autoclave on a 30-min liquid cycle to sterilize.
Then remove from autoclave and allow it to cool to room
temperature.
5. Place the 2-L Erlenmeyer flask containing 1 L LB media into
the incubator/shaker and allow it to warm to 37 C before
proceeding.
6. Once media is warmed, move media to a culture hood and add
1 mL of ampicillin (stock concentration 50 μg/mL) to the 1 L
of LB media in each flask (see Note 1).
7. Add 10 mL of starter culture to the flask containing 1 L of
sterile LB media and antibiotic to inoculate cultures.
8. Place cap on flask loosely and incubate shaking 2–2.5 h at 37 C
and 250 rpm.
9. Check the cultures OD at 600 nm (OD600) using visible light
approximately every 30 min. Blank with autoclaved sterile LB
media. When OD600 of each culture reaches between 0.45 and
0.50, induce the bacteria in each culture to express the recom-
binant galectin by addition 0.36 g of IPTG to each flask (see
Note 2).
10. Continue IPTG-induced growth for 4–5 h at 37 C and
250 rpm (see Note 3).
11. Collect bacteria by pouring 1 L of culture into two (500 mL in
each) clean 1-L plastic centrifuge bottle. Spin down at
4200 g for 30 min at 4 C.
12. Pour off the supernatant and place pelleted bacteria at 80 C.
Pellet can be stored at 80 C for up to 2 weeks.
13. Thaw pellets on ice. Add 10 mL of Lysis Buffer to each bottle

and resuspend pellets (homogenization). Let resuspension sit
at RT for 30 min or at 37–80 C for 15 min (see Note 4).
14. Pool the resuspended pellets in a 250-mL plastic centrifuge
bottle on ice and sonicate the pellets 2–3 times for 20 s per
cycle (see Note 5).
15. Spin the lysate at 13,000 g/4 C for 30 min.
16. Collect the supernatant without disrupting the pelleted cell
debris and place on ice.
3.1.1 Purification 17. Prepare the Lactosyl-Sepharose column by washing with ten
column volumes of wash buffer (see Note 6).
18. Apply the supernatant to the column (see Note 7).
19. Collect the supernatant flow through (Sample Flow
Through) (see Note 8).
20. Wash the column with ten more column volumes of Wash
Buffer (see Note 9).
21. Elute the recombinant protein by preparing three column
volumes of Elution Buffer and adding it to the column.
22. Begin to collect 1–2 mL fractions immediately following
addition of Elution Buffer.
23. Take 10 μL from each fraction and dilute tenfold (i.e., 10 μL
of each fraction in 90 μL Elution Buffer). Read the OD of
each dilution at OD at 280 nm (OD280) using a spectropho-
tometer. Blank with Elution Buffer.
24. Calculate protein concentration using the following equation:
OD280 10 extinction coefficient ratio (see Note 10).
25. Identify the peak fractions (the 3–5 fractions with the highest
concentration of protein as determined by OD280).
26. Prepare samples for an SDS PAGE using 10 μL of the “Start
Material” and “Sample Flow Through” and the volume that
equals 2 μg of total protein for each of the “Peak Fractions.”
Use 7.5 μL of 4 Loading Buffer. Bring up a total load volume
of 30 μL with dH2O. Load 30 μL into each well.
27. Use Broad Range STD Molecular Weight Markers. Total load
volume 10 μL.
28. Over a Bunsen burner, allow a pot of dH2O to come to a boil.
Place samples in the water and boil for 10 min.
29. Set up gel apparatus and fill with 1 Electrophoresis Buffer.
30. Spin down samples for 5 s before loading on Gel.
31. Load samples into a 4–20% or 16% Tris–Glycine Gel (10 Well).
32. Run Gel at 125 V for about 1.5 h.

33. Remove the gel from gel apparatus and stain the gel with
Coomassie Stain for 30 min, then destain for 2 h with destain
solution. Verify the “Peak Fractions” are the appropriate
molecular weight. Take picture of the gel and dry.
34. Once the protein identity is verified by SDS PAGE, pool all of
the fractions that contain galectin protein. Examine protein
content by measuring the OD280 of each fraction. This is
typically done by diluting each fraction tenfold (i.e., 10 μL of
each fraction in 90 μL PBS) followed by the examination of
OD280. Blank with Elution Buffer.
35. Calculate protein concentration using the following equation:
OD280 10 extinction coefficient ratio (see Note 10).
36. Make 500 μL aliquots of the purified galectin protein in 1.7-
mL microcentrifuge tubes. Store aliquots at –80 C.
3.2 Galectin 1. To remove βME, prepare a PD10 column for gel filtration by
Biotinylation equilibration with 5 column volumes of cold PBS (pH 7.4) (see
Note 11).
2. Thaw frozen stock aliquots of galectins at 4 C.
3. Add 1 mL of recombinant galectin solution (containing Elu-
tion Buffer) to the PBS re-equilibrated PD10 column and
collect 0.5 mL fractions.
4. Following complete penetration of the galectin solution into
the PD10 column, add 2 mL cold PBS.
5. Continue to collect 0.5 mL fractions while adding additional
PBS as needed to elute all protein and prevent the column from
drying.
6. To determine which fractions may have galectin protein, exam-
ine protein content by measuring the OD280 of each fraction.
This is typically done by diluting each fraction tenfold (i.e.,
10 μL of each fraction in 90 μL PBS) followed by the examina-
tion of the OD280 (see Note 10).
7. Once positive fractions are identified, pool galectin containing
fractions and re-evaluate the OD280 to determine the final
concentration of pooled galectin (see Note 12).
8. Add lactose in PBS at a final concentration of 100 mM to
facilitate the maintenance of galectin activity during the label-
ing procedure (see Note 13).
9. Add NHS biotin at the concentration recommended by the
manufacturer followed by incubation at RT for 1 h or at 4 C
for 2 h (see Note 14).
10. Remove lactose and non-reacted NHS biotin by passing the

reaction mixture over a new PD10 column, re-equilibrated
with cold PBS, using the same approach as outlined in steps
1–7 above.
11. Once pooled biotinylated galectin fractions are collected and
evaluated for concentration, add βME at a final concentration
of 14 mM to sustain galectin activity for the duration of the
additional purification steps and actual experiment (see
Note 15).
12. To separate potentially inactive galectin, from active galectin,
pass pooled biotinylated galectin over a Lactosyl-Sepharose
column re-equilibrated in PBS containing 14 mM βME.
13. Wash the column with 10 column volumes with PBS contain-
ing 14 mM βME.
14. Elute bound protein with PBS containing 14 mM βME and
100 mM lactose. Collect 0.5 mL fractions as soon as the
elution buffer is added to the column.
15. Evaluate galectin concentration within each collected fraction
by diluting each fraction tenfold in PBS containing βME (i.e.,
10 μL of each fraction in 90 μL PBS). Be sure to use PBS with
βME as a baseline measure of OD280.
16. Once protein-positive fractions are identified, pool galectin
containing fractions and re-evaluate the OD280 nm to deter-
mine the final galectin concentration.
17. The degree of biotin incorporation can be evaluated by com-
monly employed mass spectrometric analysis, Western blot
analysis using HRP-streptavidin or by examining whether
engagement of cell surface ligands by biotinylated galectin
can be detected using FITC-streptavidin using standard flow
cytometric analysis (see Note 16).
18. Once biotinylation is documented on activate recombinant
galectin, these proteins can then be analyzed on the glycan
microarray.
3.3 Microarray 1. Make TSM, TSMW, and TSMBB or bring them to room
Probing with temperature if they have made previously and stored at 4 C.
Biotinylated Galectin 2. Prepare 100 μL of sample by diluting biotin-tagged galectin
protein TSMBB with 14 mM βME added if needed to maintain
galectin activity (see Note 17).
3. Remove slide(s) from desiccator and hydrate by placing in a
glass Coplin staining jar containing 100 mL of TSMW for
5 min.
4. Remove excess liquid from slide by setting the slide upright on

a paper towel to drain the liquid off.
5. Carefully apply 70 μL of sample made in step 2 above close to
the left edge of the slide in between the black marks found on
the slide surface.
6. Slowly lower the cover slip onto the slide and gently remove
any bubbles trapped between the slide and the cover slip by soft
tapping with a pipette tip. Be sure the cover slip stays between
the black marks on the slide since this is the portion of the slide
where glycan is printed.
7. Incubate slide in a humidified tray in the dark for 1 h at RT.
8. After 1 h incubation, remove cover slip by turning the slide
onto its side and allowing the cove slip to slip off into the glass
trash/biohazard trash.
9. To wash, dip the slide 4 times into 100 mL of first TSMW and
then TSM in Coplin Jars.
10. Set the slide upright on a paper towel to remove excess TSM
buffer.
11. Add 70 μL of Streptavidin-AlexaFluor-488 and apply cover slip
as outlined above in step 6 and incubate in a humidified tray in
the dark for 1 h at RT.
12. After 1 h incubation, remove cover slip by turning the slide
onto its side and allowing the cover slip to slip off into the glass
trash/biohazard trash.
13. To wash, dip the slide 4 times into 100 mL of first TSMW
followed by TSM and then dH2O in Coplin Jars.
14. Spin slide in slide centrifuge for ~15 s to remove excess water.
15. Scan in a fluorescent scanner and obtain mean fluorescent
intensities for each glycan binding event on the array (see
Note 18).
4 Notes
1. Use the appropriate antibiotic selection protocol based on the

antibiotic resistance gene encoded in the expression vector for
each recombinant galectin.
2. IPTG can be dissolved directly into 10 mL of sterile autoclaved
media for each 1 L culture. Use the appropriate inducing agent
and concentration as outlined by the manufacture for the
expression vector used to generate recombinant galectin.
3. While induction at 0.4–0.5 OD600 followed by 4–5 h of
continued incubation yields optimal galectin production for
most galectins, variations may occur depending on expression

vector, galectin stability, and other expression parameters. In
these cases, optimal conditions should be empirically
determined.
4. Use of bacterial lytic reagent when seeking to isolate galectin-
2 will result in loss of galectin-2 activity. Instead use PBS when
isolating galectin-2. It should also be noted that several meth-
ods of bacterial lysis have been published and employed by our
lab. Each appears to yield relatively similar overall protein
amounts.
5. Although this may help bacterial lysis and appears to increase
the ultimate protein yield, this step is not absolutely required
when using the bacterial lytic reagent.
6. Assign different Lactosyl-Sepharose columns to the purifica-
tion of a single galectin family member and or mutants to avoid
potential contamination when isolating each protein.
7. Save some of the supernatant to run on an SDS-PAGE Gel as
the “Start Material.”
8. Some of the “Sample Flow Through” will also be ran on an
SDS-PAGE Gel.
9. Collect the “Wash Flow Through” in case the column capacity
is exceeded. Keep the “Wash Flow Through” separate from the
“Sample Flow Through.”
10. To extrapolate the protein concentration from the OD 280 nm
values, use the extinction coefficient for the particular galectin
being examined to calculate the actual concentration in
mg/mL. The websites http://www.basic.northwestern.edu/
biotools/proteincalc.html and http://web.expasy.org/pro
tparam/protparam-doc.html offer explanation and assistance
in calculating extinction coefficient, and this calculation is used
to determine the actual concentration of a given protein in
mg/mL, including caveats as to how these numbers may differ
from the actual number. As methods of calculating the extinc-
tion coefficient only provide estimates, alternative approaches
can be used to empirically determine these values. These
include using a Bradford assay to calculate protein concentra-
tion or simply re-equilibrating the recombinant protein
directly into water, lyophilizing, weighing, and then resus-
pending in the buffer of choice followed by empirically deter-
mining the extinction coefficients for a particular galectin
family member.
11. βME can readily inactivate NHS biotin and therefore signifi-
cantly reduce the efficacy of galectin biotin incorporation.
12. Reaction efficiency is typically a function of the biotin labeling

reagent and the concentration of the protein. We typically label
galectins at a concentration of >1 mg/mL to optimize labeling
efficiency. If the final concentration is not sufficient, galectins
can be readily concentrated using Centricon concentrating
devices (Millipore) according to the manufacturer’s protocol.
13. The galectin will pass over a column at least once more after it is
labeled. Since some galectins can be lost with each pass over a
column, it is important to start with sufficient galectin to
achieve the required amount of end product to conduct the
experiment.
14. Many different labeling strategies can be employed when seek-
ing to detect recombinant galectin binding. However, we have
found that biotinylation is least likely to induce inactivation of
the galectins following derivatization. In contrast, several
galectin readily lose carbohydrate binding activity following
direct labeling with fluorescent adducts. As a result, care should
be taken to determine the potential impact of galectin derivati-
zation on carbohydrate binding activity. The impact of labeling
concentration and duration as well as the type of label should
be empirically evaluated to assess the potential impact of the
label on galectin activity. Partially denatured, yet labeled, galec-
tin can result in nonspecific binding on array platforms and
therefore result in false-positive hits that may significantly
impact the apparent binding preferences for an individual
galectin.
15. Not all galectins are equally sensitive to oxidative inactivation.
For example, Gal-3 does not appear to display significant sen-
sitivity to oxidative inactivation in the absence of reducing
agents. In contrast, Gal-1, Gal-2, and Gal-7 can display signifi-
cant loss of activity over time. Other galectins can also lose
activity in the absence of reducing conditions, albeit at a
reduced rate compared to Gal-1. As a result, the requirement
of βME should be empirically determined for each galectin
following the labeling procedure.
16. While the first two approaches may provide a general sense of
the degree of protein biotinylation and the passage of the
protein over a column would in theory allow isolation of active
protein, the later approach simultaneously tests whether the
galectin is biotinylated and retains gross carbohydrate binding
properties.
17. Cy5-labeled Streptavidin for aligning subarray grids for analysis
can be added as a separate step on the slide before adding
galectin.
18. Although glycan microarrays allow rapid examination of GBP–

glycan interactions on a single platform, several considerations
should be taken when using this approach to study GBP bind-
ing specificity. For example, while microarray presentation of
glycans fixed on a solid support may be analogous to glycans
similarly fixed on a cell surface, they still reflect artificial pre-
sentation of potential ligands. As a result, potential alterations
in printing density, presentation, coupling formats, and meth-
ods of GBP binding detection can impact the overall binding
specificity of a particular GBP. Furthermore, some GBPs, such
as selectins, exhibit complex interactions with glycan ligands
that also include the peptide backbone and other posttransla-
tional modifications [36, 46]. In these situations, more com-
plex libraries of glycopeptides can be used to evaluate the
potential influence of non-glycan moieties on GBP–glycan
interactions [36, 46].
In addition to potential differences in glycan presentation
between individual glycan microarrays and cell surface glycans,
very few arrays possess well-defined concentrations of the gly-
cans actually coupled to the array. As a result, differences in
GBP binding to unique glycans can also be significantly influ-
enced by variation in glycan coupling efficiency. To reduce the
impact of this limitation, analysis of GBP binding over a variety
of concentrations can reduce the impact of alterations in print-
ing density by providing apparent Kd values toward individual
glycans [21, 27]. Although GBPs may be in equilibrium with
many different glycans on a microarray and therefore limit the
overall accuracy of this approach, evaluating the binding of
GBPs at different concentrations still appears to reduce bias
based on alterations in printing efficiency. This is especially
important when Bmax values between individual glycans differ
significantly [21, 27] (Figs. 1, 2, and 3). Regardless of the
microarray approach used, validation of findings using alterna-
tive methods, including evaluation of cell surface glycans, pro-
vides a useful strategy to ensure that array findings reflect real
interactions with native glycan ligands [13, 21, 22, 27, 43].
Acknowledgments

Grants R01 HL154034 to CMA.
Fig. 1 Utilization of defined glycan microarrays to elucidate GBP specificity. Libraries of well-characterized
glycans generated by release of defined glycans from glycoproteins or other natural sources or by chemical or
chemoenzymatic synthesis are used to populate well-defined glycan microarrays. Structures reflect naturally
occurring glycans and modifications of glycans not typically found in nature. Glycan libraries undergo
derivatization with a functional coupling moiety, followed by printing in a microarray format to generate the
glycan microarray. GBPs are incubated with the glycan microarrays over different concentrations and detected
by fluorescence emission if directly labeled or by a similarly labeled suitable secondary detecting agent. While
many approaches can be taken to analyze glycan array data, examination of GBP binding over a variety of
concentrations for individual glycans is shown. This research was originally published in the Journal of
Biological Chemistry. Stowell SR, Arthur CM, Slanina KA, Horton JR, Smith DF, Cummings RD. Dimeric
Galectin-8 induces phosphatidylserine exposure in leukocytes through polylactosamine recognition by the
C-terminal domain.2008 Jul 18;283(29):20547–59 © the American Society for Biochemistry and Molecular
Biology
Fig. 2 Galectins interact with glycan microarray in a concentration-dependent manner. Gal-8 recognizes
distinct classes of glycans. (a, b) Examination of the glycan microarray followed incubation of the glycan
microarray with 6 μM Gal-8 (a) or 0.3 μM Gal-8 (b). (c, d) Glycan microarray data obtained following
incubation with 6 μM Gal-8 (c) or 0.3 μM Gal-8 (d). (d) Inset: Legend of symbols for monosaccharides. This
research was originally published in the Journal of Biological Chemistry. Stowell SR, Arthur CM, Slanina KA,
Horton JR, Smith DF, Cummings RD. Dimeric Galectin-8 induces phosphatidylserine exposure in leukocytes
through polylactosamine recognition by the C-terminal domain.2008 Jul 18;283(29):20547–59 © the Ameri-
can Society for Biochemistry and Molecular Biology
Fig. 3 Examination of galectins over a range of concentrations can provide a relative affinity for individual
ligands on the glycan microarray. Trivial names followed by the structures of each glycan tested are shown.
Recognition of each representative glycan is displayed as the percent bound when compared with the highest
bound ligand at each concentration tested for each respective galectin tested in this study. Glycan recognition
of O-glycans is shown for Gal-1 (a), Gal-2 (b), and Gal-3 (c). Glycan recognition of N-glycans is shown for
Gal-1 (d), Gal-2 (e), and Gal-3 (f). (g) Legend of symbols for monosaccharides. This research was originally
published in the Journal of Biological Chemistry. Stowell SR, Arthur CM, Mehta P, Slanina KA, Blixt O, Leffler H,
Smith DF, Cummings RD Galectin-1, -2, and -3 exhibit differential recognition of sialylated glycans and blood
group antigens.J Biol Chem. 2008 Apr 11;283(15):10109–23 © the American Society for Biochemistry and
Molecular Biology
References
1. Cooper DN, Barondes SH (1999) God must and pattern recognition receptors: galectins
love galectins; he made so many of them. Gly- and glycans in innate immunity. J Clin Immu-
cobiology 9(10):979–984 nol 31(1):10–21. https://doi.org/10.1007/
2. Brewer CF, Miceli MC, Baum LG (2002) s10875-010-9494-2
Clusters, bundles, arrays and lattices: novel 4. Robinson BS, Arthur CM, Evavold B,
mechanisms for lectin-saccharide-mediated cel- Roback E, Kamili NA, Stowell CS, Vallecillo-
lular interactions. Curr Opin Struct Biol Zuniga ML, Van Ry PM, Dias-Baruffi M,
12(5):616–623 Cummings RD, Stowell SR (2019) The
3. Cerliani JP, Stowell SR, Mascanfroni ID, sweet-side of leukocytes: galectins as master
Arthur CM, Cummings RD, Rabinovich GA regulators of neutrophil function. Front
(2011) Expanding the universe of cytokines
Immunol 10:1762. https://doi.org/10. SR (2016) Key regulators of galectin-glycan

3389/fimmu.2019.01762 interactions. Proteomics 16(24):3111–3125.
5. Robinson BS, Saeedi B, Arthur CM, Owens J, https://doi.org/10.1002/pmic.201600116
Naudin C, Ahmed N, Luo L, Jones R, Neish A, 15. Teichberg VI, Silman I, Beitsch DD, Resheff G
Stowell SR (2020) Galectin-9 is a novel regula- (1975) A beta-D-galactoside binding protein
tor of epithelial restitution. Am J Pathol from electric organ tissue of Electrophorus
190(8):1657–1666. https://doi.org/10. electricus. Proc Natl Acad Sci U S A
1016/j.ajpath.2020.04.010 72(4):1383–1387
6. Liu FT, Patterson RJ, Wang JL (2002) Intra- 16. de Waard A, Hickman S, Kornfeld S (1976)
cellular functions of galectins. Biochim Bio- Isolation and properties of beta-galactoside
phys Acta 1572(2–3):263–273 binding lectins of calf heart and lung. J Biol
7. Nakahara S, Raz A (2006) On the role of Chem 251(23):7581–7587
galectins in signal transduction. Methods 17. Levi G, Teichberg VI (1981) Isolation and
Enzymol 417:273–289. https://doi.org/10. physicochemical characterization of electrolec-
1016/S0076-6879(06)17019-6 tin, a beta-D-galactoside binding lectin from
8. Dias-Baruffi M, Stowell SR, Song SC, Arthur the electric organ of Electrophorus electricus.
CM, Cho M, Rodrigues LC, Montes MA, J Biol Chem 256(11):5735–5740
Rossi MA, James JA, McEver RP, Cummings 18. Pritchett TJ, Brossmer R, Rose U, Paulson JC
RD (2009) Differential expression of immuno- (1987) Recognition of monovalent sialosides
modulatory galectin-1 in peripheral leukocytes by influenza virus H3 hemagglutinin. Virology
and adult tissues and its cytosolic organization 160(2):502–506
in striated muscle. Glycobiology 19. Ahmad N, Gabius HJ, Sabesan S, Oscarson S,
20(5):507–520 Brewer CF (2004) Thermodynamic binding
9. van Kooyk Y, Rabinovich GA (2008) Protein- studies of bivalent oligosaccharides to
glycan interactions in the control of innate and galectin-1, galectin-3, and the carbohydrate
adaptive immune responses. Nat Immunol recognition domain of galectin-3. Glycobiol-
9(6):593–601. https://doi.org/10.1038/ni. ogy 14(9):817–825. https://doi.org/10.
f.203 1093/glycob/cwh095
10. Cerri DG, Rodrigues LC, Stowell SR, Araujo 20. Brewer CF (2004) Thermodynamic binding
DD, Coelho MC, Oliveira SR, Bizario JC, studies of galectin-1, -3 and -7. Glycoconj J
Cummings RD, Dias-Baruffi M, Costa MC 19(7–9):459–465. https://doi.org/10.1023/
(2008) Degeneration of dystrophic or injured B:GLYC.0000014075.62724.d0
skeletal muscles induces high expression of 21. Stowell SR, Arthur CM, Mehta P, Slanina KA,
galectin-1. Glycobiology 18(11):842–850. Blixt O, Leffler H, Smith DF, Cummings RD
https://doi.org/10.1093/glycob/cwn079 (2008) Galectin-1, -2, and -3 exhibit differen-
11. Hirabayashi J, Hashidate T, Arata Y, Nishi N, tial recognition of sialylated glycans and blood
Nakamura T, Hirashima M, Urashima T, group antigens. J Biol Chem
Oka T, Futai M, Muller WE, Yagi F, Kasai K 283(15):10109–10123. https://doi.org/10.
(2002) Oligosaccharide specificity of galectins: 1074/jbc.M709545200
a search by frontal affinity chromatography. 22. Karmakar S, Stowell SR, Cummings RD,
Biochim Biophys Acta 1572(2–3):232–254. McEver RP (2008) Galectin-1 signaling in leu-
https://doi.org/10.1016/s0304-4165(02) kocytes requires expression of complex-type
00311-2 N-glycans. Glycobiology 18(10):770–778
12. Leffler H, Carlsson S, Hedlund M, Qian Y, 23. Fukui S, Feizi T, Galustian C, Lawson AM,
Poirier F (2004) Introduction to galectins. Chai W (2002) Oligosaccharide microarrays
Glycoconj J 19(7–9):433–440. https://doi. for high-throughput detection and specificity
org/10.1023/B:GLYC.0000014072. assignments of carbohydrate-protein interac-
34840.04 tions. Nat Biotechnol 20(10):1011–1017.
13. Carlsson S, Oberg CT, Carlsson MC, https://doi.org/10.1038/nbt735. nbt735
Sundin A, Nilsson UJ, Smith D, Cummings [pii]
RD, Almkvist J, Karlsson A, Leffler H (2007) 24. Blixt O, Head S, Mondala T, Scanlan C,
Affinity of galectin-8 and its carbohydrate rec- Huflejt ME, Alvarez R, Bryan MC, Fazio F,
ognition domains for ligands in solution and at Calarese D, Stevens J, Razi N, Stevens DJ,
the cell surface. Glycobiology 17(6):663–676. Skehel JJ, van Die I, Burton DR, Wilson IA,
https://doi.org/10.1093/glycob/cwm026 Cummings R, Bovin N, Wong CH, Paulson JC
14. Kamili NA, Arthur CM, Gerner-Smidt C, (2004) Printed covalent glycan array for ligand
Tafesse E, Blenda A, Dias-Baruffi M, Stowell profiling of diverse glycan binding proteins.
Proc Natl Acad Sci U S A 34. Yu Y, Mishra S, Song X, Lasanajak Y, Bradley

101(49):17033–17038. https://doi.org/10. KC, Tappert MM, Air GM, Steinhauer DA,
1073/pnas.0407902101 Halder S, Cotmore S, Tattersall P, Agbandje-
25. Disney MD, Seeberger PH (2004) The use of McKenna M, Cummings RD, Smith DF
carbohydrate microarrays to study (2012) Functional glycomic analysis of human
carbohydrate-cell interactions and to detect milk glycans reveals the presence of virus recep-
pathogens. Chem Biol 11(12):1701–1707. tors and embryonic stem cell biomarkers. J Biol
https://doi.org/10.1016/j.chembiol.2004. Chem 287(53):44784–44799. https://doi.
27.011. S1074-5521(04)00312-6 [pii] org/10.1074/jbc.M112.425819.
26. Stowell SR, Dias-Baruffi M, Penttila L, M112.425819 [pii]
Renkonen O, Nyame AK, Cummings RD 35. Palma AS, Feizi T, Zhang Y, Stoll MS, Lawson
(2004) Human galectin-1 recognition of AM, Diaz-Rodriguez E, Campanero-Rhodes
poly-N-acetyllactosamine and chimeric poly- MA, Costa J, Gordon S, Brown GD, Chai W
saccharides. Glycobiology 14(2):157–167. (2006) Ligands for the beta-glucan receptor,
https://doi.org/10.1093/glycob/cwh018 Dectin-1, assigned using “designer” microar-
27. Stowell SR, Arthur CM, Slanina KA, Horton rays of oligosaccharide probes (neoglycolipids)
JR, Smith DF, Cummings RD (2008) Dimeric generated from glucan polysaccharides. J Biol
Galectin-8 induces phosphatidylserine expo- Chem 281(9):5771–5779. https://doi.org/
sure in leukocytes through polylactosamine 10.1074/jbc.M511461200. M511461200 [
recognition by the C-terminal domain. J Biol pii]
Chem 283(29):20547–20559. https://doi. 36. Song X, Heimburg-Molinaro J, Dahms NM,
org/10.1074/jbc.M802495200 Smith DF, Cummings RD (2012) Preparation
28. Arthur CM, Cummings RD, Stowell SR of a mannose-6-phosphate glycan microarray
(2014) Using glycan microarrays to under- through fluorescent derivatization, phosphory-
stand immunity. Curr Opin Chem Biol lation, and immobilization of natural high-
18C:55–61. https://doi.org/10.1016/j.cbpa. mannose N-glycans and application in ligand
2013.12.017 identification of P-type lectins. Methods Mol
Biol 808:137–148. https://doi.org/10.1007/
29. Stowell SR, Arthur CM, McBride R, Berger O, 978-1-61779-373-8_9
Razi N, Heimburg-Molinaro J, Rodrigues JP,
Noll AJ, von Gunten S, Smith DF, Knirel YA, 37. Song X, Yu H, Chen X, Lasanajak Y, Tappert
Paulson JC, Cummings RD (2014) Microbial MM, Air GM, Tiwari VK, Cao H, Chokhawala
glycan microarrays define key features of host- HA, Zheng H, Cummings RD, Smith DF
microbial interactions. Nat Chem Biol (2011) A sialylated glycan microarray reveals
10(6):470–476 novel interactions of modified sialic acids with
proteins and viruses. J Biol Chem
30. Leppanen A, Stowell S, Blixt O, Cummings 286(36):31610–31622. https://doi.org/10.
RD (2005) Dimeric galectin-1 binds with 1074/jbc.M111.274217. M111.274217 [pii]
high affinity to alpha2,3-sialylated and
non-sialylated terminal N-acetyllactosamine 38. Knirel YA, Gabius HJ, Blixt O, Rapoport EM,
units on surface-bound extended glycans. J Khasbiullina NR, Shilova NV, Bovin NV
Biol Chem 280(7):5549–5562 (2014) Human tandem-repeat-type galectins
bind bacterial non-betaGal polysaccharides.
31. Sorme P, Kahl-Knutson B, Wellmar U, Nilsson Glycoconj J 31(1):7–12. https://doi.org/10.
UJ, Leffler H (2003) Fluorescence polarization 1007/s10719-013-9497-3
to study galectin-ligand interactions. Methods
Enzymol 362:504–512. https://doi.org/10. 39. Geissner A, Anish C, Seeberger PH (2014)
1016/S0076-6879(03)01033-4. Glycan arrays as tools for infectious disease
S0076687903010334 [pii] research. Curr Opin Chem Biol 18C:38–45.
https://doi.org/10.1016/j.cbpa.2013.
32. Arthur CM, Cummings RD, Stowell SR 11.013
(2014) Using glycan microarrays to under-
stand immunity. Curr Opin Chem Biol 18: 40. Stowell SR, Cho M, Feasley CL, Arthur CM,
55–61. https://doi.org/10.1016/j.cbpa. Song X, Colucci JK, Karmakar S, Mehta P,
2013.12.017 Dias-Baruffi M, McEver RP, Cummings RD
(2009) Ligand reduces galectin-1 sensitivity
33. Song X, Lasanajak Y, Xia B, Heimburg- to oxidative inactivation by enhancing dimer
Molinaro J, Rhea JM, Ju H, Zhao C, Molinaro formation. J Biol Chem 284(8):4989–4999.
RJ, Cummings RD, Smith DF (2011) Shotgun https://doi.org/10.1074/jbc.M808925200
glycomics: a microarray strategy for functional
glycomics. Nat Methods 8(1):85–90. https:// 41. Stowell SR, Qian Y, Karmakar S, Koyama NS,
doi.org/10.1038/nmeth.1540. nmeth.1540 Dias-Baruffi M, Leffler H, McEver RP, Cum-
[pii] mings RD (2008) Differential roles of galectin-
1 and galectin-3 in regulating leukocyte viabil- 44. Robinson BS, Arthur CM, Kamili NA, Stowell
ity and cytokine secretion. J Immunol SR (2018) Galectin regulation of host micro-
180(5):3091–3102 bial interactions. Trends Glycosci Glycotechnol
42. Poland PA, Rondanino C, Kinlough CL, 30(172):185–198
Heimburg-Molinaro J, Arthur CM, Stowell 45. Arthur CM, Patel SR, Mener A, Kamili NA,
SR, Smith DF, Hughey RP (2011) Identifica- Fasano RM, Meyer E, Winkler AM, Sola-
tion and characterization of endogenous galec- Visner M, Josephson CD, Stowell SR (2015)
tins expressed in Madin Darby canine kidney Innate immunity against molecular mimicry:
cells. J Biol Chem 286(8):6780–6790. https:// examining galectin-mediated antimicrobial
doi.org/10.1074/jbc.M110.179002 activity. BioEssays 37(12):1327–1337.
43. Stowell SR, Arthur CM, Dias-Baruffi M, https://doi.org/10.1002/bies.201500055
Rodrigues LC, Gourdine JP, Heimburg- 46. Leppanen A, White SP, Helin J, McEver RP,
Molinaro J, Ju T, Molinaro RJ, Rivera- Cummings RD (2000) Binding of glycosulfo-
Marrero C, Xia B, Smith DF, Cummings RD peptides to P-selectin requires stereospecific
(2010) Innate immune lectins kill bacteria contributions of individual tyrosine sulfate
expressing blood group antigen. Nat Med and sugar residues. J Biol Chem
nm.2103 1074/jbc.M005005200. M005005200 [pii]
Chapter 10
Mechanism of Mucin Recognition by Lectins: A

Thermodynamic Study
Tarun K. Dam, Jared L. Edwards, Priyanka D. Kadav, and C. Fred Brewer
Abstract
Isothermal titration microcalorimetry (ITC) can directly determine the thermodynamic binding parameters
of biological molecules including affinity constant, binding stoichiometry, heat of binding (enthalpy) and
indirectly the entropy, and free energy of binding. ITC has been extensively used to study the binding of
lectins to mono- and oligosaccharides, but limitedly in applications to lectin–glycoprotein interactions.
Inherent experimental challenges to ITC include sample precipitation during the experiment and relative
high amount of sample required, but careful design of experiments can minimize these problems and allow
valuable information to be obtained. For example, the thermodynamics of binding of lectins to multivalent
globular and linear glycoproteins (mucins) have been described. The results are consistent with a dynamic
binding mechanism in which lectins bind and jump from carbohydrate to carbohydrate epitope in these
molecules leading to increased affinity. Importantly, the mechanism of binding of lectins to mucins appears
similar to that for a variety of protein ligands binding to DNA. Recent results also show that high-affinity
lectin–mucin cross-linking interactions are driven by favorable entropy of binding that is associated with the
bind and jump mechanism. The results suggest that the binding of ligands to biopolymers, in general, may
involve a common mechanism that involves enhanced entropic effects that facilitate binding interactions.
Key words Lectins, Multivalent carbohydrates, Glycoproteins, Mucins, Thermodynamics, Bind and
jump model, Entropy, Internal diffusion
1 Introduction
1.1 General Overview Lectins are ubiquitous proteins that recognize specific carbohydrate
residues of glycoproteins and glycolipids. Lectin–glycoprotein and
lectin–glycolipid interactions are central to a variety of biological
activities including receptor-mediated endocytosis, cellular recog-
nition and adhesion [1], inflammation [2], cell growth, and metas-
tasis [3, 4]. Glycoproteins are broadly classified as N-linked and
O-linked glycoproteins, including heavily O-linked glycoproteins
known as mucins, and multidomain glycoproteins with mucin
domains (mucin-type glycoproteins).
The glycoprotein receptors for a number of lectins including
selectins, siglecs, and galectins are often mucin domains with large
169
170 Tarun K. Dam et al.
Fig. 1 Structural representations of (a) the fully carbohydrate decorated form (described in the text) of the
100-repeat 81-residue polypeptide O-glycosylation domain of PSM (Fd-PSM); (b) the 100-repeat 81-residue
polypeptide O-glycosylation domain of PSM containing only peptide linked α-GalNAc residues (Tn-PSM); (c)
the single 81-residue polypeptide O-glycosylation domain of PSM containing peptide linked a-GalNAc residues
(81-mer Tn-PSM); (d) the 38/40-residue polypeptide(s) derived from the 81-residue polypeptide O-glycosyla-
tion domain of PSM containing peptide linked α-GalNAc residues (38/40-mer Tn-PSM). The number of glycan
chains in Fd-PSM and Tn-PSM is ~2300. The number of α-GalNAc residues in 81-mer Tn-PSM is ~23, while
the number of α-GalNAc residues in 38/40-mer Tn-PSM is ~11–12
numbers of glycan epitopes [5]. Mucins are secreted by higher

organisms to protect and lubricate epithelia cell surfaces from
biological, chemical, and mechanical assaults. Mucin and mucin-
type molecules exist as soluble and transmembrane molecules that
are involved in modulating immune response, inflammation, adhe-
sion, and tumorigenesis [6–8]. The O-glycosylated domains of
mucins and mucin-type glycoproteins can contain 50–80% carbo-
hydrate (% of molecular mass) and possess expanded linear confor-
mations. Polypeptide tandem repeats are found in mucins that
contain clusters of Ser and Thr residues in high content. The
O-linked carbohydrate chains in these domains are attached via
α-GalNAc residues to Ser and Thr residues, which have been
shown to induce a threefold expansion in size of the polypeptide
chain of these molecules as determined by light scattering studies
[9]. The oligosaccharide chains on mucins have been shown to be
important for a variety of their biological properties including their
interactions with animal lectins such as the selectins and galectins,
as well as their physical properties including their extended linear
structures [10].
There are at least 17 human mucin gene products (MUC1–
MUC17) reported in the literature [7]. These gene products rep-
resent two structurally and functionally distinct classes of mucins:
Mechanism of Mucin Recognition by Lectins: A Thermodynamic Study 171
Fig. 2 Schematic representations of (a) SBA or VML binding at low density to Tn-PSM; (b) SBA or VML binding
at higher density to Tn-PSM; (c) SBA and VML binding at higher density to Tn-PSM and initiating cross-linking
of the complexes; (d) SBA cross-linked complexes with Tn-PSM under saturation binding conditions. The view
is on the end of the polypeptide chains of Tn-PSM in Fig. 1c. α-GalNAc residues extend out from the
polypeptide chains of Tn-PSM in three dimensions. Lectin tetramers are bound to four separate Tn-PSM
chains, with staggered binding down the length of the mucin chains
secreted gel-forming mucins and transmembrane mucins, although

there are a few mucin gene products that do not appear to fit into
either category. The transmembrane mucins include MUC1 whose
structure includes a cytoplasmic domain in addition to the extracel-
lular O-glycosylated polypeptide tandem repeat domains [11]. Evi-
dence indicates that the C-terminal cytoplasmic domain of MUC1
is involved in signal transduction mechanisms including T-cell acti-
vation and inhibition, and adhesion signaling responses [12, 13].
Mucins are also useful in the diagnosis of a variety of diseases
[10]. In particular, the level of expression of mucin peptide anti-
gens and the type of carbohydrate chains of mucins have proven to
be useful diagnostic markers for a variety of cancers [7, 14]. For
example, MUC1 expression as detected immunologically is
increased in colon cancers and is associated with a poor prognosis
[7]. Colon cancer–associated mucins also have differences asso-
ciated with their core carbohydrate structures, often presenting
shorter chain versions of normal mucins. Colon cancer mucins
often have increased expression of the GalNAcαThr/Ser

(Tn-antigen), Galβ3GalNAc (T or TF antigen), and
NeuAcα6GalNAc (sialyl Tn-antigen) [7]. Importantly, recent stud-
ies have shown that binding of galectin-3, an endogenous
Gal-specific animal lectin, to cancer-associated MUC1 causes
increased endothelial cancer cell adhesion [15]. Thus, the molecu-
lar recognition properties of cancer-related mucins, including their
truncated carbohydrates, are important in terms of gaining insight
into their structure–activity properties.
The binding and cross-linking of cell surface mucins and
mucin-like glycoproteins by lectins are known to lead to signal
transduction effects including cell growth and cell death
[16, 17]. For example, galectin-1 cross-linking of CD43, a trans-
membrane mucin-type glycoprotein receptor that possesses
approximately 80 O-linked chains with terminal LacNAc epitopes
[18], along with CD45 induces apoptosis in susceptible T cells
[19]. However, details of the energetics and mechanisms of lectin
binding and cross-linking of mucins and mucin-type receptors have
been lacking.
Quantitative binding studies such as isothermal titration micro-
calorimetry (ITC) often require relatively high amount of samples.
Obtaining recombinant human mucins sufficient for thermody-
namic studies is difficult. Therefore, other mammalian mucins
that show some structural similarities with human mucins can be
used in such studies. One such mucin is the porcine submaxillary
mucin (PSM), which is a physically well-characterized mucin, and
the subject of studies of the regulation of O-glycosylation with
glycosyl transferases [20]. The cDNA sequence of PSM has been
determined [21], and the 81 amino acid tandem repeat domain
that is present in 100 copies (Fig. 1a). The structures of the carbo-
hydrate chains were determined by chemical [22] and NMR tech-
niques [23]. Gerken and coworkers [23] isolated the O-
glycosylated domain of PSM that possesses a molecular mass of
~106 daltons and is fully decorated with naturally occurring carbo-
hydrates (Fd-PSM) (Fig. 2b). The O-glycosylated domain of PSM
possessing only α-GalNAc residues (Tn-PSM) was also obtained
using chemical and enzymatic treatments [24]. The GalNAcα1-O-
Ser/Thr residue(s) in Tn-PSM is the pancarcinoma carbohydrate
antigen Tn that is aberrantly expressed in mucins such as MUC1 in
adenocarcinomas [25]. The 81-mer tandem repeat domain of
Tn-PSM (81-mer Tn-PSM) and the 38/40-mer digest of this
domain (38/40-mer Tn-PSM) also have been obtained using enzy-
matic digests [24].
Historically, the relative affinities and specificities of lectins for
carbohydrates were first addressed using hemagglutination inhibi-
tion, equilibrium dialysis, and quantitative precipitation inhibition
techniques [26]. While these methods are still employed, additional
techniques are often used including gradient affinity
chromatography, nuclear magnetic resonance, and surface plasmon

resonance. However, all of the above methods either suffer from
being indirect measurements of binding constants or have special
conditions required for the measurements. For example, hemag-
glutination inhibition and quantitative precipitation inhibition
techniques are indirect binding methods, while equilibrium dialysis
typically requires radioactivity or a chromophore in the molecules.
Affinity chromatography suffers from potential matrix interactions
with ligands in solution, and covalent attachment of a receptor or
ligand to the matrix. Nuclear magnetic resonance measurements
require the binding between two molecules to be in certain kinetic
conditions. Surface plasmon resonance mandates covalent attach-
ment of one of the binding molecules on a solid-state chip. Equi-
librium constants are derived from kinetic on-off rate
measurements of a ligand in a flowing system. Thus, all of the
techniques have limitations in determining quantitative binding
constants for carbohydrate–lectin interactions in solution, and
none of these methods can provide comprehensive thermodynamic
and hence mechanistic information on lectin–glycoprotein interac-
tions. ITC has been employed extensively to study the binding
interactions of lectins with glycans and, more recently, lectins with
glycoproteins.
Using ITC measurements, the affinities of a series of galectins
for asialofetuin (ASF), a nonavalent globular glycoprotein with
terminal LacNAc residues, were observed to be ~50- to 80-fold
greater than LacNAc [27]. However, the affinity of the GalNAc-
specific soybean agglutinin (SBA) for a mucin that possesses ~2300
GalNAc residues was reported to be ~106-fold greater than the
corresponding monovalent carbohydrate [28]. Based on the ther-
modynamic data, a model called the “bind and jump mechanism”
has been proposed to provide a mechanistic basis for the higher
affinities of the lectins for these glycoproteins [28].
Previous reviews of multivalent interactions in biological sys-
tems including protein–carbohydrate interactions have been
reported [29–31]. This chapter will focus primarily on studies of
the thermodynamics of binding of lectins to mucins since the
results have provided new insights into the energetics and mechan-
isms of binding of lectins to multi- and polyvalent glycoproteins.
Furthermore, the results have suggested common energetics and
mechanisms of binding of ligands to biopolymers in general.
1.2 Affinities of SBA ITC measurements demonstrate that SBA binds to Tn-PSM pos-
and Vatairea sessing ~2300 α-GalNAc residues and a molecular mass of ~106
Macrocarpa Lectin daltons (Fig. 1b) with a Kd of 0.2 nM (Table 1), which is ~106-fold
(VML) for Mucins enhanced affinity relative to monovalent GalNAcα1-O-Ser. The
81 amino acid tandem repeat domain of Tn-PSM containing ~23
GalNAc residues (81-mer Tn-PSM) (Fig. 1c binds with ~103-fold
enhanced affinity, while the 38/40-mer fragment of Tn-PSM
Table 1
Thermodynamic binding parameters for SBA and VML at pH 7.2, 27 ˚C
Ligand (kcal/mol) Kda K(rel)b (μM) -ΔGc -ΔHd (kcal/mol) -TΔSe (kcal/mol) nf
SBA
GalNAcα1-O-Ser 170 1 5.2 7.9 2.7 1.0
38/40-mer Tn-PSM 1.4 120 8.0 32.2 24.2 0.2
81-mer Tn-PSM 0.06 2800 9.8 56.1 46.3 0.12
Tn-PSM 0.0002 850,000 13.1 4310 4297 0.0018
Fd-PSM 0.024 7100 10.4 703 693 0.008
VML
GalNAcα1-O-Ser 130 1 5.3 6.4 1.1 1.0
38/40-mer Tn-PSM 0.20 650 9.1 36.8 27.7 0.2
81-mer Tn-PSM 0.012 11,000 10.8 52.7 41.9 0.1
Tn-PSM 0.0001 1,300,000 13.6 5274 5260 0.0012
Fd-PSM 0.0014 93,000 12.1 1251 1240 0.005
a
Errors in Kd range from 1% to 7%
b
Relative to GalNAcα1-O-Ser
c
Errors in ΔG are less than 2%
d
Errors in ΔH are 1–4%
e
Errors in TΔS are 1–7%
f
Errors in n are less than 4%
containing ~11–12 GalNAc residues (38/40-mer Tn-PSM)

(Fig. 1d) shows ~102-fold enhanced affinity (Table 1). Fd-PSM,
the fully decorated form of PSM containing 40% of the core
1 blood group type A tetrasaccharide, and 58% peptide-linked
GalNAcα1-O-Ser/Thr residues, with 45% of the peptide-linked
GalNAc residues linked α(2,6) to N-glycolylneuraminic acid
(Fig. 1a), shows ~104 enhanced affinity for SBA (Table 1). VML
displays a similar pattern of affinities for the PSM analogs although
there are differences in the absolute affinities (Table 1) [28]. The
higher affinities of SBA and VML for Tn-PSM relative to Fd-PSM
indicate the importance of carbohydrate composition and epitope
density of the mucins on their affinities for the lectins. The higher
affinities of both lectins for Tn-PSM relative to its two shorter chain
analogs demonstrate that the length of a mucin polypeptide and
hence total carbohydrate valence determine the affinities of the
lectins for the three Tn-PSM analogs.
Kiessling and coworkers [32] reported a similar increase in the
inhibitory activity of synthetic polymers of increasing lengths that
possess Man residues in a hemagglutination assay with ConA. The
authors concluded that the enhanced inhibitory activities of the
longer chain polymers were “largely due to a combination of statis-
tical and chelation effects” and therefore lower dissociation rates.
1.3 Thermodynamics The ITC data also include the thermodynamics of binding of SBA
of SBA Binding Tn- and VML to the PSM analogs, including their stoichiometries of
PSM binding (Table 1) [28]. The n value for SBA binding to Tn-PSM,
which is the binding stoichiometry expressed as number of binding
sites per subunit of lectin, is 0.0018 (Table 1). The value of 1/n has
been shown to provide the functional valence of multivalent carbo-
hydrates and glycoproteins binding to lectins [27, 33]. The 1/n for
Tn-PSM binding SBA is 540, which indicates that 540 α-GalNAc
residues of the ~2300 α-GalNAc residues of Tn-PSM bind
540 monomers of SBA since each monomer binds one
GalNAcα1-O-Ser. The 1/n value indicates that the functional
valence of Tn-PSM for SBA is less than the structural valence of
Tn-PSM. The reason for the fractional binding of SBA to the
carbohydrate epitopes in Tn-PSM appears to be the size of the
SBA tetramer, which possesses a molecular mass of 120 kDa and
one N-linked Man-9 oligomannose chain per monomer
[34, 35]. Kiessling and coworkers have reported similar fractional
occupancies for the binding of ConA, a tetrameric Man-specific
lectin similar in size to SBA, to synthetic polymers containing
α-mannose residues [36]. Fractional occupancy has also been
reported for SBA binding to the nine LacNAc epitopes of
ASF [37].
SBA binding to Tn-PSM yields a ΔH of 4310 kcal/mol as
compared to 7.9 kcal/mol for GalNAcα1-O-Ser (Table 1). If the
ΔH for SBA binding to Tn-PSM is divided by the 1/n value of
540 α-GalNAc residues on Tn-PSM that bind to the lectin, the
resulting ΔH per α-GalNAc residue of Tn-PSM is 7.98 kcal/mol,
which is very similar to the ΔH of 7.9 kcal/mol for GalNAcα1-O-
Ser. This indicates that each α-GalNAc residue of Tn-PSM that
binds to SBA possesses the same enthalpy of binding as that of
GalNAcα1-O-Ser. The observed ΔH for SBA binding to Tn-PSM is
thus the sum of the individual ΔH values of the α-GalNAc binding
residues of the mucin. Similar observations have been made for the
binding of ConA and DGL to synthetic bi-, tri- and tetraantennary
glycosides [38].
The calculated TΔS for SBA binding to Tn-PSM is
4297 kcal/mol (Table 1). If TΔS is divided by the 1/n value of
540, the resulting TΔS value per α-GalNAc residue of Tn-PSM is
7.96 kcal/mol, which is more unfavorable than the 2.7 kcal/
mol for GalNAcα1-O-Ser (Table 1). If TΔS were proportional to
the number of binding epitopes in Tn-PSM, the observed TΔS
value would be 1458 kcal/mol, and ΔG would therefore be
equal to ΔH TΔS or 2852 kcal/mol, an impossibly large
value. The observation that, unlike ΔH, TΔS does not scale in
proportion to the number of binding epitopes in multivalent car-
bohydrates binding to lectins but instead is much more negative has
been previously observed in the binding of ConA and DGL to bi-,
tri-, and tetraantennary carbohydrates [38]. The

non-proportionality of TΔS is characteristic of different lectin
molecules binding to separate epitopes of a multivalent carbohy-
drate instead of a single lectin binding to multiple epitopes of a
multivalent carbohydrate [39]. In the latter case, both ΔH and TΔS
increase in proportion to the number of binding epitopes in the
multivalent ligand, with concomitantly larger increases in affinity
[39]. Nevertheless, the observed TΔS value of 4297 kcal/mol
when subtracted from the observed ΔH value of 4310 kcal/mol
for SBA binding to Tn-PSM gives a ΔG value of 13.1 kcal/mol,
which is 7.9 kcal/mol greater than the ΔG of 5.2 kcal/mol for
GalNAcα1-O-Ser (Table 1).
1.4 Thermodynamics The ITC-derived Kd value for SBA binding to 81-mer Tn-PSM is
of SBA Binding the 81- 0.06 μM. This can be compared with the value of 0.15 μM obtained
mer Tn-PSM by hemagglutination inhibition [28]. The K(rel) (enhancement of
Ka when compared with the Ka of a single GalNAcα1-O-Ser) for
81-mer Tn-PSM is 2800 as compared to GalNAcα1-O-Ser. The
affinity of SBA for 81-mer Tn-PSM is ~300-fold weaker than that
of Tn-PSM.
The n value for SBA binding to 81-mer Tn-PSM is 0.12, and
1/n ¼ 8, which suggests that ~8 α-GalNAc residues of 81-mer
Tn-PSM bind to ~8 monomers of SBA. Hence, the functional
valence of 81-mer PSM is less than its structural valence for SBA
binding, as observed for Tn-PSM.
The ΔH for SBA binding to 81-mer Tn-PSM is 56.1 kcal/
mol. If ΔH is divided by the 1/n value of 8, the number of bound
SBA monomers per ~23 α-GalNAc residues, the resulting value is
7.0 kcal/mol per α-GalNAc binding residue, which is slightly less
than the value of 7.9 kcal/mol for GalNAcα1-O-Ser. This indi-
cates that each α-GalNAc binding residue of 81-mer Tn-PSM binds
with nearly the same ΔH as that of GalNAcα1-O-Ser. This finding is
similar to that observed for SBA binding to Tn-PSM.
SBA binding to 81-mer Tn-PSM gives a TΔS of 46.3 kcal/
mol which is greater than the calculated value of 21.6 kcal/mol if
TΔS were proportional to the number α-GalNAc residues involved
in binding to SBA. Thus, TΔS for SBA binding to 81-mer Tn-PSM
does not increase in proportion to the number of α-GalNAc resi-
dues that bind, but rather has a larger negative value. Similar results
are observed for SBA binding to Tn-PSM.
1.5 Thermodynamics ITC data for the binding of SBA to 38/40-mer Tn-PSM shows a
of SBA Binding 38/40- Kd of 1.4 μM (Table 1), which can be compared with ~6 μM
mer Tn-PSM obtained by hemagglutination inhibition [28]. The Kd for 38/
40-mer Tn-PSM is reduced nearly 20-fold relative to that for
81-mer Tn-PSM. Thus, SBA shows lowest affinity for the shortest
fragment of Tn-PSM.
The n value for SBA binding to 38/40-mer Tn-PSM is 0.2, and

1/n ¼ 5. This indicates that ~5 α-GalNAc residues of 38/40-mer
Tn-PSM binds to five SBA monomers.
The ΔH and TΔS values of 32.2 kcal/mol and 24.2 kcal/
mol, respectively, are consistent with the lower affinity of SBA for
38/40-mer Tn-PSM as compared to the longer polypeptide chain
analogs. If ΔH is divided by the 1/n value of 5, then ΔH per
α-GalNAc residue of 38/40-mer is 6.44 kcal/mol, which is
somewhat lower than the 7.9 kcal/mol for GalNAcα1-O-Ser.
The lower ΔH per α-GalNAc residue may be due to the lower
affinity of SBA for 38/40-mer Tn-PSM as compared to Tn-PSM
and 81-mer Tn-PSM (discussed further below). If the observed
TΔS is divided by 1/n value of 5, the TΔS per α-GalNAc residue of
38/40-mer is 4.84 kcal/mol, which is larger than 2.7 kcal/mol
for GalNAcα1-O-Ser. Similar results are observed for SBA binding
to Tn-PSM and 81-mer Tn-PSM, suggesting similar binding
mechanisms of SBA with all three PSM analogs.
1.6 Thermodynamics The ITC-derived Kd for SBA binding to Fd-PSM is 0.024 μM

of SBA Binding Fd- (Table 1), which can be compared to the inhibition constant of
PSM 0.08 μM obtained by hemagglutination inhibition [28]. The loss in
affinity of SBA for Fd-PSM relative to Tn-PSM may be due to the
lower density of free α-GalNAc residues in Fd-PSM, due to
NeuNGl present on ~45% of the total α-GalNAc residues.
The n value for SBA binding to Fd-PSM is 0.008 (Table 1), and
1/n is 125, which is the number of α-GalNAc residues of Fd-PSM
bound to SBA monomers. This number of α-GalNAc residues is
consistent with a reduced number of binding sites for SBA on
Fd-PSM relative to the 540 α-GalNAc residues on Tn-PSM. The
factor of ~four reduction in the number of binding sites on
Fd-PSM suggests that in addition to 45% of the total α-GalNAc
residues capped with NeuNGl, binding of SBA to the non-reducing
α-GalNAc residue of the tetrasaccharide chains of PSM may alter
the accessibility of SBA binding to the single peptide-linked
α-GalNAc residues by a further factor of 2. Similar effects are
observed for VML binding to Tn-PSM and Fd-PSM (below).
The ΔH and TΔS values for Fd-PSM binding to SBA are
703 kcal/mol and 693 kcal/mol, respectively, and are lower
than the corresponding values for Tn-PSM (Table 1). If ΔH is
divided by 1/n ¼ 125, the resulting ΔH per binding α-GalNAc
residue is 5.6 kcal/mol which is somewhat lower than that for
GalNAcα1-O-Ser (Table 1). If TΔS is divided by 1/n ¼ 125, the
resulting TΔS per binding α-GalNAc residue is 5.54 kcal/mol
which is greater than that for GalNAcα1-O-Ser. Similar results are
observed for SBA binding to Tn-PSM, 81-mer Tn-PSM, and 38/
40-mer PSM.
1.7 Thermodynamics The ITC data for VML binding to Tn-PSM is similar to that for
of VML Binding Tn- SBA [28]. The main difference is the stoichiometry of binding of
PSM VML, which shows that 1/n ¼ 833 α-GalNAc residues binding to
833 monomers of VML. This can be compared with 540 α-GalNAc
residues of Tn-PSM binding to 540 monomers of SBA. Impor-
tantly, the size of the VML tetramer is similar to that of SBA [40],
and VML also possesses covalently bound carbohydrate like
SBA [40].
1.8 Thermodynamics The ITC data for VML binding to 81-mer Tn-PSM and 38/40-
of VML Binding 81-mer mer Tn-PSM in Table 1 are similar to that observed for SBA
Tn-PSM and 38/40- [28]. Differences between VML and SBA binding to the two frag-
mer Tn-PSM ments of Tn-PSM are principally in their K(rel) values which are
somewhat greater for VML. The binding stoichiometries of VML
to the two fragments are also similar to those observed for SBA.
The results support the conclusion that the affinity of VML, like
SBA, decreases with shorter fragments of Tn-PSM.
1.9 Thermodynamics The ITC data for VML binding to Fd-PSM is also similar to that for
of VML Binding Fd- SBA. Differences between the two lectins are in their K(rel) values
PSM which are greater for VML. Like SBA, VML binds with greater
affinity to Fd-PSM than to 81-mer Tn-PSM and 38/40-mer PSM,
but with less affinity to Fd-PSM than to Tn-PSM. The binding
stoichiometry of VML to Fd-PSM indicates that there are
200 α-GalNAc residues of Fd-PSM that bind to VML monomers.
This can be compared to the 125 α-GalNAc residues of Fd-PSM
that bind to SBA.
1.10 Analysis of the Using the 1/n value for Tn-PSM indicates that ~540 GalNAc
Stoichiometry of residues out of the total of ~2300 GalNAc residues of the mucin
Binding of SBA to the bind to SBA under saturation conditions. SBA is a tetramer with
Mucins four binding sites per molecule [35]. Thus, all four sites of SBA are
occupied at the end of the ITC experiment, and hence, SBA is
completely cross-linked with Tn-PSM. This is also true for the
ITC experiments for SBA binding to the shorter mucin fragments.
1.11 Mechanisms of The increasing affinities of SBA and VML for 38/40-mer Tn-PSM,
Binding of SBA and 81-mer Tn-PSM, and Tn-PSM, respectively, indicate that the
VML to PSM: The “Bind lengths of the polypeptide chains, and hence, total carbohydrate
and Jump Model” valences of these mucin analogs regulate their affinities for the
lectins. The higher affinities of SBA and VML for Tn-PSM relative
to Fd-PSM also indicates the importance of carbohydrate compo-
sition and epitope density of the mucins on their affinities for
lectins. The similarities in the ΔH per α-GalNAc binding residue
of Tn-PSM and Fd-PSM with that of GalNAcα1-O-Ser for the two
lectins, respectively, also suggest a common mechanism of binding.
Similar correlations with 81-mer Tn-PSM and 38/40-mer
Tn-PSM suggest that the binding mechanisms of these analogs are

common.
A model to explain these results is similar to that proposed for
the binding of lectins to multivalent carbohydrates [38] and glob-
ular glycoproteins [27]. Namely, binding of a lectin to a multivalent
carbohydrate or glycoprotein involves internal diffusion “jumps” of
the lectin from carbohydrate epitope to epitope before complete
dissociation. Kinetically, this has the effect of reducing the macro-
scopic off-rate of the lectin and hence increasing its affinity, since
the affinity constant is the ratio of the forward and reverse rate
constants. The forward rate constant for lectin binding may also be
enhanced as a result of the larger number of binding epitopes on
the glycoprotein.
The diffusion jump model for SBA and VML binding to
Tn-PSM and the other PSM analogs can be envisioned as occurring
with one subunit of SBA or VML bound to one α-GalNAc residue
of Tn-PSM at a time (Fig. 2a). (Two subunits of individual SBA or
VML molecules simultaneously binding to a single Tn-PSM chain
are not supported by the enhanced affinities of both lectins to 38/
40-mer Tn-PSM relative to GalNAcα1-O-Ser (Table 1). If two
subunits of an SBA tetramer bound to 38/40-mer Tn-PSM, the
affinity enhancement would be approximately the product of the
individual GalNAcα1-O-Ser dissociation constant which would be
~108 M instead of the observed 106 M.) This model provides a
molecular mechanism to account for the dependence of the affi-
nities of the two lectins on the total carbohydrate valences of
Tn-PSM, 81-mer Tn-PSM and 38/40-mer Tn-PSM. This model
further suggests that as more lectin molecules bind to a mucin
chain, the affinity of the lectin will decrease because of steric crowd-
ing and shorter diffusion distances on the polypeptide chain of the
mucin (Fig. 2b). Indeed, Hill plots for SBA binding to Tn-PSM
show evidence for negative binding cooperativity (data below).
Similar Hill plots have been interpreted as evidence for increasing
negative cooperativity in the binding of galectins to ASF [27] and
ConA and DGL to synthetic di-, tri-, and tetravalent carbohydrates
[38]. In these cases, binding of the lectins to the multivalent
carbohydrates and glycoprotein were associated with gradients of
decreasing microaffinity constants of the lectins for the multiple
epitopes of the ligands. This suggests that the observed dissociation
constants for SBA and VML binding to the four PSM analogs in
Table 1 represent a composite of the gradient of binding constants
present in each case. Such gradients have been estimated to be as
large as 3000- to 6000-fold for galectins binding to ASF
[27]. These gradient effects may explain the lower average ΔH
values per α-GalNAc residue for 81-mer Tn-PSM and 38/40-mer
PSM binding to SBA and VML.
The model for SBA and VML binding must also agree with the
final saturation density of SBA and VML molecules bound to single
PSM analogs as reflected in ITC n values. For example, the 1/n

value for SBA binding to Tn-PSM indicates 540 α-GalNAc residues
bound to 540 SBA monomers. The 1/n value for VML binding to
Tn-PSM indicates 833 α-GalNAc residues bound to 833 VML
monomers. Another requirement of the binding model is the
observation that at the end of ITC experiments, solutions of SBA
and VML with all four PSM analogs begin to precipitate out of
solution (data not shown). This indicates that lectin-mediated
cross-linking of the mucins occurs following saturation binding.
Both of these observations are accounted for in the schematic
representation in Fig. 2c, d of SBA cross-linked with Tn-PSM
under saturation conditions. The schematic shows individual SBA
molecules cross-linked to four different Tn-PSM molecules.
Importantly, in order to form the cross-linked complex shown in
Fig. 2d, lectin molecules bind to α-GalNAc residues on all four
sides of a Tn-PSM polypeptide chain. This allows staggering of
individual SBA molecules along the Tn-PSM polypeptide chain,
with concomitant reduction in steric interactions between lectin
molecules. This is important since calculations of the density of
SBA molecules bound to Tn-PSM (knowing the diameter of SBA
from x-ray crystal studies [35] and the length of the Tn-PSM
polypeptide chain) suggest that only ~300 SBA tetramers can
bind to the same side of a Tn-PSM polypeptide chain, which is
less than the 540 bound monomers of SBA and 833 bound mono-
mers of VML derived from ITC n values. The apparent steric
crowding is overcome by having lectin molecules bound to all
four sides of Tn-PSM.
In summary, the binding models first show a fraction of SBA or
VML molecules that bind to Tn-PSM and “jump” between differ-
ent α-GalNAc residues of the mucin. As the number of bound
lectin molecules increase, the affinity of the lectin decreases because
of shorter diffusion distances on the mucin chain due to steric
crowding and cross-linking by multiple bound lectin molecules.
Finally, upon saturation binding, full lectin-mediated cross-linking
of the complexes takes place. Lectin–mucin interactions play impor-
tant roles in many biological processes [41–45]. The mechanistic
insight provided by the current study will help understand those
interactions.
2 Materials
1. Soybean agglutinin (SBA) (EY Biochemicals, Inc.) (see

Note 1).
2. Vatairea macrocarpa lectin (VML) (see Note 2).
3. Guar gum column (Guar gum from Sigma Chemicals Co.).
4. Preparations of Fd-PSM, Tn-PSM, 81-mer Tn-PSM, and 38/

40-mer Tn-PSM (prepared as previously described [24]) (see
Note 3).
5. GalNAcα1-O-Ser (Sigma).
6. Rabbit erythrocytes.
7. 96-well round bottom microtiter plate.
Buffers
8. HEPES buffer: 0.1 M HEPES, 0.15 M NaCl, 1 mM CaCl2,
and 1 mM MnCl2, pH 7.2.
9. Phosphate-Buffered Saline (PBS) (Hyclone).
Special Equipment
10. VP-ITC instrument from Microcal, Inc.
(Northampton, MA).
3 Method
3.1 1. Add 0.3 mL of packed rabbit red blood cells to 9.7 mL of

Hemagglutination HEPES buffer for a final concentration of 3%.
Inhibition Assay to 2. At room temperature, prepare a twofold serial dilution of lectin
Determine the Binding sample as well as a twofold serial dilution of saccharide sample
Affinity of Mucins with PBS (see Note 4).
3. Mix 25 μL of diluted lectin samples with 25 μL of each dilution
of saccharide sample in wells of a 96-well round bottom micro-
titer plate. Final volume in each well should be 0.05 mL.
4. Add 0.05 mL of 3% erythrocyte suspension to each well and
allow erythrocytes to settle for 20–30 min.
5. Note agglutination. A positive pattern, indicating agglutina-
tion, is a uniform covering of the bottom of the round bottom
well by erythrocytes. A negative pattern, indicating no aggluti-
nation is a circular clump (button formation at the bottom of
the wells) of erythrocytes surrounded by a clear zone (Fig. 3).
6. The minimum concentration of saccharide required for com-
plete inhibition of four hemagglutination doses is determined.
3.2 ITC Studies of The thermodynamic binding properties of modified forms of por-
Lectin–Mucin cine submaxillary mucin (PSM) were investigated with SBA and
Interactions VML [28]. ITC experiments were performed using a VP-ITC
instrument from Microcal, Inc. (Northampton, MA) (see Note 5).
1. At room temperature, prepare a twofold serial dilution of lectin
sample as well as a twofold serial dilution of saccharide sample
with PBS (see Note 6).
Fig. 3 Hemagglutination of rabbit red blood cells (RBCs) by SBA. SBA was serially
diluted with buffer in the wells of a microtiter plate before adding equal volume
of RBCs. Pattern of agglutination was documented after keeping the plate
undisturbed for 30–45 min at room temperature. Agglutination activity
diminished as the dilution of SBA increased (indicated by arrows). Smear-like
appearance at the bottom of the wells (left side of the row) is the signature of
agglutination. Button formation at the bottom of the wells (far right side of the
row) is the indication of inactivity (lack of agglutination)
2. Inject 4 μL of carbohydrate solution from a computer-

controlled micro syringe at an interval of 4 min into the sample
solution of lectin (cell volume ¼ 1.43 mL) with stirring at
350 rpm. Control experiments should be performed by injec-
tion of the mucin into a cell containing only buffer.
3. Fit the experimental data to a theoretical titration curve using
software supplied by Microcal, with ΔH (binding enthalpy in
kcal mol1) (lectin monomer units), Ka (association constant),
and n (number of binding sites per monomer), as adjustable
parameters.
4. Calculate the Kd (dissociation constant) from 1/Ka. The quan-
tity c ¼ Ka Mt. (0), where Mt. (0) is the initial macromolecule
concentration, is of importance in titration microcalorimetry.
All experiments were performed with c values 1 < c < 500 (see
Note 7).
4 Notes
1. The concentration of SBA was determined spectrophotometri-

cally using A1%, 1 cm ¼ 12.8 and expressed in terms of
monomer.
2. Purified by affinity chromatography on a guargum column.
The concentration of VML was determined spectrophotomet-
rically using A1%, 1 cm ¼ 13.0 and expressed in terms of
monomer.
3. The 81-mer Tn-PSM was obtained by trypsin digestion of
Tn-PSM, and 38/40-mer Tn-PSM obtained by GluC diges-
tion of 81-mer PSM.
4. Appropriate starting concentration can be determined empiri-
cally. Blood was drawn in heparin-containing tubes from
rabbits housed in the Animal Care Facility of Michigan Tech.

Freshly drawn blood was immediately washed (4–6 times) with
PBS by centrifugation. After the final wash, packed red blood
cells accumulated at the bottom of the centrifuge tube were
used to make a 2% v/v suspension.
5. The instrument was calibrated using the calibration kit contain-
ing ribonuclease A (RNase A) and cytidine 20 -monophosphate
(20 -CMP) supplied by the manufacturer. Thermodynamic para-
meters were calculated from the Gibbs Free Energy equation,
ΔG ¼ ΔH - TΔS ¼ RT ln(Ka), where ΔG, ΔH, and ΔS are
the changes in free energy, enthalpy, and entropy of binding,
respectively. T is the absolute temperature and
R ¼ 1.98 cal mol1 K1.
6. Titrations were done at pH 7.2 using 20 mM PBS buffer. The
concentration range of the lectins should be 5–100 μM and for
the mucins should be 0.6 μM to 0.50 mM based on the dry
weight of the latter molecules. Concentration of the mucins
was based on their molecular weights.
7. Multivalent lectin–mucin interactions often generate insoluble
precipitates under proper stoichiometric condition. Such pre-
cipitation during ITC experiments affects the quality of the
binding isotherm. This problem can be minimized by choosing
proper experimental conditions, including buffers and low
concentration of reactants (lectins and mucins in this case).
ITC can be done at low sample concentration if the affinity of
interaction is high. Newer ITC instruments also possess smaller
sample volumes, thereby reducing the amount of sample
required.
Acknowledgments
Part of this work was supported by National Science Foundation

Grant 1608537 (to T.K.D.). A part of this work was supported by a
startup fund (to P.B.) and by the Research Excellence Fund (to T.K.
D.) provided by Michigan Technological University.
References
1. Drickamer K, Taylor ME (1993) Biology of 4. Akahani S, Makker-Nangia P, Inohara H et al
animal lectins. Annu Rev Cell Biol 9:237–264 (1997) Galectin-3: a novel antiapoptotic mole-
2. Liu FT (2000) Galectins: a new family of reg- cule with a functional BH1 (NHGR) domain
ulators of inflammation. Clin Immunol 97: of Bcl-2 family. Cancer Res 57:5272–5276
79–88 5. Varki A, Cummings RD, Esko JD et al (2009)
3. Konstantinov KN, Robbins BA, Liu FT (1996) Essentials of glycobiology, 2nd edn. Cold
Galectin-3, a β-galactoside-binding animal lec- Spring Harbor Laboratory Press, Cold Spring
tin, is a marker of anaplastic large-cell lym- Harbor, NY
phoma. Am J Pathol 148:25–30
6. Varki A (1993) Biological roles of oligosacchar- 21. Eckhardt AE, Timpte CS, DeLuca AW et al
ides: all of the theories are correct. Glycobiol- (1997) The complete cDNA sequence and
ogy 3:97–101 structural polymorphism of the polypeptide
7. Byrd JC, Bresalier RC (2004) Mucins and chain of porcine submaxillary mucin. J Biol
mucin binding proteins in colorectal cancer. Chem 272:33204–33210
Cancer Metastasis Rev 23:77–99 22. Carlson D (1968) Structures and immunologi-
8. Fukuda M (2002) Role of mucin-type O-gly- cal properties of oligosaccharides isolated from
cans in cell adhesion. Biochim Biophys Acta pig submaxillary mucins. J Biol Chem 243:
1573:394–405 616–626
9. Shogren RL, Gerken TA, Jentoft N (1989) 23. Gerken TA, Jentoft N (1987) Structure and
Role of glycosylation on the conformation dynamics of porcine submaxillary mucin as
and chain dimensions of O-linked glycopro- determined by natural abundance carbon-13
teins: light-scattering studies of ovine submax- nmr spectroscopy. Biochemistry 26:
illary mucin. Biochemistry 28:5526–5536 4689–4699
10. Hang HC, Bertozzi CR (2005) The chemistry 24. Gerken TA, Owens CL, Pasumarthy M (1997)
and biology of mucin-type O-linked glycosyla- Determination of the site-specific O-glycoslya-
tion. Bioorg Med Chem 13:5021–5034 tion pattern of the porcine submaxillary mucin
11. Hanisch FG, Muller S (2000) MUC1: the tandem repeat glycopeptide. J Biol Chem 272:
polymorphic appearance of a human mucin. 9709–9719
Glycobiology 10:439–449 25. Sorensen AL, Celso AR, Trap MA et al (2006)
12. Hakomori S (2002) The glycosynapse. Proc Chemoenzymatically synthesized multimeric
Natl Acad Sci U S A 99:225–232 Tn/STn MUC1 glycopeptides elicit cancer
specific anti-MUC1 antibody responses and
13. Singh PK, Hollingsworth MA (2006) Cell override tolerance. Glycobiology 16:96–107
surface-associated mucins in signal transduc-
tion. Trends Cell Biol 16:467–476 26. Goldstein IJ, Poretz RD (1986) Isolation,
physicochemical characterization, and
14. Dube DH, Bertozzi CR (2005) Glycans in carbohydrate-binding specificity of lectins. In:
cancer and inflammation - potential for thera- Liener IE, Sharon N, Goldstein IJ (eds) The
peutics and diagnostics. Nat Rev Drug Discov lectins. Academic Press, Inc., New York
4:477–488
27. Dam TK, Gabius HJ, Andre S et al (2005)
15. Yu LG, Andrews N, Zhao Q et al (2007) Galectins bind to the multivalent glycoprotein
Galectin-3 interaction with thomsen- asialofetuin with enhanced affinities and a gra-
friedenreich disaccharide on cancer-associated dient of decreasing binding constants. Bio-
MUC1 causes increased cancer cell endothelial chemistry 44:12564–12571
adhesion. J Biol Chem 282:773–781
28. Dam TK, Gerken TA, Cavada BS et al (2007)
16. Brewer CF, Miceli MC, Baum LG (2002) Binding studies of α-GalNAc specific lectins to
Clusters, bundles, arrays and lattices: novel the α-GalNAc (Tn-antigen) form of porcine
mechanisms for lectin-saccharide-mediated cel- submaxillary mucin and its smaller fragments.
lular interactions. Curr Opin Struct Biol 12: J Biol Chem 282:28256–22826
616–623
29. Kiessling LL, Pontrello JK, Schuster MC
17. Lau KS, Partridge EA, Grigorian A et al (2007) (2003) Synthetic multivalent carbohydrate
Complex N-glycan number and degree of ligands as effectors or inhibitors of biological
branching cooperate to regulate cell prolifera- processes. In: Wong CH (ed) Carbohydrate-
tion and differentiation. Cell 129:123–134 based drug discovery. Wiley-VCH, Weinheim
18. Daniels MA, Hogquist KA, Jameson SC 30. Kiessling LL, Gestwicki JE, Strong LE (2006)
(2002) Sweet ’n’ sour: the impact of differen- Synthetic multivalent ligands as probes of sig-
tial glycosylation on T cell responses. Nat nal transduction. Angew Chem Int Ed Eng 45:
Immunol 3:903–910 2348–2368
19. Pace KE, Lee C, Stewart PL et al (1999) 31. Dam TK, Brewer CF (2007) Fundamentals of
Restricted receptor segregation into membrane lectin-carbohydrate interactions. In: Kamerling
microdomains occurs on human T cells during JP (ed) Comprehensive glycoscience, vol
apoptosis induced by galectin 1. J Immunol 3. Elsevier, Ltd., Oxford
163:3801–3811
32. Kanai M, Mortell KLL (1997) Varying the size
20. Gerken TA (2004) Kinetic modeling confirms of multivalent ligands: the dependence of con-
the biosynthesis of mucin core 1 (β-Gal(1-3)- canavalin A binding on neoglycopolymer
α-GalNAc-O-Ser/Thr) O-glycan structures are length. J Am Chem Soc 119:9931–9932
modulated by neighboring glycosylation
effects. Biochemistry 43:4137–4142
33. Dam TK, Roy R, Das SK et al (2000) Binding 39. Dam TK, Brewer CF (2002) Thermodynamic
of multivalent carbohydrates to concanavalin A studies of lectin-carbohydrate interactions by
and Dioclea grandiflora lectin. Thermody- isothermal titration calorimetry. Chem Rev
namic analysis of the “multivalency effect”. J 102:387–429
Biol Chem 275:14223–14230 40. Calvete JJ, Santos CF, Mann K et al (1998)
34. Lotan R, Skutelsky E, Danon D et al (1975) Amino acid sequence, glycan structure, and
The purification, composition, and specificity proteolytic processing of the lectin of Vatairea
of the anti-T lectin from peanut (Arachis hypo- macrocarpa seeds. FEBS Lett 452:286–292
gaea). J Biol Chem 250:8518–8523 41. Burchell JM, Beatson R, Graham R et al (2018)
35. Dessen A, Gupta D, Sabesan S et al (1995) O-linked mucin-type glycosylation in breast
X-ray crystal structure of the soybean aggluti- cancer. Biochem Soc Trans 46:779–788
nin cross-linked with a biantennary analog of 42. Morozov V, Borkowski J, Hanisch FG (2018)
the blood group I carbohydrate antigen. Bio- The double face of mucin-type O-glycans in
chemistry 34:4933–4942 lectin-mediated infection and immunity. Mole-
36. Gestwicki JE, Strong LE, Cairo CW et al cules 23:1151
(2002) Cell aggregation by scaffolded receptor 43. Carlow DA, Gossens K, Naus S et al (2009)
clusters. Chem Biol 9:163–169 PSGL-1 function in immunity and steady state
37. Mandal DK, Brewer CF (1992) Cross-linking homeostasis. Immunol Rev 230:75–96
activity of the 14-kilodalton β-galactose-spe- 44. Tamura M, Tanaka T, Fujii N et al (2020)
cific vertebrate lectin with asialofetuin: compar- Potential interaction between galectin-2 and
ison with several galactose-specific plant lectins. MUC5AC in mouse gastric mucus. Biol
Biochemistry 31:8465–8472 Pharm Bull 43:356–360
38. Dam TK, Roy R, Page D et al (2002) Negative 45. Leclaire C, Lecointe K, Gunning PA et al
cooperativity associated with binding of multi- (2018) Molecular basis for intestinal mucin
valent carbohydrates to lectins. Thermody- recognition by galectin-3 and C-type lectins.
namic analysis of the “multivalency effect”. FASEB J 32(6):3301–3320
Biochemistry 41:1351–1358
Chapter 11
Examination of Whole-Cell Galectin Binding by Solid Phase

and Flow Cytometric Analysis
Anne Lepp€anen, Connie M. Arthur, Sean R. Stowell,
and Richard D. Cummings
Abstract
We have utilized simple flow cytometric and fluorescence-based solid phase assays to study the interaction of
glycan binding proteins (GBP) to cell surface glycoconjugates. These methods utilize commonly employed
flow cytometry techniques and commercially available streptavidin-coated microplates to immobilize
various biotinylated ligands, such as glycopeptides, oligosaccharides, and whole cells. Using this approach,
fluorescently labeled GBPs, in particular, members of the galectin family, can be interrogated for potential
interactions with cell surface carbohydrates, including elucidation of the potential impact of alterations in
glycosylation on carbohydrate recognition. Using these approaches, we present examples of flow cytometric
and fluorescence-based solid phase assays to study galectin–carbohydrate interactions.
Key words Galectin-1, Solid phase assay, Biotinylation, Fluorescence labeling, Immobilization, Bind-
ing affinity
1 Introduction
Glycan binding proteins (GBPs) often regulate a wide variety of

biological processes through recognition of highly modifiable ter-
minal residue recognition. However, in addition to the potential
impact of terminal modifications, the core backbone of many gly-
can structures can influence terminal glycan binding [1–7]. As a
result, binding data obtained following analysis on glycan micro-
arrays that primarily contain relatively simple terminal glycan motifs
may not fully illustrate the impact of core glycan structure on
overall binding interactions [8, 9]. To overcome these limitations,
various approaches have employed natural carbohydrates harvested
from glycoproteins or glycolipids [8, 10–15]. However, in addition
to the glycan backbone, cell surface presentation of glycans can also
significantly impact galectin–carbohydrate interactions
[16, 17]. Engagement of distinct glycan ligands may be responsible
for the unique outcomes that can occur following engagement by
187
188 €nen et al.
Anne Leppa
distinct members of the galectin family [18–23]. As a result, in

addition to examining galectin–carbohydrate specificity using
defined arrays populated with synthetic or naturally occurring gly-
cans, interrogating galectin binding specificity toward intact cells
can provide a complementary tool to investigate the potential
impact of cell surface glycan alterations on the binding interactions
of various galectin family members.
Although previous studies have employed a variety of formats
to examine galectin binding toward cell surface carbohydrates, flow
cytometric analysis of galectin interactions with mammalian cells or
microbes likely reflects the most commonly employed method
[24, 25]. This approach enables a relatively high-throughput
method of evaluating significant changes in galectin engagement
following genetic or enzymatic alterations in cell surface glycans
(Fig. 1) [16, 26–32]. As different cell populations may express
distinct galectin ligands [33], this approach also allows examination
of potential galectin interactions with distinct populations of cells
within a complex mixture, providing a useful tool to not only
examine the potential impact of changes in glycosylation on galec-
tin binding and function but also characterize previously unrecog-
nized cellular subsets that may express distinct carbohydrate ligand
for various members of the galectin family.
While flow cytometric approaches provide a useful tool to
examine general interactions between galectins and cell surface
carbohydrates, the concentrations of galectins utilized in this for-
mat rarely saturate cell surface counter receptors prior to inducing
significant agglutination. In contrast to antibody engagement
against most antigen targets used in flow cytometry [34–36],
galectin-induced agglutination can induce artifacts during flow
cytometric analysis secondary to cell fragmentation. Although
most antibodies used in flow cytometry fail to induce cellular
aggregation [37–40], some antibodies, antibodies capable of
agglutinating cells, can likewise induce cellular fragmentation
[41], which can likewise alter the cellular profile and apparent
density of ligands. Although sub-agglutinating concentrations of
galectins are often employed to avoid this limitation, these
sub-saturating concentrations also limit complete analysis of bind-
ing as saturation can rarely be achieved in order to fully appreciate
the impact of ligand density and other potential changes on binding
affinity. In addition, examination of galectin binding in the linear
range of ligand engagement can result in significant inter-assay
variability as a result of subtle differences in the concentration of
galectins employed in different assay conditions. By contrast, most
flow cytometric analyses purposely employ monoclonal antibodies
at saturating levels to ensure complete engagement of target ligands
in order to adequately identify potential alteration in ligand density
between cellular subsets and following potential ligand modifica-
tions [42–44]. In addition, flow cytometric analysis does not
Examination of Whole-Cell Galectin Binding by Solid Phase and Flow. . . 189
Fig. 1 Gal-8N and Gal-8C display differential recognition of cell surface glycans. (a) Schematic representation
of full-length Gal-8 and individual domains. (b) Binding of Gal-8N toward HL60 cells with or without incubation
20 mM TDG or 20 mM sucrose. (c) Binding of Gal-8N toward HL60 cells treated with A. ureafaciens
neuraminidase. (d) Binding of Gal-8C toward HL60 cells treated with A. ureafaciens neuraminidase. (e)
Geometric mean fluorescent intensities (GeoMFI), a measure of mean fluorescent intensity on logarithmic
scales, of Gal-8N binding before and after treatment of cells with A. ureafaciens neuraminidase. (f) GeoMFI of
Gal-8C binding before and after treatment of cells with A. ureafaciens neuraminidase. (g) Comparison of
Gal-8, Gal-8N, or Gal-8C binding toward HL60 cells following treatment with A. ureafaciens neuraminidase.
Bars represent the percent change in cell surface binding when compared with the mean fluorescent intensity
of non-treated cells SD. (This research was originally published in the Journal of Biological Chemistry.
Stowell SR, Arthur CM, Slanina KA, Horton JR, Smith DF, Cummings RD. Dimeric Galectin-8 induces
phosphatidylserine exposure in leukocytes through polylactosamine recognition by the C-terminal
domain.2008 Jul 18;283(29):20547–59 © the American Society for Biochemistry and Molecular Biology)
190 €nen et al.
Anne Leppa
Fluorescently-labeled
galectin-1
HL-60
cell
Biotin with spacer
Immobilized HL-60 cells
Fig. 2 Galectin-1 binding to immobilized HL-60 cells as an example of

fluorescence-based solid phase assay to study GBP–glycoconjugate
interactions on streptavidin-coated microplates
necessarily provide a measurement of affinity to determine whether,

in addition to alterations in ligand density, the actual affinity of
galectins toward carbohydrate ligands may differ between unique
cell populations decorated with different glycan ligands.
To overcome inherent limitations associated with flow cyto-
metric analysis, we have also utilized a simple fluorescence-based
solid phase assay on a microplate format to study interactions
between galectins and cell surface glycoconjugates. This approach
appears to avoid potential artifacts that may be introduced when
using saturating concentrations of galectins [45–47]. To accom-
plish this, biotinylated human HL-60 cells or T cells were immo-
bilized onto the streptavidin-coated microplates and probed with
fluorescently labeled galectins (Fig. 2) [26, 46]. A portion of cells
were also treated sequentially with different glycosidases before
immobilization into the microtiter plates to gather structural infor-
mation regarding the impact of carbohydrate modification on cell
surface binding [26, 46]. Parallel binding experiments were per-
formed with fluorescently labeled plant lectins with known binding
specificity to control for the effect of glycosidases on cells [26, 46,
48, 49]. By using serially diluted galectins in solid phase assay with
immobilized cells, apparent binding affinity of galectin to cell sur-
face glycoconjugates was obtained [46, 47]. Thus, this approach
not only facilitates examination of the impact of alterations in
glycosylation on galectin binding but also provides a useful
approach to determine the relative affinity of galectins toward
various cell surface-associated glycans.
Regardless of the approach used, galectin typically require

labeling using a variety of fluorescent probes that react with differ-
ent functional groups of a protein to facilitate flow cytometric or
flourimeter detection following potential binding reactions. Fluo-
rescent probes can be covalently attached in a variety of ways,
including via primary amines, reduced cysteine residues, or oxi-
dized carbohydrate residues. The fluorescent probe should be
selected carefully, because covalent derivatization of amino acid
residues involved in ligand binding may result in galectin inactiva-
tion. Therefore, it is important to test the activity of the labeled
galectins before using it in an assay. If a monoclonal antibody is
available to a galectin under study, an antibody can be fluorescently
labeled and used to detect the bound galectin in the assay. Alterna-
tively, if a recombinant galectin has been produced as an IgG fusion
protein [50], commercial fluorescently labeled anti-IgG monoclo-
nal antibody can be used for detection. Taken together, these
methods provide a variety of approaches to elucidate galectin inter-
actions with highly modifiable cell surface glycans.
2 Materials
2.1 Biotinylation and 1. HL-60 cells (other types of cells can also be used).
Fixation of HL-60 Cells 2. Phosphate-buffered saline (PBS) and PBS, pH 8.0.
3. 8% paraformaldehyde in PBS, pH 7.2 (prepare fresh or store at
20 C).
4. EZ-Link™ Sulfo-NHS-LC-Biotin (Pierce).
5. Burker chamber (or suitable equipment) for counting cells.
2.2 Glycosidase 1. Arthrobacter ureafaciens neuraminidase (e.g., Roche Diagnos-

Treatments of tics, Mannheim, Germany).
Biotinylated and Fixed 2. Escherichia freundii endo-β-galactosidase (e.g., V-Labs, Inc.,
HL-60 Cells Covington, LA).
3. Jack bean β-galactosidase (e.g., Glyko).
4. PBS (standard) pH 7.4.
5. PBS, pH 5.0 containing 1 mg/mL BSA.
2.3 Fluorescence 1. Recombinant human dimeric Galectin-1.

Labeling of Gal-1 2. Alexa Fluor 488 carboxylic acid succinimidyl ester
Through Primary (Invitrogen).
Amines
3. 0.1 M lactose in PBS.
4. 14 mM β-mercaptoethanol in PBS.
192 €nen et al.
Anne Leppa
5. PD-10 (or equivalent) small gel filtration column

(GE Healthcare).
6. α-Lactosyl-agarose (Sigma) (1–2 mL volume).
2.4 Fluorescence 1. Recombinant human dimeric Galectin-1.

Labeling of Gal-1 2. Alexa Fluor 488 C5-maleimide (Invitrogen).
Through Cysteines
3. 0.1 M lactose in PBS.
4. 14 mM β-mercaptoethanol in PBS.
(GE Healthcare).
6. α-Lactosyl-agarose (Sigma) (1–2 mL volume).
2.5 Fluorescence 1. Lycopersicon esculentum (tomato) agglutinin (LEA) (Vector

Labeling of Tomato Laboratories).
Lectin (LEA) Through 2. Sodium m-periodate.
Carbohydrates 3. Alexa Fluor 488 hydrazide (Invitrogen), prepare fresh solution
2 mg/mL in PBS.
4. PBS (standard) pH 7.4.
(GE Healthcare).
2.6 Gal-1, LEA, and 1. Streptavidin-coated high binding capacity 96-well microtiter
GS-II Binding to plates (Pierce).
Immobilized HL-60 2. PBS (standard), pH 7.4.
Cells
3. 1% BSA in PBS.
4. Fluorescently labeled lectins: Alexa Fluor 488-labeled Gal-1,
Alexa Fluor 488-labeled LEA, and Fluorescein-labeled Griffo-
nia simplicifolia lectin II (GS-II) (EY Laboratories Inc., San
Mateo, CA).
5. Fluorescence microtiter plate reader with suitable filters (for
Alexa Fluor 488 and Fluorescein excitation at 485 nm and
emission at 535 nm).
6. Eight-channel (or 12-channel) manual pipet (250–300 μL).
2.7 Determination of 1. PBS.

Gal-1 Binding Using 2. 1% BSA in PBS.
Flow Cytometric
3. Fluorescently labeled lectins: Alexa Fluor 488-labeled Gal-1,
Analysis
Alexa Fluor 488-labeled LEA, and Fluorescein-labeled Griffo-
nia simplicifolia lectin II (GS-II) (EY Laboratories Inc., San
Mateo, CA).
4. BD FACSCaliber.
2.8 Determination of 1. Streptavidin-coated high binding capacity 96-well microtiter

Apparent Binding plates (Pierce).
Affinity of Gal-1 for 2. PBS (standard), pH 7.4.
Immobilized HL-60
3. 1% BSA in PBS.
Cells
4. 1% BSA and 20 mM lactose in PBS.
5. Alexa Fluor 488-labeled Gal-1.
6. Fluorescence microtiter plate reader with suitable filters (for
Alexa Fluor 488 and Fluorescein excitation at 485 nm and
emission at 535 nm).
7. Eight-channel (or 12-channel) manual pipet (250–300 μL).
8. Computer software capable to calculate nonlinear curve fitting
(e.g., SigmaPlot).
3 Methods
3.1 Biotinylation and 1. Wash HL-60 cells three times with PBS (see Note 1).
Fixation of HL-60 Cells 2. Suspend cells at a concentration of 20 106 cells/mL in PBS,
pH 8.0.
3. Add 1.3 mg EZ-Link Sulfo-NHS-LC-Biotin per mL of cells
and incubate at RT for 30 min.
4. Wash cells three times with ice-cold PBS.
5. Add 8% paraformaldehyde in PBS to a final concentration of 2%
and incubate at RT for 30 min.
7. After final wash, suspend cells to 1 mL of PBS and count cells.
3.2 Fixation Alone of 1. Wash cells three times with ice-cold PBS.
HL-60 Cells (for 2. Add 8% paraformaldehyde in PBS to final concentration of 2%
Enzymatic Treatment and incubate at RT for 30 min.
Followed by Flow
Cytometric Analysis)
4. After final wash, suspend cells to 1 mL of PBS and count cells.
3.3 Glycosidase 1. Suspend biotinylated and fixed HL-60 cells at the concentra-
Treatments of tion of 5 106 cells/mL in PBS. Take a portion of cells for
Biotinylated and Fixed neuraminidase treatment (proceed to step 2) and leave the rest
(or Fixed Alone) HL-60 of cells untreated (proceed directly to step 4).
Cells 2. Add Arthrobacter ureafaciens neuraminidase to a final concen-
tration of 100 milliunits/mL of cells in PBS and incubate at
37 C for 3 h.
3. Wash neuraminidase-treated cells three times with PBS and
count cells.
194 €nen et al.
Anne Leppa
4. Suspend neuraminidase-treated and -untreated cells (from step

1) into PBS, pH 5.0, containing 1 mg/mL BSA at the concen-
tration of 2–5 106 cells/mL and divide each into three equal
aliquots.
5. Add Escherichia freundii endo-β-galactosidase (250 milliunits/
mL, final concentration) to the first sample, Jack bean
β-galactosidase (100 milliunits/mL, final concentration) to
the second sample, and buffer alone (PBS, pH 5.0, containing
1 mg/mL BSA) to the third sample.
6. Incubate overnight at 37 C.
7. Glycosidase-treated cells and control cells can be immobilized
directly into streptavidin-coated microtiter plates (Subheading
3.6) (see Note 2).
3.4 Fluorescence 1. Human galectin-1 (Gal-1) (1–2 mg) can be labeled through
Labeling of Gal-1 primary amines using Alexa Fluor 488 carboxylic acid succini-
Through Primary midyl ester according to manufacturer’s instructions with
Amines minor modifications as described in the following steps.
2. Incubate Gal-1 with reactive dye in PBS containing 0.1 M
lactose for 1 h at room temperature or by continue incubation
overnight at 4 C with stirring.
3. Remove free dye and lactose from the labeled Gal-1 using a
PD-10 column equilibrated in PBS containing 14 mM
β-mercaptoethanol.
4. To separate functionally active Gal-1 from inactive protein, pass
labeled Gal-1 over a small lactosyl-agarose column (1–2 mL
volume) in PBS. Wash unbound (inactive) Gal-1 from the
column with 3–5 column volumes of PBS. Bound (active)
Gal-1 is eluted with 0.1 M lactose in PBS. Before each experi-
ment, lactose must be removed using a PD-10 column equili-
brated in PBS containing 14 mM β-mercaptoethanol.
5. Gal-1 samples labeled with Alexa Fluor 488 carboxylic acid
succinimidyl ester can be stored in 0.1 M lactose in PBS at
+4 C for few months. Lactose must be removed using a
PD-10 column in PBS containing 14 mM β-mercaptoethanol
before each experiment (see Note 3).
3.5 Fluorescence 1. Accessible cysteine residues on human galectin-1 (Gal-1)

Labeling of Gal-1 (1–1.5 mg) can be labeled using thiol reactive Alexa Fluor
Through Cysteines 488 C5-maleimide according to manufacturer’s instructions
with minor modifications as described in the following steps.
2. Incubate Gal-1 with tenfold molar excess of thiol reactive Alexa
Fluor 488 C5-maleimide in PBS containing 0.1 M lactose
overnight at 4 C under stirring.
3. Remove free dye and lactose from the labeled Gal-1 using a
PD-10 column in PBS containing 14 mM β-mercaptoethanol.
4. To separate functionally active Gal-1 from inactive protein, pass
labeled Gal-1 over a small lactosyl-agarose column (1–2 mL
volume) in PBS. Wash unbound (inactive) Gal-1 from the
column with 3–5 column volumes of PBS. Bound (active)
Gal-1 is eluted with 0.1 M lactose in PBS. Before each experi-
ment, lactose must be removed using a PD-10 column equili-
brated in PBS containing 14 mM β-mercaptoethanol.
5. Gal-1 samples labeled with Alexa Fluor 488 C5-maleimide can
be stored in PBS containing 14 mM β-mercaptoethanol at 4 C
for at least 3 months without losing activity (see Note 4).
3.6 Fluorescence 1. Dissolve Lycopersicon esculentum (tomato) agglutinin (LEA) in

Labeling of Tomato PBS to the final concentration of 4 mg/mL.
Lectin (LEA) Through 2. Weigh out solid sodium m-periodate (to obtain final concen-
Carbohydrates tration of 100 mM in the sample).
3. Add LEA in PBS to the tube containing solid sodium m-
periodate and incubate for 30 min at room temperature in
the dark to oxidize cis-diols of carbohydrates to aldehydes.
4. Remove sodium m-periodate using a PD-10 gel filtration col-
umn in PBS. Collect 0.5 mL fractions and measure absorbance
at 280 nm for each fraction. Pool fractions containing protein
(¼oxidized LEA).
5. Add Alexa Fluor 488 hydrazide solution (2 mg/mL in PBS) to
oxidized LEA sample to the final concentration of 100 μg/mg
lectin.
6. Incubate for 1.5–2 h at room temperature under stirring.
7. Remove free dye using a PD-10 column in PBS. Collect 0.5 mL
fractions and measure absorbance at 280 nm and 494 nm for
each fraction. Pool fractions containing protein (see Note 5).
3.7 Gal-1, LEA, and 1. To immobilize equal amount of biotinylated HL-60 cells in
GS-II Binding to microtiter well, adjust cell density in each sample to 2 106
Immobilized HL-60 cells/mL using PBS (see Note 6).
Cells 2. Wash streptavidin-coated microtiter plates three times with
200 μL of PBS.
3. Immobilize glycosidase-digested and -non-digested HL-60
cells to streptavidin-coated microtiter wells at equivalent den-
sities (100,000 cells/well) in 50 μL of PBS for 1.5 h at room
temperature.
4. Wash the wells three times with 200 μL of PBS containing 1%
BSA. After last wash, remove buffer carefully by tapping the
plate upside down gently against paper towel. Do not let the
plate dry (see Note 7).
196 €nen et al.
Anne Leppa
5. Add 50 μL of fluorescently labeled Gal-1 (40 μg/mL), LEA

(100 μg/mL), or GS-II (100 μg/mL) in PBS containing 1%
BSA and incubate for 1 h at room temperature.
6. Wash the wells four times with 250–300 μL of PBS containing
1% BSA (see Note 7).
7. Add 100 μL of PBS to each well and read the fluorescence by
fluorescence microplate reader. Figure 2 shows results on the
binding of Gal-1, LEA, and GS-II to immobilized HL-60 cells.
3.8 Determination of 1. Prepare dilutions 0.04, 0.08, 0.15, 0.31, 0.75, 1.25, 2.5, and
Gal-1 Binding to HL-60 5.0 μM of fluorescently labeled Gal-1 in PBS with 1% BSA and
Cells Using Flow in PBS with 1% BSA and 20 mM lactose (see Note 8).
Cytometric Analysis 2. Add 50 μL of fluorescently labeled Gal-1 dilutions in PBS with
1% BSA or PBS with 1% BSA and 20 mM lactose to each sample
containing 1 106 cells/mL (final concentration) and incu-
bate for 1 h at room temperature. When analyzing non-fixed
live cells, incubate the indicated concentrations of fluorescently
labeled Gal-1 with 1 106 cells/mL for 30 min at 4 C to
avoid Gal-1 internalization or other alterations that may occur
at room or higher temperatures during the binding assay.
1% BSA. After last wash, remove buffer carefully by tapping the
plate upside down gently against paper towel (see Note 7).
4. Following incubation, evaluate cells by microscope to ensure
that the cells are not agglutinated.
5. Add 300 μL of PBS to each well and examine bound galectin by
evaluating the mean fluorescence intensity of appropriately
gated cells by using commonly employed flow cytometric anal-
ysis techniques. Figure 3 shows results on the determination of
the binding of Gal-1 to HL-60 cells by flow cytometry.
3.9 Determination of 1. Prepare dilutions 0.0625, 0.125, 0.25, 0.5, 1.0, 2.0, 4.0, 8.0,
Apparent Binding 12.0, and 20.0 μM of fluorescently labeled Gal-1 in PBS with
Affinity of Gal-1 for 1% BSA and in PBS with 1% BSA and 20 mM lactose (see Note
Immobilized HL-60 9).
Cells 2. Wash streptavidin-coated microtiter plates three times with
200–300 μL of PBS.
3. Immobilize neuraminidase-digested and -non-digested HL-60
cells to streptavidin-coated microtiter wells at equivalent den-
sity (100,000 cells/well) in 50 μL of PBS for 1.5 h at room
temperature.
4. Wash the wells three times with 200 μL of PBS containing 1%
BSA. After last wash, remove buffer carefully by tapping the
plate upside down gently against paper towel. Do not let the
plate dry (see Note 7).
40000 A Gal-1
no treatment
30000
endo-β-galactosidase
β-galactosidase
RFU
20000
10000
0
B LEA
200000
150000
RFU
100000
50000
0
C GS-II
100000
80000
RFU
60000
40000
20000
0
Desialylated HL-60 cells
HL-60 cells
Fig. 3 Binding of Gal-1, LEA, and GS-II to immobilized desialylated and non-
treated HL-60 cells. A Portion of biotinylated and fixed HL-60 cells were first
desialylated. Nontreated and desialylated HL-60 cells were treated with
endo-β-galactosidase or β-galactosidase, and immobilized on streptavidin-
coated microtiter wells (100,000 cells/well). Fluorescently labeled A, Gal-1
(40 μg/mL); B, LEA (100 μg/mL); and C, GS-II (100 μg/mL) were incubated
with the immobilized cells. All assays were performed in triplicate, and the
198 €nen et al.
Anne Leppa
5. Add 50 μL of fluorescently labeled Gal-1 dilutions in PBS with

1% BSA or PBS with 1% BSA and 20 mM lactose to each well
and incubate for 1 h at room temperature.
1% BSA. After last wash, remove buffer carefully by tapping the
plate upside down gently against paper towel (see Note 7).
7. Add 100 μL of PBS to each well and read the fluorescence by
fluorescence microplate reader.
8. Calculate the apparent dissociation constants (Kd) using non-
linear curve fitting program. Figure 4 shows results on the
determination of the binding affinity of Gal-1 for immobilized
HL-60 cells.
4 Notes
1. When planning the experiment, it should be considered that

relatively large number of cells are required for solid phase
assay. Here we use 100,000 cells/well on 96-well plate. Initial
amount of harvested cells should be considerably larger
because a number of centrifugations during the course of the
experiment result in loss of cells.
2. In the present experiments, we treated biotinylated and fixed
cells with glycosidases before capturing cells on streptavidin
plates. We did not observe significant differences in the binding
assays, if glycosidase treatments were performed on the plate
after capturing biotinylated and fixed cells.
3. Alternatively, labeled Gal-1 can be stored without lactose in
PBS containing 14 mM β-mercaptoethanol at 4 C for a
few days.
4. Gal-1 labeled with Alexa Fluor 488 C5-maleimide is more
stable during long-term storage than Gal-1 labeled with Alexa
Fluor 488 carboxylic acid succinimidyl ester.
5. The protein concentration and degree of labeling can be calcu-
lated using absorbance readings at 280 nm and 494 nm accord-
ing to Alexa Fluor 488 hydrazide manufacturer’s instructions.
The degree of labeling of LEA was significantly higher by
labeling through carbohydrates than in commercial fluores-
cently labeled LEA.
ä
Fig. 3 (continued) results are the means SD of three determinations. (This

research was originally published in the Journal of Biological Chemistry.
Lepp€anen A, Stowell S, Blixt O, Cummings RD. Dimeric galectin-1 binds with
high affinity to alpha2,3-sialylated and non-sialylated terminal N-acetyllactosa-
mine units on surface-bound extended glycans.2005 Feb 18;280(7):5549–62.
© the American Society for Biochemistry and Molecular Biology)
40000
A
Desialylated HL-60 cells
30000
- Lactose
+ Lactose
RFU
20000
Kd = 2.6 ± 0.3 μM
10000
0
0 4 8 12 16 20
Gal-1 [μM]
15000 B
HL-60 cells
- Lactose
10000 + Lactose
RFU
Kd = 5.9 ± 0.7 μM
5000
0
0 4 8 12 16 20
Gal-1 [μM]
Fig. 4 Binding affinity of Gal-1 for immobilized desialylated and nontreated

HL-60 cells. Biotinylated, fixed, desialylated HL-60 cells (a) and biotinylated,
fixed, nontreated HL-60 cells (b) were immobilized on streptavidin-coated
microtiter wells (100,000 cells/well). Various concentrations of Gal-1 were
incubated with the immobilized cells in buffer with or without 20 mM lactose.
All assays were performed in duplicate, and the results are the average of two
determinations. (This research was originally published in the Journal of
Biological Chemistry. Lepp€anen A, Stowell S, Blixt O, Cummings RD. Dimeric
galectin-1 binds with high affinity to alpha2,3-sialylated and non-sialylated
terminal N-acetyllactosamine units on surface-bound extended glycans.2005
Feb 18;280(7):5549–62. © the American Society for Biochemistry and Molecular
Biology)
200 €nen et al.
Anne Leppa
6. Samples should be prepared at least in triplicate and the results

should be calculated as a mean SD.
7. Microtiter plate washing machines should be avoided and
microtiter wells should be washed manually using eight-
channel (or 12-channel) pipet. Fill wells gently with wash
buffer using pipet and empty wells into the waste by inverting
the plate. After the last wash remove buffer carefully by gently
tapping the plate against paper towel. Do not let the wells to
dry at any point during the experiment.
8. Examination of potential Gal-1 interactions over a range of
concentrations will allow a linear and sub-agglutinating range
of concentrations to be established. Comparison between dif-
ferent potential cellular subsets, enzymatic treatments, or
genetic changes in glycosylation should be done by examining
potential differences in binding at the same concentration of
galectin. As subtle changes in concentration of galectin may
result in some degree of inter-assay variability, differences in
binding between cell subsets within an assay may be best
enumerated by examining percent changes in comparison to
control cells.
9. Samples should be prepared at least in triplicate and the results
should be calculated as an average of three determinations.
References
1. Johannes L, Jacob R, Leffler H (2018) Galec- 6. Kamili NA, Arthur CM, Gerner-Smidt C,
tins at a glance. J Cell Sci 131(9):jcs208884. Tafesse E, Blenda A, Dias-Baruffi M, Stowell
https://doi.org/10.1242/jcs.208884 SR (2016) Key regulators of galectin-glycan
2. Liu FT, Rabinovich GA (2010) Galectins: reg- interactions. Proteomics 16(24):3111–3125.
ulators of acute and chronic inflammation. Ann https://doi.org/10.1002/pmic.201600116
N Y Acad Sci 1183:158–182. https://doi.org/ 7. Liu FT, Yang RY, Hsu DK (2012) Galectins in
10.1111/j.1749-6632.2009.05131.x acute and chronic inflammation. Ann N Y Acad
3. Cerliani JP, Stowell SR, Mascanfroni ID, Sci 1253:80–91. https://doi.org/10.1111/j.
Arthur CM, Cummings RD, Rabinovich GA 1749-6632.2011.06386.x
(2011) Expanding the universe of cytokines 8. Arthur CM, Cummings RD, Stowell SR
and pattern recognition receptors: galectins (2014) Using glycan microarrays to under-
and glycans in innate immunity. J Clin Immu- stand immunity. Curr Opin Chem Biol
nol 31(1):10–21. https://doi.org/10.1007/ 18C:55–61. https://doi.org/10.1016/j.cbpa.
s10875-010-9494-2 2013.12.017
4. van Kooyk Y, Rabinovich GA (2008) Protein- 9. Blixt O, Head S, Mondala T, Scanlan C,
glycan interactions in the control of innate and Huflejt ME, Alvarez R, Bryan MC, Fazio F,
adaptive immune responses. Nat Immunol Calarese D, Stevens J, Razi N, Stevens DJ,
9(6):593–601. https://doi.org/10.1038/ni. Skehel JJ, van Die I, Burton DR, Wilson IA,
f.203 Cummings R, Bovin N, Wong CH, Paulson JC
5. Poland PA, Rondanino C, Kinlough CL, (2004) Printed covalent glycan array for ligand
Heimburg-Molinaro J, Arthur CM, Stowell profiling of diverse glycan binding proteins.
SR, Smith DF, Hughey RP (2011) Identifica- Proc Natl Acad Sci U S A 101(49):
tion and characterization of endogenous galec- 17033–17038. https://doi.org/10.1073/
tins expressed in Madin Darby canine kidney pnas.0407902101
cells. J Biol Chem 286(8):6780–6790. 10. Song X, Xia B, Stowell SR, Lasanajak Y, Smith
https://doi.org/10.1074/jbc.M110.179002 DF, Cummings RD (2009) Novel fluorescent
glycan microarray strategy reveals ligands for Van Ry PM (2020) Treatment with galectin-1
galectins. Chem Biol 16(1):36–47. https:// improves myogenic potential and membrane
doi.org/10.1016/j.chembiol.2008.11.004 repair in dysferlin-deficient models. PLoS One
11. Song X, Lasanajak Y, Olson LJ, Boonen M, 15(9):e0238441. https://doi.org/10.1371/
Dahms NM, Kornfeld S, Cummings RD, journal.pone.0238441
Smith DF (2009) Glycan microarray analysis 21. Rodrigues LC, Kabeya LM, Azzolini A, Cerri
of P-type lectins reveals distinct phosphoman- DG, Stowell SR, Cummings RD, Lucisano-
nose glycan recognition. J Biol Chem 284(50): Valim YM, Dias-Baruffi M (2019) Galectin-1
35201–35214. https://doi.org/10.1074/jbc. modulation of neutrophil reactive oxygen spe-
M109.056119 cies production depends on the cell activation
12. de Boer AR, Hokke CH, Deelder AM, Wuhrer state. Mol Immunol 116:80–89. https://doi.
M (2007) General microarray technique for org/10.1016/j.molimm.2019.10.001
immobilization and screening of natural gly- 22. Stowell SR, Cho M, Feasley CL, Arthur CM,
cans. Anal Chem 79(21):8107–8113. https:// Song X, Colucci JK, Karmakar S, Mehta P,
doi.org/10.1021/ac071187g Dias-Baruffi M, McEver RP, Cummings RD
13. Song X, Lasanajak Y, Rivera-Marrero C, (2009) Ligand reduces galectin-1 sensitivity
Luyai A, Willard M, Smith DF, Cummings to oxidative inactivation by enhancing dimer
RD (2009) Generation of a natural glycan formation. J Biol Chem 284(8):4989–4999.
microarray using 9-fluorenylmethyl chlorofor- https://doi.org/10.1074/jbc.M808925200
mate (FmocCl) as a cleavable fluorescent tag. 23. Stowell SR, Dias-Baruffi M, Penttila L,
Anal Biochem 395(2):151–160. https://doi. Renkonen O, Nyame AK, Cummings RD
org/10.1016/j.ab.2009.08.024 (2004) Human galectin-1 recognition of
14. Liu Y, Palma AS, Feizi T (2009) Carbohydrate poly-N-acetyllactosamine and chimeric poly-
microarrays: key developments in glycobiology. saccharides. Glycobiology 14(2):157–167.
Biol Chem 390(7):647–656. https://doi.org/ https://doi.org/10.1093/glycob/cwh018
10.1515/BC.2009.071 24. Kohatsu L, Hsu DK, Jegalian AG, Liu FT,
15. Song X, Lasanajak Y, Xia B, Heimburg- Baum LG (2006) Galectin-3 induces death of
Molinaro J, Rhea JM, Ju H, Zhao C, Molinaro Candida species expressing specific beta-1,2-
RJ, Cummings RD, Smith DF (2011) Shotgun linked mannans. J Immunol 177(7):
glycomics: a microarray strategy for functional 4718–4726
glycomics. Nat Methods 8(1):85–90. https:// 25. Karmakar S, Stowell SR, Cummings RD,
doi.org/10.1038/nmeth.1540 McEver RP (2008) Galectin-1 signaling in leu-
16. Patnaik SK, Potvin B, Carlsson S, Sturm D, kocytes requires expression of complex-type
Leffler H, Stanley P (2006) Complex N-glycans. Glycobiology 18(10):770–778
N-glycans are the major ligands for galectin-1, 26. Stowell SR, Arthur CM, Mehta P, Slanina KA,
-3, and -8 on Chinese hamster ovary cells. Gly- Blixt O, Leffler H, Smith DF, Cummings RD
cobiology 16(4):305–317. https://doi.org/ (2008) Galectin-1, -2, and -3 exhibit differen-
10.1093/glycob/cwj063 tial recognition of sialylated glycans and blood
17. Arthur CM, Baruffi MD, Cummings RD, group antigens. J Biol Chem 283(15):
Stowell SR (2015) Evolving mechanistic 10109–10123. https://doi.org/10.1074/jbc.
insights into galectin functions. Methods Mol M709545200
Biol 1207:1–35. https://doi.org/10.1007/ 27. Stowell SR, Arthur CM, Slanina KA, Horton
978-1-4939-1396-1_1 JR, Smith DF, Cummings RD (2008) Dimeric
18. Robinson BS, Arthur CM, Kamili NA, Stowell Galectin-8 induces phosphatidylserine expo-
SR (2018) Galectin regulation of host micro- sure in leukocytes through polylactosamine
bial interactions. Trends Glycosci Glycotechnol recognition by the C-terminal domain. J Biol
30(172):SE185–SE198 Chem 283(29):20547–20559. https://doi.
19. Robinson BS, Saeedi B, Arthur CM, Owens J, org/10.1074/jbc.M802495200
Naudin C, Ahmed N, Luo L, Jones R, Neish A, 28. Stowell SR, Arthur CM, Dias-Baruffi M,
Stowell SR (2020) Galectin-9 is a novel regula- Rodrigues LC, Gourdine JP, Heimburg-
tor of epithelial restitution. Am J Pathol Molinaro J, Ju T, Molinaro RJ, Rivera-
190(8):1657–1666. https://doi.org/10. Marrero C, Xia B, Smith DF, Cummings RD
1016/j.ajpath.2020.04.010 (2010) Innate immune lectins kill bacteria
20. Vallecillo-Zuniga ML, Rathgeber MF, Poulson expressing blood group antigen. Nat Med
PD, Hayes S, Luddington JS, Gill HN, 16(3):295–301. https://doi.org/10.1038/
Teynor M, Kartchner BC, Valdoz J, nm.2103
Stowell C, Markham AR, Arthur C, Stowell S,
202 €nen et al.
Anne Leppa
29. Bax M, Garcia-Vallejo JJ, Jang-Lee J, North SJ, 37. Patel SR, Bennett A, Girard-Pierce K, Maier
Gilmartin TJ, Hernandez G, Crocker PR, CL, Chonat S, Arthur CM, Zerra PE,
Leffler H, Head SR, Haslam SM, Dell A, van Mener A, Stowell SR (2018) Recipient priming
Kooyk Y (2007) Dendritic cell maturation to one RBC alloantigen directly enhances
results in pronounced changes in glycan subsequent alloimmunization in mice. Blood
expression affecting recognition by siglecs and Adv 2(2):105–115. https://doi.org/10.
galectins. J Immunol 179(12):8216–8224 1182/bloodadvances.2017010124
30. Carlsson S, Oberg CT, Carlsson MC, 38. Sullivan HC, Gerner-Smidt C, Nooka AK,
Sundin A, Nilsson UJ, Smith D, Cummings Arthur CM, Thompson L, Mener A, Patel SR,
RD, Almkvist J, Karlsson A, Leffler H (2007) Yee M, Fasano RM, Josephson CD, Kaufman
Affinity of galectin-8 and its carbohydrate rec- RM, Roback JD, Lonial S, Stowell SR (2017)
ognition domains for ligands in solution and at Daratumumab (anti-CD38) induces loss of
the cell surface. Glycobiology 17(6):663–676. CD38 on red blood cells. Blood 129(22):
https://doi.org/10.1093/glycob/cwm026 3033–3037. https://doi.org/10.1182/
31. Stowell SR, Arthur CM, McBride R, Berger O, blood-2016-11-749432
Razi N, Heimburg-Molinaro J, Rodrigues JP, 39. Sullivan HC, Arthur CM, Thompson L, Patel
Noll AJ, von Gunten S, Smith DF, Knirel YA, SR, Stowell SR, Hendrickson JE, Lazarus AH
Paulson JC, Cummings RD (2014) Microbial (2018) Anti-RhD reduces levels of detectable
glycan microarrays define key features of host- RhD antigen following anti-RhD infusion.
microbial interactions. Nat Chem Biol 10(6): Transfusion 58(2):542–544. https://doi.org/
470–476 10.1111/trf.14452
32. Stowell SR, Arthur CM, McBride R, Berger O, 40. Zerra PE, Arthur CM, Chonat S, Maier CL,
Razi N, Heimburg-Molinaro J, Rodrigues LC, Mener A, Shin S, Allen JWL, Baldwin WH,
Gourdine JP, Noll AJ, von Gunten S, Smith Cox C, Verkerke H, Jajosky RP, Tormey CA,
DF, Knirel YA, Paulson JC, Cummings RD Meeks SL, Stowell SR (2020) Fc gamma recep-
(2014) Microbial glycan microarrays define tors and complement component 3 facilitate
key features of host-microbial interactions. anti-fVIII antibody formation. Front Immunol
Nat Chem Biol 10(6):470–476. https://doi. 11:905. https://doi.org/10.3389/fimmu.
org/10.1038/nchembio.1525 2020.00905
33. Robinson BS, Arthur CM, Evavold B, 41. Liepkalns JS, Hod EA, Stowell SR, Cadwell
Roback E, Kamili NA, Stowell CS, Vallecillo- CM, Spitalnik SL, Zimring JC (2012) Biphasic
Zuniga ML, Van Ry PM, Dias-Baruffi M, clearance of incompatible red blood cells
Cummings RD, Stowell SR (2019) The through a novel mechanism requiring neither
sweet-side of leukocytes: galectins as master complement nor Fcgamma receptors in a
regulators of neutrophil function. Front murine model. Transfusion 52(12):
Immunol 10:1762. https://doi.org/10. 2631–2645. https://doi.org/10.1111/j.
3389/fimmu.2019.01762 1537-2995.2012.03647.x
34. Mener A, Arthur CM, Patel SR, Liu J, Hen- 42. Maier CL, Mener A, Patel SR, Jajosky RP,
drickson JE, Stowell SR (2018) Complement Bennett AL, Arthur CM, Hendrickson JE,
component 3 negatively regulates antibody Stowell SR (2018) Antibody-mediated
response by modulation of red blood cell anti- immune suppression by antigen modulation is
gen. Front Immunol 9:676. https://doi.org/ antigen-specific. Blood Adv 2(21):2986–3000.
10.3389/fimmu.2018.00676 https://doi.org/10.1182/bloodadvances.
35. Mener A, Patel SR, Arthur CM, Chonat S, 2018018408
Wieland A, Santhanakrishnan M, Liu J, Maier 43. Zerra PEPS, Jajosky RP, Arthur CM, McCoy
CL, Jajosky RP, Girard-Pierce K, Bennett A, JW, Allen JWL, Chonat S, Fasano R, Roback
Zerra PE, Smith NH, Hendrickson JE, Stowell JD, Josephson CD, Hendrickson JE, Stowell
SR (2018) Complement serves as a switch SR (2021) Marginal zone B cells mediate a
between CD4+ T cell-independent and CD4 T cell dependent extrafollicular antibody
-dependent RBC antibody responses. JCI response following RBC transfusion in mice.
Insight 3(22):e121631. https://doi.org/10. Blood 138(8):706–721
1172/jci.insight.121631 44. Arthur CM, Allen JWL, Verkerke H, Yoo J,
36. Mener A, Patel SR, Arthur CM, Stowell SR Jajosky RP, Girard-Pierce K, Chonat S,
(2019) Antibody-mediated immunosuppres- Zerra P, Maier C, Rha J, Fasano R, Josephson
sion can result from RBC antigen loss indepen- CD, Roback JD, Stowell SR (2021) Antigen
dent of Fcgamma receptors in mice. density dictates RBC clearance, but not antigen
Transfusion 59(1):371–384. https://doi.org/ modulation, following incompatible RBC
10.1111/trf.14939 transfusion in mice. Blood Adv 5(2):527–538.
https://doi.org/10.1182/bloodadvances. 48. Lyer PN, Wilkinson KD, Goldstein LJ (1976)

2020002695 An -N-acetyl-D-glycosamine binding lectin
45. Leffler H, Carlsson S, Hedlund M, Qian Y, from Bandeiraea simplicifolia seeds. Arch Bio-
Poirier F (2004) Introduction to galectins. chem Biophys 177(1):330–333
Glycoconj J 19(7–9):433–440. https://doi. 49. Merkle RK, Cummings RD (1987) Relation-
org/10.1023/B:GLYC.0000014072. ship of the terminal sequences to the length of
34840.04 poly-N-acetyllactosamine chains in asparagine-
46. Leppanen A, Stowell S, Blixt O, Cummings linked oligosaccharides from the mouse lym-
RD (2005) Dimeric galectin-1 binds with phoma cell line BW5147. Immobilized tomato
high affinity to alpha2,3-sialylated and lectin interacts with high affinity with glyco-
non-sialylated terminal N-acetyllactosamine peptides containing long poly-N-acetyllactosa-
units on surface-bound extended glycans. J mine chains. J Biol Chem 262(17):8179–8189
Biol Chem 280(7):5549–5562 50. Cedeno-Laurent F, Barthel SR, Opperman MJ,
47. Stowell SR, Qian Y, Karmakar S, Koyama NS, Lee DM, Clark RA, Dimitroff CJ (2010)
Dias-Baruffi M, Leffler H, McEver RP, Cum- Development of a nascent galectin-1 chimeric
mings RD (2008) Differential roles of galectin- molecule for studying the role of leukocyte
1 and galectin-3 in regulating leukocyte viabil- galectin-1 ligands and immune disease modu-
ity and cytokine secretion. J Immunol 180(5): lation. J Immunol 185(8):4659–4672.
3091–3102 h t t p s : // d o i . o r g / 1 0 . 4 0 4 9 / j i m m u n o l .
1000715
Chapter 12
Dissecting Context-Specific Galectin Binding Using

Glycoengineered Cell Libraries
Mathias Ingemann Nielsen and Hans H. Wandall
Abstract
The family of galectins has critical functions in a wide range of biological processes, primarily based on their
broad interactions with proteins carrying β-galactoside-containing glycans. To understand the diversity of
functions governed by galectins, it is essential to define the binding specificity of the carbohydrate
recognition domain (CRDs) of the individual galectins. The binding specificity of galectins has primarily
been examined with glycoarrays, but now the ability to probe and dissect binding to defined glycans in the
context of a cellular membrane is facilitated by the generations of glycoengineered cell libraries with defined
glyco-phenotypes. The following section will show how galectin specificities can be probed in the natural
context of cellular surfaces using glycoengineered cell libraries, and how binding to glycoproteins can be
measured in solution with fluorescence anisotropy.
Key words Galectin, Substrate specificity, Carbohydrate binding proteins, Glycoengineering, Glyco-
genes, Gene editing, Glycosyltransferases, Glycosylation
1 Introduction
Galectins are carbohydrate binding proteins with broad implica-

tions in inflammation, immunity, and cancer [1, 2]. Given their
involvement in many key biological processes, there is a continuous
interest in understanding the binding specificities of the individual
galectin members, especially in a natural context.
Mammalian galectins are divided into three subgroups includ-
ing prototypical, chimeric, and the tandem-repeat galectins, based
on the structural organization of their carbohydrate recognition
domains (CRDs). This interaction is facilitated through at least one
structurally conserved carbohydrate-recognition domain (CRD)
comprising ~130 amino acids [3]. The galectin CRD is a bent
β-sandwich, with a binding groove sufficiently long to accommo-
date a linear tetrasaccharide, and accordingly divided into
subsites A, B, C, and D. Interaction with a β-galactose unit in site
C is essential for the glycan binding, and carbohydrate residues on
205
206 Mathias Ingemann Nielsen and Hans H. Wandall
either side of the β-galactose may either increase or decrease bind-

ing, giving each galectin CRD its fine specificity for larger
saccharides [3].
Prototypical galectins contain a single carbohydrate recogni-
tion domain and have the ability to form homodimers (galectins -1,
-2, -5, -7, -10, -13, -14, and -15) [4]. The second group, chimeric
galectins, has only one member in vertebrates, galectin-3. Galectin-
3 possess a single C-terminal carbohydrate recognition domain and
a unique non-lectin N-terminal domain [5]. The last subgroup, the
tandem-repeat galectins, carry two structurally and functionally
distinct carbohydrate recognition domains, bridged by a small
linker domain (galectins -4, -6, -8, -9, and -12) [4, 6]. Several
studies have collectively contributed to our current understanding
of galectin specificity. Some of the methods that have been used to
define galectin binding determinants include frontal affinity chro-
matography [7], fluorescence polarization [5, 8, 9], and glycan
arrays [10–12].
These studies have for the most part revealed that the majority
of galectins bind β-galactose residues widely distributed on glyco-
proteins and glycosphingolipids, while individual galectins require
additional distinct structures for high-affinity binding. For exam-
ple, it has been shown that galectin-1, -3, and -8 bind to N-glycans
especially through poly-LacNAc extensions, while other galectins,
such as galectin-4, only bind weakly to N-linked glycans, and
instead bind predominantly to glycosphingolipids [12–16]. Also,
terminal sialylation of the terminal galactose residues seems to
control binding, with α2-6 sialic acid blocking galectin binding
[7, 12, 16], while for example galectin-8N has a high affinity for
galactose residues with a terminal α2-3 sialic acid [9, 17].
Collectively, these results have provided essential insights into
the galectin affinity to specific isolated glycan epitopes; however, we
have been missing methods to test galectin specificity in the context
of the cellular membrane. Based on recent advances in precise gene
editing and insight into the biosynthetic glycosylation pathways, it
has now been possible to develop glycoengineered cell libraries
comprised of isogenic cells genetically engineered to display distinct
features of the glycome. By the systematic and combinatorial
knock-out/knock-in of glycosyltransferase genes in mammalian
cell lines, it is possible to produce large libraries of stable isogenic
cells that each display the cellular glycome with loss or gain of
distinct glycosylation features. The same cell lines can naturally be
used to produce recombinant proteins with homogenous and well-
defined glycan structures. Different glycoengineered cell libraries
have been created in CHO cells [18], HEK293 cells [19], and more
recently in cells that can be used to create organotypic tissue models
[20, 21]. The use of such glycoengineered cell libraries, with the
corresponding possibility to produce recombinant proteins with
defined glycan structures, provides an important supplement to
Dissecting Context-Specific Galectin Binding Using Glycoengineered Cell. . . 207
existing binding assays used to probe glycan specificities of glycan

binding proteins [22].
The purpose of this chapter is to provide a detailed protocol for
a cell-based detection of galectin binding to specific glycan struc-
tures in the context of a cell surface and to offer support with
troubleshooting based on our experience and studies. The concept
and methods described here could easily be transferred to other
kinds of posttranslational modifications (Fig. 1).
2 Materials
2.1 Galectin 1. Lyophilized recombinant galectin CRDs (produced as

Preparation described in Salomonsson et al. [5]).
2. 20 mM HEPES buffer.
3. NHS-fluorescein (Thermo Fisher Scientific).
4. Dimethyl sulfoxide, DMSO.
5. PD10 column (GE Healthcare).
6. Phosphate-buffered saline pH 7.4.
7. Quick Start™ Bradford Protein Assay.
2.2 Culturing CHO 1. CHOZN GS/ wild-type and glycogene knockouts (see
Wild-Type and Note 1).
KO Cells 2. EX-CELL CHO CD Fusion serum-free media (Sigma).
3. GlutaMAX™ Supplement (Gibco).
4. TPP TubeSpin® Bioreactors (Sigma).
5. 36.5 C Incubator (5% CO2) with orbital shaker (200 rpm).
2.3 Cell Surface 1. 15-mL Falcon tubes.

Binding: Flow 2. 50-mL Falcon tubes.
Cytometry
3. Trypan blue staining solution, 0.4%.
4. U-bottom 96-well plate.
5. Phosphate-buffered saline (PBS), pH 7.4 (with Ca2+ and
Mg2+, pH 7.4).
6. 100 mM Lactose in PBS.
7. Fluorescently labeled galectin as prepared in Subheading 2.1.
8. Propidium Iodide (PI), 1 mg/mL in water (Invitrogen).
9. LSRII cytometer (BD Bioscience).
2.4 Single-Molecule 1. PD10 columns (GE Healthcare).

Binding: Fluorescence 2. Pierce BCA Protein Assay Kit (Thermo Fisher Scientific).
Anisotropy
3. 384-well plates (Corning).
Fig. 1 Applications of glycoengineered cells in galectin specificity studies. (a) Glycans produced by CHO cells
can be altered through rational manipulation of the repertoire of expressed glycosyltransferases. The initial
genetically engineered cell libraries were produced using zinc-finger nuclease (ZFN)-based knockout system
[18], but more recent and extensive collection of cells were constructed using CRISPR/Cas9 [19, 21]. Specific
alterations of glycogenes within the glycosylation pathway will influence the structure of all glycans related to
the specific pathway, both on the cell surface and on secreted glycoproteins. (b) Galectin binding to cell
surfaces is monitored using flow cytometry with measurement of fluorescence intensity at different
4. Phosphate-buffered saline, pH 7.4.

5. Lyophilized recombinant galectin CRDs.
6. Recombinant purified erythropoietin (EPO) glycovariants in
PBS (produced as described in Yang et al. [18]).
7. Fluorescein-tagged saccharide probes (for details see Nielsen
et al. [22]).
8. PheraStarFS plate reader (BMG Labtech).
3 Methods
3.1 Galectin 1. Lyophilized recombinant galectin CRD is dissolved in 20 mM

Preparation HEPES (see Note 2) and passed through a PD10 column that
has been preconditioned with 20 mM HEPES (see Note 3).
2. The protein concentration of the eluted solution is measured
with a BCA assay, and the final galectin concentration is
adjusted to 2 mg/mL with 20 mM HEPES.
3. A tenfold molar excess of NHS-fluorescein is calculated, and
the appropriate amount is weighed and dissolved in DMSO for
a final concentration of 10 mg/mL.
4. The calculated amount of NHS-fluorescein is then added to the
2 mg/mL galectin solution in 20 mM HEPES.
5. The mixture is incubated for 1 h at room temperature while
kept protected from light.
6. The reaction mixture is then passed through a PD10 column
that has been preconditioned with PBS (see Note 4).
Fig. 1 (continued) concentrations of fluorescently labeled galectin. Addition of lactose works as an effective
inhibitor that can be used to reveal unspecific non-carbohydrate binding. (c) The glyco-engineered cells can
also be used as production platforms for recombinant glycoproteins. The glycans on the glycoprotein will
match the structures presented on the surface of the cell. (d) The interactions between purified recombinant
glycoproteins and galectins can be measured in solution by fluorescence anisotropy. Here, increasing
concentrations of glycoprotein is titrated against a fixed concentration of galectin-bound saccharide probe.
Replacement of the fluorescent saccharide probe with the glycoprotein will result in an increase in anisotropy.
The resulting inhibition curve can be used to calculate binding constants of the galectin–glycoprotein
interaction. (e) Flow cytometry data showing different galectin CRDs (1.5 μM) binding to wild-type and
mgat1 knockout CHO cells. Mgat1 (alpha-1,3-mannosyl-glycoprotein 2-beta-N-acetylglucosaminyltransfer-
ase) initiates complex N-linked carbohydrate formation and is essential for the conversion of high-mannose to
hybrid and complex N-glycans. Mgat1 knockout cells, therefore, lack complex hybrid and N-linked glycans.
Gal-1, -3, -9C, and -9N binding is abolished in mgat1 knockout cells showing that binding of these galectin
CRDs are dependent on complex N-glycans. The ability of Gal-8Nto bind O-glycans are here illustrated by the
fact that Gal-8N still retain some binding to the mgat1 knockout cells. (f) Fluorescence anisotropy data
similarly shows that all galectin CRD binding to mgat1/ EPO disappears, except for Gal-8 N that can still
bind to the O-glycan on EPO. For more details, see Nielsen et al. [22]
7. The protein concentration can then be measured using a Brad-

ford assay, and the degree of labeling can then be measured
using manufacturer’s instructions (see Note 5).
3.2 Culturing CHO 1. CHOZN GS/ wild-type and glycogene knockout clones are
KO Cells grown as suspension culture in EX-CELL CHO CD Fusion
serum-free media, supplemented with 4 mM L-glutamine.
2. Cell culture is grown in 50 mL TPP TubeSpin Bioreactors at
200 rpm in incubator at 36.5 C (5% CO2).
3.2.1 Cell Surface 1. Grow cell suspension cultures to a density of 0.5–3 106cells/
Binding: Flow Cytometry mL and ensure a cell viability of >90% (see Note 6) with Trypan
blue staining (or other viability stains).
2. Harvest appropriate number of cells for desired number of
experiments (0.4 106 cells/experiment) and transfer to
50-mL Falcon tube.
3. Centrifuge at 300 g for 3 min. Discard the supernatant.
4. Wash the pellet with 10 mL PBS.
5. Centrifuge again at 300 g for 3 min and discard the
supernatant.
6. Wash cell pellet in 3 mL PBS with 100 mM Lactose added and
transfer to 15-mL Falcon tube (see Note 7).
7. Centrifuge at 300 g for 3 min.
8. Carefully remove all the lactose wash by pipette, and wash cells
with 5 mL PBS.
9. Centrifuge at 300 g for 3 min and discard the supernatant.
10. Resuspend cells in a volume of PBS that results in cell density of
2 106 cells/mL and transfer 200 μL cell suspension to each
well of a U-bottom 96-well plate. Keep cells on ice throughout
experiment from this point on.

11. Centrifuge 96-well plate at 400 g for 5 min (4 C) and
remove PBS with multi-channel pipette.
12. To each well, add 100 μL PBS 200 mM Lactose and 100 μL
PBS with different concentrations of labeled galectin (1, 2,
4, 8, and 16 μM), resulting in final galectin concentrations of
0, 5, 1, 2, 4, and 8 μM 100 mM lactose.
13. Keep in dark and incubate at 4 C for 1 h with gentle agitation
at 30 rpm on an orbital shaker.
14. Carefully wash each well with 200 μL ice-cold PBS. Repeat
twice.
15. Add PI to a final concentration of 40–50 μg/mL to each well
immediately before data acquisition.
16. Galectin cell surface binding is measured with flow cytometry

on a BD LSR-II cytometer (see Note 8).
3.2.2 Data Analysis: Flow 1. FITC/fluorescein geometrical mean fluorescence intensity

Cytometry (GMFI) (see Note 9) of sample cell populations is calculated
with FlowJo software.
2. Cell aggregates identified in an FSC-H vs. FSC-A scatter plot
and PI-positive cell populations are excluded from analyses.
3. Data points for binding curves can be calculated as: (GMFI-
Lactose)-(GMFI+Lactose). This calculation excludes back-
ground fluorescence and non-carbohydrate-specific galectin
binding (see Note 10).
3.2.3 Single-Molecule 1. Recombinant human EPO glycovariants (see Note 11) were
Binding: Fluorescence produced in CHO cells and purified as described in Yang et al.
Anisotropy [18].
2. Purified EPO is buffer exchanged to PBS using PD10 columns,
and protein concentration is determined using BCA assay.
3. Serial 1:2 dilutions of EPO glycoforms in PBS are mixed with a
fixed concentration of galectin CRD (0.4–1.5 μM depending
on the galectin and its affinity to the saccharide probe) and
fluorescein-tagged saccharide probe (final concentration
0.02 μM) (see Note 12) in a 386-well plate at a final volume
on 20 μL.
4. The fluorescence anisotropy of the tagged probe is measured at
20 C on a PheraStarFS plate reader. Excitation and emission
wavelengths were set to 485 and 520 nm, respectively.
3.2.4 Data Analysis: 1. Data points, representing fluorescence anisotropy inhibition at

Fluorescence Anisotropy different concentrations of EPO, are calculated as percent inhi-
bition and represented as mean values SE. All negative
inhibition values are set to 0.
2. Inhibition curves can then be drawn as best-fit curves, where
Y ¼ Ymin + (Ymax Ymin)/(1 + (XHill slope/IC50Hill slope)).
3. The IC50 values for the EPO–galectin CRD interaction are
calculated as described in Sörme et al. [8].
4 Notes
1. CHO cells were used in this binding study, and glycogene

knockouts were made using zinc finger nuclease knockout
(for details, see Yang et. al [18]). Similar cell libraries have
since been made using human cell lines such as HEK293
(Narimatsu et al. 2019 [19]) and N/TERT-1 (Dabelsteen
et al. 2020 [21]).
2. Besides 20 mM HEPES, other compatible coupling buffers

include 50 mM borate, pH 8.5; 20 mM sodium phosphate,
0.15 M NaCl; 100 mM carbonate/bicarbonate.
3. The first PD10 column clean-up removes leftover salts, LPS,
lactose, and EDTA from lyophilized galectins produced in
E. coli and purified on a Lactosyl-Sepharose column.
4. The PD10 column clean-up removes unreacted dye and buffer
exchanges to PBS.
5. BCA is not well-suited post-labeling, since there is a substantial
overlap between the emission spectrum of fluorescein and the
wavelength used as readout of the assay. Method to determine
degree of labeling can be found in protocol under related
documents on Thermo’s product webpage for
NHS-Fluorescein (Catalog number 46410).
6. A high cell viability is needed to maintain homogenous glycan
structures on the cell surface. Excessive cell death will release
enzymes to culture media that can alter glycan structures. The
same applies if the cells are used for recombinant glycoprotein
production.
7. A lactose wash is necessary to remove endogenously bound
galectins that would otherwise limit the binding of labeled
galectin.
8. Ideally, cells should be kept cold throughout the data acquisi-
tion by flow cytometry. If this is not a possibility, the experi-
ment should be divided into several 96-well plates.
9. If using PI, it is necessary to perform compensation on the
instrument to correct for the spectral overlap between fluores-
cein and PI.
10. The presence of lactose has not been shown to affect back-
ground fluorescence.
11. EPO is used as a stable knock-in reporter glycoprotein, which is
easily produced and purified from CHO cell culture medium.
EPO contains three N-glycans and one O-glycan, which con-
tribute to a large proportion of the molecular weight of the
protein (up to 40%).
12. Optimal probes were used for each galectin CRD that allows
lowest possible concentration of galectin while still obtaining a
sufficient change in anisotropy.
References
1. Thiemann S, Baum LG (2016) Galectins and 2. Sciacchitano S, Lavra L, Morgante A,

immune responses-just how do they do those Ulivieri A, Magi F, De Francesco GP,
things they do? Annu Rev Immunol 34: Bellotti C, Salehi LB, Ricci A (2018)
243–264. https://doi.org/10.1146/annurev- Galectin-3: one molecule for an alphabet of
immunol-041015-055402 diseases, from A to Z. Int J Mol Sci
19(2):379. https://doi.org/10.3390/ 12. Stowell SR, Arthur CM, Mehta P, Slanina KA,
ijms19020379 Blixt O, Leffler H, Smith DF, Cummings RD
3. Leffler H, Carlsson S, Hedlund M, Qian Y, (2008) Galectin-1, -2, and -3 exhibit differen-
Poirier F (2002) Introduction to galectins. tial recognition of sialylated glycans and blood
Glycoconj J 19(7–9):433–440. https://doi. group antigens. J Biol Chem
org/10.1023/B:GLYC.0000014072. 283(15):10109–10123. https://doi.org/10.
34840.04 1074/jbc.M709545200
4. Cummings RD, Liu FT, Vasta GR (2015) 13. Sorme P, Arnoux P, Kahl-Knutsson B,
Galectins. In: Varki A, Cummings RD et al Leffler H, Rini JM, Nilsson UJ (2005) Struc-
(eds) Essentials of glycobiology. Cold Spring tural and thermodynamic studies on cation-Pi
Harbor Laboratory Press, Cold Spring Harbor, interactions in lectin-ligand complexes: high-
NY, pp 469–480. https://doi.org/10.1101/ affinity galectin-3 inhibitors through fine-
glycobiology.3e.036 tuning of an arginine-arene interaction. J Am
5. Salomonsson E, Carlsson MC, Osla V, Chem Soc 127(6):1737–1743. https://doi.
Hendus-Altenburger R, Kahl-Knutson B, org/10.1021/ja043475p
Oberg CT, Sundin A, Nilsson R, Nordberg- 14. Patnaik SK, Potvin B, Carlsson S, Sturm D,
Karlsson E, Nilsson UJ, Karlsson A, Rini JM, Leffler H, Stanley P (2006) Complex
Leffler H (2010) Mutational tuning of N-glycans are the major ligands for galectin-1,
galectin-3 specificity and biological function. J -3, and -8 on Chinese hamster ovary cells. Gly-
Biol Chem 285(45):35079–35091. https:// cobiology 16(4):305–317. https://doi.org/
doi.org/10.1074/jbc.M109.098160 10.1093/glycob/cwj063
6. Yang RY, Rabinovich GA, Liu FT (2008) 15. Bum-Erdene K, Leffler H, Nilsson UJ, Blan-
Galectins: structure, function and therapeutic chard H (2015) Structural characterization of
potential. Expert Rev Mol Med 10:e17. human galectin-4 C-terminal domain: eluci-
h t t p s : // d o i . o r g / 1 0 . 1 0 1 7 / dating the molecular basis for recognition of
S1462399408000719 glycosphingolipids, sulfated saccharides and
7. Hirabayashi J, Hashidate T, Arata Y, Nishi N, blood group antigens. FEBS J
Nakamura T, Hirashima M, Urashima T, 282(17):3348–3367. https://doi.org/10.
Oka T, Futai M, Muller WE, Yagi F, Kasai K 1111/febs.13348
(2002) Oligosaccharide specificity of galectins: 16. Leppanen A, Stowell S, Blixt O, Cummings
a search by frontal affinity chromatography. RD (2005) Dimeric galectin-1 binds with
Biochim Biophys Acta 1572(2–3):232–254. high affinity to alpha2,3-sialylated and
https://doi.org/10.1016/s0304-4165(02) non-sialylated terminal N-acetyllactosamine
00311-2 units on surface-bound extended glycans. J
8. Sorme P, Kahl-Knutson B, Wellmar U, Nilsson Biol Chem 280(7):5549–5562. https://doi.
UJ, Leffler H (2003) Fluorescence polarization org/10.1074/jbc.M412019200
to study galectin-ligand interactions. Methods 17. Ideo H, Matsuzaka T, Nonaka T, Seko A,
Enzymol 362:504–512. https://doi.org/10. Yamashita K (2011) Galectin-8-N-domain rec-
1016/S0076-6879(03)01033-4 ognition mechanism for sialylated and sulfated
9. Carlsson S, Oberg CT, Carlsson MC, glycans. J Biol Chem 286(13):11346–11355.
Sundin A, Nilsson UJ, Smith D, Cummings https://doi.org/10.1074/jbc.M110.195925
RD, Almkvist J, Karlsson A, Leffler H (2007) 18. Yang Z, Wang S, Halim A, Schulz MA,
Affinity of galectin-8 and its carbohydrate rec- Frodin M, Rahman SH, Vester-Christensen
ognition domains for ligands in solution and at MB, Behrens C, Kristensen C, Vakhrushev SY,
the cell surface. Glycobiology 17(6):663–676. Bennett EP, Wandall HH, Clausen H (2015)
https://doi.org/10.1093/glycob/cwm026 Engineered CHO cells for production of
10. Song X, Xia B, Stowell SR, Lasanajak Y, Smith diverse, homogeneous glycoproteins. Nat Bio-
DF, Cummings RD (2009) Novel fluorescent technol 33(8):842–844. https://doi.org/10.
glycan microarray strategy reveals ligands for 1038/nbt.3280
galectins. Chem Biol 16(1):36–47. https:// 19. Narimatsu Y, Joshi HJ, Nason R, Van Coillie J,
doi.org/10.1016/j.chembiol.2008.11.004 Karlsson R, Sun L, Ye Z, Chen YH, Schjoldager
11. Arthur CM, Baruffi MD, Cummings RD, KT, Steentoft C, Furukawa S, Bensing BA,
Stowell SR (2015) Evolving mechanistic Sullam PM, Thompson AJ, Paulson JC,
insights into galectin functions. Methods Mol Bull C, Adema GJ, Mandel U, Hansen L, Ben-
Biol 1207:1–35. https://doi.org/10.1007/ nett EP, Varki A, Vakhrushev SY, Yang Z, Clau-
978-1-4939-1396-1_1 sen H (2019) An atlas of human glycosylation
pathways enables display of the human glycome
by gene engineered cells. Mol Cell
75(2):394–407.e5. https://doi.org/10. Clausen H, Vakhrushev SY, Bagdonaite I, Wan-

1016/j.molcel.2019.05.017 dall HH (2020) Essential functions of glycans
20. Bagdonaite I, Pallesen EM, Ye Z, Vakhrushev in human epithelia dissected by a CRISPR-
SY, Marinova IN, Nielsen MI, Kramer SH, Cas9-engineered human organotypic skin
Pedersen SF, Joshi HJ, Bennett EP, model. Dev Cell 54(5):669–684.e7. https://
Dabelsteen S, Wandall HH (2020) O-glycan doi.org/10.1016/j.devcel.2020.06.039
initiation directs distinct biological pathways 22. Nielsen MI, Stegmayr J, Grant OC, Yang Z,
and controls epithelial differentiation. EMBO Nilsson UJ, Boos I, Carlsson MC, Woods RJ,
Rep 21(6):e48885. https://doi.org/10. Unverzagt C, Leffler H, Wandall HH (2018)
15252/embr.201948885 Galectin binding to cells and glycoproteins
21. Dabelsteen S, Pallesen EMH, Marinova IN, with genetically modified glycosylation reveals
Nielsen MI, Adamopoulou M, Romer TB, galectin-glycan specificities in a natural context.
Levann A, Andersen MM, Ye Z, Thein D, Ben- J Biol Chem 293(52):20249–20262. https://
nett EP, Bull C, Moons SJ, Boltje T, doi.org/10.1074/jbc.RA118.004636
Chapter 13
Method for Identifying Galectin Ligands on Lymphocyte

Membrane Glycoproteins
Kashyap R. Patel, Adam W. Barb, and Sean R. Stowell
Abstract
Glycosylation is one of the most common protein posttranslational modifications. Most human lymphocyte
membrane receptors are modified by diverse glycan structures, and functional studies have indicated that a
family of glycan-binding proteins, galectins, can significantly modulate lymphocyte development and
function by interacting with these glycans. Several galectins have a varying degree of affinity for the
N-acetyllactosamine (LacNAc) disaccharide, and some critical lymphocyte receptors can be modified by
glycan structures carrying this motif. However, the site-specific glycan composition on primary lymphocyte
membrane receptors in healthy individuals is largely limited. The main reason for the limitation is low
abundance of available material from a single donor and compositional heterogeneity in glycan structures
that can potentially modify a protein. Donor-dependent variability in N-glycan structures on CD16a
isolated from primary NK cells of healthy human donors was recently reported. NK cell CD16a is
glycosylated at five N-glycosylation sites, and two of the five sites are modified, almost exclusively, by
N-glycans with multiple LacNAc repeats which can serve as ligands for endogenous galectins. Thus, the
protocol described in this section can be utilized to identify galectin ligands at specific glycosylation sites of
endogenous membrane receptor from circulating primary human lymphocytes.
Key words Glycoproteomics, Mass spectrometry, N-glycosylation, Natural killer cell, CD16a,
N-acetyllactosamine (LacNAc)
1 Introduction
Galectins regulate several critical functions in lymphocyte biology

including modulating cell surface receptor clustering [1–3]. All
galectins (11 distinct human galectins) have conserved
carbohydrate-recognition domains (CRDs) which can bind
β-galactose-containing asparagine(N)-linked or O-linked (oxygen
atom of serine or threonine) glycan structures present on extracel-
lular glycoproteins [4]. Several human galectins (Gal-1, Gal-3,
Gal-8, and Gal-9) implicated in regulation of lymphocyte function
have differential affinities to glycan structures containing multiple
N-acetyllactosamine (LacNAc) units with distinct terminal and
branch modifications [1, 5–11].
215
216 Kashyap R. Patel et al.
In addition to recognizing glycan motifs on the surface of

microbes [12–14], galectins are naturally structured into multiva-
lent configurations which allows their CRD to theoretically bind
multiple glycan moieties present on distinct membrane receptors
and cross-link them to elicit a functional response (Fig. 1) [15]. For
example, T-cell receptor clustering was enhanced in mice T-cells
lacking β1,6 N-acetylglucosaminyltransferase V (Mgat5), a key
enzyme in biosynthesis of LacNAc repeats [16]. Higher mobility
of the T-cell receptor on the surface of the cell led to lower anti-
genic threshold for cell activation and increased incidence of auto-
immune disease in the mouse model. Gal-3 was identified as the
protein interacting with the T-cell receptor on cell surface to form a
multivalent lattice through its interaction with LacNAc structures
likely expressed on T-cell receptor. Similarly, Gal-3 mediated lattice
formation on the surface of CD8+ mice T-cells restricted interac-
tion between T-cell receptor carrying Mgat5 derived glycan struc-
tures and CD8 co-receptor, subsequently increasing antigenic
threshold for cell activation in a mouse model [17]. In B-cells
isolated from human tonsillar tissue, Gal-9 preferentially bound
linear poly-LacNAc repeats on cell surface CD45 to regulate naı̈ve
B-cell activation but failed to bind I-branched LacNAc structures
expressed on germinal B-cell [3]. Moreover, multiple lymphocyte
surface receptors are identified as binding partners for secreted
galectins; therefore, galectins and their ligands represent an impor-
tant class of immunoregulatory molecules [2].
Most membrane-bound extracellular receptors are glycosylated
at one or more amino acid residues. Biosynthesis of glycan struc-
tures depends on availability of activated donor substrates and
interaction of the oligosaccharide precursor with up to roughly
200 glycosidases and glycosyltransferases present in the secretory
pathway [18]. The stochastic nature of the modification can gener-
ate structures with enormous diversity; in mammals, each potential
glycosylation site can be modified by 7000 identified structures
[18]. Moreover, the location of potential N-glycosylation site in the
folded polypeptide can also affect glycan processing; interactions
between glycan residues and the polypeptide backbone can prevent
modification reactions even if full set of glycan modifying enzymes
are expressed in the cell [19, 20]. This interaction results in mini-
mally processed forms at certain sites on cell surface glycoproteins
while unrestricted access to the N-glycosylation site and glycan
residues by the glycan-modifying enzymes can result in highly
processed glycan structures with multiple LacNAc repeats and
extensive terminal modifications. Thus, diverse glycoforms can be
expected at different glycosylation sites on the same protein mole-
cule. Additionally, abundance and localization of glycan modifying
enzymes vary depending on the cell type; therefore, the structures
present on the same protein expressed in different cell can be largely
different [21, 22]. As a result, a comprehensive site-specific and
Method for Identifying Galectin Ligands on Lymphocyte Membrane Glycoproteins 217
A Complex-type
Oligomannose Hybrid-type LacNAc
oligosaccharyl Fucose Mannose

transferase N-Acetylglucosamine Galactose
dol-dP
protein N-Acetylneuraminic acid
B N-glycan with
LacNAc
Transmembrane
receptor
Plama
membrane
Galectin
Fig. 1 (a) Illustration of N-glycan remodeling in a mammalian cell. Precursor N-glycans can be modified into
three main types: oligomannose-, hybrid-, and complex-type N-glycans. Complex-type N-glycans can contain
multiple LacNAc units which serve as ligands for extracellular galectins. (b) Extracellular galectins can interact
with transmembrane receptors by binding to N-glycan containing LacNAc motif. Multivalency of galectin
promotes binding to multiple receptors at the plasma membrane and enhance receptor clustering
cell-type-specific glycan characterization of a single type of endog-

enous membrane protein is necessary to understand any existing
structure–function relationship including potential interactions
with carbohydrate binding proteins such as galectins [23].
Variability in protein N-glycosylation is very well studied for
soluble serum proteins, primarily IgG, due to the ease of availability
and their abundance in human body; however, identification of the
glycosylated sites and corresponding glycan structures on an
endogenous membrane protein remains challenging [24–
26]. Nearly nothing is known about donor-dependent variability
in site-specific glycosylation of endogenous receptor isolated from
circulating primary lymphocytes mainly due to their relatively low
abundance. By one estimate about 1 liter of whole blood from a
single donor is necessary to isolate enough material to characterize
glycan structures at five N-glycosylation sites on CD16a from
natural killer (NK) cells [27]. Leukocyte reduction system (LRS)
filters are an attractive alternate source of primary lymphocytes
from a single donor since the lymphocytes are specifically enriched
in these filters during apheresis, and lymphocytes isolated from LRS
filters can be used for site-specific glycosylation analysis of endoge-
nous receptor [28, 29]. Additionally, a high-resolution analysis of
low abundance glycoproteins is also possible due to recent devel-
opment in glycopeptide enrichment and separation methods in
addition to advancement in mass spectrometry instruments which
are capable of collision induced dissociation of glycopeptides to
reveal compositional details [30–32]. This chapter describes a
method to characterize N-glycopeptides from a membrane bound
receptor, Fc γ receptor IIIa (CD16a), immunoprecipitated from
primary NK cells isolated from a single healthy human donor.
CD16a is glycosylated at five N-glycosylation sites and glycan
structures with galectin ligands were identified on four of the five
sites [29]. Characterized glycan structures at each site can be rela-
tively quantitated for a donor-dependent comparative analysis.
2 Materials
2.1 NK Cell Isolation 1. Leukoreduction system (LRS) chamber, stored at 25 C.

and Validation (See 2. 70% ethanol.
Note 1)
3. Sterile phosphate-buffered saline (PBS), pH 7.4 (Thermo
Fisher Scientific).
4. Cell isolation buffer: sterile PBS with 2% heat-inactivated,
IgG-depleted fetal bovine serum (FBS), pH 7.4.
5. Sterile blunt end 22G needle (VWR).
6. Sterile 10-mL syringe (VWR).
7. 0.4% Trypan blue solution (Thermo Fisher Scientific).
8. Instrument or apparatus for cell counting and determining

viability.
9. RosetteSep human NK Cell enrichment reagent (Stemcell
Technologies).
10. Lymphoprep density gradient medium (Stemcell
Technologies).
11. RBC lysis buffer: 155 mM ammonium chloride, 12 mM
sodium bicarbonate, 0.1 mM EDTA, pH 7.4.
12. Sterile 50-mL and 1.5-mL tubes (VWR).
13. Sterile serological pipettes.
14. Centrifuge with swinging bucket rotor.
15. Flow cytometer (e.g., BD FACSCanto, BD Biosciences).
2.2 NK Cell Lysis 1. Tris buffer: 100 mM Tris, 100 mM sodium chloride, pH 8.0.
2. 0.5 M ethylenediaminetetraacetic acid (EDTA).
3. 10% dodecyl maltoside (DDM) solution, stored at 20 C.
4. Oxidized glutathione.
5. 10 mM potassium ferricyanide.
6. 1 M 4-(2-aminoethyl) benzenesulfonyl fluoride (AEBSF),
stored at 20 C.
7. 1.5-mL tubes and suitable refrigerated centrifuge.
2.3 CD16a 1. Anti-CD16 antibody (3G8)-coupled agarose resin (see

Immunoprecipitation Note 2).
2. Protein G Sepharose resin (GE healthcare).
3. Wash buffer 1: 100 mM Tris, 100 mM sodium chloride, 5 mM
EDTA, 10 mg/mL DDM, pH 8.0 (prepare fresh before each
immunoprecipitation).
4. Wash buffer 2: 50 mM Tris, 100 mM sodium chloride, pH 8.0.
5. Wash buffer 3: 50 mM ammonium carbonate, pH 8.0.
6. Elution buffer: 54.9% acetonitrile (ACN), 0.1% trifluoroacetic
acid (TFA) in water (prepare fresh before each
immunoprecipitation).
7. Neutralization buffer: 1 M ammonium carbonate, pH 8.0.
8. Mini Bio-Spin Chromatography Columns (Bio-Rad).
9. 1.5-mL tubes and suitable refrigerated centrifuge.
10. Bath sonicator (VWR).
11. Tube rotator (VWR).
12. Freeze dryers for lyophilization, FreeZone Freeze Dryers
(Labconco).
13. Protein gel electrophoresis and Western blot apparatus

(Bio-Rad).
2.4 CD16a 1. Digestion buffer: 50 mM ammonium carbonate, pH 8.0.

Proteolysis and 2. HPLC grade methanol.
Glycopeptide
3. Sequencing grade Chymotrypsin, 0.5 mg/mL in digestion
Enrichment
buffer, stored at 20 C (Millipore Sigma).
4. Sequencing grade GluC, 0.5 mg/mL in digestion buffer,
stored at 20 C (Promega).
5. 1 M dithiothreitol (DTT).
6. 0.5 M iodoacetamide (IAM) solution (prepare fresh before
each digestion).
7. HPLC grade water.
8. ProteoExtract Glycopeptide enrichment kit (Millipore Sigma)
(see Note 3).
9. 1.5-mL microcentrifuge tubes and suitable centrifuge.
10. Heat block.
11. Freeze dryers for lyophilization, FreeZone Freeze Dryers
(Labconco).
2.5 Liquid 1. Sample resuspension buffer: 5% ACN, 0.1% TFA in water.

Chromatography and 2. Buffer A: 0.1% formic acid in water.
Mass Spectrometry
3. Buffer B: 80% ACN and 0.1% formic acid in water.
(See Note 4)
4. Nano-C18 column, 75 μm 20 cm fused silica capillary col-
umn/emitter packed with 3 μm C18 medium (Next Advance)
(see Note 5).
5. Nano-flow UHPLC system, EASY nLC-1200 LC system with
a Nanospray FlexIon source (Thermo Fisher Scientific).
6. Mass spectrometer capable of collision induced dissociation, Q
Exactive Hybrid Quadrupole-Orbitrap Mass Spectrometer
with an HCD fragmentation cell (Thermo Fisher Scientific).
2.6 Data Analysis 1. MS data viewer software capable of exporting m/z and intensity
values, Xcalibur Qual Browser (Thermo Fisher Scientific).
2. R script to report intensities from extracted m/z values and
peak intensities (see Note 6).
3. R script to generate image files from reported m/z values and
peak intensities (see Note 6).
4. Software capable of running R scripts, R (www.r-project.org).
3 Method
3.1 NK Cell Isolation 1. Cells must be isolated under aseptic conditions. Prewarm cell
isolation buffer and RBC lysis buffer to 37 C and perform
following centrifugation steps at room temperature.
2. Spray the LRS filter with 70% ethanol before placing it in the
biological safety cabinet. Typical LRS filters contain a high
number of erythrocytes so the undrained filter appears red
(see Note 7).
3. Mix contents of the chamber by inverting it gently for 30 s. Cut
the filter tube at one end and place it a 50 mL tube. Cut the
filter tube at opposite end to drain content in the 50 mL tube
(see Note 8).
4. Collect residual material from the chamber by rinsing with
20 mL cell isolation buffer. Inject the buffer into the chamber
using a 22G blunt end needle and 10-mL syringe.
5. Bring the final volume of collected cell suspension up to
approximately 30 mL with cell isolation buffer and transfer
30 mL in two new 50-mL tubes, 15 mL in each tube.
6. Add 750 μL RosetteSep human NK Cell enrichment reagent in
each tube and mix by pipetting gently. Incubate for 20 min at
room temperature.
7. Add 15 mL cell isolation buffer to each tube and carefully layer
the cell suspension over 15 mL lymphoprep density gradient
media in two new 50-mL tubes. Centrifuge both tubes at
1200 g for 20 min with the brake off.
8. Aspirate and discard most of the top layer without disturbing
the layer of enriched cells at plasma-density gradient medium
interface. Use a serological pipette to transfer enriched cells
into a new 50-mL tube (see Note 9).
9. Add 20 mL cell isolation buffer to the collected cells and
centrifuge at 600 g for 8 min. Discard the supernatant.
10. Resuspend the cell pellet in 45 mL cell isolation buffer and
centrifuge at 600 g for 8 min. Discard the supernatant.
11. Resuspend the pellet in 7 mL RBC lysis buffer and incubate at
room temperature for 10 min.
12. Add 20 mL cell isolation buffer and centrifuge at 350 g for
10 min. Discard the supernatant.
13. Resuspend the cell pellet in 10 mL PBS. Determine the cell
count and viability using a cell counter (or an alternate appara-
tus) by mixing a small fraction of cells with equal volume of
0.4% trypan blue solution.
14. Collect cells to determine relative percentage of NK cells in

total isolated cells (see Note 10).
15. Centrifuge the cell suspension at 600 g for 8 min and discard
the supernatant. Store the cell pellet at 80 C.
3.2 NK Cell Lysis 1. Prepare fresh lysis buffer by supplementing Tris buffer with 1%
DDM, 5 mM EDTA, 5 mM oxidized glutathione, 1 mM
AEBSF, and 10 μM potassium ferricyanide (see Note 11).
2. Add 250 μL cold lysis buffer per 1 107 frozen cells immedi-
ately after removing cells stored at 80 C. Gently dissociate
the pellet by pipetting and incubate the lysate on ice for 20 min.
Nuclei of the cell remain intact during lysis using DDM and can
be removed by centrifugation described in next step. Vigorous
pipetting should be avoided since release of the nuclear content
will increase the viscosity of the lysate and interfere with protein
immunoprecipitation (see Note 12).
3. Centrifuge the lysate at 10,000 g for 10 min at 4 C. Transfer
the supernatant in a new 1.5-mL tube and freeze at 80 C.
3.3 CD16a 1. Thaw the lysate on ice and centrifuging at 10,000 g for
Immunoprecipitation 10 min at 4 C. Transfer the supernatant into 1.5-mL tube
containing prewashed Protein G Sepharose resin (use 5 μL
Protein G Sepharose resin per 200 μL clarified lysate and pre-
wash the resin with 2 mL freshly prepared lysis buffer) (see
Notes 13 and 14).
2. Incubate for 60 min at 4 C on a tube rotor. Pellet the resin by
centrifuging at 500 g for 5 min at 4 C and transfer the
supernatant to new 1.5-mL tube.
3. Sonicate the lysate in a bath sonicator for a total of 1 min with
intermittently incubating on ice every 15 s. Centrifuge the
lysate at 10,000 g for 10 min at 4 C. Collect 10 μL super-
natant for western blot analysis (see Notes 15 and 16).
4. Transfer the supernatant to a new 1.5-mL tube with prewashed
anti-CD16 antibody (3G8)-coupled agarose resin (use 10 μL
3G8 agarose resin per 200 μL clarified lysate and prewash the
resin with 2 mL freshly prepared lysis buffer) (see Note 13).
5. Incubate overnight at 4 C on a tube rotor. Centrifuge the
lysate at 500 g for 5 min at 4 C. Collect 10 μL supernatant
for western blot analysis and discard the supernatant (see
Note 15).
6. Gently resuspend the resin pellet in 600 μL cold wash buffer
1. Centrifuge at 500 g for 5 min at 4 C and discard the
supernatant.
7. Repeat the wash step one more time with wash buffer 1 fol-
lowed by successive two washes with cold wash buffer 2 and
wash buffer 3. Washing the resin with ammonium carbonate
buffer (wash buffer 3) is necessary to decrease the salt content

in elution fractions since ammonium carbonate can sublime
during lyophilization.
8. Elute CD16a by adding 200 μL elution buffer and transfer the
suspension into a mini Bio-Spin chromatography column.
Place the column in a 1.5-mL tube containing 26 μL neutrali-
zation buffer and spin at 100 g for 15 s.
9. Repeat the elution step four more time, each time adding
200 μL elution buffer to the resin bed before transferring the
mini Bio-spin column into fresh 1.5-mL tube containing neu-
tralization buffer.
10. Discard the resin and pool the eluted fractions. Collect 5% of
total volume in a separate tube for CD16a Western blot analysis
and lyophilize overnight. Remaining eluate should be lyophi-
lized and stored at 20 C till quantity of CD16a in eluted
samples is determined by Western blot analysis (see Note 15).
3.4 CD16a 1. Resuspend the lyophilized eluate in 120 μL digestion buffer

Proteolysis and and supplement the solution with methanol (10% v/v final
Glycopeptide concentration). Methanol can assist in protein unfolding and
Enrichment promotes efficient proteolysis [33].
2. Boil the sample at 95 C for 5 min. Immediately place the tube
on ice.
3. Add sequencing grade Chymotrypsin and GluC at an enzyme:
protein molar ratio of 1:20 and 1:30, respectively. Incubate at
25 C for 18 h.
4. Add 75 ng of both enzymes and further incubate at 37 C for
4 h. Additional enzyme is added to promote complete diges-
tion of CD16a.
5. Reduce the digested protein by adding DTT (5 mM final
concentration) and incubate at 37 C for 1 h.
6. Alkylate by adding IAM (15 mM final concentration) and
incubate at 25 C for 1 h in dark.
7. Centrifuge the sample at 10,000 g for 3 min. Transfer the
supernatant in a new 1.5-mL tube and lyophilize overnight.
8. Resuspend the sample in 10 μL HPLC grade water and follow
the recommended protocol in the ProteoExtract Glycopeptide
enrichment kit. Elute three additional times with HPLC grade
water to maximize glycopeptide recovery (see Note 17).
9. Combine the elution fractions and lyophilize overnight.
3.5 Liquid 1. Resuspend the lyophilized sample in 5 μL sample resuspension

Chromatography and buffer. Vortex thoroughly and briefly centrifuge before inject-
Mass Spectrometry ing into a pre-equilibrated nano-C18 column attached to an
EASY nLC-1200 LC system with a Nanospray FlexIon source.

Inject the sample at flow rate of 3 μL/min.
2. Elute with a 0–35% gradient of Buffer B over 60 min at 300 nl/
min. Set electrospray voltage to 2650 V.
3. After each elution step, wash the column with a 35–70% gradi-
ent of Buffer B over 10 min followed by 100% Buffer B over
next 10 min and equilibrate by passing 20 μL Buffer A at
400 bar. Perform a blank run after every sample run by inject-
ing 5 μL Buffer A instead of the sample. The blank run mini-
mizes carry over from previous sample run.
4. Collect MS spectra in positive ion mode on a Q Exactive
Hybrid Quadrupole Orbitrap Mass Spectrometer with an
HCD fragmentation cell. Acquire MS1 spectra within scan
range of 600–2500 m/z. Set parameters to trigger MS2 event
for 20 most abundant ions eluted over retention time range of
80 ms and fragment glycopeptides with stepped normalized
collision energy of 17, 27, and 37 eV (see Note 18).
3.6 Data Analysis 1. Use Xcalibur Qual Browser to determine retention time range
for different glycopeptides generated by expected protease
cleavage pattern of Chymotrypsin and GluC (expect cleavage
at C-terminal to phenylalanine, tryptophan, tyrosine, and leu-
cine for Chymotrypsin and C-terminal to glutamic acid for
Glu-C). Glycopeptides are identified by the presence of b-
and y-type glycopeptide fragment ions, and characteristic oxo-
nium ion species in MS2 spectra (Fig. 2) (see Note 19).
2. Manually identify all glycoforms associated with a peptide
within predetermined retention time range. Different glyco-
forms are identified by comparing precursor ion’s observed
mass to calculated mass and fragmented glycopeptide ions in
MS2 spectra (Fig. 2) (see Note 20).
3. Create a glycopeptide mass list comprising of all the identified
glycan structures associated with each peptide along with
potential glycoform variations expected based on identified
structures. Convert the mass list into .cvs file containing unique
peptide identifier and associated glycoform in the first column,
neutral mass for that glycopeptide in the second column, and
charge in the third column. Do not assign labels to the columns
(see Note 21).
4. Create a separate .cvs file with exported m/z and intensity
values of all ions within the predetermined retention time for
individual peptides. Label the first and second columns with m/
z and int, respectively.
5. Specify folder location, file with exported m/z values, and file
with peptide-specific mass list in R script to calculate summed
intensity. Run the script using R to get output .cvs file
MS1 spectrum for N74 glycopeptide Observed mass
in the MS1 spectra
+4
+4
+2 +4 +2
+3
+4
+1 IAM
Pep+4 +4
1564.37 RCQTNLSTLSDPVQLE
100 Pep
1655.65
+4 +4
obs. (M+4H)+4 - 1564.368
exp. (M+4H)+4 - 1564.370
Relative Intensity (%)
Pep+4 +( )
1473.09
+( ) Pep+4 +4
+( ) 1746.94
+5 +4
+( )
Pep+4
1381.81
+( ) +6
Pep+4
1838.24
Pep+4
1929.75
0
1350 1400 1450 1500 1550 1600 1650 1700 1750 1800 1850 1900
m/z
MS2 spectrum for single N74 glycopeptide
Characteristic Fragmented glycopeptide ions

oligosaccharide Pep+1
oxonium ions +1 2063.97
+2
+2 +1
+3
+2 Pep +2
100 657.23 1917.31 Pep+2
454.15 Pep+2 Pep+2
1961.30
Relative Intensity (%)
Pep+2 1880.34 2143.40 Pep+2 +1

+3
1815.25 2210.07 +2 +4
+1 +3
528.19 +1 Pep+2
1998.32 Pep+2
Pep+2 2326.98
Pep+2 2289.94
1734.27
569.21
731.27
1022.36 1700 1800 1900 2000 2100 2200 2300 2400 2500
0
400 600 800 1000 1200 1400 1600 1800 2000 2200 2400
m/z
Fig. 2 Identification of glycopeptide in mass spectrometry data. MS1 spectrum (top) showing elution pattern of
glycosylated peptides spanning N74 residue of CD16a isolated from primary human NK cells. The glycopeptide
is modified by iodoacetamide at the cysteine residue (represented as red IAM) and elute between 49.2 and
61.1 min. Glycopeptides are identified by comparing observed mass to expected mass of the parent ion in the
MS1 spectra; difference in observed and expected mass of a quadruple charged N74 glycopeptide modified
with tetraantennary complex-type N-glycan with two repeating LacNAc units is shown. Low abundant species
within comparable retention time range are identified by predictable mass differences corresponding to
monosaccharides or disaccharides, for example, addition of single fucose (146.05 Da) or single LacNAc unit
(365.12 Da) is identified in this MS1 spectrum. MS2 spectrum (bottom) of glycopeptide highlighted in the top
section contains characteristic oligosaccharide oxonium ion species (ion corresponding to mass of
1022.36 Da is indicative of LacNAc repeats in N-glycan structure) and ions generated by fragmented glycan
with intact peptide (ion corresponding to peptide with one GlcNAc is frequently the most intense). The glycan
structures indicate one possible configuration based on composition. MS1 and MS2 spectra are extracted
from two different datasets. (Image of spectra was adapted from a version first published in Molecular &
Cellular Proteomics. Patel KR, Nott JD, Barb AW. Primary Human Natural Killer Cells Retain Proinflammatory
IgG1 at the Cell Surface and Express CD16a Glycoforms with Donor-dependent Variability. 2019 Aug;18:
2178–2190)
containing the unique identifier, neutral mass, charge, calcu-

lated, and observed m/z values for first seven isotopologue
peaks and their respective intensities.
6. Specify the output file from the first script and the file with
exported m/z values in R script to create image files with
calculated m/z values overlayed on MS1 spectra. Run the R
script in R to get an image file corresponding to each unique
peptide label reported in output file from first script.
7. The spectral image files from second script facilitate manual
verification of theoretical and observed m/z values, peak inten-
sity and isotopologue distribution of all identified glycopep-
tides. Further verify identity of glycopeptide by confirming the
presence of relevant glycopeptide fragments and oligosaccha-
ride specific oxonium ions in MS2 spectra (if available). MS2
spectra for glycopeptides containing LacNAc can have distinc-
tive oxonium ion species (Fig. 2 and Table 1).
8. Sum the intensities of the reported isotopologue peaks for
validated glycopeptides and sum the intensities of the same
glycopeptide with different charge states to calculate relative
abundance of all identified glycoforms on a single peptide.
Table 1
Calculated m/z values of observed and theoretical glycan oxonium ion species
Monosaccharide composition Calculated m/z of protonated species

Hex 163.060
HexNAc 204.087
NeuAc-H2O 274.092
NeuAc 292.103
Hex+HexNAc 366.139
Hex+NeuAc 454.156
2Hex+HexNAc 528.192
Hex+2HexNAc 569.219
Hex+HexNAc+NeuAc 657.235
2Hex+2HexNAc 731.272
a
2Hex+2HexNAc+NeuAc 1022.367
a
a
a
Oxonium ions indicate the presence of LacNAc modification in parent glycopeptide
4 Notes
1. Lab personnel should have relevant safety training before work-

ing with human blood. Minimum requirements are laboratory
safety training and training for working safely with HIV and
other bloodborne pathogens.
2. An aglycosylated variant of 3G8 should be used to prevent
Fc-mediated binding of 3G8 to CD16a. Details about anti-
body expression, purification, and coupling to agarose resin are
described elsewhere [29]. Alternatively, use of commercially
available biotinylated version of 3G8 can be optimized for
CD16 immunoprecipitation.
3. In case ProteoExtract kit is unavailable, an alternate method
can be used [34].
4. Use HPLC grade solvents for this step.
5. Capillary C18 column can be bought commercially or packed
manually. Details regarding manually packing the column is
described elsewhere [29].
6. Both the R scripts can be found in the supplementary section of
previously published report [29].
7. Erythrocytes are required for successful negative selection.
Antibody mixture in negative selection kit cross-links
unwanted cell types and erythrocytes to form cell rosettes
which precipitates in the density gradient centrifugation step.
8. A small fraction of cells can be collected at this point if evalua-
tion of serum proteins or donor genotyping is desired.
9. The content of LRS filter is fractionated into four distinct
layers; top layer will appear yellow in color and consist mostly
of plasma and cell isolation buffer, enriched cells at the plasma-
density gradient medium interface will appear as thin white
layer or white clumps attached to the wall of the tube, density
gradient medium will form a colorless layer below the enriched
cells and bottom red layer will be erythrocytes cross-linked
with unwanted cells. Slowly aspirate the enriched cell to maxi-
mize the yield without disturbing the lowermost erythrocyte
layer.
10. Stain NK cells with antiCD56 and antiCD16 antibodies to
determine relative proportion of NK cells in the isolated cells.
Staining protocol is described elsewhere [27].
11. Maintaining oxidizing condition during NK cell lysis is essen-
tial for successful CD16a immunoprecipitation. The CD16a-
binding epitope of 3G8 is lost during protein denaturation,
thus it is possible that reductants released during cell lysis can
reduce the disulfide bonds in CD16a and diminish its affinity to
3G8. Oxidized glutathione and potassium ferricyanide are

added to the lysis buffer to maintain oxidizing condition dur-
ing cell lysis.
12. A minimum number of NK cells necessary to characterize the
glycan structure at all five potential N-glycosylation sites of the
membrane bound CD16a should be lysed. Using the method
described in this chapter, approximately 64 106 viable NK
cells from single donor is sufficient.
13. Wash Protein G Sepharose resin and 3G8 coupled agarose resin
to remove the storage buffer. Add 1 mL lysis buffer to required
amount of resin and centrifuge at 500 g for 5 min at 4 C.
Discard the supernatant and repeat the wash with 1 mL lysis
buffer.
14. The lysate is incubated with Protein G Sepharose resin to
deplete human IgG carried over with the isolated cells. IgG
and 3G8 bind to the second immunoglobulin-like domain of
CD16a; furthermore, 3G8 is known to inhibit IgG binding to
CD16a, thus the presence of IgG can potentially reduce the
CD16a-3G8 interaction [35].
15. Efficiency of CD16a immunoprecipitation and final protein
yield can be determined by Western blot analysis as described
elsewhere [27]. CD16a depletion of over 60% in flow-through
as compared to crude lysate is indicative of efficient immuno-
precipitation. Amount of CD16a in 5% of eluted sample is
compared to standard quantity of recombinant CD16a to esti-
mate approximate yields.
16. Lysate sonication increases the final CD16a yield possibly by
destabilizing large detergent micelles and releasing CD16a into
smaller protein-detergent complex.
17. Glycopeptide enrichment is often a necessary step to prevent
ion suppression of glycopeptides since unmodified peptides
ionize relatively better than glycosylated peptides. Moreover,
residual detergent carried over from the immunoprecipitation
step can reduce mass spectrometer sensitivity. This step
removes residual detergent from the glycopeptide fraction.
Unmodified peptides can be analyzed separately after washing
the pooled wash and flow-through fractions with water
saturated ethyl acetate [29, 36].
18. Detailed description of mass spectrometer instrument para-
meters and a sample method file can be found elsewhere
[29, 37]. MS2 scan range can be pre-set by the user to be
consistent for all MS2 events or can be variable and data
dependent. Setting the lower MS2 scan range to 125 m/z can
allow acquisition of data for oxonium ions corresponding to
breakdown product of monosaccharides.
19. Multiple missed proteolytic cleavages were frequently observed

in glycopeptide modified by large glycan structure with multi-
ple LacNAc units. Thus, identifying large glycopeptides cre-
ated due to as many as four missed cleavage should be
considered. Multiple glycopeptides can span a single N-glyco-
sylation site, and glycan structures identified on all possible
glycopeptides should be characterized to avoid potential bias
introduced by proteolysis affected by type of glycan; e.g., core
fucose likely prevented complete proteolysis of peptide con-
taining NK cell-CD16a N45 glycan [29]. Ion corresponding
to peptide with one GlcNAc is generally the most abundant
fragmented glycopeptide ion in MS2 spectra and should be
used to identify the parent ion. Presence of glycan oxonium
ions in MS2 spectra is indicative of glycosylated peptide and
can reveal information about glycan structure.
20. Computational tools such as NIST Glycan Mass calculator,
GlycoMod, and pGlyco are very useful in determining glycan
structures based on mass of parent ion and ions in MS2 spectra
[38, 39]. Low abundance glycopeptide ions which do not
trigger MS2 can be identified with predictable difference in
mass corresponding to mono- or disaccharides within same
retention time range in MS1 spectra, e.g., a difference of
365.13 Da indicates difference of a disaccharide LacNAc unit.
21. An example of the format of glycopeptide mass list .cvs file.
Unique identifier contains N-glycosylation site, peptide num-
ber, monosaccharide composition, and charge.
Unique identifier Exact neutral mass Charge

N38_1_10-11-0-4-0_4 7390.842 4
N38_1_10-11-0-4-0_5 7390.842 5
N38_1_10-11-1-2-0_4 6954.7091 4
Acknowledgments

Grants T32 HL 066987 to KRP, U01 AI148114 to AWB, and
U01 CA242109 to SRS.
References
1. Arthur CM, Baruffi MD, Cummings RD, 2. Rabinovich GA, Toscano MA (2009) Turning
Stowell SR (2015) Evolving mechanistic ‘sweet’ on immunity: galectin-glycan interac-
insights into galectin functions. Methods Mol tions in immune tolerance and inflammation.
Biol 1207:1–35. https://doi.org/10.1007/ Nat Rev Immunol 9(5):338–352. https://doi.
978-1-4939-1396-1_1 org/10.1038/nri2536
3. Giovannone N, Smith LK, Treanor B, Dimitr- Visner M, Josephson CD, Stowell SR (2015)
off CJ (2018) Galectin-glycan interactions as Innate immunity against molecular mimicry:
regulators of B cell immunity. Front Immunol examining galectin-mediated antimicrobial
9:2839. https://doi.org/10.3389/fimmu. activity. BioEssays 37(12):1327–1337.
2018.02839 https://doi.org/10.1002/bies.201500055
4. Varki A, Cummings RD et al (eds) (2015) 13. Stowell SR, Arthur CM, Dias-Baruffi M,
Essentials of glycobiology. Cold Spring Harbor Rodrigues LC, Gourdine JP, Heimburg-
Laboratory Press, Cold Spring Harbor, NY Molinaro J, Ju T, Molinaro RJ, Rivera-
5. Kamili NA, Arthur CM, Gerner-Smidt C, Marrero C, Xia B, Smith DF, Cummings RD
Tafesse E, Blenda A, Dias-Baruffi M, Stowell (2010) Innate immune lectins kill bacteria
SR (2016) Key regulators of galectin-glycan expressing blood group antigen. Nat Med
interactions. Proteomics 16(24):3111–3125. 16(3):295–301. https://doi.org/10.1038/
https://doi.org/10.1002/pmic.201600116 nm.2103
6. Nielsen MI, Stegmayr J, Grant OC, Yang Z, 14. Stowell SR, Arthur CM, McBride R, Berger O,
Nilsson UJ, Boos I, Carlsson MC, Woods RJ, Razi N, Heimburg-Molinaro J, Rodrigues LC,
Unverzagt C, Leffler H, Wandall HH (2018) Gourdine JP, Noll AJ, von Gunten S, Smith
Galectin binding to cells and glycoproteins DF, Knirel YA, Paulson JC, Cummings RD
with genetically modified glycosylation reveals (2014) Microbial glycan microarrays define
galectin-glycan specificities in a natural context. key features of host-microbial interactions.
J Biol Chem 293(52):20249–20262. https:// Nat Chem Biol 10(6):470–476. https://doi.
doi.org/10.1074/jbc.RA118.004636 org/10.1038/nchembio.1525
7. Giovannone N, Liang J, Antonopoulos A, 15. Di Lella S, Sundblad V, Cerliani JP, Guardia
Geddes Sweeney J, King SL, Pochebit SM, CM, Estrin DA, Vasta GR, Rabinovich GA
Bhattacharyya N, Lee GS, Dell A, Widlund (2011) When galectins recognize glycans:
HR, Haslam SM, Dimitroff CJ (2018) from biochemistry to physiology and back
Galectin-9 suppresses B cell receptor signaling again. Biochemistry 50(37):7842–7857.
and is regulated by I-branching of N-glycans. https://doi.org/10.1021/bi201121m
Nat Commun 9(1):3287. https://doi.org/10. 16. Demetriou M, Granovsky M, Quaggin S, Den-
1038/s41467-018-05770-9 nis JW (2001) Negative regulation of T-cell
8. Stowell SR, Arthur CM, Mehta P, Slanina KA, activation and autoimmunity by Mgat5
Blixt O, Leffler H, Smith DF, Cummings RD N-glycosylation. Nature 409(6821):733–739.
(2008) Galectin-1, -2, and -3 exhibit differen- https://doi.org/10.1038/35055582
tial recognition of sialylated glycans and blood 17. Smith LK, Boukhaled GM, Condotta SA,
group antigens. J Biol Chem Mazouz S, Guthmiller JJ, Vijay R, Butler NS,
283(15):10109–10123. https://doi.org/10. Bruneau J, Shoukry NH, Krawczyk CM,
1074/jbc.M709545200 Richer MJ (2018) Interleukin-10 directly inhi-
9. Stowell SR, Arthur CM, Slanina KA, Horton bits CD8(+) T cell function by enhancing
JR, Smith DF, Cummings RD (2008) Dimeric N-glycan branching to decrease antigen sensi-
Galectin-8 induces phosphatidylserine expo- tivity. Immunity 48(2):299–312.e5. https://
sure in leukocytes through polylactosamine doi.org/10.1016/j.immuni.2018.01.006
recognition by the C-terminal domain. J Biol 18. Moremen KW, Tiemeyer M, Nairn AV (2012)
Chem 283(29):20547–20559. https://doi. Vertebrate protein glycosylation: diversity, syn-
org/10.1074/jbc.M802495200 thesis and function. Nat Rev Mol Cell Biol
10. Stowell SR, Cho M, Feasley CL, Arthur CM, 13(7):448–462. https://doi.org/10.1038/
Song X, Colucci JK, Karmakar S, Mehta P, nrm3383
Dias-Baruffi M, McEver RP, Cummings RD 19. Lee LY, Lin CH, Fanayan S, Packer NH,
(2009) Ligand reduces galectin-1 sensitivity Thaysen-Andersen M (2014) Differential site
to oxidative inactivation by enhancing dimer accessibility mechanistically explains
formation. J Biol Chem 284(8):4989–4999. subcellular-specific N-glycosylation determi-
https://doi.org/10.1074/jbc.M808925200 nants. Front Immunol 5:404. https://doi.
11. Stowell SR, Dias-Baruffi M, Penttila L, org/10.3389/fimmu.2014.00404
Renkonen O, Nyame AK, Cummings RD 20. Subedi GP, Falconer DJ, Barb AW (2017)
(2004) Human galectin-1 recognition of Carbohydrate-polypeptide contacts in the anti-
poly-N-acetyllactosamine and chimeric poly- body receptor CD16A identified through solu-
saccharides. Glycobiology 14(2):157–167. tion NMR spectroscopy. Biochemistry
https://doi.org/10.1093/glycob/cwh018 56(25):3174–3177. https://doi.org/10.
12. Arthur CM, Patel SR, Mener A, Kamili NA, 1021/acs.biochem.7b00392
Fasano RM, Meyer E, Winkler AM, Sola-
21. Goh JB, Ng SK (2018) Impact of host cell line CD16a glycoforms with donor-dependent
choice on glycan profile. Crit Rev Biotechnol variability. Mol Cell Proteomics
07388551.2017.1416577 1074/mcp.RA119.001607
22. Zeck A, Pohlentz G, Schlothauer T, Peter- 30. Illiano A, Pinto G, Melchiorre C,
Katalinic J, Regula JT (2011) Cell type-specific Carpentieri A, Faraco V, Amoresano A (2020)
and site directed N-glycosylation pattern of Protein glycosylation investigated by mass
FcgammaRIIIa. J Proteome Res spectrometry: an overview. Cell 9(9):1986.
10(7):3031–3039. https://doi.org/10.1021/ https://doi.org/10.3390/cells9091986
pr1012653 31. Ruhaak LR, Xu G, Li Q, Goonatilleke E, Leb-
23. Robinson BS, Arthur CM, Evavold B, rilla CB (2018) Mass spectrometry approaches
Roback E, Kamili NA, Stowell CS, Vallecillo- to glycomic and glycoproteomic analyses.
Zuniga ML, Van Ry PM, Dias-Baruffi M, Chem Rev 118(17):7886–7930. https://doi.
Cummings RD, Stowell SR (2019) The org/10.1021/acs.chemrev.7b00732
sweet-side of leukocytes: galectins as master 32. Shajahan A, Heiss C, Ishihara M, Azadi P
regulators of neutrophil function. Front (2017) Glycomic and glycoproteomic analysis
Immunol 10:1762. https://doi.org/10. of glycoproteins-a tutorial. Anal Bioanal Chem
3389/fimmu.2019.01762 409(19):4483–4505. https://doi.org/10.
24. Gudelj I, Lauc G, Pezer M (2018) Immuno- 1007/s00216-017-0406-7
globulin G glycosylation in aging and diseases. 33. Sun L, Zhu G, Li Y, Yang P, Dovichi NJ (2012)
Cell Immunol 333:65–79. https://doi.org/ Coupling methanol denaturation, immobilized
10.1016/j.cellimm.2018.07.009 trypsin digestion, and accurate mass and time
25. Pucic M, Knezevic A, Vidic J, Adamczyk B, tagging for liquid-chromatography-based
Novokmet M, Polasek O, Gornik O, Supraha- shotgun proteomics of low nanogram amounts
Goreta S, Wormald MR, Redzic I, of RAW 264.7 cell lysate. Anal Chem
Campbell H, Wright A, Hastie ND, Wilson 84(20):8715–8721. https://doi.org/10.
JF, Rudan I, Wuhrer M, Rudd PM, Josic D, 1021/ac3019608
Lauc G (2011) High throughput isolation and 34. Xiao H, Chen W, Smeekens JM, Wu R (2018)
glycosylation analysis of IgG-variability and An enrichment method based on synergistic
heritability of the IgG glycome in three isolated and reversible covalent interactions for large-
human populations. Mol Cell Proteomics scale analysis of glycoproteins. Nat Commun
10(10):M111.010090. https://doi.org/10. 9(1):1692. https://doi.org/10.1038/
1074/mcp.M111.010090 s41467-018-04081-3
26. Wuhrer M, Stam JC, van de Geijn FE, Koele- 35. Tamm A, Schmidt RE (1996) The binding
man CA, Verrips CT, Dolhain RJ, Hokke CH, epitopes of human CD16 (Fc gamma RIII)
Deelder AM (2007) Glycosylation profiling of monoclonal antibodies. Implications for ligand
immunoglobulin G (IgG) subclasses from binding. J Immunol 157(4):1576–1581
human serum. Proteomics 7(22):4070–4081. 36. Yeung YG, Stanley ER (2010) Rapid detergent
https://doi.org/10.1002/pmic.200700289 removal from peptide samples with ethyl ace-
27. Patel KR, Roberts JT, Subedi GP, Barb AW tate for mass spectrometry analysis. Curr Pro-
(2018) Restricted processing of CD16a/Fc toc Protein Sci Chapter 16:Unit 16.12.
gamma receptor IIIa N-glycans from primary https://doi.org/10.1002/0471140864.
human NK cells impacts structure and func- ps1612s59
tion. J Biol Chem 293(10):3477–3489. 37. Patel KR, Roberts JT, Barb AW (2020)
https://doi.org/10.1074/jbc.RA117.001207 Allotype-specific processing of the CD16a
28. Neron S, Thibault L, Dussault N, Cote G, N45-glycan from primary human natural killer
Ducas E, Pineault N, Roy A (2007) Character- cells and monocytes. Glycobiology
ization of mononuclear cells remaining in the 30(7):427–432. https://doi.org/10.1093/
leukoreduction system chambers of apheresis glycob/cwaa002
instruments after routine platelet collection: a 38. Cooper CA, Gasteiger E, Packer NH (2001)
new source of viable human blood cells. Trans- GlycoMod--a software tool for determining
fusion 47(6):1042–1049. https://doi.org/10. glycosylation compositions from mass spectro-
1111/j.1537-2995.2007.01233.x metric data. Proteomics 1(2):340–349.
29. Patel KR, Nott JD, Barb AW (2019) Primary h t t p s : // d o i . o r g / 1 0 . 1 0 0 2 / 1 6 1 5 - 9 8 6 1
human natural killer cells retain proinflamma- (200102)1:2<340::AID-PROT340>3.0.
tory IgG1 at the cell surface and express CO;2-B
39. Liu MQ, Zeng WF, Fang P, Cao WQ, Liu C, comprehensive quality control and one-step
Yan GQ, Zhang Y, Peng C, Wu JQ, Zhang XJ, mass spectrometry for intact glycopeptide
Tu HJ, Chi H, Sun RX, Cao Y, Dong MQ, identification. Nat Commun 8(1):438.
Jiang BY, Huang JM, Shen HL, Wong CCL, https://doi.org/10.1038/s41467-017-
He SM, Yang PY (2017) pGlyco 2.0 enables 00535-2
precision N-glycoproteomics with
Chapter 14
Transformation of Agrocybe cylindracea Galectin

into αGalNAc-Specific Lectin
Dan Hu and Jun Hirabayashi
Abstract
A multi-specific fungal galectin from the mushroom Agrocybe cylindracea (ACG) binds a broad range of
β-galactosides, as well as their derivative GalNAcα1-3Gal. Site-directed mutagenesis of the hydrophilic
residues His, Asn, Arg, and Glu, involved in carbohydrate recognition, abolished the binding affinity of the
derived mutants to β-galactosides, whereas only N46A caused increased affinity to GalNAcα1-3Gal-con-
taining oligosaccharides and loss of β-galactoside-binding activity. Detailed structural analysis revealed that
Pro45, the preceding residue of Asn46 of the wild-type ACG, takes the cis imide conformation to tether
Asn46 onto a loop region to make new hydrogen bonds with β-galactosides and to compensate for the lack
of evolutionarily conserved Asn. In contrast, in the N46A mutant, Pro45 takes the more stable trans
conformation, resulting in “switched” specificity to αGalNAc. Such an altered recognition system in the
binding specificity of galectins can be observed in other lectin molecules not only in nature but will also be
observed in those engineered in the future.
Key words Agrocybe cylindracea, Cis-trans tautomerism, Proline, Imide bond, Extra loop,
β-Galactoside, αGalNAc, Evolution, Lectin engineering, Transformation
1 Introduction
Site-directed mutagenesis is a useful method to identify the key

amino acid residues involved in protein functions. Its application to
galectins has been shown to be valid in previous research, before
crystallographic studies identified sugar-binding residues and their
binding features, including hydrogen bonding and van der Waals
contacts. Such an approach sometimes produces an unexpected
finding: alanine substitution at one of the sugar-binding residues
resulted in enhanced affinity to a particular group of oligosacchar-
ides, while all other mutants lost substantial sugar-binding
activity [1].
The galectin known as ACG (a multi-specific fungal galectin)
was originally isolated from the mushroom Agrocybe cylindracea by
Yagi et al. as a soluble lectin with strong interactions with
233
234 Dan Hu and Jun Hirabayashi
glycoconjugates containing Neu5Acα2-3Galβ1-3GlcNAc/GalNAc

sequences [2]. Later, it was found to be a member of the galectin
family [3]. ACG is a prototype galectin, with the 16,500-Da sub-
units, which is somewhat larger than other prototype galectins (i.e.,
14,000–15,000 Da). Its size is owing to the presence of an addi-
tional β-strand, designated F0, as well as a redundant loop region
between the F2 and S3 strands (Fig. 1) [4]. The physiological
function of ACG remains unclear; however, a galectin called
CCG2, from another mushroom, Coprinopsis cinerea, has been
shown to have anti-nematode activity arising through the recogni-
tion of Galβ1-4Fuc structures, which form an N-glycan core in
Caenorhabditis elegans [5]. Antitumor activity has been reported
in another mushroom galectin, i.e., Agrocybe aegerita galectin
(AAG). AAG shows striking similarity to ACG, with 90% amino
acid identity, and both the F0 β-strand and the redundant loop
mentioned above are conserved (Fig. 2). However, the details of
the sugar-binding specificity of AAG remain to be investigated [6].
In this chapter, we describe the engineering of an ACG mutant,
N46A, which has acquired an altered carbohydrate specificity by a
single amino acid substitution targeting hydrophilic amino acid
residues involved in β-galactoside binding ([7]; for a recent review,
see [1]). Evolutionarily conserved hydrophilic residues play an
essential role in the sugar-binding function [8]. These residues are
His44, Asn46, Arg48, Asn61, Glu71, and Arg73 (residue numbers
are those of human galectin-1, where the initial methionine is
numbered 0). There are, however, a significant number of excep-
tions in naturally occurring galectins, in which substituted sugar-
binding residues produce a reduction or loss of the sugar-binding
function. Such loss-of-function events can be compensated for by
the appearance of a “rescue” residue, which becomes transiently
involved in the binding function in place of the mutated residue. In
the case of ACG, the presence of a redundant loop region consist-
ing of Gly-Ser-Pro45-Asn46-Asn was found to play a key role in
compensating for the lack of the evolutionarily conserved Asn
(corresponding to Asn61 in human galectin-1): Asn46 contributes
to this rescue-type hydrogen bonding to the 4-OH group of βGal.
For this phenomenon to occur, however, the imide group of the
preceding Pro45 must take part in the cis conformation, to tether
the side chain of Asn46 to form hydrogen bonds with the 4-OH of
Gal (Fig. 3).
The resulting N46A mutant totally lost its sugar-binding func-
tion toward all β-galactosides, while retaining the αGalNAc-bind-
ing activity [7]. N46A was therefore no longer a galectin under the
original definition; galectins are defined as an evolutionarily related
group of β-galactoside-binding proteins [9]. Structural analysis
revealed a significant conformational change in the redundant
loop region of the N46A mutant, where Pro45 takes a normal
trans conformation to facilitate the contact of another set of
Transformation of Agrocybe cylindracea Galectin into αGalNAc. . . 235
Fig. 1 A schematic drawing of the secondary structure of ACG. Amino acid residues involved in the lactose-
binding function are numbered and shown in red. Pro45 takes a cis conformation when binding to
β-galactoside (e.g., lactose) while it has a trans conformation when it binds to αGalNAc (e.g., Forssman
disaccharide). Redundant loops between F2 and S3 β-strands and between S4 and S5 β-strands are shown in
light blue
Fig. 2 Alignment of amino acid sequences of ACG and related galectins in mushroom. Evolutionarily conserved
sugar-binding residues are shaded and numbered above the sequence of ACG and below that of human
galectin-1. Pro45-Asn46 (PN), which are highlighted with a red background, are involved in the βGal-binding
function when the imide bond of Pro45 takes the cis conformation. F0-F5 and S1-S6 β-strands are shown in
blue and red horizontal bars, respectively. Abbreviations; ACG: Agrocybe cylindracea galectin; AAG: Agrocybe
aegerita galectin; CCG2: Coprinopsis cinerea galectin-2; human galectin-1 (hGal-1)
Fig. 3 Summary of the binding feature of ACG to N-acetyllactosamine. A lack of evolutionarily conserved Asn
(Ala64 in ACG), leads to insufficient Asn to recognize the 4-OH group of the non-reducing terminal galactose.
To compensate for this lack, the Asn46 side chain comes close, to contact the 4-OH group. This happens only
when the imide bond of Pro45 takes the cis conformation. Data are originally from Kuwabara et al. [10]
residues with αGalNAc, including the side chain of evolutionarily

conserved tryptophan (corresponding to Trp89 of human galectin-
1) with the 6-OH group of αGalNAc, as well as side chains of
Asn47 and Ala49, to make van der Waals contact with the N-acetyl
group of αGalNAc [10]. These findings led to the conclusion that
Pro residues located in a redundant loop region can transform
galectin specificity from β-galactoside to others via cis-trans
tautomerism.
In the following sections, polymerase chain reaction (PCR)-
aided site-directed mutagenesis experiments targeting Asn46 of
ACG are described as a model case. There is a possibility that cis-
trans tautomerism of a particular Pro residue causes transformation
of the sugar-binding function in other galectins, such as AAG, as
well as other carbohydrate-binding proteins [7].
2 Materials
Site-directed mutagenesis is performed using a pair of custom-

synthesized primers: forward primer contains a mutagenic site
(N46A), while the reverse primer lacks this site, but overlaps the
50 -region of the former primer (Fig. 4; also see Notes 1 and 2). The
method is conducted as described by Nøhr and Kristiansen [11],
and pET27b-Flag-ACG is used as a template [7].
2.1 Primers 1. Mutagenic forward primer (N46A-F): 50 -CT GCA GGT GCG
GGT AGT CCG GCT AAT ACG GCC TTG-30 coding for
“Ala-Gly-Ala-Gly-Ser-Pro-Ala (46)-Asn-Thr-Ala-Leu,” where
mutagenic sites are underlined.
Fig. 4 Scheme showing procedures for PCR-based site-directed mutagenesis. The expression plasmid
pET27b-FLAG-ACG is used as a template for PCR amplification of the plasmid pET27b-FLaG-N46D. Asterisks
(*) denote methylation sites for DpnI, which cleaves E. coli-derived plasmid, but not PCR-amplified mutant
plasmids
2. Overlapping reverse primer: 50 -CGG ACT ACC CGC ACC

TGC AGA CAG GTT TAA C-30 antisense coding for “Leu-
Asn-Leu-Ser-Ala-Gly-Ala-Gly-Ser-Pro.”
2.2 PCR 1. Thermal cycler (GeneAmp® PCR System 9700, Applied

Biosystems).
2. 0.2-mL polypropylene tube.
3. AccuPrime Taq DNA polymerase (5 units/μL, Invitrogen).
4. 10 AccuPrimer PCR buffer I (Invitrogen).
2.3 Transformation A pET expression vector is used to express the mutant recombinant
and Bacterial protein N46A in BL21 (DE3) E. coli strains.
Expression
1. pET27b-FLAG-ACG (used as a template for mutagenic PCR;
see Note 3).
2. Dpn I (approx. 10 units/μL; New England Biolabs).
3. Plasmid purification kit (Qiagen).
4. DH5α competent cells (Takara).
5. SOC medium.
6. LB medium.
7. LB-agar plate containing kanamycin (50 μg/mL).
8. 0.1 M isopropyl β-D-1-thiogalactopyranoside (IPTG)
solution.
9. BL21 (DE3) E. coli strains (Novagen).
2.4 Glycoconjugate Glycoconjugate microarray analysis is recommended for rapid

Microarray profiling of the glycan-binding properties of mutant clones using
bacterial lysates [12]. It enables sensitive analysis with a panel of
around 100 glycoconjugates (see Note 4).
1. Evanescent-type scanner: GlycoStation™ Reader 1200 (Gly-
coTechnica, Ltd., Yokohama, Japan; see Note 5).
2. Glycoconjugate microarray consisting of 96 glycoconjugates:
serially glycosidase-treated common glycoproteins and a variety
of synthetic glycan-polyacrylamide conjugates (GlycoTech;
http://www.glycotech.com/glycoconjugates.html), immobi-
lized on an epoxy-activated glass slide and prepared as
described by Tateno et al. ([12]; also see Note 6).
3. BugBuster® HT Protein Extraction Reagent (Novagen).
4. Probing buffer: 25 mM Tris–HCl, pH 7.5, 140 mM NaCl
(TBS) containing 2.7 mM KCl, 1 mM CaCl2, 1 mM MnCl2,
and 1% Triton X-100.
2.5 Purification of 1. Lactosyl-Sepharose (Cytiva, formerly GE Healthcare Life

N46A Mutant by Sciences).
Affinity 2. Polypropylene-type disposable column (1.5 12 cm,
Chromatography BIO-RAD).
3. 1 PBS (): 10 mM Na-phosphate, 137 mM NaCl, 2.7 mM
KCl, pH 7.4.
4. Protease inhibitor cocktail (Sigma-Aldrich).
5. Lysis buffer: PBS () containing 1 mM EDTA, 0.1% (w/v)
TritonX-100, and protease inhibitor cocktail (diluted
100 fold).
6. Wash buffer: PBS () containing 1 mM EDTA.
7. Elution buffer: PBS () containing 0.1 M lactose.
8. Sonicator: Sonifier (Branson).
2.6 Frontal Affinity Frontal affinity chromatography (FAC) has been used to evaluate
Chromatography (FAC) the sugar-binding properties of galectins [13–15], producing dis-
sociation constants (Kd’s) to a large (>100) panel of fluorescently
labeled oligosaccharides in a precise and sensitive manner ([16–19];
see Note 7).
1. Frontal Affinity Chromatography (FAC) system: An automated
FAC system (Shimadzu, Kyoto) (see Note 8).
2. Labeled oligosaccharides: various oligosaccharides can be used
for this purpose (see Note 9). Prepare 1–10 nM fluorescently
labeled (e.g., pyridylaminated; PA), or 1–100 μM UV-sensitive
(e.g., p-nitrophenyl and p-methoxyphenyl) saccharides. The
latter is often used for concentration-dependence analysis to
determine the column capacity (Bt).
3. Tris-buffered saline (TBS): 10 mM Tris–HCl, pH 7.4, and
0.8% NaCl.
4. Recombinant ACG galectin (N46A): prepared as a 4 mg/mL
solution.
5. N-Hydroxysuccinimide (NHS)-activated Sepharose-4FF:
prepared as 100–200 μL of 50% suspension (Cytiva, formerly
GE Healthcare Life Sciences).
6. Coupling buffer: 0.2 M NaHCO3, 0.5 M NaCl, pH 8.3.
7. 1 mM ice-cold HCl.
8. Blocking buffer: 0.5 M monoethanolamine, pH 8.3,
0.5 M NaCl.
9. Capsule-type miniature column (inner diameter, 2 mm; length,
10 mm; bed volume, 31.4 μL, Shimadzu).
3 Methods
3.1 Site-Directed PCR-based site-directed mutagenesis is performed using the whole

Mutagenesis vector pET27b-FLAG-ACG as a template, with the forward
primer, N46A-F, containing the desired mutations and the reverse
primer without mutations (Fig. 4; see also Note 10).
3.1.1 Amplification of the 1. Prepare a 0.2-mL polypropylene tube.

Whole Vector Using 2. Add 1 μL of a template DNA (pET27b-FLAG-ACG, approxi-
Mutagenic Primers mately 10 ng/μL; see Note 1), 1 μL each of forward and reverse
(both 10 μM) primers, 5 μL of 10 AccuPrimer PCR buffer I,
0.2 μL of AccuPrime Taq DNA polymerase (5 units/μL), and
42 μL of distilled water (total volume, 50 μL).
3. Set a thermal cycler program as follows:
(a) Initial denaturation: 94 C, 2 min.
(b) 25 cycles.
l Denaturation: 94 C, 15 s.
l Annealing: 60 C, 30 s.
l Elongation: 68 C, 6 min.
(c) Final extension: 68 C, 10 min.
4. Next, 1 μL (10 units) of DpnI solution is added and incubated
at 37 C for 1 h to completely digest the parental plasmid.
3.1.2 Transformation of 1. Thaw 50 μL of DH5α competent cells on ice.

Competent Cells 2. Add 5 μL of the DpnI-treated PCR solution (step 4 of Sub-
heading 3.1.1) and gently flick the tube three times before
incubating on ice for 30 min.
3. Heat shock the cells by incubation at 42 C for about 45 s.
4. Incubate on ice for more than 2 min.
5. SOC media (450 μL) is added to the cells, which are then
agitated at 200 rpm and 37 C for 1 h.
6. A 200 μL aliquot is spread on a prewarmed LB-agar plate
containing kanamycin (50 μg/mL).
3.1.3 Plasmid Extraction 1. Pick 3–5 bacterial colonies and grow them overnight in 5 mL
and Sequencing LB medium.
2. Purify plasmid using a Qiagen plasmid purification kit.
3. Sequencing of DNA is used to confirm that the plasmid con-
tains the desired mutation, and to obtain the target plasmid
pET27b-FLAG-ACG (N46A).
3.2 Analysis of 1. Transform E. coli strain BL21-CondonPlus (DE3)-RIL with

Sugar-Binding the sequence-confirmed pET27b-FLAG-ACG (N46A).
Functions of ACG 2. Induce expression of N-terminal Flag-tag N46A by adding
Mutants by 0.1 M IPTG to a final concentration of 1 mM after E. coli
Glycoconjugate culture reaches OD600 at 0.5–0.6.
Microarray Using 3. Incubate at 20 C for 24 h with shaking (200 rpm).
Crude Samples
4. Wash harvested cells with PBS () three times before lysis by
suspension in 1/5 culture volume of BugBuster® HT Protein
Extraction Reagent.
6. Dilute 1 μL of the supernatant in 100 μL probing buffer.
7. Add 0.5 μL of anti-Flag M2 antibody (4.6 mg/mL).
8. Incubate at 4 C for 20 min.
9. Add 2 μL of Cy3-labeled goat anti-mouse antibody (1 mg/
mL).
10. Incubate for another 20 min at 4 C.
11. Apply the labeled solution onto a glycoconjugate microarray.
12. Incubate at 20 C for 20 h.
13. Wash out the unbound proteins with 100 μL probing buffer
three times.
14. Detect fluorescence intensity by an evanescent-type scanner
under Cy3 mode.
Typical results are shown in Fig. 5 (see ref. 7).
3.3 Purification of 1. Induce expression of N-terminal Flag-tagged N46A as

N46A Mutant by described in steps 1–3 of Subheading 3.2.
Affinity 2. Harvest the cells and suspend them in 20 mL PBS ().
Chromatography
3. Wash the cells with PBS () three times.
4. Suspend the cells in PBS () containing 1 mM EDTA, 0.1%
TritonX-100, and protease inhibitor cocktail (diluted
100 fold).
5. Lyse cells by sonication using Sonifier at 4 C for 2 min (output
2, duty cycle 50).
6. Repeat step 4 after 1 min interval.
7. Centrifuge the lysate at 150,000 g for 30 min.
8. Collect the supernatant and apply it onto a Lactosyl-Sepharose
(2.5 mL beads) column equilibrated with wash buffer.
9. Wash the column with 12 mL of wash buffer.
10. Elute N46A with elution buffer.
11. Collect the elute and dialyze extensively against PBS () over-
night at 4 C.
Fig. 5 Typical result of glycoconjugate microarray analysis showing different glycan-binding profiles between
wild-type AGC (top) and mutant N46A (bottom). In the analysis, it is clear that wild-type ACG shows substantial
binding to a broad range of glycans including conventional β-galactosides and non-conventional αGalNAc-
containing glycans. On the other hand, N46A mutant shows only a selected set of glycans composed of
αGalNAc, such as A-tetrasaccharide and Forssman disaccharide
3.4 Quantitative 1. Add 100 μL of NHS-activated Sepharose Fast Flow (50% sus-
Analysis of Sugar- pension) to a 1.5-mL polypropyrene conical tube.
Binding Specificity of 2. Wash the beads with 1 mL of 1 mM ice-cold HCl.
N46A by Frontal
3. Discard the supernatant.
Affinity
Chromatography (FAC)
4. Add 500 μL of 4 mg/mL of recombinant N46A dissolved in
coupling buffer.
Using Purified
Samples 5. Incubate the suspension at room temperature for 2 h.
6. Discard the supernatant and add 1 mL of blocking buffer.
7. Incubate the suspension at room temperature for 1 h.
8. Discard the supernatant and wash the beads with TBS.
9. Suspend the beads in 1 mL of TBS.
10. Pack the lectin-immobilized gel into a capsule-type miniature
column.
11. Connect the miniature column to an automated FAC system
(FAC-1).
12. Inject a series of fluorescently labeled oligosaccharides succes-

sively into the column using the auto-sampling system.
13. Perform elution by fluorescence. In the case of PA oligosac-
charides, excitation and emission wavelengths are 310 nm and
380 nm, respectively.
14. Analyze the elution font volume (V) as described by Arata
et al. [20].
15. Determine the Kd value for each oligosaccharide according to
the following equation established by Kasai [16]: Kd ¼ Bt/
(V V0),where Bt is the column content (mol) defined by a
series of concentration-dependency analyses using 1–100 μM
UV-sensitive saccharides (e.g., see ref. 21).
4 Notes
1. The forward primer should be 30–35 nucleotides in length and

contain a target site for mutation in the center, with correct
sequences on both sides. The reverse primer is reverse comple-
mentary to approximately half the 50 region of the forward
primer, so it does not contain the mutation site (Fig. 4).
2. Primers should have a minimum GC content of 40% and
terminate in one or more C or G.
3. The ACG-coding region flanked by NdeI and XhoI is PCR
amplified, and the PCR fragment is treated with NdeI and
XhoI and inserted into pET27b-FLAG (Novagen) with an N-
terminal FLAG tag. Alternatively, to request plasmids, consult
either of the authors (D. H. in Jinan University, Guangzhou, or
J. H. in the National Institute of Advanced Industrial Science
and Technology, Tsukuba).
4. The method is effectively performed with an evanescent-field-
activated fluorescence detection instrument (also see Note 5).
It enables direct detection of Flag-tagged mutant lectins with-
out purification, with the aid of anti-FLAG M2 antibody
(mouse), to which Cy3-labeled anti-mouse antibody is
bound, where only the bound Cy3-labeled material is detect-
able under the evanescent field (<100 nm).
5. GlycoLite™ 2200 of the same marker with restricted functions
(e.g., excitation wavelength selective for Cy3) can also be used
for this purpose. For commercial information, see https://
www.glycotechnica.com/en/tools/.
6. Such an array will also be commercially available from either
Sumitomo Bakelite (Tokyo) or GlycoTechnica (Yokohama).
7. Other analytical methods are also available for this purpose;
however, many of them have basic difficulties in the accurate
determination of Kds, particularly for oligosaccharides of com-

plex structures. For detailed procedures, readers are advised to
refer to the previous edition of this series by Iwaki and
Hirabayashi [20].
8. For details, see ref. 20. Briefly, the typical system is composed of
a controller, a pair of isocratic pumps, each connecting to a
capsule-type miniature column (2 mm 10 mm) packed with
galectin-immobilized resin, a column oven (25 C), an auto-
sampler, and either fluorescence (excitation: 310 nm, emission:
380 nm) or UV (absorbance: 280 nm) detectors. The flow rate
for both sample injection and analysis is maintained at
0.125 mL/min to ensure a dynamic equilibrium state in the
column, where the linear flow rate is as low as 0.66 mm/s.
Signal intensity vs. elution time is used to process the elution
profile to precisely calculate the elution front (mL), according
to the methods described by Arata et al. [20].
9. A series of PA oligosaccharides are commercially available from
Masuda Chemical Industries (Takamatsu), Seikagaku Corpora-
tion (Tokyo), and Takara Bio Incorporation, Kusatsu).
pNP-labeled oligosaccharides are purchased from Sigma, Cal-
biochem, Funakoshi Co., Ltd., and Toronto Research
Chemicals, Inc.
10. To ensure that no extra mutations are introduced in the vector
backbone, subcloning of the sequenced insert into a vector
that has not been amplified by PCR should be performed.
Alternatively, the whole vector can be sequenced.
References
1. Hirabayashi J, Hu D, Tateno H et al (2018) 6. Yang N, Li DF, Feng L et al (2009) Structural
Carbohydrate recognition mechanism of the basis for the tumor cell apoptosis-inducing
mushroom galectin ACG. Trends Glycosci activity of an antitumor lectin from the edible
Glycotechnol 30(172):SE75–SE88 mushroom Agrocybe aegerita. J Mol Biol
2. Yagi F, Miyamoto M, Abe T et al (1997) Puri- 387(3):694–705
fication and carbohydrate-binding specificity of 7. Hu D, Tateno H, Sato T et al (2013) Tailoring
Agrocybe cylindracea lectin. Glycoconj J GalNAcalpha1-3Galbeta-specific lectins from a
14(2):281–288 multi-specific fungal galectin: dramatic change
3. Yagi F, Hiroyama H, Kodama S (2001) Agro- of carbohydrate specificity by a single amino-
cybe cylindracea lectin is a member of the galec- acid substitution. Biochem J 453(2):261–270
tin family. Glycoconj J 18(10):745–749 8. Hirabayashi J, Kasai K (1991) Effect of amino
4. Ban M, Yoon HJ, Demirkan E et al (2005) acid substitution by sited-directed mutagenesis
Structural basis of a fungal galectin from Agro- on the carbohydrate recognition and stability
cybe cylindracea for recognizing sialoconjugate. of human 14-kDa beta-galactoside-binding
J Mol Biol 351(4):695–706 lectin. J Biol Chem 266(35):23648–23653
5. Butschi A, Titz A, Walti MA et al (2010) Cae- 9. Barondes SH, Castronovo V, Cooper DN et al
norhabditis elegans N-glycan core beta- (1994) Galectins: a family of animal beta-
galactoside confers sensitivity towards nemato- galactoside-binding lectins. Cell
toxic fungal galectin CGL2. PLoS Pathog 6(1): 76(4):597–598
e1000717 10. Kuwabara N, Hu D, Tateno H et al (2013)
Conformational change of a unique sequence
in a fungal galectin from Agrocybe cylindracea phenomenon-based research tool for analyzing
controls glycan ligand-binding specificity. weak interactions between biomolecules.
FEBS Lett 587(22):3620–3625 Methods Mol Biol 2132:29–37
11. Nohr J, Kristiansen K (2003) Site-directed 17. Kasai K (2021) Frontal affinity chromatogra-
mutagenesis. Methods Mol Biol 232:127–131 phy: an excellent method of analyzing weak
12. Tateno H, Mori A, Uchiyama N et al (2008) biomolecular interactions based on a unique
Glycoconjugate microarray based on an principle. Biochim Biophys Acta Gen Subj
evanescent-field fluorescence-assisted detec- 1865(1):129761
tion principle for investigation of glycan- 18. Nakamura-Tsuruta S, Uchiyama N, Kominami
binding proteins. Glycobiology J et al (2007) Chapter 10 - Frontal affinity
18(10):789–798 chromatography: systematization for quantita-
13. Hirabayashi J, Hashidate T, Arata Y et al tive interaction analysis between lectins and
(2002) Oligosaccharide specificity of galectins: glycans/lectins. Elsevier Science B.V, pp
a search by frontal affinity chromatography. 239–266
Biochim Biophys Acta Gen Subj 1572 19. Tateno H, Nakamura-Tsuruta S, Hirabayashi J
(2–3):232–254 (2007) Frontal affinity chromatography:
14. Iwaki J, Tateno H, Nishi N et al (2011) The sugar–protein interactions. Nat Protoc
Galbeta-(syn)-gauche configuration is required 2(10):2529–2537
for galectin-recognition disaccharides. Biochim 20. Arata Y, Hirabayashi J, Kasai K (2001) Appli-
Biophys Acta 1810(7):643–651 cation of reinforced frontal affinity chromatog-
15. Iwaki J, Hirabayashi J (2018) Carbohydrate- raphy and advanced processing procedure to
binding specificity of human galectins: an over- the study of the binding property of a Caenor-
view by frontal affinity chromatography. habditis elegans galectin. J Chromatogr A 905
Trends Glycosci Glycotechnol 31(172): (1–2):337–343
SE137–SE153 21. Iwaki J, Hirabayashi J (2015) Evaluation of
16. Kasai K (2020) Frontal affinity chromatogra- galectin binding by frontal affinity chromatog-
phy: a highly suitable retardation raphy (FAC). Methods Mol Biol 1207:63–74
Chapter 15
What Happens If a Human Galectin Enters the Endoplasmic

Reticulum?
Tanja J. Kutzner, Alonso M. Higuero, Martina Süßmair, Michael Hingar,
Herbert Kaltner, Ingo Lindner, Jürgen Kopitz, José Abad-Rodrı́guez,
Dietmar Reusch, and Hans-Joachim Gabius
Abstract
Mammalian galectins have no signal peptide, and it is not known what would happen if a galectin is directed
to take the classical export route. The corresponding engineering of galectin-specific cDNA will answer
questions on the fate of a signal peptide-bearing protein variant after its entry into the endoplasmic
reticulum (ER). Affinity chromatography and mass-spectrometric analysis of occupancy of potential
N-glycosylation sites for the galectin, binding and functional assays with cells as well as subcellular
fractionation by density gradient ultracentrifugation and immunocytochemical colocalization with
ER/Golgi markers report on aspects of the consequences of letting a galectin enter new territory. Applying
these methods will help to clarify why galectins are leaderless and thus produced by free ribosomes.
Key words Autophagy, Calnexin, N-glycosylation, Routing, Secretion
1 Introduction
Natural molecular tagging of a protein by a signal peptide guarantees

efficient export of (glyco)proteins from cells via the classical secretory
pathway, i.e., along the endoplasmic reticulum (ER)/Golgi route.
Interestingly, presence of such a leader sequence is not obligatory to
enable a protein to eventually reach the cell surface: several categories
of proteins, among them annexins, certain chromatin-associated
proteins such as high mobility group box 1 (HMGB1), cytokines,
heat shock proteins, or thioredoxin find alternative ways to leave a
cell [1–3]. Their activity profiles can thus include both intra- and
extracellular functions, which gives a physiological meaning to their
synthesis on free ribosomes.
This potential for multifunctionality operative at different sites
within a cell, at its surface or in the extracellular space is also
characteristic for mammalian galectins that invariably are devoid
247
248 Tanja J. Kutzner et al.
of a signal peptide [4–9]. Their emerging capacity to engage in

specific contacts with glycans, versatile molecular signals for func-
tional pairing with lectins [10, 11], and with peptide motifs of
distinct proteins, e.g., involved in apoptosis (e.g., Bcl-2) or
immune-status regulation (e.g., CXCL12) and in lysosome repair
or removal by autophagy (e.g., Alix, NDP52, TAK1/AMPK, or
TRIM16) [12–17], appears to be ideal for multipurpose and -site
use in cell (patho)physiology. Intriguingly, analogies to serve as
alarmin have been delineated between galectins and several proteins
that share the mode of unconventional exit such as the mentioned
HMGB1 and the cytokine interleukin-1α [18–20]. Thus, the
steadily increasing knowledge on the diversity of galectin activities
gives reason to weave more and more elements into a compelling
whole, with the aim to explain why galectins are leaderless.
At the same time, these insights raise our curiosity to learn what
will happen, if a galectin is deliberately directed to enter the
ER. Will it acquire lectin activity in this oxidative environment?
Will a galectin be N-glycosylated (for an overview on common
presence of N-sequons in human galectins, see Table 1)? Will pres-
ence of an N-glycan then affect aspects of functionality? Will the
galectin travel along the entire classical route and, if that is the case,
then be secreted?
Answers to these questions will be gained by a strategic combi-
nation of molecular and cell biological experiments [21]. Engineer-
ing using galectin-specific cDNA and suited vectors to make this
type of tagging possible, expression of signal peptide-bearing galec-
tin in eukaryotic systems and (glyco)protein purification
(as galectin or as fusion protein), N-glycan detection and analysis,
cell binding (using labeled protein), and functional assays as well as
subcellular fractionation and marker-based colocalization are the
methods to be applied toward this end, as graphically illustrated in
overview in Fig. 1. Examples for representative data, for example,
presence of an N-glycan and sequon occupancy (Fig. 2) as well as
determination of the position of galectin in density gradient centri-
fugation (Fig. 3a) and in immunocytochemistry relative to markers
(Fig. 3b), are given. With perspective for testing practical applica-
tions, initial data presented previously [21] point to a potential of a
glycosylated galectin (or a monoPEGylated mutant [22]) to
become a functional antagonist in certain circumstances and to
make it possible to segregate aspects of intra- from extracellular
activities by expressing a galectin exclusively with a signal peptide.
2 Materials
2.1 Ligation 1. Photometer (NanoPhotometer® P-Class P 300; Implen,

Munich, Germany).
What Happens If a Human Galectin Enters the Endoplasmic Reticulum? 249
Table 1
Presence of the sequon identifying potential sites for N-glycosylation in human galectins
Sequon
Galectin Position Sequence Spatial position*

Gal-1 95 NLT F41
Gal-2 123 NMS F11
Gal-3 3 NFS NTS
Gal-4 242 NGT L (S4/S5)
252 NGS L (S5/S6a)
Gal-7 28 NAS L (F2/S3)
Gal-8 51 NGS S31
Gal-8S/L 254/296 NIT L (S6b/F3)
Gal-8 L 208 NHT Linker
Gal-9 33 NGT F2
78 NGS L (S5/S6a)
136 NGS S21
Gal-10 135 NVS F11
GRP 19 NNS NTE
Gal-1 (PDB ID: 1GZW); Gal-2 (PDB ID: 5EWS); Gal-4 (PDB ID: 5CBL); Gal-7 (PDB ID: 3ZXF); Gal-8N (PDB ID:
4BMB); Gal-8C (PDB ID: 4GXL); Gal-9 (PDB ID: 3LSD); Gal-10 (PDB ID: 5XRG); GRP (galectin-related protein);
*S/F: β-strand, L: loop; 1loop start within two amino acids; NTS: N-terminal segment; NTE: N-terminal extension
(reprinted from [21], with permission from Elsevier, Copyright 2019)
2. Alkaline phosphatase and buffer (rAPid Alkaline Phosphatase;

Roche Diagnostics, Mannheim, Germany).
3. Thermomixer comfort (Eppendorf, Hamburg, Germany).
4. T4 DNA ligase and 10 buffer (Promega, Walldorf,
Germany).
2.2 Transformation 1. Chemically competent E. coli cells (One Shot® TOP10; Invi-
of Chemically trogen, Dreieich, Germany).
Competent E. coli One 2. Thermomixer comfort (Eppendorf).
Shot® TOP10 Cells
3. LB medium (Roth, Karlsruhe, Germany).
4. Microcentrifuge (Model 5415D; Eppendorf).
5. LB agar plates (Roth).
6. Incubator (Type BM100; Memmert, Schwabach, Germany).
7. Orbital shaker with incubation hood (KS-15, TH-15; Edmund
Bühler, Bodelshausen, Germany).
Fig. 1 Flow diagram illustrating the analytical as well as molecular and cell biological experiments to answer
the question on what happens to a galectin when directed to start the ER/Golgi route. The schematic
8. Plasmid preparation kit (Invisorb® Spin Plasmid Mini Two;

Invitek Molecular, Berlin, Germany).
9. Antibiotics (e.g., ampicillin, final concentration: 50 μg/mL or
zeocin, final concentration: 25 μg/mL).
2.3 Cloning of 1. Galectin-coding cDNA of interest.

Galectin-Coding cDNA 2. Phusion® High-Fidelity DNA polymerase and 5 buffer (New
into a Pichia England Biolabs, Frankfurt, Germany).
Expression Vector with
3. dNTPs (10 mM; Promega).
Signal Sequence
(pPICZα A)
4. Primer set containing a forward primer with a XhoI restriction
site, a reverse primer with a XbaI restriction site, each with an
overhang of at least 4 bp at the 50 -ends (Metabion, Planegg,
Germany).
5. Thermocycler (Eppendorf).
6. LE Agarose (EEO 0.09–0.13; Biozym Scientific, Hessisch Ol-
dendorf, Germany).
7. TAE buffer (5 mM ethylenediaminetetraacetic acid (EDTA),
40 mM Tris, 20 mM acetic acid, pH 8.5).
8. Gel documentation system (ChemiDoc™ Touch Imaging Sys-
tem; Bio-Rad, Munich, Germany).
9. DNA extraction kit (Invisorb® Fragment CleanUp; Invitek
Molecular).
10. XhoI, XbaI, and provided 10 buffer (all reagents supplied by
Thermo Fisher Scientific, Dreieich, Germany).
11. Pichia expression vector (pPICZα A; Invitrogen).
12. Thermomixer comfort (Eppendorf).
2.4 Modified 1. NetNGlyc 1.0 Server Software (http://www.cbs.dtu.dk/

QuikChange™ Site- services/NetNGlyc/).
Directed Mutagenesis 2. Mutagenesis primer (Metabion).
Protocol
3. PfuTurbo DNA polymerase and 10 buffer (Agilent Technol-
ogies, Waldbronn, Germany).
5. Nuclease-free water.
Fig. 1 (continued) representation of the sequential processing steps starts with cloning and mutagenesis of
the galectin-coding cDNA of interest and transfection of HEK293T cells/electroporation of Pichia (top).
Subsequently, detection in subcellular membrane vesicles/HEK293T cells (middle left) or expression and
purification of the galectin of interest (middle right) are performed. The recombinant galectin of interest is
analyzed by mass spectrometry and functional assays (bottom half) (PDB ID: 1GZW, human galectin-1; parts of
this figure are reprinted from [21], with permission from Elsevier, Copyright 2019)
Fig. 2 LC-ESI MS-MS extracted ion current chromatogram of wild-type human Gal-1 with signal peptide
expressed in P. pastoris after PNGase F treatment. Hydrolytic removal of N-glycan results in N-to-D conversion
with a retention time of 33.274 min for the D-containing peptide compared to the retention time of 32.148 min
for the non-glycosylated N-containing peptide. The sequon occupancy is at a ratio of 85%:15% of the N95
position (for details, see [21]; reprinted with permission from Elsevier, Copyright 2019)
6. Bacterial expression vector containing the galectin-coding

cDNA of interest.
8. DpnI (Thermo Fisher Scientific).
2.5 Electroporation 1. LB medium (Roth) containing zeocin (25 μg/mL) (Invivo-

of Pichia Strain X-33 Gen, San Diego, USA).
2.5.1 Preparation of 2. Chemically competent E. coli cells (One Shot® TOP10; Invi-
Plasmid DNA for trogen) transformed with Pichia expression vector (pPICZα;
Electroporation Invitrogen) containing the respective galectin-coding cDNA
(prepared in Subheading 3.3 or 3.4).
Bühler).
Invitek Molecular).
5. SacI, 10 buffer, and 100 BSA solution (all reagents supplied
by Promega).
Molecular).
8. Vacuum concentrator (Type BaVaCo Mini-30; Bachofer, Reu-
tlingen, Germany).
2.5.2 Preparation of 1. Yeast extract peptone dextrose medium (YPD, composed of 1%

Electrocompetent Pichia X- (w/v) yeast extract (AppliChem, Darmstadt, Germany), 2%
33 Cells (w/v) peptone, 2% (w/v) dextrose, and 100 μg/mL ampicillin
(Roth)).
Fig. 3 Subcellular localization of human Gal-1 without or with signal peptide (sp). (a) Fractionation of
membrane vesicles of postnuclear extracts from HEK293T cells expressing wild-type Gal-1 or Gal-1 contain-
ing a signal peptide (sp-Gal-1; visible as two bands, which represent glycosylated (higher molecular weight
band) and non-glycosylated protein) in density gradients from 10% to 30% iodixanol solution and then
analyzed by Western blotting. Levels of Gal-1 cofractionation with markers for light membranes (early
endosomes A1, EEA1), the cis-Golgi region (Golgi marker 130, GM130), and the ER (calnexin, Cnx) were
evaluated (reprinted from [21], with permission from Elsevier, Copyright 2019). (b) Immunocytochemical
colocalization profiles in HEK293T cells (shown as single confocal z-planes) of GM130 (green; top, left) and
calnexin (green; bottom, left) with tagged sp-Gal-1 (red; central column). Merged channels are shown in the
right column, where occurrence of yellow color indicates extent of overlap. White arrowheads draw attention
to the scarce extent of overlap between Gal-1 and GM130, compared to the sp-Gal-1/calnexin case with
marked colocalization highlighted by white arrows. Scale bar: 10 μm
2. Pichia strain X-33 (Invitrogen).

3. Sterile 50-mL centrifuge tubes.
4. Sterile gauze (Hartmann, Heidenheim, Germany).
Bühler).
6. Photometer (NanoPhotometer® P-Class P 300; Implen).
7. Baffled culture flask, 1 L.
8. Cellulose plug (Roth).
9. Refrigerated centrifuge (Model 5804R; Eppendorf).
10. Sterile ice-cold water.
11. Sterile ice-cold 1 M sorbitol solution (Roth).
2.5.3 Electroporation 1. Linearized Pichia expression vector containing the respective

Procedure of Pichia X-33 galectin-coding cDNA (see Subheading 3.5.1).
Cells 2. Electrocompetent Pichia X-33 cells (see Subheading 3.5.2).
3. Electroporation cuvette (2 mm; Biozym Scientific).
4. Electroporator (Gene Pulser® II; Bio-Rad).
5. Sterile ice-cold 1 M sorbitol solution (Roth).
7. Sterile gauze (Hartmann).
9. Yeast extract peptone dextrose sorbitol (YPDS) agar plates
(composed of 1% (w/v) yeast extract (AppliChem), 2% (w/v)
peptone, 2% (w/v) dextrose, 1 M sorbitol, 2% (w/v) agar)
containing zeocin (100 μg/mL; InvivoGen) and 100 μg/mL
ampicillin (Roth)).
10. Incubator (Type INE200; Memmert).
2.6 Analysis of 1. Pichia X-33 colonies transformed with Pichia expression vector
Colonies (pPICZα A; Invitrogen) containing the cDNA coding for the
galectin of interest (prepared in Subheading 3.5).
2. Minimal agar plates (1.34% (w/v) yeast nitrogen base (Invitro-
gen), 4 105% (w/v) biotin, 2% (w/v) agar) containing 2%
(w/v) dextrose or 0.5% (v/v) methanol.
3. YPD agar plates (composed of 1% (w/v) yeast extract (Appli-
Chem), 2% (w/v) peptone, 2% (w/v) dextrose, 2% (w/v) agar)
containing 100 μg/mL, 500 μg/mL or 1000 μg/mL zeocin
(InvivoGen).
4. Incubator (Type INE200; Memmert).
Bühler).
6. YPD (see Subheading 2.5.2, item 1) containing 100 μg/mL
zeocin (InvivoGen).
10. 50 mM EDTA, pH 8.0.
11. Lyticase from Arthrobacter luteus (Sigma-Aldrich, Darmstadt,
Germany).
13. Wizard® genomic DNA purification kit (Promega).
14. Isopropanol (Roth).
15. 70% (v/v) Ethanol (Roth).
16. TopTaq™ DNA polymerase and 10 buffer (including MgCl2

and Q solutions as provided by the manufacturer; Qiagen,
Hilden, Germany) (see Note 1).
18. Primer set of 50 AOX (50 -GACTGGTTCCAATTGACAAGC-
30 ) and 30 AOX (50 - GCAAATGGCATTCTGACATCC -30 )
(Metabion).
20. LE Agarose (EEO 0.09–0.13; Biozym Scientific).
21. TAE buffer (see Subheading 2.3, item 7).
tem; Bio-Rad).
2.7 Expression of 1. Pichia X-33 (Mut+ phenotype) colonies transformed with

Galectin in P. pastoris Pichia expression vector (pPICZα A; Invitrogen) containing
the galectin-coding cDNA of interest (see Subheading 3.6).
2. Buffered complex medium containing glycerol (BMGY, com-
posed of 1% (w/v) yeast extract (AppliChem), 2% (w/v) pep-
tone, 100 mM potassium phosphate, pH 6.0, 1.34% (w/v)
yeast nitrogen base (Invitrogen), 4 105% (w/v) biotin, 1%
(v/v) glycerol, and 100 μg/mL ampicillin (Roth)).
3. Baffled flasks, 250 mL and 1000 mL.
Bühler).
6. Centrifuge (Model 5804R; Eppendorf).
7. Buffered complex medium containing methanol (BMMY,
composed of 1% (w/v) yeast extract (AppliChem), 2% (w/v)
peptone, 100 mM potassium phosphate, pH 6.0, 1.34% (w/v)
yeast nitrogen base (Invitrogen), 4 105% (w/v) biotin,
0.5%–2% (v/v) methanol and 100 μg/mL ampicillin (Roth)).
8. Cellulose plug (Roth).
9. Filter-sterilized 125 mM dithiothreitol (DTT; Roth).
10. Filter-sterilized 100% methanol.
12. Refrigerated centrifuge (Avanti™ J-25; Beckman Coulter, Kre-
feld, Germany).
2.8 Purification of 1. Lactosylated Sepharose 4B (home-made after activation with

Galectin from P. divinyl sulfone [23]).
pastoris Supernatant 2. 20 mM phosphate-buffered saline (PBS: 20 mM KH2PO4/
by Affinity Na2HPO4, 150 mM NaCl, pH 7.2) containing 4 mM
Chromatography on β-mercaptoethanol and 2 mM EDTA.
Lactosylated 3. Supernatant of P. pastoris cell culture (see Subheading 3.7).
Sepharose 4B Resin
4. β-Mercaptoethanol (Roth).
5. Roller mixer.
6. Chromatography column (1–2 mL).
7. PBS (pH 7.2).
8. PBS (pH 7.2) containing 100 mM lactose and 20 mM iodoa-
cetamide (Sigma-Aldrich).
9. Protein assay dye reagent concentrate for Bradford assay
(Bio-Rad).
10. PD-10 desalting column (GE Healthcare, Munich, Germany).
11. 20 C freezer.
12. Freeze dryer for lyophilization.
2.9 Electrophoresis 1. Protein sample.

and Transfer 2. Standard proteins of known molecular weight such as
β-galactosidase (116 kDa), bovine serum albumin (66 kDa),
carbonic anhydrase (45 kDa), β-lactoglobulin (18.4 kDa), lyso-
zyme (14.2 kDa) (all proteins supplied by Sigma-Aldrich).
3. Protein sample loading buffer: 0.25 M Tris–HCl pH 6.8
(Roth), 10% (w/v) sodium dodecyl sulfate (SDS; Roth), 50%
(w/v) sucrose (Roth), 20% (v/v) β-mercaptoethanol (Roth),
0.1% (w/v) bromophenol blue (Serva, Heidelberg, Germany).
4. Thermoblock (Test Tube Thermostat TCR 100; Roth).
5. Microcentrifuge (Biofuge 13; Heraeus, Hanau, Germany).
6. SDS polyacrylamide gel: 15% separating gel (4.5 mL acrylam-
ide (37.5:1 acrylamide/bis-acrylamide solution; Bio-Rad),
3 mL distilled water, 2.5 mL 1.5 M Tris–HCl pH 8.8 (Roth)
with 0.4% (w/v) SDS (Roth), 0.1 mL 10% (w/v) ammonium
persulfate (Roth), 0.01 mL N,N,N0 ,N0 -tetramethylethylene-
diamine (Roth)) and 4% stacking gel (0.65 mL acrylamide
(37.5:1 acrylamide/bis-acrylamide solution; Bio-Rad),
3.1 mL distilled water, 1.25 mL 0.5 M Tris–HCl pH 6.8
(Roth) with 0.4% (w/v) SDS (Roth), 0.06 mL 10% (w/v)
ammonium persulfate (Roth), 0.01 mL N,N,N0 ,N0 -tetra-
methylethylenediamine (Roth)).
7. Gel running buffer: 25 mM Tris (Roth), 192 mM glycine
(Roth), resulting in a pH of 8.6.
8. Electrophoresis and blotting module (Mini-PROTEAN® Tetra
Vertical Electrophoresis Cell, Mini Trans-Blot® Cell; Bio-Rad).
9. Power supply (Power Pac™ Basic; Bio-Rad).
10. Transfer buffer: 25 mM Tris (Roth), 192 mM glycine (Roth),
20% (v/v) methanol, resulting in a pH of 8.3 (see Note 2).
11. Blotting paper (Rotilabo® Blotting paper, 0.35 mm; Roth).
12. Blotting fiber pads (Bio-Rad).

13. Nitrocellulose membrane (Roti® NC, 2.0 μm; Roth).
14. Magnetic stirring bars and stirrer.
15. Tris-buffered saline with Tween® 20 (TBS-T): 50 mM Tris
(Roth), 150 mM NaCl (Roth) adjust to pH 7.5 with 12 M
HCl, 0.1% (v/v) Tween® 20 (Roth).
16. Ponceau S solution: 0.1% (w/v) Ponceau S (Merck, Darm-
stadt, Germany) in 1% (v/v) acetic acid.
2.10 Western 1. Nitrocellulose membrane (see Subheading 3.9).

Blotting 2. TBS-T (see Subheading 2.9, item 15).
3. Blocking solution: 5% (w/v) milk powder (Roth) in TBS-T.
4. Rabbit non-cross-reactive (among galectins) polyclonal anti-
galectin immunoglobulin G (IgG) (home-made, [24, 25]).
5. PBS/BSA: PBS containing 1% (w/v) bovine serum albumin
(BSA).
6. Goat polyclonal anti-rabbit IgG conjugated with horseradish
peroxidase (Sigma-Aldrich).
7. Enhanced chemiluminescence (ECL) solution: 2 mL 0.1 M
Tris–HCl, pH 8.6 with 1.41 mM luminol (Sigma-Aldrich),
0.2 mL 6.7 mM p-coumaric acid (Sigma-Aldrich) in dimethyl
sulfoxide and 0.6 μL hydrogen peroxide.
tem; Bio-Rad).
2.11 Lectin Blotting 1. Sodium bicarbonate buffer (0.1 M NaHCO3 buffer, pH 8.0,
containing 150 mM NaCl and 20 mM D-mannose to protect
the carbohydrate recognition site).
2. Lyophilized concanavalin A (Vector Laboratories, Burlingame,
CA).
3. N-Succinimidyl ester derivative of biotin (biotin NHS ester;
Sigma-Aldrich).
4. Dimethylformamide (DMF; Roth).
5. Nitrocellulose membrane (see Subheading 3.9).
6. Tris-buffered saline with Tween® 20 (TBS-T): 50 mM Tris
(Roth), 150 mM NaCl (Roth) adjust to pH 7.5 with 12 M
HCl, 0.1% or 0.5% (v/v) Tween® 20 (Roth).
7. Blocking solution: 3% (w/v) BSA in TBS-T (0.1% (v/v)
Tween® 20).
8. Blocking solution containing 1 mM CaCl2 and 0.1 mM
MnCl2.
9. Blocking solution containing 1 mM CaCl2, 0.1 mM MnCl2,

0.2 M D-mannose, and 0.02 M EDTA (pH 8.0).
10. Peroxidase-conjugated avidin (Sigma-Aldrich or home-made).
11. Enhanced chemiluminescence (ECL) solution (see Subheading
2.10, item 7).
tem; Bio-Rad).
2.12 Sample 1. Digestion buffer: 10 mM sodium phosphate buffer, pH 7.2.

Preparation of 2. Thermomixer.
Galectins for
Electrospray Mass
Spectrometry (ESMS)
Analysis
2.12.1 Sample
Preparation for Analysis of
Protein
2.12.2 Sample 1. Denaturation buffer: 0.4 M Tris–HCl, 6.0 M guanidinium

Preparation with Prior chloride, pH 8.5.
Protein Deglycosylation 2. 100 mM Tris-(2-carboxyethyl)-phosphin (TCEP) in water.
3. Digestion buffer: 10 mM sodium phosphate buffer, pH 7.2.
4. N-Glycosidase F (100 units/0.1 mL; Roche).
5. Micro Bio-Spin™ P-6 Gel Columns, Tris buffer (Bio-Rad).
6. Thermomixer.
7. Microcentrifuge.
2.13 ESMS Analysis 1. Eluent A: 0.1% (v/v) formic acid in water.

of Galectin 2. Eluent B: 0.1% (v/v) formic acid in acetonitrile.
3. ACQUITY UPLC BEH C4 column (300 Å, 1.7 μm,
2.2 50 mm; Waters Corporation, Milford, USA).
4. ACQUITY UPLC System (Waters Corporation).
5. Synapt G2 HDMS mass spectrometer (equipped with a Lock-
Spray ESI source; Waters Corporation).
6. MassAnalyzer software (Roche).
2.14 Sample 1. Rapid PNGase F (New England Biolabs).

Preparation for LC/ 2. Rapid PNGase F Buffer (5) (New England Biolabs).
MSMS Analysis of
3. Thermomixer.
Galectin
2.14.1 Deglycosylation
2.14.2 Tryptic Digestion 1. Denaturation buffer: 0.4 M Tris–HCl, 6.0 M guanidinium

chloride, pH 8.5.
2. 100 mg/mL DTT solution.
3. 111.6 mg/mL iodoacetamide solution.
4. Digestion buffer: Tris–HCl, pH 7.0.
5. Zeba™ Spin Desalting Columns (7 k MWCO, 0.5 mL;
Thermo Fisher Scientific).
6. Trypsin solution: 0.25 μg/μL, sequencing grade (Roche).
7. 10% (v/v) Trifluoroacetic acid (TFA) solution.
8. Thermomixer.
9. Microcentrifuge.
2.15 LC/MSMS 1. Eluent A: 0.1% (v/v) formic acid in water.

Analysis of Galectin 2. Eluent B: 0.1% (v/v) formic acid in acetonitrile.
3. ACQUITY UPLC CSH C18 column (130 Å, 1.7 μm,
2.1 150 mm; Waters Corporation).
4. VANQUISH UHPLC system (Thermo Fisher Scientific).
5. Triple TOF 6600 mass spectrometer with an ESI source (Sciex,
Framingham, USA).
6. Software Peak View 2.0 with Master View add-on (Sciex).
7. Software Protein Pilot 5.0 (Sciex).
2.16 Radioiodination 1. Pierce Iodination Beads (Iodobeads™; Thermo Fisher

of Galectin Scientific).
2. Sodium [125I]iodide (>74 TBq/mmol) (Hartmann Analytic,
Braunschweig, Germany).
3. 50 mM sodium phosphate buffer (SPB), pH 7.0: 0.7744 g
Na2HPO4 7 H2O, 0.2913 g NaH2PO4 H2O (BioXtra
>99%; Sigma-Aldrich) in 100 mL H2O.
4. 1 M D-Lactose (Sigma-Aldrich) in SPB.
5. Phosphate-buffered saline (PBS) (137 mM NaCl, 2.7 mM
KCl, 10 mM Na2HPO4, 1.8 mM KH2PO4, pH 7.4).
6. Disposable PD-10 Desalting Columns (GE Healthcare).
7. 1 mg BSA (Sigma-Aldrich)/mL PBS.
8. Ultima Gold™ liquid scintillation cocktail (PerkinElmer®).
9. Liquid scintillation counter Tricarb 2900TR (PerkinElmer®).
10. NanoDrop 1000 Spectrophotometer (Thermo Fisher
Scientific).
2.17 Cell Culture and 1. Human SK-N-MC neuroblastoma cells obtained from the
Cell Binding Studies American Type Culture Collection (ATCC HTB-19).
2. 96-Well cell culture plates (Falcon Primaria™; Falcon, Heidel-
berg, Germany).
3. Eagle’s Minimum Essential Medium (MEM; Thermo Fisher
Scientific).
4. Fetal bovine serum (FBS; Thermo Fisher Scientific).
5. Penicillin–streptomycin (10,000 U/mL; Thermo Fisher
Scientific).
6. 25 mM HEPES (Sigma-Aldrich), pH 7.4, and 0.01 mg/mL
BSA (Sigma-Aldrich).
7. 150 mM D-Lactose (Sigma-Aldrich) and 0.5 mg/mL asialofe-
tuin (home-made) in MEM, 25 mM HEPES, and 0.01 mg/
mL BSA.
125
8. I-Labeled galectins, specific radioactivities 200–300 kBq/μg
protein (prepared as described in Subheading 3.16).
9. Phosphate-buffered saline (PBS) (137 mM NaCl, 2.7 mM
10. 0.2 M NaOH.
11. Ultima Gold™ liquid scintillation cocktail (PerkinElmer®).
12. Scintillation vials (Maxi Vial 18 mL; PerkinElmer®).
13. Liquid scintillation counter (Tricarb 2900TR; PerkinElmer®).
2.18 Proliferation 1. Human SK-N-MC neuroblastoma cells obtained from the

Assay American Type Culture Collection (ATCC HTB-19).
2. 96-Well cell culture plates (Falcon Primaria™; Falcon).
3. Eagle’s Minimum Essential Medium (MEM; Thermo Fisher
Scientific).
4. FBS (Thermo Fisher Scientific).
5. Penicillin–streptomycin (10,000 U/mL; Thermo Fisher
Scientific).
6. Recombinant galectins.
7. CellTiter 96™ dye solution (Promega).
8. CellTiter96™ solubilization/stop solution (Promega).
9. Microplate reader (GENios™; Tecan, Crailsheim, Germany).
1. Galectin-coding cDNA of interest.

2. Phusion® High-Fidelity DNA polymerase and 5 buffer (New
England Biolabs).
2.19 Cloning of 4. Primer set containing a forward and a reverse primer, each
Galectin-Coding cDNA extended by an appropriate restriction site and an overhang of
into Mammalian at least 4 bp at the 50 -ends (Metabion).
Expression Vectors 5. Thermocycler (Eppendorf).
(pSecTag2 B, 6. LE Agarose (EEO 0.09–0.13; Biozym Scientific).
pcDNA3.1+/C-HA) with
7. TAE buffer (see Subheading 2.3, item 7).
Signal Peptide
tem; Bio-Rad).
Molecular).
10. pSecTag2 B mammalian expression vector (Invitrogen).
11. Restriction enzymes including provided 10 buffers (Thermo
Fisher Scientific).
12. pcDNA3.1+/C-HA mammalian expression vector
(Invitrogen).
2.20 Culture of 1. Cell culture vessels (Petri dishes and multiwell plates).
Human Embryonic 2. Standard 90 mm bacteriological petri dishes with lid (Thermo
Kidney 293T Cells Fisher Scientific).
(HEK293T)
3. Acid-washed glass cover slips (#1.5, 170 μm in thickness)
2.20.1 Preparation of (Marienfeld Superior, Lauda-Königshofen, Germany).
Plastic and Glass Substrata 4. Small forceps (e.g., Dumont n 5; Fine Science Tools, Heidel-
for Cell Culture berg, Germany).
5. Poly-L-lysine (30,000–70,000 mw) (1 mg/mL in 0.1 M borate
buffer, pH 8.5).
6. Sterile tissue culture grade water (Sigma Aldrich).
2.20.2 Cell Cultures for 1. HEK293T human embryonic kidney cell line (ATCC/CRL-
Biochemical Analysis or for 3216).
Immunocytochemical 2. Dulbecco’s Modified Eagle Medium (DMEM), high glucose
Staining (4.5 g/L glucose).
3. FBS (Thermo Fisher Scientific).
4. L-Glutamine (200 mM).
5. Penicillin/streptomycin (P/S) (10,000 U/mL).
6. Trypsin–EDTA solution (0.25% (w/v)).
7. Opti-MEM I reduced serum medium (GIBCO® Thermo
Fisher Scientific).
Invitek Molecular).
9. Purified galectin-expressing mammalian expression vector (see

Note 3).
10. Lipofectamine LTX with Plus Reagent (Invitrogen).
11. 15-mL and 50-mL conical sterile centrifuge tubes.
12. Cell counter.
13. Water bath.
14. Standard table-top centrifuge.
15. Biosafety cell culture cabinet.
16. Incubator for cell culture.
17. Inverted light microscope.
2.21 Density 1. PBS (pH 7.4).

Gradient Fractionation 2. HEPES buffer (25 mM HEPES, pH 7.4, 2 mM EGTA,
of Subcellular 150 mM NaCl, 1 mM DTT) supplemented with a protease
Membrane Vesicles inhibitor cocktail (Sigma-Aldrich).
2.21.1 Preparation of 3. Plastic syringes (1 or 2 mL) and needles (22 gauge).
Total Membrane Extracts 4. Cell scrapers.
from HEK293T Cells for
5. Cold plate or flat tray with crushed ice.
Subcellular Fractionation
6. Refrigerated table-top centrifuge.
7. Table-top ultracentrifuge with rotor for small volumes (e.g.,
Optima™ MAX-XP Benchtop Ultracentrifuge with TL-110
rotor; Beckman Coulter).
8. Thick-wall ultracentrifuge tubes for small volumes (suggested
volume at 3.2 mL, Open-Top Thickwall Polypropylene Tube,
13 56 mm; Beckman Coulter).
2.21.2 Preparation of 1. Iodixanol concentrated solution (OptiPrep™; Sigma-Aldrich).

Density Gradients, 2. HEPES buffer (25 mM).
Membrane Fractionation,
3. Ultracentrifuge with swinging-bucket rotor (suggested SW40
and Fraction Protein
or SW41 rotors; Beckman Coulter).
Precipitation
4. Ultracentrifuge tubes (suggested volume at 14 mL, Open-Top
Thinwall Ultra-Clear or Polypropylene Tubes, 14 95 mm;
Beckman Coulter).
5. High density foam (2- to 3-mm-thick slabs) (see Subheading
3.21.2, step 1 and Note 4).
6. Puncher (10 mm diameter).
2.21.3 Precipitation of 1. Methanol.

Proteins from Fractions 2. Chloroform.
from the Density Gradient
3. Table-top refrigerated centrifuge.
2.21.4 Analysis of 1. Precipitated proteins from density gradient fractionation (see

Fractions by Western Subheading 3.21.3).
Blotting 2. Laemmli buffer (60 mM Tris–HCl pH 6.8, 2% (w/v) SDS, 10%
(v/v) glycerol, 5% (v/v) β-mercaptoethanol, 0.01% (w/v)
bromophenol blue).
3. Running buffer (25 mM Tris–HCl, 200 mM glycine, 0.1%
(w/v) SDS; pH 8.3).
4. Transfer buffer (20% (v/v) methanol in running buffer).
5. 8% and 15% Tris–glycine polyacrylamide gels, 0.75 mm thick
(Bio-Rad).
6. Power supply, vertical polyacrylamide gel running system,
(suggested Bio-Rad minigel system), and wet electrotransfer
system (Mini Trans-Blot Cell; Bio-Rad).
7. Nitrocellulose membrane (0.2 μm pore size; GE Healthcare).
8. TBS-T (see Subheading 2.9, item 15).
9. Blocking solution (5% (w/v) skimmed milk powder in TBS-T).
10. Primary antibodies: monoclonal anti-EEA1, anti-GM130, and
anti-calnexin antibodies (all from BD Biosciences, Heidelberg,
Germany), rabbit polyclonal anti-galectin-1 (home-made), and
rat monoclonal anti-HA antibody (Roche).
11. Second-step antibodies: anti-species (mouse, rat)-specific anti-
bodies conjugated with horseradish peroxidase (Invitrogen).
12. Chemiluminescence detection system (Western Lightning™
plus-ECL; PerkinElmer®, Rodgau, Germany).
13. Chemiluminescence reader.
2.22 Immunocyto- 1. Gal-1-transfected HEK293T cells.

chemistry (ICC) 2. Phosphate-buffered saline (PBS) (137 mM NaCl, 2.7 mM
3. 4% (w/v) paraformaldehyde in PBS, pH 7.4 (freshly prepared).
4. Quenching solution (50 mM NH4Cl in PBS).
5. 0.1% (v/v) Triton X-100 in PBS.
6. Antigen retrieval solution (6 M urea in 0.1 M Tris–HCl,
pH 9.0).
7. Blocking solution (2% (v/v) FCS, 2% (w/v) BSA, and 0.2%
(w/v) gelatin in PBS).
8. Primary antibodies (diluted in blocking solution): monoclonal
anti-GM130 and anti-calnexin antibodies (BD Biosciences,
Heidelberg, Germany), rabbit polyclonal anti-galectin-1
[24, 25] and rat monoclonal anti-HA antibody (Roche).
9. Fluorescent second-step antibody diluted in blocking solution.
10. Mounting media with an antifading reagent (Vectashield; Vec-

tor Laboratories).
11. Confocal microscope (sp5; Leica Microsystems, Wetzlar,
Germany).
3 Methods
3.1 Ligation 1. Determine the concentration (OD260) and purity (ratio

OD260/OD280) of the DNA samples for ligation.
2. Add 1 unit of alkaline phosphatase and 2 μL of buffer to 17 μL
of a solution of the linearized vector DNA and incubate the
mixture at 37 C for 10 min. Stop the enzyme reaction by an
incubation step at 75 C for 2 min.
3. Mix solutions of 50 ng vector DNA from step 2 with the
digested PCR product in a molar ratio of 1:3, add 2 units of
T4 DNA ligase, 1 μL ligase buffer and distilled water up to
10 μL.
4. Incubate the reaction mixture at 25 C for 3 h, then store at
20 C.
3.2 Transformation 1. Transfer 5 μL of the reaction mixture from ligation or muta-

of Chemically genesis to a suspension of 50 μL of competent E. coli cells and
Competent E. coli One incubate on ice for 30 min.
Shot® TOP10 Cells 2. Incubate the cells at 42 C for 50 s and cool on ice.
3. Add 600 μL of LB medium and incubate at 37 C for 1 h, while
shaking gently (350 rpm).
4. Centrifugate for 1 min at 16,000 g, discard the supernatant,
resuspend the cell pellet in 100 μL of fresh medium, spread the
suspension on a LB agar plate containing the appropriate anti-
biotics and incubate at 37 C overnight.
5. Inoculate 4 mL of LB medium containing the appropriate
antibiotics with a single colony and grow bacteria in medium
under gentle shaking (180 rpm) in an orbital shaker at 37 C
overnight.
6. Prepare plasmid DNA using a plasmid preparation kit and
verify correct insertion of cDNA by enzymatic treatment with
restriction endonuclease and sequencing of inserted cDNA.
3.3 Cloning of 1. Amplify the galectin-coding cDNA of interest (bacterial

Galectin-Coding cDNA expression vector as template) in the setup as follows: Phusion®
into a Pichia High-Fidelity DNA polymerase, polymerase buffer (5),
Expression Vector with dNTPs, an oligonucleotide-sense primer flanked with the rec-
Signal Sequence ognition sequence for XhoI and an anti-sense primer with the
(pPICZα A) recognition sequence for XbaI (see Note 5). Thermocycler
program: initial step of denaturation at 98 C for 30 s, then

30 cycles each consisting of a denaturation step at 98 C for
10 s, an annealing step at 50–65 C for 25 s, an elongation step
at 72 C for 25 s, finally an elongation step at 72 C for 10 min.
2. Purify the amplification products from agarose gel slices (visua-
lized with a gel documentation system) using a DNA
extraction kit.
3. Add the required amount of restriction enzymes XhoI and
XbaI as well as respective buffer to DNA-containing solutions.
Incubate at 37 C overnight. 10 Units (20 units) XhoI and
20 units (40 units) XbaI for the amplified DNA products
(pPICZα A vector DNA) were applied.
4. Purify linearized vector DNA from agarose gel slices or the
PCR product directly from the restriction digestion mixture
using a DNA extraction kit.
5. Perform ligation (see Subheading 3.1) in-frame with the
α-factor signal sequence of the Pichia expression vector
(pPICZα A).
6. Transformation of competent E. coli cells as described in
Subheading 3.2.
3.4 Modified 1. Find sequons as potential sites for N-glycosylation using the
QuikChange™ Site- NetNGlyc 1.0 Server software to generate loss-of-function
Directed Mutagenesis mutants or to identify appropriate positions to establish new
Protocol sequon(s) (putative gain-of-function mutants) by nucleotide
exchange(s) (see Table 1 for occurrence of the acceptor
sequence in human galectins).
2. According to the modified QuikChange™ site-directed muta-
genesis protocol [26], design a pair of DNA oligonucleotides
with the following properties: 25–45 bp, Tm 78 C, both
DNA oligonucleotides (sense and anti-sense) must contain the
desired mutation at the same position (see Note 6).
3. Perform a two-step PCR, the first step consists of two reaction
mixtures with 5 μL of polymerase buffer (10), 1 μL of 10 mM
dNTPs, 1 μL of template (50–200 ng bacterial expression
vector DNA with inserted galectin-coding cDNA which should
be changed by mutations), 1 μL PfuTurbo DNA polymerase
and 37 μL of nuclease-free water each. One reaction mixture
contains 5 μL of 10 μM forward primer, while the other reac-
tion mixture contains 5 μL of 10 μM reverse primer. Thermo-
cycler program: initial step of denaturation at 95 C for 30 s,
then three cycles each consisting of a denaturation step at 95 C
for 30 s, an annealing step at 55 C for 1 min, and an elonga-
tion step at 68 C for 8 min. In the second step, mix 25 μL of
each reaction of step 1, add 1 μL of PfuTurbo DNA polymerase
and perform the PCR program: initial step of denaturation at
95 C for 30 s, then 18 cycles each consisting of a denaturation

step of 95 C for 30 s, an annealing step at 55 C for 1 min, and
an elongation step at 68 C for 5 min (see Note 7) [26].
4. Add 10 units DpnI and incubate at 37 C for at least 2 h.
5. Inactivate DpnI at 80 C for 20 min, cool solution to RT
followed by transformation of competent E. coli cells (see
Subheading 3.2).
6. For cloning in-frame with the α-factor sequence (Pichia expres-
sion vector pPICZα A) amplify the mutated galectin-coding
cDNA as described in Subheading 3.3 (see Note 8).
3.5 Electroporation 1. Inoculate 50 mL of LB medium containing 25 μg/mL zeocin

of Pichia Strain X-33 with a single colony of E. coli cells, which are transformed with
the Pichia expression vector containing the inserted cDNA
3.5.1 Preparation of
coding for the galectin of interest with or without mutation
Plasmid DNA for
(generated as outlined in Subheading 3.3 or 3.4). Incubate on
Electroporation
an orbital shaker at 37 C and 225 rpm overnight.
2. Prepare plasmid DNA of the Pichia expression vector with the
inserted galectin cDNA of interest using a plasmid preparation
kit and the cell culture grown in the previous step (see Note 9).
3. Mix 20 μg of the Pichia expression vector containing galectin-
coding cDNA (prepared in step 2) with 90 units of SacI and
incubate at 37 C overnight (see Notes 10 and 11).
4. Purify the linearized vector by using a DNA extraction kit and
concentrate to a volume of 10 μL using a vacuum concentrator.
3.5.2 Preparation of 1. Inoculate 5 mL of YPD with a single colony of Pichia X-33

Electrocompetent Pichia X- strain in a 50-mL centrifuge tube covered with gauze and
33 Cells incubate at 29 C for 18 h while shaking.
2. Inoculate 100 mL of YPD with the overnight culture (prepared
as described in step 1) to yield a starting OD600 of 0.005 in a
1 L baffled flask covered with a cellulose plug and incubate at
29 C for 17–18 h.
3. Grow overnight to a final OD600 of 1.3–1.5 and harvest X-33
cells by centrifugation at 4 C, 1500 g for 10 min.
4. Resuspend the X-33 cell pellet successively in 100 mL and
50 mL ice-cold sterile water as well as in 4 mL ice-cold 1 M
sorbitol solution, in each case followed by a centrifugation step
as in step 3.
5. Finally, resuspend the X-33 cell pellet in 200 μL ice-cold sorbi-
tol (1 M) solution. The cells are ready for electroporation.
3.5.3 Electroporation 1. Add 90 μL of the competent X-33 cell suspension (see Sub-
Procedure of Pichia X-33 heading 3.5.2) to a solution of 10 μg/10 μL Pichia expression
Cells vector with the galectin-coding cDNA of interest (prepared as
described in Subheading 3.5.1), mix and incubate on ice for
5 min.
2. Transfer the reaction mix to a cold 2 mm electroporation
cuvette and perform electroporation applying conditions as
follows: 1.5 kV, 25 μF, and 200 Ω for 5 s.
3. Immediately add 1 mL ice-cold 1 M sorbitol solution, transfer
the mixture to a 15-mL centrifuge tube covered with gauze and
incubate at 29 C for 2 h.
4. Transfer the mixture to an Eppendorf tube, centrifugate sus-
pension of transformed X-33 cells at 1500 g at RT for 5 min,
discard the supernatant, resuspend the cell pellet in 200 μL of
1 M sorbitol solution and spread 100 μL each on YPDS agar
plates with 100 μg/mL zeocin.
5. Incubate at 29 C for 4–6 days until colonies appear and
analyze colonies as outlined in Subheading 3.6.
3.6 Analysis of 1. Determine the phenotype by streaking each colony (prepared

Colonies as outlined in Subheading 3.5) on minimal dextrose or minimal
methanol agar plates. Incubate the plates at 29 C for 2 days
(see Note 12).
2. Check plates: MutS (methanol utilization slow) results in little
or no growth on minimal methanol agar plates while Mut+
(methanol utilization plus) leads to normal growth on both
minimal agar plates.
3. Select colonies on YPD plates with increasing concentrations of
zeocin (100 μg/mL, 500 μg/mL, or 1000 μg/mL) and incu-
bate at 29 C for 2 days (see Note 13).
4. Inoculate 5 mL of YPD medium with a transformed X-33
colony (Mut+, selected as described in steps 2 and 3) in a
50-mL centrifuge tube covered with gauze and incubate at
29 C for 18 h while shaking (270 rpm).
5. Centrifugate 2 mL of suspension of transformed X-33 cells at
1500 g for 5 min and discard the supernatant.
6. Resuspend the cell pellet in 146 μL of 50 mM EDTA (pH 8.0),
add 3.75 μL lyticase solution (75 U/μL), and incubate at 37 C
for 1 h (see Note 14).
7. Cool to RT, centrifugate for 2 min at maximum speed and
discard supernatant.
8. Resuspend the pellet in 300 μL of nucleic acid solution, add
100 μL of protein precipitation solution (both solutions from
Wizard® genomic DNA purification kit), mix for 20 s, incubate
on ice for 5 min, and centrifugate for 3 min at maximum speed.
9. Carefully mix the supernatant from step 8 with 300 μL of

isopropanol, centrifugate as in step 7, and discard the
supernatant.
10. Wash DNA pellet with 300 μL of 70% (v/v) ethanol, centrifu-
gate as in step 7, discard supernatant and air-dry the pellet for
15 min.
11. Add 50 μL of DNA rehydration solution and 1.5 μL of RNase
A (both solutions from Wizard® genomic DNA purification
kit) and incubate at 37 C for 15 min.
12. Rehydrate genomic DNA by incubation at 65 C for 1 h with
moderate shaking.
13. Confirm integration of insert of interest performing PCR
applying the following setup: TopTaq™ DNA polymerase,
polymerase buffer (10), MgCl2, Q solution, and primer pair
(50 AOX and 30 AOX). Thermocycler program: initial denatur-
ation at 95 C for 3 min, then 30 cycles of denaturation at
95 C for 30 s, annealing at 54 C for 1 min, and elongation at
72 C for 2 min, final elongation step at 72 C for 10 min.
14. Analyze amplification products by agarose gel electrophoresis
(visualize with a gel documentation system). For calculation of
size, 588 bp (part of the vector) must be added to the expected
size (bp) of the galectin-coding cDNA. Mut+ integrants show a
second band (AOX1 gene), which has a size of approximately
2.2 kb.
3.7 Expression of 1. Inoculate a colony (selected as outlined in Subheading 3.6)

Galectin in P. pastoris into 25 mL BMGY medium filled in a 250-mL baffled flask
covered with gauze and incubate at 29 C overnight while
shaking at 270 rpm.
2. Centrifugate the overnight culture for 10 min at 1500 g and
3. Resuspend the cell pellet in 100 mL of BMMY using a 1-L
baffled flask covered with a cellulose plug (see Note 15). Add
methanol for induction of protein expression to a final concen-
tration of 0.5–2% (v/v) and, if necessary, DTT to a final con-
centration of 1.25 mM (see Note 16).
4. Incubate at 29 C under gentle shaking at 270 rpm for
56–72 h.
5. Add methanol (0.5–2% (v/v)) and, if necessary, DTT (to reach
the final concentration of 1.25 mM) every 24 h.
6. Harvest cell culture supernatant at a final OD600 of approxi-
mately 70 by centrifugation at 4 C and 1500 g for 10 min.
7. For optimal yields, supernatant should be processed directly.
3.8 Purification of 1. Pre-equilibrate 1–2 mL slurry of home-made lactosylated

Galectin from P. Sepharose 4B resin with 20 mM PBS containing 4 mM
pastoris Supernatant β-mercaptoethanol and 2 mM EDTA.
by Affinity 2. Add β-mercaptoethanol to the yeast cell supernatant (prepared
Chromatography on as described in Subheading 3.7) to a final concentration of
Lactosylated 4 mM.
Sepharose 4B Resin 3. Mix the yeast cell supernatant with the pre-equilibrated lacto-
sylated Sepharose 4B resin and batch-bind by shaking the slurry
gently on a roller mixer at 4 C for 1 h.
4. Fill the slurry of supernatant and lactosylated Sepharose 4B
resin slowly into a chromatography column, wait until the
beads have settled down and preclude exposure of resin to air
by adding some supernatant (see Note 17).
5. Wash the column with 25–50 mL of step 1 buffer.
6. Wash the column with 5–20 mL of 20 mM PBS (always test for
presence of protein in the pass-through buffer).
7. Elute the recombinant protein with 20 mM PBS containing
100 mM lactose and 20 mM iodoacetamide (additive protects
sulfhydryl groups and thus galectin-1 from inactivation by
oxidation) collecting the eluate in 500 μL fractions (see
Note 18).
8. Analyze fractions for protein content by the Bradford assay and
combine fractions which contain protein.
9. Exchange buffer against PBS (5 or 20 mM) by using a PD-10
desalting column. Determine the protein concentration apply-
ing the Bradford assay.
10. Divide the protein-containing fraction into aliquots and store
the aliquots at 20 C for lyophilization.
3.9 Electrophoresis 1. Mix protein sample (between 50 and 200 ng for Western
and Transfer blotting, 1 and 2 μg for lectin blotting) or the protein standard
mixture with the corresponding volume of the protein sample
loading buffer (5).
2. Incubate solution of protein sample in protein sample loading
buffer at 95 C for 5 min followed by a centrifugation step for
3 min at 15,000 g.
3. Set up the SDS gel into the electrophoresis module, and then
fill the inner and outer chamber with gel running buffer.
4. Load mixtures of protein-containing sample with loading
buffer (10–20 μL) into slots of the gel.
5. Run electrophoresis at 200 V for approximately 45 min.
6. Thereafter, equilibrate two blotting fiber pads, six blotting
papers, a nitrocellulose membrane, and the processed SDS gel
in transfer buffer.
7. Prepare the blotting sandwich (as recommended by the manu-

facturer): place the cassette with the gray side down, then place
step wisely one pre-wetted blotting fiber pad, three blotting
papers, the equilibrated gel, the pre-wetted nitrocellulose
membrane, three pre-wetted blotting papers, and, in the final
step, one blotting fiber pad on each other.
8. Remove air bubbles from nitrocellulose-gel contact area, close
the gel cassette, and fix it in the module.
9. Add the frozen cooling unit, place it in the tank, fill the tank
with transfer buffer to the “blotting” mark on the tank and add
a magnetic stirring bar.
10. Put on the lid, connect the cables with the power supply and
run the blot under constant stirring on a magnetic stirrer at
100 V. The run is completed at a current of 350 mA (estimated
period of time: 120 min).
11. Remove the membrane from the cassette and stain it with
Ponceau S solution for approximately 5 min to check the
quality of the transfer and mark the bands of the standard
proteins with a pencil.
12. Wash nitrocellulose membrane with TBS-T (0.1% (v/v)
Tween® 20), until the transferred proteins are free of
Ponceau S.
3.10 Western 1. Immerse the membrane (see Subheading 3.9) in blocking solu-
Blotting tion at RT for 1 h.
2. Dissolve lyophilized anti-galectin antibody in PBS/BSA (1 μg/
μL) and add the antibody solution to 3 mL of blocking solu-
tion, yielding a final concentration of 0.5–1 μg/mL. Incubate
the membrane with this antibody-containing solution at 4 C
overnight.
3. Wash the membrane three times with TBS-T each for 10 min.
4. Dilute 1.5 μL of solution containing anti-rabbit IgG conju-
gated with horseradish peroxidase (1 μg/μL) with 3 mL of
blocking solution and incubate the membrane with this solu-
tion at RT for 2 h.
5. Wash the membrane three times with TBS-T each for 10 min.
6. Prepare fresh ECL solution and incubate the membrane for
1–2 min in the solution.
7. Remove the excess of ECL solution and detect the bands with a
gel documentation system.
3.11 Lectin Blotting 1. Dissolve lyophilized concanavalin A in sodium bicarbonate

buffer to a final concentration of 2–5 mg/mL.
2. Add 50 μL of DMF to 1 mg biotin NHS derivative (prepare

freshly).
3. Add 15 μL (0.3 mg biotin NHS derivative) of the mixture from
step 2 per mg concanavalin A and incubate the mixture for 4 h
at 4 C.
4. Use a Sephadex G25 PD-10 desalting column (GE Healthcare)
to remove the biotinylated protein from non-bound, free
reagent and fractionate biotinylated concanavalin A solution
in aliquots (containing 50 μg) for lyophilization.
5. Immerse the membrane (see Subheading 3.9) in blocking solu-
tion at 4 C overnight.
6. Prepare 3 mL of blocking solution containing 1 μg/mL bioti-
nylated concanavalin A, 0.1 mM CaCl2, and 0.01 mM MnCl2.
7. For inhibition experiments add 0.2 M D-mannose and 0.02 M
EDTA to the solution of step 2 and incubate at RT for 2 h.
8. Incubate the membrane in the solution either of step 2 or the
mixture of step 3 at RT for 2 h.
9. Wash the membrane three times with TBS-T (0.5% (v/v)
Tween® 20), each time for 10 min.
10. Incubate the membrane with 3 mL of blocking solution con-
taining 1 μg/mL peroxidase-conjugated avidin at RT for 2 h.
11. Wash the membrane three times with TBS-T (0.5% (v/v)
Tween® 20) each for 10 min.
12. Incubate the membrane with fresh ECL solution for 1–2 min.
13. Remove the excess of the ECL solution and detect the bands
with a gel documentation system.
3.12 Sample 1. Dissolve 100 μg of galectin in 100 μL of digestion buffer.

Preparation for ESMS
Analysis for Galectins
3.12.1 Sample
Preparation for Analysis of
Protein
3.12.2 Sample 1. Dissolve 100 μg of galectin in 28.4 μL of distilled water.

Preparation with 2. Add 45 μL of denaturation buffer.
Deglycosylation
3. Add 13 μL of 100 mM TCEP (see Note 19).
4. Incubate the prepared sample at 37 C for 1 h (see Note 20).
5. For desalting load sample on Micro Bio-Spin™ P-6 Gel Col-
umn (see Note 21).
6. Remove bottom plug of Micro Bio-Spin™ P-6 Gel Column
and spin in a microcentrifuge for 1 min at 1500 g.
7. Equilibrate column by spinning three times with 300 μL of

phosphate buffer (pH 7.2) for 1 min at 1500 g.
8. Transfer sample to the empty column and spin for 2 min at
1000 g.
9. Add 100 μL of 10 mM phosphate buffer (pH 7.2).
10. Transfer sample to an Eppendorf test tube.
11. Add 5.5 μL of N-Glycosidase F-containing solution (see
Note 22).
12. Incubate the enzyme-containing solution at 37 C for 16 h (see
Note 23).
3.13 ESMS Analysis 1. Install the ACQUITY UPLC BEH C4 column at the
ACQUITY ULPC System (see Note 24).
2. Equilibrate the column with eluent A.
3. Set the column temperature to 60 C.
4. Set the eluent flow to 50 μL/min.
5. Tune and calibrate the mass spectrometer according to the
manufacturer’s specifications (see Notes 25–27).
6. Hyphenate the column outlet of the ACQUITY UPLC System
to the Synapt G2 HDMS mass spectrometer (see Note 28).
7. Inject 10 μL of the solution with degyclosylated or mock-
treated (glyco)protein on to the column.
8. For separation, a linear gradient from 0% eluent B to 100%
eluent B in 22 min is applied (see Note 29).
9. Acquire mass spectra in the positive mode over a range of m/z
700–2000.
10. Deconvolute the mass spectra by using MassAnalyzer software
from Roche (see Note 30).
11. Assign peaks and calculate mass differences.
3.14 Sample 1. Dissolve 100 μg of galectin in 100 μL of distilled water.

Preparation for LC/ 2. To 36 μL of the sample solution add 10 μL 5 Rapid buffer.
MSMS Analysis of
3. Incubate at 80 C for 2 min (see Note 31).
Galectin
4. Add 4 μL of Rapid PNGase F (see Note 32).
3.14.1 Deglycosylation
3.14.2 Tryptic Digestion 1. Add 50 μL of denaturation buffer to the deglycosylated sample.

2. To reduce the disulfide bridges, add 2 μL of DTT solution
(100 mg/mL).
4. Allow sample to cool down to RT.
5. For protection of the sulfhydryl groups of the reduced disulfide

bridges add 4 μL of iodoacetamide-containing solution
(111.6 mg/mL).
6. Incubate at RT for 30 min in the dark.
7. For buffer exchange, remove bottom plug of Zeba Spin col-
umn and spin in a microcentrifuge for 1 min at 1500 g (see
Note 33).
8. Equilibrate columns by spinning three times with 300 μL of
digestion buffer for 1 min at 1500 g.
9. Transfer sample to the empty column and spin for 2 min at
1000 g.
10. Transfer sample (100 μL) to an Eppendorf test tube.
11. Add 2 μL of trypsin-containing solution (see Note 34).
12. Incubate at 37 C for 18 h.
13. Stop digestion by adding 2.5 μL of 10% TFA (v/v) solution (see
Notes 35 and 36).
3.15 LC/MSMS 1. Install the ACQUITY UPLC CSH C18 column at the VAN-
Analysis of Galectin QUISH UHPLPC System (see Note 37).
2. Set the eluent flow to 50 μL/min.
3. Set the column temperature to 60 C.
4. Equilibrate the column with eluent A.
5. Calibrate the mass spectrometer according to the manufac-
turer’s specifications (see Notes 26, 38, and 39).
6. Hyphenate the column outlet of the VANQUISH UHPLC
System to the Triple TOF mass spectrometer equipped with
an ESI source (see Note 28).
7. Set column temperature to 60 C.
8. Inject 10 μL of the sample after trypsin treatment onto the
column.
9. For separation, a linear gradient from 0% eluent B to 100%
eluent B in 60 min is applied (see Note 29).
10. Acquire mass spectra in the positive MS/MS mode over a range
of m/z 200–2000.
11. By using PeakView software: generate Extracted Ion Chroma-
tograms for the separated and analyzed peptides, then integrate
the resulting peaks with Master View add-on (see Note 40).
12. Perform database search using the software Protein Pilot 5.0
based on FASTA-files containing the sequences of the analyzed
galectin (Search algorithm: Thorough Peptide Con-
fidence>95% Cys alkylation Iod E / Iod A) (see Note 41).
3.16 Radioiodination 1. To the PD-10 Desalting Columns for separation of the iodi-
of Galectin nated galectins, flush columns with 10 mL of SPB, followed by
1 mL of BSA (1 mg/mL) solution (blocking of nonspecific
protein binding sites on the Sepharose matrix of the columns),
and additional 10 mL of SPB.
2. Dissolve 100 μg of galectin in 100 μL of SPB.
3. Add 10 μL of the 1 M D-lactose solution (final lactose concen-
tration of 90.9 mM) to protect the carbohydrate recognition
domain of the galectins.
4. Add 20 μL of sodium [125I]iodide (3.7 GBq/mL) (see
Note 42).
5. Start the reaction by the addition of two Iodobeads and incu-
bate at RT for 15 min (see Note 43).
6. Transfer the whole reaction mixture to another vial, leaving the
beads behind.
7. Wash the beads with 100 μL of SPB and merge the wash with
the reaction mix.
8. For separation of the protein from unreacted iodine and lac-
tose, apply the whole mixture onto a PD-10 Desalting Column
(prepared as described above), elute with PBS by gravity flow,
and collect 1-mL fractions. The radioiodinated proteins will
typically elute in fractions 2–4, while free iodine and lactose will
start to elute in fraction 6.
9. Monitor the elution by liquid scintillation counting: mix ali-
quots (1 μL) of the eluted fractions with 10 mL of UltimaGold
Scintillation cocktail (PerkinElmer®) and count in a liquid
scintillation counter. The counting range is set from 0 to
70 keV with automatic dpm calculation using the tSIE quench
index (see Notes 44 and 45).
10. Combine fractions 2–4 and count aliquots of the pool by liquid
scintillation counting (as described above), determine the pro-
tein concentration by a Nanodrop spectrophotometer, and
calculate the radioactivity per μg of protein (specific radioactiv-
ity). Typical specific radioactivities range between 200 and
300 KBq/μg protein (see Notes 46 and 47).
3.17 Neuroblastoma 1. Seed neuroblastoma cells (strain SK-N-MC) at an initial density

Cell Culture and of 104 cells/well in 96-well plates and culture the cells in
Binding Studies Eagle’s minimal essential medium (MEM) supplemented with
10% (v/v) fetal calf serum, penicillin (100 IU/mL), strepto-
mycin (100 μg/mL) in an atmosphere of 95% air and 5% CO2
at 37 C for 5 days.
2. Change medium to MEM without fetal calf serum for addi-
tional 16 h.
3. For the binding assay, change medium to serum-free MEM

(100 μL/well) supplemented with 25 mM HEPES, pH 7.4,
0.01% (w/v) BSA to block non-specific protein-binding sites,
and the iodinated proteins at non-saturating concentrations,
then incubate at 37 C for 2 h (see Note 48).
4. For quantitation of carbohydrate-independent adsorption of
labeled probes, determine binding in the presence of a mixture
of 150 mM lactose/0.5 mg/mL asialofetuin that blocks
glycan-dependent binding by galectin.
5. After 2 h, remove the radioactive medium, then wash the cells
three times with PBS. No radioactivity should be detectable in
the third wash by liquid scintillation counting.
6. Solubilize the cell layer by adding 200 μL of 0.2 M NaOH
per well.
7. After 2 h, transfer the solution to a liquid scintillation vial.
8. Wash wells with an additional 200 μL of 0.2 M NaOH and
transfer the solution to the corresponding scintillation vial.
9. Add 10 mL of UltimaGold scintillation cocktail and count
radioactivity events in a Tricarb scintillation counter with the
following settings (see Notes 44, 45 and 49):
(a) Assay type: dpm.
(b) Nuclide: Iodine-125.
(c) Channel setting: 0–70 keV.
(d) Count time: 60 min.
(e) Count termination: 2 s% ¼ 0.5.
(f) Quench indicator: tSIE.
10. Calculate the amount of bound galectin from the specific
radioactivity of the applied galectins.
11. Plot the amount of galectin that was included per well versus
the amount of bound galectin to establish a binding curve.
12. Plot the reciprocal value of the galectin amount that was
included per well versus the reciprocal value of bound galectin
to establish a Scatchard plot (see Note 50).
13. Calculate the binding constant (KD) and number of binding
sites (Bmax) from the plot.
3.18 Proliferation 1. Seed neuroblastoma cells (strain SK-N-MC) at an initial density

Assay of 104 cells/well in 96-well plates and culture the cells in
Eagle’s minimal essential medium (MEM) supplemented with
10% (v/v) fetal calf serum, penicillin (100 IU/mL), strepto-
mycin (100 μg/mL) in an atmosphere of 95% air and 5% CO2
at 37 C (see Note 51).
2. Allow the cells to attach to the wells for 16 h.
3. Change medium to MEM containing supplements (see step 1)

and the test concentration of the galectin. In parallel, prepare
controls: (1) with wells containing medium and supplement,
but no cells; (2) incubations of the cells without galectin
addition.
4. At the desired time points (e.g., 24 h, 48 h, or 72 h), after the
addition of galectin, add 15 μL of CellTiter 96™ dye solution
per well in the culture medium.
5. Stop the reaction after 2 h by the addition of 100 μL of
CellTiter 96™ solubilization/stop solution and incubate for
1 h.
6. Read the absorbance in a microplate reader at 570 nm (see
Notes 52–55).
7. Calculation of the result: subtract the absorbance of the wells
containing medium without cells (control 1) from all measure-
ments. Compare the corrected absorbance readings obtained
from the incubation without galectin addition to the results
from the galectin treated cells. Express the result as % growth
inhibition or % increased growth.
3.19 Cloning of 1. Amplify the galectin-coding cDNA of interest from a bacterial

Galectin-Coding cDNA expression vector as template using Phusion® High-Fidelity
into Mammalian DNA polymerase and buffer (10), dNTPs, a sense primer
Expression Vectors and an anti-sense primer each extended by an appropriate
(pSecTag2 B, restriction site and an overhang of at least 4 bp at the 50 -ends
pcDNA3.1+/C-HA) with (see Note 56). Thermocycler program: denaturation at 98 C
Signal Peptide for 30 s, then 30 cycles of denaturation at 98 C for 10 s,
annealing at 60–68 C for 20 s, and elongation at 72 C for
15 s, finally an elongation step at 72 C for 10 min were
performed.
2. Use a DNA extraction kit to purify the PCR products from gel
slices (bands for DNA presence visualized with a gel documen-
tation system) and perform digestion at 37 C overnight with
the appropriate restriction enzymes.
3. In parallel, linearize a sufficient amount of pSecTag2 B
vector DNA.
4. After an additional purification step, perform ligation and
transformation, as described in Subheadings 3.1 and 3.2.
5. Amplify the cDNA coding the galectin of interest in-frame with
leader sequence of the murine Igκ chain using Phusion® High-
Fidelity DNA polymerase and buffer, with a sense primer and
an anti-sense primer each with an appropriate recognition
sequence for a restriction enzyme and the pSecTag2 B vector
containing the galectin cDNA of interest in-frame with leader
sequence (prepared in steps 1–4) as template (see Note 57).
Thermocycler program: denaturation at 98 C for 30 s, then

30 cycles of denaturation at 98 C for 10 s, annealing at
60–68 C for 30 s, and elongation at 72 C for 30 s, finally
an elongation step at 72 C for 10 min were performed.
6. Use a DNA extraction kit to purify the PCR products from gel
slices (visualized with a gel documentation system).
7. Expose purified cDNAs and the mammalian expression vector
DNA (pcDNA3.1+/C-HA) to the specified restriction
enzymes and incubate mixture at 37 C overnight.
8. After a further purification step using a DNA extraction kit,
proceed with ligation and transformation steps (see Subhead-
ings 3.1 and 3.2).
3.20 Culture of 1. Prepare a working solution of 0.1 mg/mL poly-L-lysine (PLL)

Human Embryonic by diluting the 1 mg/mL stock PLL solution in sterile tissue-
Kidney 293T Cells culture-grade water.
(HEK293T) 2. Add 0.5 mL/cm2 of 0.1 mg/mL PLL to the distinct plastic
culture surfaces. Rock dishes gently to ensure an even coating
Plastic and Glass Substrata
distribution. Place in a humidity incubator at 37 C overnight.
for Cell Culture 3. In the case of glass cover slips, place the appropriate number of
cover slips in standard 90 mm bacteriological petri dishes using
small forceps. Make sure cover slips are evenly distributed
across the dish and do not come into contact with each other.
4. Add 1 mg/mL PLL solution that will completely cover the
cover slip’s surface (see Note 58). Keep in cell culture incubator
at 37 C overnight.
5. The following day, remove the PLL solution and thoroughly
rinse the culture surfaces with sterile tissue-culture-grade
water. Repeat this step three times.
6. Allow culture surfaces to dry before use or being stored (see
Note 59).
3.20.2 Cell Cultures for 1. Prepare complete growth medium (DMEM, 10% (v/v) FBS,
Biochemical Analysis or for 1% (v/v) P/S, 2 mM L-glutamine) and warm to 37 C in a
Immunocytochemical water bath.
Staining 2. Trypsinize a maintenance dish of HEK293T cells (see
Note 60).
3. Count cells and seed the required number of PLL-coated
dishes or wells at a density of 5 104 cells/cm2 in complete
growth medium. Incubate cell cultures overnight.
4. Before transfection, examine the cells under a light microscope
to make sure that the culture density is between 50% and 70%
confluency (see Note 61). Optional: Replace the medium with
fresh complete growth medium at least 1 h before transfection.
5. To transfect cells, dilute 250 ng of plasmid DNA (purified by

using a plasmid preparation kit) in 50 μL of warm Opti-MEM I
for every cm2 of cell culture to be transfected.
6. Likewise, dilute 0.5–1 μL of lipofectamine LTX Reagent in
another tube containing 50 μL of warm Opti-MEM I for
every cm2 of cell culture to be transfected.
7. Mix the contents of both tubes and incubate the mixture to
form DNA-lipofectamine complexes at RT for 30 min.
8. Carefully add 100 μL of the DNA-lipofectamine mixture for
every cm2 of corresponding dish or well containing the cells.
Gently rock the dishes or plates back and forth and from side to
side to evenly distribute the transfection complexes. The media
containing the transfection mixture does not have to be
removed following transfection.
9. Incubate the cells in a cell culture incubator at 37 C for 24 h.
3.21 Density 1. Place culture dishes with control or transfected HEK293T cells
Gradient Fractionation on a cold tray.
of Subcellular 2. Wash cell monolayers with cold PBS (2 mL/60 mm dish or
Membrane Vesicles 5 mL/100 mm dish).
3.21.1 Preparation of 3. Add HEPES buffer (25 mM, pH 7.4) to culture dishes
Total Membrane Extracts (0.5 mL/60 mm dish or 1 mL/100 mm dish), scrape cells,
from HEK293T Cells for and transfer them to Eppendorf test tubes.
Subcellular Fractionation 4. Incubate cell suspensions on ice for 15 min (to induce osmotic
cell breakage).
5. Homogenize with a syringe (22G) up and down, 10 times.
6. Centrifugate in a table-top microcentrifuge (4 C, 1500 g,
5 min).
7. Discard pellets.
8. Centrifugate postnuclear supernatants in a table-top ultracen-
trifuge (4 C, average centrifuge field 100,000 g, 60 min).
9. Resuspend membrane pellets in cold HEPES buffer by pipet-
ting up and down using a syringe (22G), until no pellet clumps
are detectable. These suspensions are ready for immediate
fractionation.
3.21.2 Preparation of 1. Cut disks (10 mm diameter) from high-density foam original
Density Gradients, slab using a puncher. They have to fit and float smoothly inside
Membrane Fractionation, the centrifuge tubes as each step of the gradient is overlayed (see
and Fraction Protein Note 4).
Precipitation 2. Prepare ten iodixanol solutions from 10% to 30% (2.5% con-
centration increment) by diluting original 60% (w/v) iodixanol
solution in 25 mM HEPES buffer (pH 7.4).
3. Load 1 mL of the highest density (30%) at the bottom of the

ultracentrifuge tube, then drop the foam disk into the tube
making sure that it floats flat on the top of the solution.
4. Load the rest of the step gradients (1 mL each) one by one
from the highest density to the lowest. Pipet slowly, with the tip
near above the foam disk, and smoothly release the gradient-
step solutions (see Note 62).
5. Load the sample (1 mL volume; see Note 63) at the top of the
gradient.
6. Remove the foam disk carefully before centrifugation.
7. Centrifugate at 35,000 rpm (average centrifuge field
151,194 g) in a SW 41-Ti rotor at 4 C for 16 h.
8. Repeatedly pipet 1-mL fractions from the top of the gradient
and transfer them one by one to plastic tubes. Care should be
taken to avoid going too deep with the tip of the pipette, trying
always to keep it at the gradient surface, doing a slow circular
movement while pipetting (see Note 64).
3.21.3 Precipitation of 1. For 1-mL fractions, add sequentially:

Proteins from Fractions (a) 4 mL of methanol and vortex.
from the Density Gradient
(b) 1 mL of chloroform and vortex.
(c) 3 mL of distilled water and vortex.
2. Centrifugate for 1 min at 9000 g, RT.
3. Carefully remove the upper phase and discard.
4. Precipitate proteins from the lower chloroform phase by add-
ing 3 mL of methanol and vortex.
5. Centrifugate for 5 min at 9000 g, RT.
6. Aspirate supernatant and dry pellet with a stream of air or
leaving the tubes open in the fume hood for 1 h.
7. Store frozen at 20 C or take up in Laemmli buffer and
incubate for SDS-PAGE at 95 C for 5 min (proceed with
steps in Subheading 3.21.4).
3.21.4 Analysis of 1. Load precipitated proteins from gradient fractions and protein
Fractions by Western molecular mass markers in:
Blotting (a) 8% polyacrylamide gels for the detection of calnexin,
GM130 and EEA1.
(b) 15% polyacrylamide gels for the detection of galectin-1 or
galectin-4.
2. Run PAGE-SDS at 20 mA/gel (constant current).
3. Transfer to 0.2 μm nitrocellulose at 100 V (constant voltage) at
4 C for 1 h, and block it in blocking solution at room temper-
ature for 1 h.
4. Incubate on a tumbling rocker plate at 4 C for 16 h as follows:

(a) Membranes from 8% gels with monoclonal anti-EEA1,
anti-GM130 or anti-calnexin antibody dissolved and
diluted in blocking solution (see Note 65).
(b) Membranes from 15% gels with polyclonal anti-Gal-1 or
anti-Gal-4 or rat monoclonal anti-HA antibody diluted in
blocking solution.
Concentrations of all primary antibodies (a,b) were 1 μg
per mL blocking solution (1:1000).
5. Wash three times with cold TBS-T.
6. Incubate on a tumbling rocker plate at 4 C for 1 h as follows:
(a) Anti-mouse antibodies conjugated with horseradish per-
oxidase diluted in blocking solution (0.2 μg/mL;
1:5000).
(b) Anti-rabbit or anti-rat antibodies conjugated with horserad-
ish peroxidase diluted in blocking solution (0.2 μg/mL;
1:5000).
7. Wash three times with cold TBS-T, then rinse in TBS to elimi-
nate detergents.
8. Develop protein bands with a chemiluminescence reagent, and
detect/quantify them using a chemiluminescence reader.
3.22 Immunocyto- 1. Wash cells twice with PBS at RT (see Note 66).
chemistry 2. Fix cells with fixative solution (4% (w/v) paraformaldehyde in
PBS, pH 7.4) at RT for 15 min.
3. Gently wash cells three times with PBS for 5 min.
4. Quench free aldehyde groups with 50 mM NH4Cl in PBS at
RT for 15 min.
5. Wash cells with PBS for 5 min.
6. Permeabilize cells with 0.1% (v/v) Triton X-100 in PBS at RT
for 5 min.
7. Select the cover slips that will be used for detection of ER
markers and continue with step 8 to perform an antigen-
retrieval step (see Note 67). For the other cover slips, proceed
to step 12.
8. Preheat the antigen retrieval solution to 80 C in a borosilicate
glass petri dish or a horizontal staining jar, which is placed in a
water bath or oven. Several dishes can be prepared to check for
optimal experimental conditions.
9. Using a pair of forceps, carefully immerse the appropriate cover
slips with the cells facing up in the corresponding dishes con-
taining the preheated antigen retrieval solution.
10. Place the dishes or jars containing the cover slips in a water bath
or oven at 80 C for 10 min.
11. Carefully remove the cover slips and place them in multi-well
dishes containing PBS.
12. Rinse the cells with PBS three times.
13. Block sites for nonspecific binding of protein by incubating
cells with blocking solution at RT for 1 h.
14. Place cover slips in a humid incubation box and incubate with
primary antibody (concentrations: anti-GM 130, 0.625 μg/
mL; anti-calnexin, 1.25 μg/mL; anti-HA, 0.25 μg/mL)
diluted in 150 μL of blocking solution overnight at 4 C.
15. Wash three times with PBS each for 15 min.
16. Incubate cover slips with the appropriate fluorescent second-
step antibody (4 μg/mL) diluted in 150 μL of blocking solu-
tion at RT for 1 h in a humid incubation box.
17. Wash three times with PBS each for 15 min.
18. Pipet 10 μL of mounting media on a clean glass slide.
19. Rinse the cover slip with distilled water and mount with the
cells facing down on the glass slides.
20. Store at 4 C, keep in the dark and document by confocal
microscopy.
4 Notes
1. It is recommended that the reaction mixture containing Top-

Taq™ DNA polymerase (Qiagen) and the provided standard
buffer is supplemented with Q solution and additional MgCl2.
2. Use chilled transfer buffer to improve heat dissipation.
3. The use of highly purified plasmid preparations that are phe-
nol- and endotoxin-free is recommended.
4. Floating disks can be obtained cutting them from wine oak
corks, as described previously [27].
5. When using the restriction enzyme XhoI for cloning, it is
possible to express the protein with its native N-terminus by
direct fusion of the cDNA sequence of the galectin to the Kex2
and Ste13 cleavage sites. For this reasoning, the sense primer
contains the Kex2 and the Ste13 cleavage site as well as the first
nucleotides of the galectins cDNA sequence. Exemplarily, in
the case of human galectin-1, the designed DNA oligonucleo-
tide is: 50 -GGCTCGAGAAAAGAGAGGCTGAAGCT ATG
GCTTGTGGTCTGGTCGC -30 (restriction site underlined,
Kex2 and Ste13 cleavage sites in bold).
6. For optimal annealing and elongation, it is recommended that

mutations are introduced in the middle of the primer with
10–15 bases of unchanged, thus fully base-pairing sequence
flanking the site of mutation.
7. Determine the rate of synthesis of the used commercial DNA
polymerase preparation and adjust elongation time for the used
template.
8. A two-step PCR procedure can be an alternative as mutagenesis
protocol for rare cases, in which the QuikChange™ site-
directed mutagenesis does not work. Amplify two PCR frag-
ments with two sets of primers using the Phusion® High-
Fidelity DNA polymerase. The first set consists of a sense
primer (with restriction site) combined with an anti-sense
primer containing the mutagenic nucleotides and 10 bp of
correct sequence on both sides. The second set consists of a
sense primer containing the site of the intended mutation and
10 bp of correct sequence on both sides in combination with an
anti-sense primer with a restriction site. Thermocycler pro-
gram: denaturation at 98 C for 30 s, then 30 cycles of dena-
turation at 98 C for 10 s, annealing at 60–68 C for 15 s, and
elongation at 72 C for 10 s, finally an elongation step at 72 C
were performed for 10 min. Purify PCR products and use them
as template in a second PCR with the sense primer of set 1 and
the anti-sense primer of set 2. Thermocycler program: denatur-
ation at 98 C for 30 s, then 30 cycles of denaturation at 98 C
for 10 s, annealing at 60–68 C for 15 s, and elongation at
72 C for 20 s, finally an elongation step at 72 C for 10 min.
After purification and enzymatic digestion, the PCR amplicons
can be ligated into a vector and used for transformation of
competent E. coli cells.
9. The plasmid preparation should yield 20 μg to compensate for
losses during further working steps and to have a minimum
amount of 10 μg available for electroporation.
10. Be sure to check your cDNA sequence that there is no SacI
restriction site.
11. Check whether the plasmid DNA is completely linearized.
Sufficient yields in electroporation with circular plasmid DNA
could only be obtained if you use 5- to 10-fold higher amounts
compared to linearized DNA.
12. Be aware that you first streak colonies on minimal methanol
plates followed by this procedure on minimal dextrose plates.
13. First spread colonies on plates containing 100 μg/mL zeocin
and then on plates containing increasing concentrations of
zeocin. The manufacturer’s advice appears to be a “direct way
to select putative multi-copy recombinants.”
14. Lyticase in solution is stable for only 4–6 h.

15. The volume of the P. pastoris cell culture should not exceed
30% of the baffled flask volume to guarantee optimal supply
with air.
16. To obtain optimal yields, check different concentrations of
methanol for induction (0.5–2% (v/v)) and test whether the
presence of DTT is necessary.
17. It is essential to take care to perform affinity chromatography
very thoroughly according to recommendations of manufac-
turers. It is especially advised to prevent the beads from run-
ning dry, otherwise binding capacity will decrease dramatically
by clumping.
18. Incubate the beads with elution buffer for 15 min after collect-
ing the first fraction. Most of the bound protein will then elute
in a single narrow peak, resulting in a small number of collected
fractions, thus avoiding a diluted solution.
19. Handle TCEP with care, wear eye and face protection.
20. Any type of thermomixer can be used.
21. Using the Micro Bio-Spin™ P-6 Gel Column stick carefully to
the recommendations of the supplier, any other columns that
are able to desalt may also be used.
22. N-Glycosidase F from other suppliers may also be used. Be
careful to use the same amount of activity units.
23. Incubation time with N-glycosidase F may be lowered, but it
may be convenient to incubate overnight.
24. C4 columns from other manufacturers can also be used.
25. Use calibration solutions that are able to cover the mass range
of m/z 700–2000 according to the recommendations of the
mass spectrometer’s supplier.
26. Use the standard settings of your mass spectrometer and tune
the instrument for an optimal signal and low fragmentation.
27. Any mass spectrometer that is equipped with an electrospray
source and is able to acquire mass spectra in the positive mode
over a range of m/z 700–2000 from other suppliers like
Thermo, Bruker, or Waters may be used.
28. Make sure that there are no leaks when you hyphenate your
column to the mass spectrometer.
29. Equilibrate carefully to 0% eluent A before you analyze the next
sample.
30. Suppliers of mass spectrometers provide their own deconvolu-
tion software. These deconvolution programs may also
be used.
31. Any type of thermomixer and microcentrifuge can be used.
32. N-Glycosidase F from other suppliers may also be used. How-

ever, activity units, duration of incubation, and temperature
have to be carefully adjusted, and the completeness of the
deglycosylation has to be checked.
33. Using the Zeba Spin column, stick carefully to the recommen-
dations of the supplier.
34. Any other type of commercial trypsin may be used. It is recom-
mended to stick to the same amount of trypsin, if you use
trypsin from a different supplier.
35. Handle TFA with care, wear eye and face protection.
36. Stopping the tryptic digestion with TFA solution may be
avoided, if you analyze the sample immediately after the diges-
tion procedure.
37. C18 columns from other manufacturers can also be used.
38. Use calibration solutions that are able to cover the mass range
of m/z 200–2000 according to the recommendations of the
mass spectrometer’s supplier.
39. Any mass spectrometer that is equipped with an electrospray
source and is able to acquire MS/MS mass spectra in the
positive mode over a range of m/z 200–2000 from other
suppliers like Thermo, Bruker or Waters may be used.
40. There are several types of software available especially from the
mass spectrometer suppliers. All software that are able to pro-
duce Extracted Ion Chromatograms and to integrate the
resulting peaks may be used.
41. Any databases that allow to be accessed and searched for the
sequences of the analyzed galectins may be used. The search
algorithm has to be adapted accordingly.
42. All steps handling the radioactive material have to be carried
out behind a lead shielding, following the local rules for radia-
tion protection.
43. The oxidation reaction between the Iodobeads and [125I]
iodide may generate volatile [125I] iodine. Therefore, the reac-
tion should be carried out in a well-ventilated fume hood. In
some countries (e.g., Germany), fume hoods with active coal
filters are mandatory. Comply with the local radiation protec-
tion rules.
44. Any type of Tricarb (PerkinElmer®) liquid scintillation counter
ranging from models 1600TR to 5110TR can be used at the
indicated settings.
45. Instead of liquid scintillation counting, also γ-counters (e.g.,
Wizard II from PerkinElmer® or Multi Crystal LB 2111 from
Berthold, Bad Wildbad, Germany) may be used with slightly
reduced counting efficiency as compared to liquid scintillation
counting.
46. Effective protection of integrity of the carbohydrate-binding

sites of the galectin during the labeling reaction may be
checked by binding of the labeled product to lactose-Sepharose
4B (for protocol of preparation, see [23]). Binding of radioac-
tivity as measured by liquid scintillation counting should be
above 90%.
47. Typical specific radioactivity range between 200 and
300 KBq/μg protein, documenting maintained activity as
lectin.
48. Due to the increased buffer capacity of the cell culture medium
with 25 mM HEPES, no additional CO2 supply is necessary
during the incubation with the radiolabeled galectin.
49. If other liquid scintillation cocktails are used, miscibility with
the sample has to be checked.
50. When concentrations for Scatchard plots are set, it should be
kept in mind that the reciprocal value of the concentration is
plotted. In order to get evenly distributed points in the plot,
concentrations should be adapted to the reciprocal graphical
application (for examples of Scatchards plots for radioactive
galectins, see [28, 29]).
51. The assay had also been successfully applied to measure effects
of galectins on cell growth in various colorectal cancer cell lines
and is generally applicable to any cell line.
52. A reference wavelength between 630 and 750 nm may be used
for the absorbance measurements.
53. The produced blue dye is stable at 4 C for several days. Thus,
the plates can be stored in a refrigerator before measurement.
54. Absorbance readings should be below 2000 mOD. If the
absorbance readings exceed 2000 mOD, the incubation time
with the dye can be reduced to obtain more accurate readings.
55. If the absorbance is too close to the control 1 (medium only),
the incubation time with the dye can be increased to up to 4 h.
56. Take particular care to ensure that galectin-coding cDNA is
amplified without start codon to ensure obtaining a single
open reading frame for the sequence of the fusion protein
sequence.
57. Ensure that the anti-sense primer is devoid of a stop codon to
have a continuous open reading frame including the
C-terminal HA tag sequence.
58. A round glass cover slip of 15 mm diameter can be coated with
150 μL of PLL solution.
59. PLL-coated glass cover slips and plastic dishes can be kept at
4 C or 20 C for up to a month. However, it is preferable to
coat the cover slips/plates the day before cell culture is done.
60. HEK293T cells are in the category of biosafety level 2 (accord-

ing to the instruction provided by ATCC collection) and
should be handled accordingly.
61. Confluency can be estimated using a brightfield image of the
cultured cells. Measure the area occupied by the cells within the
image of known dimensions using the segmentation option of
an image processing software such as ImageJ, and calculate the
percentage area covered by the cells.
62. It is important to leave enough empty volume at the top of the
gradient (1.5 mL) to load the sample. The prepared gradient
should not at all be disturbed by tilting or shaking.
63. In order to avoid overloading the gradient, leading to a poor
fractionation, a maximum of 1 mg of total protein should be
loaded in the described system.
64. Alternatively, a peristaltic pump connected to a capillary tube
can be used to collect fractions from the bottom of the gradi-
ents, as described previously.
65. If necessary, after separating proteins in an 8% polyacrylamide
gel and transferring them to nitrocellulose, Cnx (90 kDa),
GM130 (130 kDa), and EEA1 (180 kDa) can be detected on
the same membrane separately by cutting it transversally with a
scalpel at 100 and at 150 kDa, using the positions of the
molecular mass markers as reference. Cnx is detected in the
stripe below 100 kDa, GM130 in that between 100 and
150 kDa, and EEA1 in that over 150 kDa.
66. Exert utmost care when washing the cells with PBS, as they can
easily detach. Slowly pour all solutions or buffers along the
walls of the wells or dishes containing the cover slips. Never
aspirate the buffer solution directly over the cover slips.
Instead, hold the dishes at an angle and carefully aspirate the
liquid along the walls of the wells or dishes.
67. ER chaperones form large protein complexes, which often
hinder their detection by immunological methods. In order
to improve the detection of the ER marker calnexin by ICC, an
antigen-retrieval step is necessary.
References
1. Nickel W, Seedorf M (2008) Unconventional 103(5):1327–1343. https://doi.org/10.
mechanisms of protein transport to the cell 1002/jcb.21513
surface of eukaryotic cells. Annu Rev Cell Dev 3. Popa SJ, Stewart SE, Moreau K (2018) Uncon-
Biol 24:287–308. https://doi.org/10.1146/ ventional secretion of annexins and galectins.
annurev.cellbio.24.110707.175320 Semin Cell Dev Biol 83:42–50. https://doi.
2. Prudovsky I, Tarantini F, Landriscina M, org/10.1016/j.semcdb.2018.02.022
Neivandt D, Soldi R, Kirov A, Small D, Kathir 4. Wilson TJ, Firth MN, Powell JT, Harrison FL
KM, Rajalingam D, Kumar TK (2008) Secre- (1989) The sequence of the mouse 14 kDa
tion without Golgi. J Cell Biochem β-galactoside-binding lectin and evidence for
its synthesis on free cytoplasmic ribosomes. endocytic vesicles. Trends Glycosci Glycotech-
Biochem J 261(3):847–852. https://doi.org/ nol 30(172):SE179–SE184. https://doi.org/
10.1042/bj2610847 10.4052/tigg.1733.1SE
5. Hughes RC (1999) Secretion of the galectin 16. Eckardt V, Miller MC, Blanchet X, Duan R,
family of mammalian carbohydrate-binding Leberzammer J, Duchene J, Soehnlein O,
proteins. Biochim Biophys Acta Megens RT, Ludwig A-K, Dregni A,
1473(1):172–185. https://doi.org/10.1016/ Faussner A, Wichapong K, Ippel H,
s0304-4165(99)00177-4 Dijkgraaf I, Kaltner H, Doring Y,
6. Cooper DNW (2002) Galectinomics: finding Bidzhekov K, Hackeng TM, Weber C, Gabius
themes in complexity. Biochim Biophys Acta H-J, von Hundelshausen P, Mayo KH (2020)
1572(2–3):209–231. https://doi.org/10. Chemokines and galectins form heterodimers
1016/s0304-4165(02)00310-0 to modulate inflammation. EMBO Rep 21(4):
7. Kaltner H, Toegel S, Garcı́a Caballero G, Man- e47852. https://doi.org/10.15252/embr.
ning JC, Ledeen RW, Gabius H-J (2017) 201947852
Galectins: their network and roles in immu- 17. Jia J, Claude-Taupin A, Gu Y, Choi SW,
nity/tumor growth control. Histochem Cell Peters R, Bissa B, Mudd MH, Allers L,
Biol 147(2):239–256. https://doi.org/10. Pallikkuth S, Lidke KA, Salemi M, Phinney B,
1007/s00418-016-1522-8 Mari M, Reggiori F, Deretic V (2020)
8. Sato S (2018) Cytosolic galectins and their Galectin-3 coordinates a cellular system for
release and roles as carbohydrate-binding pro- lysosomal repair and removal. Dev Cell
teins in host-pathogen interaction. Trends Gly- 52(1):69–87. https://doi.org/10.1016/j.
cosci Glycotechnol 30(172):SE199–SE209. devcel.2019.10.025
https://doi.org/10.4052/tigg.1739.1SE 18. Sato S, St-Pierre C, Bhaumik P, Nieminen J
9. Garcı́a Caballero G, Kaltner H, Kutzner TJ, (2009) Galectins in innate immunity: dual
Ludwig A-K, Manning JC, Schmidt S, functions of host soluble β-galactoside-binding
Sinowatz F, Gabius H-J (2020) How galectins lectins as damage-associated molecular patterns
have become multifunctional proteins. Histol (DAMPs) and as receptors for pathogen-
Histopathol 35(6):509–539. https://doi.org/ associated molecular patterns (PAMPs).
10.14670/HH-18-199 Immunol Rev 230(1):172–187. https://doi.
org/10.1111/j.1600-065X.2009.00790.x
10. Gabius H-J, Roth J (2017) An introduction to
the sugar code. Histochem Cell Biol 19. Bertheloot D, Latz E (2017) HMGB1, IL-1α,
147(2):111–117. https://doi.org/10.1007/ IL-33 and S100 proteins: dual-function alar-
s00418-016-1521-9 mins. Cell Mol Immunol 14(1):43–64.
https://doi.org/10.1038/cmi.2016.34
11. Kaltner H, Abad-Rodrı́guez J, Corfield AP,
Kopitz J, Gabius H-J (2019) The sugar code: 20. Yang D, Han Z, Oppenheim JJ (2017) Alar-
letters and vocabulary, writers, editors and mins and immunity. Immunol Rev
readers and biosignificance of functional 280(1):41–56. https://doi.org/10.1111/imr.
glycan-lectin pairing. Biochem J 12577
476(18):2623–2655. https://doi.org/10. 21. Kutzner TJ, Higuero AM, Süßmair M,
1042/BCJ20170853 Kopitz J, Hingar M, Dı́ez-Revuelta N, Garcı́a
12. Haudek KC, Spronk KJ, Voss PG, Patterson Caballero G, Kaltner H, Lindner I, Abad-
RJ, Wang JL, Arnoys EJ (2010) Dynamics of Rodrı́guez J, Reusch D, Gabius H-J (2020)
galectin-3 in the nucleus and cytoplasm. Bio- How presence of a signal peptide affects
chim Biophys Acta 1800(2):181–189. https:// human galectins-1 and -4: clues to explain
doi.org/10.1016/j.bbagen.2009.07.005 common absence of a leader sequence among
adhesion/growth-regulatory galectins. Bio-
13. Harazono Y, Nakajima K, Raz A (2014) Why chim Biophys Acta 1864(1):129449. https://
anti-Bcl-2 clinical trials fail: a solution. Cancer doi.org/10.1016/j.bbagen.2019.129449
Metastasis Rev 33(1):285–294. https://doi.
org/10.1007/s10555-013-9450-8 22. Kopitz J, Fı́k Z, André S, Smetana K Jr, Gabius
H-J (2013) Single-site mutational engineering
14. Arthur CM, Baruffi MD, Cummings RD, and following monoPEGylation of the human
Stowell SR (2015) Evolving mechanistic lectin galectin-2: effects on ligand binding,
insights into galectin functions. Methods Mol functional aspects, and clearance from serum.
Biol 1207:1–35. https://doi.org/10.1007/ Mol Pharm 10(5):2054–2061. https://doi.
978-1-4939-1396-1_1 org/10.1021/mp4000629
15. Hong M-H, Weng I-C, Liu F-T (2018) Galec- 23. Gabius H-J (1990) Influence of type of linkage
tins as intracellular regulators of cellular and spacer on the interaction of β-galactoside-
responses through the detection of damaged binding proteins with immobilized affinity
ligands. Anal Biochem 189(1):91–94. https:// Akt2, phosphatidylinositol 3-kinase, and

doi.org/10.1016/0003-2697(90)90050-j PTEN are in lipid rafts of intestinal cells: role
24. Kaltner H, Seyrek K, Heck A, Sinowatz F, in absorption and differentiation. Gastroenter-
Gabius H-J (2002) Galectin-1 and galectin-3 ology 126(1):122–135. https://doi.org/10.
in fetal development of bovine respiratory and 1053/j.gastro.2003.10.061
digestive tracts. Comparison of cell type- 28. Kopitz J, von Reitzenstein C, Burchert M,
specific expression profiles and subcellular Cantz M, Gabius H-J (1998) Galectin-1 is a
localization. Cell Tissue Res 307(1):35–46. major receptor for ganglioside GM1, a product
https://doi.org/10.1007/s004410100457 of the growth-controlling activity of a cell sur-
25. Saal I, Nagy N, Lensch M, Lohr M, Manning face ganglioside sialidase, on human neuroblas-
JC, Decaestecker C, André S, Kiss R, Salmon I, toma cells in culture. J Biol Chem
Gabius H-J (2005) Human galectin-2: expres- 273(18):11205–11211. https://doi.org/10.
sion profiling by RT-PCR/ 1074/jbc.273.18.11205
immunohistochemistry and its introduction as 29. Ludwig A-K, Michalak M, Xiao Q, Gilles U,
a histochemical tool for ligand localization. Medrano FJ, Ma H, FitzGerald FG, Hasley
Histol Histopathol 20(4):1191–1208. WD, Melendez-Davila A, Liu M, Rahimi K,
https://doi.org/10.14670/HH-20.1191 Kostina NY, Rodriguez-Emmenegger C,
26. Wang WY, Malcolm BA (1999) Two-stage Möller M, Lindner I, Kaltner H, Cudic M,
PCR protocol allowing introduction of multi- Reusch D, Kopitz J, Romero A, Oscarson S,
ple mutations, deletions and insertions using Klein ML, Gabius H-J, Percec V (2019)
QuikChange™ site-directed mutagenesis. Bio- Design-functionality relationships for
Techniques 26(4):680–682. https://doi.org/ adhesion/growth-regulatory galectins. Proc
10.2144/99264st03 Natl Acad Sci U S A 116(8):2837–2842.
27. Li X, Leu S, Cheong A, Zhang H, Baibakov B, https://doi.org/10.1073/pnas.1813515116
Shih C, Brinbaum MJ, Donowitz M (2004)
Chapter 16
Investigation of Galectins in Frozen Tissue and Mammalian

Cell Culture Using Confocal Miccroscopy
Daniel Giuliano Cerri, Lilian Cataldi Rodrigues, Marise Lopes Fermino,
Marcelo Papoti, Richard D. Cummings, Sean R. Stowell,
and Marcelo Dias-Baruffi
Abstract
Galectins are multifunctional glycan-binding proteins present in various tissues that participate in multiple
physiological and pathological processes and are considered as not only biomarkers of human diseases but
also molecular targets for treating cancer and inflammatory illnesses in many organs. In the glycobiology
field, it is crucial to determine the pattern of galectin expression and location in cells and tissues. Confocal
microscopy is a powerful imaging technology that represents a unique approach to investigate the expres-
sion and location of biomolecules in various tissues and cells. The confocal microscope acquires images of
the specimen through the reflected or fluorescent light from the objective’s focal plane, using laser light
focused on a small spot inside the tissue or cell. This technique provides high-resolution and high-contrast
images without artifacts generated by conventional microscopy and enables reconstruction of virtual
tridimensional images by acquiring multiple sections from several focal planes, which makes it possible to
obtain the precise spatial location of any cellular structure or molecule. Furthermore, confocal microscopy
is a non-invasive tissue imaging strategy used in clinical practices. We describe herein the immunofluores-
cence confocal method for examining galectins in frozen tissue sections and mammalian cell culture.
Key words Confocal microscopy, Galectins, Tissue and cell culture, Immunofluorescence
1 Introduction
Galectins are mammalian β-galactoside-binding proteins bearing a

conserved amino acid sequence named carbohydrate recognition
domain (CRD) that accounts for most of their functions [1–
3]. Galectins are expressed in various tissues and immune cells,
such as macrophages, neutrophils, and dendritic cells, and are
found in both intracellular (cytoplasm and nucleus) and extracellu-
lar (cell surface and serum) compartments [4, 5]. Interestingly, the
galectin interaction with endogenous glycans can mediate cell-cell,
cell-matrix, or cell-pathogen binding, which play many roles in
physiological and pathological processes [6–9]. Galectins also
289
290 Daniel Giuliano Cerri et al.
display intracellular activities involving pre-mRNA splicing and a

multi-task lectin that can trigger innate immune modulation sig-
naling [3, 10].
Although 15 members of the galectin family have been
described [3], galectin-1 is the most ubiquitously expressed mem-
ber of this family. Mammalian galectin-1 was detected in skeletal
and striated muscle, liver, lung, brain, kidney, spleen, intestine, and
arterial walls, cytosol, and nucleus of cultured human endothelial
cells [11, 12]. Most of the studies that identified galectin members
were performed using the well-known confocal methodology, an
established microscopy for imaging fluorescently labeled cells or
tissues that enabled a more accurate three-dimensional structure
analysis [11–16].
Confocal microscopy consists of directly or indirectly targeting
the desired molecule with a fluorescent probe (for a schematic view
of the process, see Fig. 1a). The conventional wide-field microscope
(epifluorescent) has a significant drawback for image capture: the
out-of-focus light is not blocked and hazes the captured image,
resulting in a blurred low resolution image [17, 18]. Laser Scan-
ning Confocal Microscopy (LSCM) adds the possibility to collect
light from the objective’s focal plane and eliminates blurring caused
by out-of-focus light, which enables visualization of finer details
than conventional systems [17]. Confocal software makes it possi-
ble to set up the bottom and top regions of the probed biological
specimen, as well as to take several images between the Z-axis (also
named Z-stack), which enables reconstitution of three-dimensional
structure [17]; a simplified scheme of LSCM advantages is depicted
in Fig. 1b. Moreover, the biological sample can be probed with
multiple fluorochromes. LSCM can manage multiple laser beams,
emit light bandwidths, and be used for quantitative measurements
in biological specimens [19] and time-resolving experiments that
provide dynamic information about cellular physiology. It is worth
to note that several new techniques associated with confocal
microscopy, per se, have been developed to improve and refine
data analysis, such as STimulated Emission Depletion microscopy
(STED). STED adds a new perspective to optical microscopy: a
nanoscale resolution that has been used for capturing micrographs
from immunofluorescence or fluorescence protein procedures [20–
24]. The confocal microscope may also be coupled to a Raman
spectrometer to greatly expand the variety of data acquired [25]. In
vivo confocal microscopy is another excellent update that consists
of a non-invasive clinical approach for inflamed tissues, such as
human cornea [26], focusing on their histological architecture.
Besides, bacterial biofilms matrix has been characterized by confo-
cal microscopy [27].
The use of immunofluorescence followed by LSCM image
capture for analyzing the location and expression of galectin family
proteins may broaden the knowledge on Glycobiology, including
Investigation of Galectins in Frozen Tissue and Mammalian Cell Culture. . . 291
Fig. 1 Simplified schematic view of Immunofluorescence assay (a) and Microscope image capture (b). In (a), a
biological specimen is processed to probe the desired protein/cell structure. In (b) wide-field microscope
(at left) is compared with LSCM (at right); the best image qualities are presented, as well as the possibilities of
three-dimensional reconstruction of the structure
studies covering either physiological or pathological conditions.

Here, we report galectin-1 location in brain, cardiac muscle, liver,
and stomach (Fig. 2), and galectin-3 in myoblast primary cell
culture (Fig. 3). Some galectins have more selective expression
patterns, such as galectin-7 and -12, which are found in epithelial
and adipose tissue, respectively [9, 28, 29]. Galectin-4 is mainly
expressed in intestinal epithelial cells, and its downregulation is
associated with colorectal cancer prevalence [30]. On the other
hand, galectin-4 is an important biomarker for pancreatic ductal
adenocarcinoma, suggesting diversity in its expression and location
[31]. Hence, the development of confocal microscopy increased
and improved the understanding of location, trafficking, and inter-
actions of galectins and several biomolecules in cells and tissues
[17, 32, 33].
Nevertheless, it is worth highlighting that slight modifications
in the protocol can markedly alter the results. For example, it is well
known that there should be an intricate balance between a well-
preserved cell/tissue structure and its chemical composition, and
that the best antigenic site preservation is controversially faced to
cell morphology structure [34]. As a typical example of this last fact
mentioned above, we present the results of an assay performed in
Fig. 2 Gal-1 exhibits diffuse cytosolic location in various adult tissues. Frozen sections of porcine tissue were
subjected to confocal analysis using anti-Gal-1 primary (αhGal-1) and secondary antibody (αhGal-1 + 2 ).
α-hGal-1 followed by secondary stain in (a) Liver (HP, hepatic plate; P, portal space); (b) Stomach (Ep,
epithelium cells forming gastric glands; MC, mucous cell); (c) cardiac muscle; and (d) Brain (N, neuropil and
glia; NC, nerve cells). Bar ¼ 25 μm. (The data here reported were originally published in Dias-Baruffi M,
Stowell SR, Song SC, Arthur CM, Cho M, Rodrigues LC, Montes MA, Rossi MA, James JA, McEver RP,
Cummings RD. Differential expression of immunomodulatory galectin-1 in peripheral leukocytes and adult
tissues and its cytosolic organization in striated muscle. Glycobiology. 2010 May; 20(5):507–20. Oxford
University Press [12])
human muscle with and without chemical fixation (Fig. 4), demon-
strating that attention should be paid to outlining the correct
method to maximize image quality and gather fruitful results.
LSCM has multiple possible configurations due to the recent
advances, ranging from the unique location of a single protein/
structure to non-invasive applications in medical care. This current
chapter presents material and methods for preparing immunofluo-
rescence samples for detecting galectins in the tissue-frozen sec-
tions and mammalian cell culture.
Fig. 3 Galectin-3 location in myoblast-myotube cell culture. Triple fluorescence for Nucleus (a), Transcription
muscle factor—MyoD (b), and Galectin-3 (c) in murine myoblast primary cell culture. The images from (d–f)
were merged two by two (d—Nuclei and MyoD; e—Galectin-3 and MyoD; f—Galectin-3 and nuclei) for better
understanding, and a final merged image with the three staining is depicted (f). See Note 26. Bar ¼ 25 μm
Fig. 4 Comparison among confocal images from human muscle frozen section.
For dystrophin, lipid droplets and nuclei—without chemical fixation (a and b) and
treated with Zamboni’s fixative (c, d, and e). The traditional method resulted in
no BODIPY staining (a) but preserved better dystrophin immunostaining (b, in
red). The use of Zamboni’s fixative resulted in a well-organized sarcoplasmic
pattern staining (c, d and Zoom in e—in green) probed with BODIPY. See Note
27. (a–d, Bar ¼ 25 μm and e, Bar ¼ 10 μm)
2 Materials
2.1 Tissue Technique 1. Gelatin from bovine source, Bloom 225 (Sigma, St
Louis, MO).
Glass Slides 2. ddH2O.
3. Chromium (III) potassium sulfate (chrome alum (Sigma, St

Louis, MO)).
4. General-purpose filter paper (Sigma, St Louis, MO).
5. Glass slides (Knitel, Braunschweig, Germany).
6. Poly-L-lysine 0.1% (w/v) in water (Sigma, St Louis, MO).
7. (Optional) Fresh Zamboni’s fixative: 0.73% Formaldehyde plus
0.035% picric acid in 142 mM Na2HPO4 and 40 mM
NaH2PO4 (see Note 1).
2.1.2 Processing of 1. Surgical material for tissue sampling (tweezers, scissors, etc.).
Tissue 2. PBS 1 (137 mM NaCl, 2.7 mM KCl, 4.4 mM Na2HPO4, and
1.4 mM KH2PO4 pH ¼ 7.3) and PBS 10.
3. Paraformaldehyde 4% in PBS (Sigma, St Louis, MO).
4. Sucrose 30–50% in PBS (w/v) (Fisher, Hampton, NH).
2.1.3 Cryogenic 1. OCT Compound—Tissue-Tek (Sakura Finetek, Netherlands).

Technique 2. (Isopentane (Fisher, Hampton, NH) and liquid nitrogen) or
(acetone (Fisher, Hampton, NH) and dry ice).
3. Polyethylene embedding cassettes (Fisher, Hampton, NH) or
cryogenic vials (Corning Incorporated Life Sciences,
Tewksbury, MA).
2.1.4 Preparation of 1. OCT Compound—Tissue-Tek (Sakura Finetek, Netherlands).

Sample Sectioning and 2. Pencil.
Mounting
3. Cryostat (Leica CM 1850, Wetzlar, Germany).
4. Gelatin-coated glass slides (Subheading 3.1.1).
2.1.5 Tissue 1. 0.25 M glycine in PBS 1.

Immunofluorescence 2. Hydrophobic barrier pen (Vector Laboratories,
Burlingame, CA).
3. Bovine Serum Albumin—Fraction V (BSA) (Sigma, St
Louis, MO).
4. DAPI—4,6-diamidino-2-phenylindole, dihydrochloride
(Molecular Probes/Life Technologies, Grand Island, NY).
5. Triton X-100 (Fisher, Hampton, NH).
6. Tween™ 20 (Fisher, Hampton, NH).
7. Fc receptor Blocker—ChromPure whole molecule
non-immune IgG from the same animal source as your second-
ary antibody (Jackson ImmunoResearch, West Groove, PA).
8. Mounting medium (ProLong Antifade Mountants series—
Thermo Fisher Scientific Waltham, Massachusetts, EUA,
Fluoromount G—Electron Microscopy Sciences, Hatfield, PA
or any similar product).
9. Primary antibody.
10. Secondary antibody.
11. PBS 1 (see Note 2).
12. Refrigerated microcentrifuge.
2.2 Cell Culture 1. 12- or 13-mm coverslips.

Immunofluorescence 2. Glass slides (Knitel, Braunschweig, Germany).
Technique
3. 24-Well, sterile flat-bottom plate.
4. Ultra-Thin Fine Tip Curved Tweezers, Stainless Steel.
5. (Optional) Matrix substratum (e.g., Matrigel®, Fibronectin or
other available).
6. Fixative: 2–4% fresh formaldehyde solution.
(a) (Optional) Additives to Fixative: 50 μm Paclitaxel plus
50 mm EGTA in 2–4% fresh paraformaldehyde solution
(see Note 3).
(b) Optional Fixatives: coagulant fixatives such as organic
solvents (Methanol or acetone at 20 C) may be used
instead of traditional formaldehyde solution (see Note 4).
7. Blocking solution: Bovine Serum Albumin—Fraction V (BSA)
(Sigma, St Louis, MO).
8. Nuclear staining: DAPI-4,6-diamidino-2-phenylindole, dihy-
drochloride (Molecular Probes/Life Technologies, Grand
Island, NY).
(a) Nuclear staining for non-permeabilized cells: HOESCHT
33342 (BisBenzimide H 33342 trihydrochloride—Merck
KGaA, Darmstadt, Germany).
9. Detergent solution for permeabilization: Triton X-100 (Fisher,
Hampton, NH).
(a) (Optional detergents): Saponin, Nonidet P-40, Tween-
20, Brij 35, etc. (see Note 5).
10. Fc receptor Blocker—ChromPure whole molecule
non-immune IgG from the same animal source as your second-
ary antibody (Jackson ImmunoResearch, West Groove, PA).
11. Mounting medium (ProLong Antifade Mountants series—
Thermo Fisher Scientific Waltham, Massachusetts, EUA,
Fluoromount G—Electron Microscopy Sciences, Hatfield,
PA, or any similar product).
12. Primary antibody.
13. Secondary antibody.
14. PBS 1 (see Note 2).
15. Refrigerated microcentrifuge.

16. Multidispenser pipette with syringe (e.g., Eppendorf Repeater
M4 pipette, 1 μL–10 mL Dispensing Volume—Eppendorf,
Hamburg, Germany).
17. Vacuum pump coupled to a reservoir to aspirate solutions in
between all the steps (see Note 6).
3 Methods
3.1 Tissue Technique 1. Clean glass slides thoroughly with ddH2O and identify indi-
vidual slides using a pencil (see Note 7).
Glass Slides 2. Prepare the gelatin solution by dissolving 2.5 g of gelatin into
500 mL of ddH2O heated to 60 C to make a 0.5% solution
(Do not exceed 60 C).
3. After gelatin dissolution, add 0.25 g of Chromium (III) potas-
sium sulfate (chrome alum) to make 0.05% chrome alum solu-
tion (see Note 8).
4. Filter the solution using general-purpose filter paper.
5. Dip the slide once into the gelatin-coating solution for few
seconds.
6. Place gelatinized glass slides at approximately an 80 angle to
drain the excess solution onto a paper towel. Cover the slides
with aluminum foil to keep off dust.
7. Allow the slides to dry at room temperature for 2 days or
overnight at 37 C. After drying, the slides can be stored in
the refrigerator (4–8 C).
8. Discard gelatin solution after use.
9. Alternatively, the glass slides may be prepared by coating with
0.1% Poly-L-lysine (see Note 9).
3.1.2 Processing of 1. Tissue sampling: Excise the desired tissue and wash the biopsy
Tissue briefly in PBS. If freezing tissue on the same day as tissue
excision, then proceed directly to Subheading 3.1.3 Cryogenic
technique. Otherwise proceed to step two (see Note 10).
2. Tissue fixation: Place excised tissue into a tissue cassette and fix
in fresh 4% paraformaldehyde in PBS for 4 h at room tempera-
ture (see Notes 11–13).
3. Post-fixation treatment: To improve the cryopreservation of
paraformaldehyde-fixed tissue, these materials should be incu-
bated in sucrose solution (5–30%) and optimal cutting temper-
ature (OCT) compound Tissue-Tek (see Note 14).
3.1.3 Cryogenic 1. Cryogenic technique—Samples should be placed into a poly-

Technique ethylene embedding cassette (alternatively, cryogenic tubes can
be used) filled with OCT. Be sure that the tissue biopsy is in the
correct orientation for future sectioning (transversal, longitu-
dinal, or oblique) (see Notes 15 and 16).
2. Freeze the embedded sample using one of the two freezing
techniques described below: (see Note 17).
(a) Acetone plus dry ice: Place dry ice into a single
250–500 mL Becker and very slowly add acetone to
cover the ice. Wait for 5 min for acetone to chill and
then, with the aid of tweezers, dip the cassette
(or cryogenic tube) and wait for at least 2 min to ensure
the sample has frozen completely (see Note 18).
(b) Liquid nitrogen plus isopentane: Fill a polystyrene con-
tainer (Nalgene, Rochester, NY) with liquid nitrogen to
approximately ¾ of its total volume. Use tongs to lower a
stainless steel or polypropylene container of isopentane
into the liquid nitrogen. Begin chilling isopentane, at
least 30 min before freezing samples. As isopentane
nears, freezing it will become opaque. With the aid of
tweezers, dip the cassette (or cryogenic tube) into the
chilled isopentane and wait for at least 2 min to ensure
the sample has frozen completely.
3.1.4 Preparation of 1. Carefully detach the frozen sample block from the cassette and
Sample Sectioning and identify it with a small piece of paper or other appropriate
Mounting marking.
2. Handle the cryostat according to the manufacturer guidelines.
3. Fasten the block into the block holder of the cryostat and
adjust the tissue thickness for 5–10 μm.
4. Slice 5–10 μm repeatedly until a ribbon includes tissue forms
(see Note 19).
5. Place a previously gelatin-coated slide into position so that the
ribbon containing the desired tissue section settles onto the
slide as it is sliced.
6. If your primary antibody is sensitive to paraformaldehyde fixa-
tion, antigen retrieval may be required. Otherwise, proceed
directly to Subheading 3.1.5 (see Note 20).
3.1.5 Tissue 1. If tissue was fixed in formaldehyde/paraformaldehyde (step

Immunofluorescence 2 of Subheading 3.2), then block free formaldehyde/parafor-
maldehyde residues by placing tissue in 0.25 M glycine in PBS
for 5 min at room temperature. Otherwise, proceed as follows:
2. Determine appropriate antibody conditions (see Notes
21–24).
3. With a hydrophobic barrier pen, manually circumvent the tis-

sue to restrain the antibody solution around the tissue during
the incubation period.
4. All subsequent steps should be performed in a moisture cham-
ber at room temperature.
5. Block tissue sections in 500 μL PBS containing 2% BSA and
0.5% (v/v) Triton X-100 plus Fc blocker (we used 50 μg/mL
donkey IgG) for 45 min.
6. Primary antibody should be prepared in PBS with 2% BSA.
7. Remove blocking buffer and add primary antibody at appro-
priate concentration to slide. Incubate for 1 h.
8. Wash slides two times for 15 min in PBS plus 0.1% (v/v) Triton
X-100 and then two times in PBS plus 0.05% (v/v) Tween 20.
9. Secondary antibody should be prepared in PBS with 2% BSA
and 50 ng/mL DAPI. The optimal concentration of secondary
antibody should be provided by the manufacturer.
10. Incubate slide with secondary antibody for 1 h.
11. Wash slides four times for 15 min each in PBS plus 0.05% (v/v)
Tween 20.
12. Mounting slides: Dip the slide briefly in Milli-Q water. With
the aid of a soft paper towel, dry around the sample very
carefully.
13. To mount the coverslip over the tissue section, gently add
20–30 μL of Fluoromount, directly to the slide and carefully
lay the coverslip over the tissue section taking care to avoid
bubbles.
14. Dry gently around the coverslip and seal with colorless nail
polish (with no formaldehyde in its formulation) to prevent
drying and displacement of biological specimen.
15. Observe the prepared tissue slide using a conventional fluores-
cence microscope to determine whether fluorescence is detect-
able in the sample.
16. In the case of positive results, place them in a LCSM and
proceed as indicated by manufacturer (see Note 25).
3.2 Cell Culture 1. Seed 1.0 up to 5.0 104 cells onto 13 mm round coverslips
Immunofluorescence (pretreatment with some extracellular matrix component is
Technique optional).
2. In case of several wells, when changing any solution, it is
strongly recommended to use the pump system (Subheading
2.2) adjusted in a moderate velocity of aspiration—the dispos-
able tip used in the silicon/latex hose should be put on the
edge of the well; the new solution should be given out carefully
by a multidispenser (Subheading 2.2).
3. Perform your assay as previously outlined (with some drug/

protein/transfection treatment in cell culture or not).
4. Fix cells with fixative for 20 min at 37 C.
5. Permeabilized with 0.3% Triton in PBS for 10 min at RT.
(a) (Optional) For analyzing antigens present only on the cell
surface this step should be skipped.
6. Incubate with 0.1% Glycine in PBS for 5 min at RT to eliminate
any formaldehyde residue.
7. Add 250 μL of blocking solution and incubate for 45 min.
8. Wash cells with PBS1x for 5 min—Repeat twice.
9. Pipet the primary(ies) antibody(ies) diluted in PBS + 1% BSA
(Sigma-Aldrich) for 1 h at RT.
10. Rinse thoroughly in PBS for 5 min. Repeat 4 times.
11. Pipet the secondary(ies) antibody(ies) diluted in PBS plus
0.4 μM DAPI. For nuclei staining on non-permeabilized
cells, use 1 μg/mL Hoechst 33342.
12. Repeat step 9.
13. In glass slides, annotate the assay’s identification and pipet
20 μL of mounting medium on each one.
14. With the aid of a fine tweezer, remove the coverslip from the
well and rinse it quickly in dH2O. Remove the excess of water
by touching the edge of coverslip in a tissue paper.
15. Pour delicately the coverslip with cells faced to the glass.
16. Place a tissue paper to absorb the excess of mounting medium.
17. Seal the edges of coverslip with nail polish without
formaldehyde.
4 Notes
1. Picric acid when added to formaldehyde solution aids cell

membrane structure preservation by coagulation of proteins
into positively charged salts, retaining their antigenicity [33].
2. Alternatively, you may use TBS (10 mM Tris and 100 mM
NaCl pH ¼ 8.0).
3. When staining microtubules or some structure depending on
it, you may add paclitaxel to the fixative for stabilizing the
microtubules tridimensional structure; EGTA will complex
with calcium, diminishing the action of calcium-dependent
proteases.
4. Organic solvents are strong fixatives, coagulating proteins and
extracting carbohydrates and lipids. They also offer an “one-
step” fixation and permeabilization, so if you choose this
method, the use of detergent should be obliterated. Remember

that when using acetone, the coverslips should be transferred
to a silicone-supported cover-glass to avoid plastic dissolution
of the 24-well plate during the fixation period.
5. The best detergent for your application should be determined
experimentally or by searching in the literature.
6. The vacuum pump must be set to moderate suction; the dis-
posable tip could be a P200 tip or a glass bulb/dropper.
7. When marking glass slides, pencil should always be used. Mark-
ings made using pens are frequently erased during immunoflu-
orescence techniques.
8. Chrome Alum will positively charge the slides allowing them to
attract negatively charged tissue sections.
9. For coating glass slides with poly-L-lysine, drop 10 μL of
solution in the middle of a glass slide and distribute it evenly
over the surface by the aid of another moving glass slide.
Air-dry for 5 min at room temperature.
10. As this chapter focuses on confocal analysis of galectin localiza-
tion within tissue, unique considerations should be employed
when examining galectin localization in cells in suspension.
Cells in suspension can be spread on a glass slide using routine
blood smear techniques or overlaid on a poly-L-lysine coated
coverslip [35, 36]. When using this approach, evaluation of
antibody engagement over a range of concentrations using
flow cytometric examination may be beneficial to insure ade-
quate saturation of the target antigen when seeking to ascertain
potential changes under different conditions [36, 37]. If fixa-
tion is required, this process typically follows adhesion of cells
to the slide or coverslip followed by employing similar consid-
erations as are utilized when staining a tissue section
[38]. However, as cells may not adhere to the slide or coverslip
as well as a tissue section, great care should be taken when
incubating and washing the slide to not remove the cells.
11. Several galectins, including galectin-1, galectin-2, and galectin-
3 can undergo rapid oxidation following removal from a reduc-
ing environment [4, 35, 39]. Galectin oxidation results in
significant conformational changes that may preclude antibody
recognition depending on the antibody employed and the
galectin being examined [12, 35]. As a result, significant care
should be taken to rapidly process tissue immediately after
harvesting to reduce the potential impact of galectin oxidation
on detection and localization of these proteins.
12. To prepare 100 mL of 8% paraformaldehyde (2) solution,
add 100 μL 1.0 M NaOH into a glass beaker containing 80 mL
of water. Add 8.0 g of paraformaldehyde with agitation and
heating to 60 C (be sure to keep a cap on or heat under a fume

hood). Wait until complete solubilization (clear solution).
Cool solution in ice until it reaches room temperature. Add
10 mL of PBS 10 and complete to 100 mL with water. Mix
and filter the resulting solution with a 0.2 μm pore size filter.
Prepare a fresh 4% paraformaldehyde solution by diluting 8%
paraformaldehyde in PBS (1). The time of incubation may
vary according to the kind and size of tissue (over 4 h to several
days). Search literature to determine the ideal time for tissue
fixation.
13. The desired concentration and content of paraformaldehyde
solution may vary depending on the chosen protein or struc-
ture that is being observed. (e.g., for microtubules visualiza-
tion 4% paraformaldehyde plus 50 mM EGTA and 50 μM
Taxol preserves cytoskeleton structure). Always consult litera-
ture for appropriate conditions.
14. The post-fixation step in sucrose solution should not be per-
formed if your tissue was not fixed in formaldehyde/parafor-
maldehyde, to avoid shrinking of non-fixed tissue. Following
chemical fixation, tissue samples are sequentially immersed in
solutions of sucrose (5, 10, 20, and 30%) in PBS, pH 7.2. The
transition from a concentration of sucrose solution to another
should occur after chemical tissue fixation. Then transfer the
tissue sample into a cassette (or cryogenic tube) filled with
OCT. This procedure is indicated for tissues such as brain
and skeletal muscle.
15. Make sure to mark the facilities identification and correct ori-
entation of particular tissue pieces or surfaces during embed-
ding and subsequent microscopical examination. Cassettes
should be marked before embedding to establish orientation.
Tissue blocks are simply marked after freezing by cutting a
notch on the reverse side of the block face to be sectioned, or
by trimming the block to a particular shape.
16. For some special tissues such as bones, wait for at least 1 h at
4 C in OCT for best infiltration, before freezing.
17. It is important to freeze biological sample until it has been
completely frozen. Freezing will begin on the outside and
move inward. The process will be complete when tissue plus
OCT becomes milky-white in appearance.
18. Be careful when adding acetone directly to dry ice. It will
“boil” fast, splashing acetone in all directions. Do not permit
the acetone to contact your sample directly.
19. Be sure to obtain additional sections for hematoxylin and eosin
staining using commonly employed approaches to provide a
histological reference for each slide [40].
20. Fixation should immobilize antigens while retaining cellular

and subcellular structure and permeabilize samples to allow
for access of antibodies to all cells and subcellular compart-
ments. Cross-linking reagents such as paraformaldehyde serve
this purpose, but may reduce the antigenicity of some cell
components as the cross-linking can obstruct antibody bind-
ing. As a result, antigen retrieval techniques may be required.
We use an additional acetone fixation if antigen retrieval is
needed. For acetone fixation, dip the slide into 20 C acetone
for 10 min. Before proceeding with immunofluorescence,
rehydrate sections with PBS (1). Remember that acetone
fixation will promote cell permeabilization.
21. To confirm antibody specificity, perform immunoblotting with
its specific purified protein and/or tissue/cell extract to verify
if the protein recognized is of the expected size. In addition,
antibody specificity may be demonstrated by performing the
immunofluorescence stain with primary antibody that has been
previously neutralized with its specific purified protein.
22. Centrifuging the diluted solutions of antibodies at maximum
speed (~20,000 g) for 30 min at 4 C before using them can
help to avoid undesired background during the immunofluo-
rescence procedure.
23. It is important to use the minimal concentration of primary
antibody to achieve a good staining. As a result, it is recom-
mended to perform some test staining with multiple concen-
trations. If the antibody is a purchased antibody, begin with the
minimum concentration indicated on the product data-sheet.
Usually, we see good staining by using 10 μg/mL of purified
antibody.
24. Negative controls such as absence of primary antibody or use of
an isotype control should be included in the staining protocol
in order to achieve interpretable results.
25. When using the confocal microscope, select the best para-
meters for your samples, such as a low laser power and a
quick scan, to avoid photobleaching. The pinhole should be
set to the optimal size for each objective. Care should be taken
to avoid overlapping signals between fluorochromes by setting
each channel excitation power and emission bandwidth cor-
rectly. The PMT (Photo Multiplier Tube) gain should be high
enough to have a bright image without saturated pixels. If the
signal is weak, increasing laser power at maximum may be
useless, because you probably will create a false positive image.
26. The authorization for using murine muscle as showed in Fig. 3
was approved by the Ethics Committee on Animal Use
(CEUA) of University of São Paulo in the process number
2014.1.478.53.4.
27. The authorization for using human muscle as showed in Fig. 4

was approved by the Ethics Committee on Human use of
University of São Paulo in the process CAAE
60300316.6.0000.5659.
Acknowledgments
We thank the technicians Maria Vani Alves (Department of Cell

Biology and Pathogenic agents from Faculty of Medicine of
Ribeirão Preto—University of São Paulo), Márcia Sirlene Zardin
Graeff (Research Center-Faculty of Odontology of Bauru—Uni-
versity of São Paulo), Eduardo Tozatto (School of Pharmaceutical
Sciences of Ribeirão Preto, University of São Paulo, Ribeirão Preto,
SP, Brazil), and Danilo Rodrigues Bertucci (Scholl of Physical
education and Sport of Ribeirão Preto—University of São Paulo)
for their invaluable assistance. We also thank to the physicians
Dr. Marcelo Riberto (Department of Biomechanics, Medicine
and Rehabilitation of the locomotor system from Faculty of Medi-
cine of Ribeirão Preto—University of São Paulo) and Dr. Marcello
Teixeira Castiglia (Orthopedic surgeon) whom performed human
biopsy. We are thankful to the Brazilian National Council for Scien-
tific and Technological Development (CNPq - #312606/2019-
2 for M.D.B.).
References
1. Barondes SH, Castronovo V, Cooper DNW, 5. Vasta GR, Feng C, González-Montalbán N,
Cummings RD, Drickamer K, Felzi T, Gitt Mancini J, Yang L, Abernathy K, Frost G,
MA, Hirabayashi J, Hughes C, Kasai K et al Palm C (2017) Functions of galectins as ‘self/
(1994) Galectins: a family of animal non-self’-recognition and effector factors.
β-galactoside-binding lectins. Cell Pathog Dis 75(5):ftx046. https://doi.org/10.
76(4):597–598. https://doi.org/10.1016/ 1093/femspd/ftx046
0092-8674(94)90498-7 6. Thiemann S, Baum LG (2016) Galectins and
2. Cooper DNW, Barondes SH (2000) immune responses-just how do they do those
Galectins. In: Carbohydrates in chemistry and things they do? Annu Rev Immunol 34:
biology, vol 2000, pp 625–647. https://doi. 243–264. https://doi.org/10.1146/annurev-
org/10.1002/9783527618255.ch77 immunol-041015-055402
3. Varki A, Cummings RD, Esko JD, Stanley P, 7. Murphy P, André S, Gabius H-J (2013) The
Hart GW, Aebi M, Darvill AG, Kinoshita T, third dimension of Reading the sugar code by
Packer NH, Prestegard JH, Schnaar RL, See- lectins: design of glycoclusters with cyclic scaf-
berger PH (eds) (2015–2017) Essentials of folds as tools with the aim to define correla-
glycobiology [Internet], 3rd edn. Cold Spring tions between spatial presentation and activity.
Harbor Laboratory Press, Cold Spring Molecules 18(4):4026–4053. https://doi.
Harbor, NY org/10.3390/molecules18044026
4. Stowell SR, Qian Y, Karmakar S, Koyama NS, 8. Davicino RC, Eliçabe RJ, Di Genaro MS, Rabi-
Dias-Baruffi M, Leffler H, McEver RP, Cum- novich GA (2011) Coupling pathogen recog-
mings RD (2008) Differential roles of nition to innate immunity through glycan-
Galectin-1 and Galectin-3 in regulating leuko- dependent mechanisms. Int Immunopharma-
cyte viability and cytokine secretion. J Immu- col 11(10):1457–1463. https://doi.org/10.
nol 180(5):3091–3102. https://doi.org/10. 1016/j.intimp.2011.05.002
4049/jimmunol.180.5.3091
9. Di Lella S, Sundblad V, Cerliani JP, Guardia Eggeling C, Hell SW (2006)

CM, Estrin DA, Vasta GR, Rabinovich GA Macromolecular-scale resolution in biological
(2011) When galectins recognize glycans: fluorescence microscopy. Proc Natl Acad Sci
from biochemistry to physiology and back U S A 103(31):11440–11445. https://doi.
again. Biochemistry 50(37):7842–7857. org/10.1073/pnas.0604965103
https://doi.org/10.1021/bi201121m 19. Jonkman J, Brown CM, Wright GD, Anderson
10. Cooper D, Iqbal AJ, Gittens BR, Cervone C, KI, North AJ (2020) Tutorial: guidance for
Perretti M (2012) The effect of galectins on quantitative confocal microscopy. Nat Protoc
leukocyte trafficking in inflammation: sweet 15(5):1585–1611. https://doi.org/10.1038/
or sour? Ann N Y Acad Sci 1253:181–192. s41596-020-0313-9
https://doi.org/10.1111/j.1749-6632.2011. 20. Leung BO, Chou KC (2011) Review of super-
06291.x resolution fluorescence microscopy for biology.
11. Cerri DG, Rodrigues LC, Stowell SR, Araujo Appl Spectrosc 65(9):967–980. https://doi.
DD, Coelho MC, Oliveira SR, Bizario JCS, org/10.1366/11-06398
Cummings RD, Dias-Baruffi M, Costa MCR 21. Dyba M, Jakobs S, Hell SW (2003) Immuno-
(2008) Degeneration of dystrophic or injured fluorescence stimulated emission depletion
skeletal muscles induces high expression of microscopy. Nat Biotechnol
Galectin-1. Glycobiology 18(11):842–850. 21(11):1303–1304. https://doi.org/10.
https://doi.org/10.1093/glycob/cwn079 1038/nbt897
12. Dias-Baruffi M, Stowell SR, Song S-C, Arthur 22. Willig KI, Kellner RR, Medda R, Hein B,
CM, Cho M, Rodrigues LC, Montes MAB, Jakobs S, Hell SW (2006) Nanoscale resolution
Rossi MA, James JA, McEver RP et al (2010) in GFP-based microscopy. Nat Methods
Differential expression of immunomodulatory 3(9):721–723. https://doi.org/10.1038/
galectin-1 in peripheral leukocytes and adult nmeth922
tissues and its cytosolic organization in striated 23. Roobala C, Ilanila IP, Basu JK (2018) Applica-
muscle. Glycobiology 20(5):507–520. tions of STED fluorescence nanoscopy in unra-
https://doi.org/10.1093/glycob/cwp203 velling nanoscale structure and dynamics of
13. Bayguinov PO, Oakley DM, Shih C-C, Gea- biological systems. J Biosci 43(3):471–484.
non DJ, Joens MS, Fitzpatrick JAJ (2018) https://doi.org/10.1007/s12038-018-
Modern laser scanning confocal microscopy. 9764-3
Curr Protoc Cytom 85(1):e39. https://doi. 24. Badawi Y, Nishimune H (2020) Super-
org/10.1002/cpcy.39 resolution microscopy for analyzing neuromus-
14. Rodrigues LC, Kabeya LM, Azzolini AECS, cular junctions and synapses. Neurosci Lett
Cerri DG, Stowell SR, Cummings RD, 715:134644. https://doi.org/10.1016/j.
Lucisano-Valim YM, Dias-Baruffi M (2019) neulet.2019.134644
Galectin-1 modulation of neutrophil reactive 25. Gomes da Costa S, Richter A, Schmidt U,
oxygen species production depends on the cell Breuninger S, Hollricher O (2019) Confocal
activation state. Mol Immunol 116:80–89. Raman microscopy in life sciences. Morpholo-
https://doi.org/10.1016/j.molimm.2019. gie 103(341):11–16. https://doi.org/10.
10.001 1016/j.morpho.2018.12.003
15. Lingblom C, Andersson K, Wennerås C (2021) 26. Patel DV, Zhang J, McGhee CN (2019) In
Kinetic studies of galectin-10 release from eosi- vivo confocal microscopy of the inflamed ante-
nophils exposed to proliferating T cells. Clin rior segment: a review of clinical and research
Exp Immunol 203:230–243. https://doi. applications. Clin Exp Ophthalmol
org/10.1111/cei.13540 47(3):334–345. https://doi.org/10.1111/
16. Cerri DG, Rodrigues LC, Alves VM, ceo.13512
Machado J, Bastos VAF, Kettelhut IC, Alberici 27. Schlafer S, Meyer RL (2017) Confocal micros-
LC, Stowell SR, Costa MCR, Cummings RD copy imaging of the biofilm matrix. J Microbiol
et al (2021) Endogenous galectin-3 is required Methods 138:50–59. https://doi.org/10.
for skeletal muscle repair. bioRxiv Cell Biol 1016/j.mimet.2016.03.002
1–41. https://doi.org/10.1101/2020.10.01.
322461 28. Gendronneau G, Sidhu SS, Delacour D,
Dang T, Calonne C, Houzelstein D,
17. Conchello J-A, Lichtman JW (2005) Optical Magnaldo T, Poirier F (2008) Galectin-7 in
sectioning microscopy. Nat Methods 2: the control of epidermal homeostasis after
9 2 0 – 9 3 1 . h t t p s : // d o i . o r g / 1 0 . 1 0 3 8 / injury. Mol Biol Cell 19(12):5541–5549.
nmeth815 https://doi.org/10.1091/mbc.e08-02-0166
18. Donnert G, Keller J, Medda R, Andrei MA,
Rizzoli SO, Lührmann R, Jahn R,
29. Yang R-Y, Havel P, Liu F-T (2012) Galectin- formation. J Biol Chem 284(8):4989–4999.
12: a protein associated with lipid droplets that https://doi.org/10.1074/jbc.M808925200
regulates lipid metabolism and energy balance. 36. Stowell SR, Liepkalns JS, Hendrickson JE,
Adipocyte 1(2):96–100. https://doi.org/10. Girard-Pierce KR, Smith NH, Arthur CM,
4161/adip.19465 Zimring JC (2013) Antigen modulation con-
30. Kim SW, Park KC, Jeon SM, Ohn TB, Il Kim T, fers protection to red blood cells from antibody
Kim WH, Cheon JH (2013) Abrogation of through Fcγ receptor ligation. J Immunol
galectin-4 expression promotes tumorigenesis 191(10):5013–5025. https://doi.org/10.
in colorectal cancer. Cell Oncol 4049/jimmunol.1300885
36(2):169–178. https://doi.org/10.1007/ 37. Girard-Pierce KR, Stowell SR, Smith NH,
s13402-013-0124-x Arthur CM, Sullivan HC, Hendrickson JE,
31. Hu D, Ansari D, Zhou Q, Sasor A, Said Zimring JC (2013) A novel role for C3 in
Hilmersson K, Andersson R (2019) Galectin antibody-induced red blood cell clearance and
4 is a biomarker for early recurrence and antigen modulation. Blood
death after surgical resection for pancreatic 122(10):1793–1801. https://doi.org/10.
ductal adenocarcinoma. Scand J Gastroenterol 1182/blood-2013-06-508952
54(1):95–100. https://doi.org/10.1080/ 38. Hernandez JD, Nguyen JT, He J, Wang W,
00365521.2018.1561937 Ardman B, Green JM, Fukuda M, Baum LG
32. Dings R, Miller M, Griffin R, Mayo K (2018) (2006) Galectin-1 binds different CD43 Gly-
Galectins as molecular targets for therapeutic coforms to cluster CD43 and regulate T cell
intervention. Int J Mol Sci 19(3):905. https:// death. J Immunol 177(8):5328–5336.
doi.org/10.3390/ijms19030905 https://doi.org/10.4049/jimmunol.177.8.
33. Malatesta M (2016) Histological and histo- 5328
chemical methods - theory and practice. Eur J 39. Stowell SR, Karmakar S, Stowell CJ, Dias-
Histochem 60:2639. https://doi.org/10. Baruffi M, McEver RP, Cummings RD
4081/ejh.2016.2639 (2007) Human galectin-1, -2, and -4 induce
34. Login GR, Schnitt SJ, Dvorak AM (1987) surface exposure of phosphatidylserine in acti-
Rapid microwave fixation of human tissues for vated human neutrophils but not in activated T
light microscopic immunoperoxidase identifi- cells. Blood 109(1):219–227. https://doi.
cation of diagnostically useful antigens. Lab org/10.1182/blood-2006-03-007153
Investig 57:585–591 40. Rogerio AP, Cardoso CR, Fontanari C, Souza
35. Stowell SR, Cho M, Feasley CL, Arthur CM, MA, Afonso-Cardoso SR, Silva ÉVG, Koyama
Song X, Colucci JK, Karmakar S, Mehta P, NS, Basei FL, Soares EG, Calixto JB et al
Dias-Baruffi M, McEver RP et al (2009) (2007) Anti-asthmatic potential of a D-
Ligand reduces galectin-1 sensitivity to oxida- galactose-binding lectin from Synadenium car-
tive inactivation by enhancing dimer inatum latex. Glycobiology 17(8):795–804.
https://doi.org/10.1093/glycob/cwm053
Chapter 17
Exploring the Galectin Network by Light and Fluorescence

Microscopy
Gabriel Garcı́a Caballero, Joachim C. Manning, Adele Gabba,
Donella Beckwith, Forrest G. FitzGerald, Tanja J. Kutzner,
Anna-Kristin Ludwig, Herbert Kaltner, Paul V. Murphy,
Mare Cudic, and Hans-Joachim Gabius
Abstract
Dynamic changes of a cell’s glycophenotype are increasingly interpreted as shifts in the capacity to interact
with tissue (endogenous) lectins. The status of glycan branching or chain length (e.g., core 1 vs core
2 mucin-type O-glycans and polyLacNAc additions) as well as of sialylation/sulfation has been delineated
to convey signals. They are “read” by galectins, for example regulating lattice formation on the membrane
and cell growth. Owing to the discovery of the possibility that these effectors act in networks physiologically
resulting in functional antagonism or cooperation, their detection and distribution profiling need to be
expanded from an individual (single) protein to the—at best—entire family. How to work with non-cross-
reactive antibodies and with the labeled tissue-derived proteins (used as probes) is exemplarily documented
for chicken and human galectins including typical activity and specificity controls. This description intends
to inspire the systematic (network) study of members of a lectin family and also the application of tissue
proteins beyond a single lectin category in lectin histochemistry.
Key words Glycocluster, Glycoprotein, Glycosylation, Histochemistry, Sialylation
1 Introduction
Traditionally, lectins have been used as tools for descriptive glyco-

phenotyping. Hereby, cell type- and malignancy-associated differ-
ences as well as a dynamic nature of structural changes among
glycans were unraveled [1–7]. Obviously, the profile of sugars of
glycoconjugates (glycome) is an equivalent of a molecular finger-
print. This has been an incentive to refer to cell surface carbohy-
drates as “molecules in search of a function” [8] and to figure out
whether they are “a whim of Nature or a multipurpose tool” [9].
Gabriel Garcı́a Caballero and Joachim C. Manning contributed equally to this work.
307
308 Gabriel Garcı́a Caballero et al.
The growing realization (a) of the capacity of oligosaccharides

to encode biological information and (b) of the role of tissue
(endogenous) lectins to “read” these signals has established the
concept of the sugar code, that is the impact of structural comple-
mentarity between glycans and lectins on a wide range of cellular
activities [10–12]. For example, whether and how a glycoconjugate
is sialylated is thus thereafter no longer considered as a purely
structural phenomenon. It is instead viewed as a molecular message
“written” with carbohydrates as letters, and its recognition by a
lectin that leads to post-binding events is the functional link to
understand, e.g., growth regulation [13–15].
From their detection by lactose-inhibitable hemagglutination
onwards, galectins have intrigued researchers by a selective ability
to bridge counterreceptors on cells and between cells as well as by a
marked modulation of expression during development [16–
21]. The diversity among glycans led to expect a similar sophistica-
tion on the level of their receptors. The stepwise exploration of the
size of the galectin family in man and animals (for its current status,
please see Fig. 1) as well as of their presence in and outside of cells
then raised the exciting perspective that these proteins can have
individual distribution and activity profiles. Supported by initial
immunohistochemical detection of three galectins in a study on
breast cancer [23] and by systematic RT-PCR profiling of galectin
type-specific mRNA production, e.g., in 61 human tumor cell lines
of diverse histogenetic origin [24], the hypothesis of an expression
as network took shape [22, 25].
What’s more, initial cases of functional analysis with two or
three galectins in comparison or even in mixtures disclosed cases of
antagonism, cooperation, and non-redundant activity profiles
between human galectins on tumor/immune cell growth regula-
tion or in pro-degradative/-inflammatory activity of osteoarthritic
chondrocytes in vitro [26–29]. With focus on all seven chicken
galectins (CGs) as proof-of-principle model and corroborating
evidence on murine galectins, the so far obtained results encourage
to pursue this network (systems biology) approach on a broad scale
[30–33]. With these proteins themselves in hand, their application
as histochemical tools becomes possible: in comparative analysis
and in colocalization studies together with galectin immunohisto-
chemistry, as documented first for human galectin-1 in sections of
human peripheral nerves [34] and exemplarily illustrated in Fig. 2.
This chapter describes the preparation of non-cross-reactive
anti-galectin antibodies and of labeled galectins as well as their
application in light and fluorescence microscopy. An example of
the synthesis of a potent inhibitor for specificity controls, i.e., a
glycocluster, and of an activity assay that reports on binding capacity
of each carbohydrate recognition domain, i.e., isothermal titration
calorimetry (ITC), is added, here also broadening the perspective of
galectin testing by using an engineered variant (a tetramer) along
Exploring the Galectin Network by Light and Fluorescence Microscopy 309
Fig. 1 The galectin family in selected organisms at different positions of the phylogenetic tree. Evidence for
presence of galectins on the level of the gene (Roman number), the mRNA (Arabic number), and the protein
(numerical denotation) is summarized and given as number in each case. Special cases are highlighted:
GRIFIN with its species-dependent variability of lectin activity (+) and the presence/absence polymorphism of
Gal-6 exclusively found in certain mouse strains (*) as well as the tetrameric nature of two distinct oyster
galectins (#) (Reprinted from [22], with permission from the editor, copyright 2020)
with the homodimeric wild-type protein [37, 38]. The flow diagram
presented in Fig. 3 presents a graphical overview. This type of tool
can further find applications in confocal laser scanning microscopy
and cytofluorimetry, described previously in this series
Fig. 2 Illustration of examples for immunohistochemical localization of chicken galectins (CGs; a), accessible
binding sites for labeled CGs (b), immunohistochemical colocalization of two chicken proteins (c), and
detection of colocalization of endogenous galectins and accessible binding sites (d) by light and fluorescence
microscopy, as described in Subheadings 3.6 and 3.7. (a) Detection of CG-1A in cross sections of the anterior
part of a fixed developing chicken eye at Hamburger Hamilton stage 25 (4.5–5 days of embryonal develop-
ment [35, 36]) by light microscopy (see Subheading 3.6.1) using Vector®Red alkaline phosphatase substrate
kit (see item 12 in Subheading 2.6). Weak to moderate staining is detected in the developing lens and stronger
intensity was found in cells of the developing ciliary margin, as also detected by fluorescence microscopy
(inset to a, see Subheading 3.7.1). (b) Biotinylated CG-3 (see Subheading 3.3) bound to a cross section of fixed
adult chicken kidney at the apical part of the epithelial cells in a secondary branch of the ureter; signal for
detection of bound galectin was generated using Vectastain® ABC solution (see item 11 in Subheading 2.6)
and Vector®Red alkaline phosphatase substrate (see item 12 in Subheading 2.6) and its distribution analyzed
[39, 40]. Powered by application of these cyto- and histochemical

methods, our understanding of the interplay between galectins as
well as a galectin and its counterreceptor(s) in cells and tissues will be
noticeably advanced.
Beyond galectins, these experimental protocols can readily be
followed with other types of tissue lectins. This will broaden the
scope to detect and to localize distinct glyco-targets, e.g., α2,6-
sialylated branch ends of N-glycans by siglec-2 (CD22) [41] or
(sialyl)Tn-antigen (CD175(s)), a tumor-associated antigen
[42, 43], and/or LacdiNAc (GalNAcβ1,4GlcNAc) by the macro-
phage galactose(-binding C)-type lectin (MGL, CD301,
CLEC10A) [44].
2 Materials
2.1 Preparation of 1. Polyclonal immunoglobulin G (IgG) fraction raised in rabbit

Non-cross-Reactive (home-made) and directed against the galectin of interest [45].
Antibodies 2. Recombinant galectin proteins: in the case of the family of
CGs, the seven proteins CG-1A/-B, -2, -3, -8, C-GRIFIN
and C-GRP.
3. Econo-column® chromatography columns (1.0 cm or 1.5 cm
in diameter, 5 cm in length; Bio-Rad, Munich, Germany) and
fittings for tube connection to the pump or any other glass
columns (also for batch binding and chromatography with
gravity flow).
4. EP-1 Econo Pump (Bio-Rad) or any other peristaltic pump for
low-pressure chromatography.
5. Affi-Gel® 10 and 15 activated affinity media (presenting
N-hydroxysuccinimide ester for conjugation; addition of pro-
tein results in displacement of N-hydroxysuccinimide and for-
mation of a stable amide bond; Bio-Rad).
6. 0.1 M MOPS buffer (pH 7.5; coupling buffer).
ä
Fig. 2 (continued) by light microscopy (see Subheading 3.6.2). An equivalent signal pattern was observed
after incubation with fluorescent CG-3 (with Alexa Fluor® 488, see Subheading 3.4) and analyzed with
fluorescence microscopy (see Subheading 3.7.2). (c) Localization of CG-8 (using fluorescent goat anti-rabbit
IgG (Alexa Fluor® 568) as second-step reagent; red color) with the B-cell marker chB6 (Bu-1) using mouse
monoclonal AV20 antibody and FITC-labeled goat anti-mouse IgG1 as second reagent step in a cryosection of
chicken bursa of Fabricius by fluorescence microscopy (see Subheading 3.7.1). Channels for monitoring
signals for CG-8 presence, B-cell marker, and DAPI staining are given. (d) Localization of endogenous CG-8
(using fluorescent goat anti-rabbit IgG (Alexa Fluor® 568) as second-step reagent; red color) and accessible
binding sites for CG-8 labeled with Alexa Fluor® 488 (green signal) by fluorescence microscopy (see
Subheading 3.7.3) in sections of fixed adult chicken retina. Concentrations of probes were 1.0 μg/mL for
anti-CG-1A, 5.0 μg/mL for biotinylated and fluorescently labeled CG-3, 4.0 μg/mL for anti-CG-8, 2.5 μg/mL for
the AV20 antibody, and 2.0 μg/mL for labeled CG-8. Scale bars are 20 μm and 50 μm (for gray-scale images)
Fig. 3 Flow diagram presenting the experimental approaches for the immuno- and galectin histochemical
study of the galectin network described in this chapter
7. Lactose (Sigma-Aldrich, Munich, Germany).

8. Solution containing 1 M glycine ethyl ester (pH 8.0).
9. 20 mM phosphate-buffered saline (PBS: 20 mM KH2PO4/
Na2HPO4, 150 mM NaCl, pH 7.2).
10. Cold distilled water (4 C).
11. Bio-Rad Protein Assay Dye Reagent (Bradford assay; Bio-Rad).
12. Stuart™ roller mixer SRT9D.
2.2 Measurement of 1. Pipettors and tips (10 μL, 200 μL, 1000 μL).
Galectin Activity 2. Analytical scale.
3. pH meter.
4. Sample buffer: 20 mM PBS (pH 7.2), 10–150 mM sodium
chloride (NaCl, see Note 1), 10 mM β-mercaptoethanol (BME,
see Note 2) and Milli-Q water.
2.2.1 Preparation of 1. Eppendorf vials (0.5 mL and 1.5 mL) and 50 mL falcon tubes.
Solutions of Galectins 2. Disposable syringes (5 mL and 10 mL) and 18 gauge needles.
3. Magnetic stirring bars and 1 L dialysis beakers.
4. Dialysis cassettes (Slide-A-Lyzer™, 7 K MWCO, 66710;
Thermo Fisher Scientific, Dreieich, Germany).
5. Ultrafiltration equipment (Pall Microsep™ Advanced, 10 K
Omega, MCP010C41; Port Washington, NY, USA).
6. Coldroom or workplace at 4 C.
7. Stir plate.
8. Refrigerated centrifuge.
9. HPLC grade water.
10. Sample buffer (see Subheading 2.2, item 4).
11. UV-Vis spectrophotometer for protein concentration determination
12. Galectins (see Note 3).
2.2.2 Preparation of 1. HPLC grade water.

Solutions of Carbohydrate 2. Carbohydrate ligands: Lactose (Lac, Fisher Scientific, Wal-
Ligands tham, MA), N-acetyllactosamine (LacNAc) and 2-acetamido-
2-deoxy-3-O-(β-D-galactopyranosyl)-D-galactose
(TF) (Toronto Research Chemicals, Ontario, Canada).
2.2.3 Instrumentation 1. Microcalorimeter (model: MicroCal PEAQ-ITC; Malvern

and Software Panalytical, Malvern, UK).
2. MicroCal PEAQ-ITC analysis software.
3. Instrument-specific Hamilton syringe.
4. Sample buffer (see Subheading 2.2, item 4).
5. PCR tubes.
6. ITC detergent (20% Contrad™ 70).
7. Milli-Q water.
8. Methanol (MeOH), HPLC grade.
2.3 Preparation of 1. Sodium bicarbonate buffer (0.1 M NaHCO3 buffer, pH 8.3,

Biotinylated Galectins containing 150 mM NaCl; reconstitution buffer).
2. Lyophilized galectin (in addition to wild-type proteins,
mutants or engineered variants can be used).
3. N-Succinimidyl ester derivative of biotin (biotin NHS ester;
Sigma-Aldrich).
4. 99.8% Dimethylformamide (DMF; Carl Roth, Karlsruhe,
Germany).
2.4 Preparation of 1. Lyophilized galectin.

Fluorescent Galectins 2. Sodium bicarbonate buffer (0.1 M NaHCO3 buffer, pH 8.3,
containing 150 mM NaCl; reconstitution buffer for fluores-
cence labeling).
3. Fluorophores Alexa Fluor® (488, 555, 647) as NHS ester
(succinimidyl ester; Invitrogen by Thermo Fisher Scientific)
for labeling.
4. Dimethylformamide (DMF).
5. Photometer (NanoPhotometer® P-Class P300; Implen,

Munich, Germany).
2.5 Fixation, 1. 40 mM PBS (40 mM KH2PO4/Na2HPO4, 300 mM NaCl,

Embedding, and pH 7.2–7.4), only for paraformaldehyde (PFA) fixative (see
Tissue Sectioning below).
2. 4% (w/v) PFA in PBS: First, prepare an 8% (w/v) PFA (Appli-
Chem, Darmstadt, Germany) solution in distilled water. For
this, heat PFA to 60–65 C and then add 1 M NaOH dropwise
while stirring, just until the solution is clear. Finally, dilute with
40 mM PBS 1:2 and adjust pH to 7.2–7.4.
3. Bouin’s solution: 71% (v/v) saturated picric acid (~1.3%, VWR,
Darmstadt, Germany), 24% (v/v) formaldehyde solution (37%;
AppliChem, Darmstadt, Germany), 5% (v/v) glacial acetic acid.
4. Ethanol, isopropanol, xylene (analytical grade).
5. Fume hood.
6. Paraffin wax.
7. Embedding molds (stainless steel, L W H, internal
23 37 6/12 mm; R. Langenbrinck GmbH, Emmendin-
gen, Germany) and plastic standard/biopsy embedding cas-
settes (L W, 4.2 2.9 cm, Langenbrinck GmbH).
8. Embedding workstation (HistoCore Arcadia H and HistoCore
Arcadia C, Leica Microsystems GmbH, Wetzlar, Germany).
9. Rotary microtome (pfm Rotary 3006 EM, pfm medical AG,
Köln, Germany).
10. Microscope slides SuperFrost ® Plus (25 75 1.0 mm,
R. Langenbrinck GmbH) Emmendingen, Germany).
11. Drying oven.
12. Glass slide rack and glass staining dishes.
2.6 Light Microscopy 1. 20 mM PBS (pH 7.2).

2. 20 mM PBS containing 1% (w/v) carbohydrate free bovine
serum albumin (BSA; Sigma-Aldrich).
3. 0.1 M TRIS (Sigma-Aldrich) buffer (pH 8.4).
4. Glass slide rack and glass staining dish.
5. Humid incubation box.
6. ImmEdge® Hydrophobic Barrier PAP Pen (Vector Labora-
tories, Burlingame, CA, USA).
7. Normal goat serum (Dianova, Hamburg, Germany).
8. Working solutions of home-made non-cross-reactive poly-
clonal antibodies (for preparation details, see Subheadings 2.1
and 3.1).
9. Biotinylated galectins (for biotinylation protocol, see Subhead-

ings 2.3 and 3.3).
10. Goat anti-rabbit IgG (whole molecule)-alkaline phosphatase
conjugate (Sigma-Aldrich); for an alternative procedure using
a peroxidase-conjugate detection system (see Note 4).
11. Vectastain® ABC (Avidin Biotin Complex) kit AK5000 (Vector
Laboratories); for an alternative detection procedure (see
Note 4).
12. Vector®Red alkaline phosphatase substrate kit SK5100 (Vector
Laboratories). For alternative detection procedure (see
Note 4).
13. Hemalaun solution (hematoxylin acid according to MAYER;
Morphisto, Frankfurt am Main, Germany).
14. Small metal forceps (105 mm).
15. Quick-hardening mounting medium (Eukitt®; Sigma-
Aldrich).
16. Cover slips.
17. Axio Imager.M1 microscope equipped with AxioCam MRc
color digital camera and AxioVision software version 4.9.
(Zeiss, Göttingen, Germany) (see Note 5).
2.7 Fluorescence 1. 20 mM PBS containing 0.1% (v/v) Tween® 20 (PBS-T; wash-

Microscopy (See ing buffer) and/or 1% (w/v) BSA (Sigma-Aldrich) (PBS-T/1%
Note 6) BSA; blocking buffer).
2. Humid incubation box.
3. Normal goat serum (Dianova; see Note 7).
4. Non-cross-reactive IgG preparations (for details on removal of
cross-reactivity, see Subheadings 2.1 and 3.1).
5. Goat anti-rabbit IgG (H + L) labeled by fluorophores Alexa
Fluor® 488, 568 and/or 647 (Thermo Fisher Scientific) (see
Note 8).
6. Fluorescent (fluorophores Alexa Fluor® 488, 555, or 647)
galectin (for details on preparation, see Subheadings 2.4 and
3.4); for an alternative detection procedure (applying
fluorophore-avidin conjugates) (see Note 9).
7. 40 ,6-Diamidino-2-phenylindole (DAPI; Roche Diagnostics,
Mannheim, Germany).
8. Fluorescence mounting medium with anti-fading agent (Dako,
Hamburg, Germany).
9. Axio Imager.M1 microscope equipped with AxioCam MRm
black and white digital camera, fluorescence filter set, objectives
compatible with fluorescence microscopy and AxioVision soft-
ware version 4.9 (Zeiss) (see Note 5).
2.8 Synthesis of a Reagents for synthesis were purchased from Sigma-Aldrich, unless
Glycocluster otherwise stated.
1. Aluminum sheets precoated with silica gel 60 (HF254;
E. Merck, for thin layer chromatography (TLC)), H2SO4-
EtOH (1:20) for charring TLC plates.
2. CH2Cl2, MeOH, H2O, and MeCN solvents (Fisher Scientific
and Sigma-Aldrich).
3. Rotary evaporator (Buchi Rotavapor® R-300).
2.8.1 Reagents for Click This method is widely used for conjugation [46].
Chemistry
1. Copper (II) sulfate pentahydrate (CAS number 7758-99-8).
2. Sodium ascorbate (CAS number 134-03-2).
3. Silica gel 60 (0.040–0.630 mm; Sigma-Aldrich).
4. Microwave reactor (CEM instrument).
5. Reactants 1 (azide) and 2 (dialkyne) prepared as described
previously [38, 47].
2.8.2 Reagents for 1. Metallic sodium cubes in mineral oil (CAS Number 7440-23-
Zemplén Deacetylation and 5).
Purification of Glycocluster 2. Glacial acetic acid (CAS Number: 64-19-7).
3. C18-silica gel (Sigma-Aldrich).
4. Anhydrous solvents (MeOH, THF) used as obtained from a
Pure Solv™ Solvent Purification System.
5. Freeze dryer.
2.9 Evaluation of the 1. Biotinylated or fluorescent galectin (see Subheadings 2.3

Inhibitory Capacity of or 2.4).
Glycocluster in 2. Materials for light or fluorescence microscopy (see Subheadings
Galectin 2.6 or 2.7).
Histochemistry 3. Bi- and tetravalent glycoclusters presenting lactose as
headgroup.
4. Free lactose.
3 Methods
3.1 Preparation of A polyclonal IgG fraction against a galectin is first tested by screen-
Non-cross-Reactive ing against other family members (e.g., by Western blotting or
Antibodies ELISAs) for presence of cross-reactivity. Removal of these activities
is then performed by affinity chromatography using galectin-
presenting resin as follows.
1. Select resin type for galectin coupling to batches of Affi-Gel® 10 or

15. When coupling is performed (as here with MOPS buffer) at
pH 7.5, Affi-Gel® 10 is recommended for proteins with a pI of 6.5
to 11 (e.g., CG-3 with a pI of 8.79), and Affi-Gel® 15 is recom-
mended for proteins with a pI below 6.5 (e.g., CG-1A with a pI of
4.95) to achieve highest possible coupling efficiency.
2. Wash an aliquot (ca. 1–2 mL) of affinity gel (stored in isopro-
panol) with three bed volumes of cold distilled water and three
bed volumes of cold MOPS buffer (see Note 10).
3. Transfer the resin to a 15 mL test tube and add the galectin-
containing solution (10–12 mg galectin in 1–2 mL MOPS
buffer containing 50 mM lactose per 1 mL beads) (see Notes
10 and 11).
4. Gently agitate the slurry of resin and galectin solution on a tilt/
roller mixer at 20 rpm for 1 h at RT or 4 h at 4 C.
5. Block remaining active esters by adding 0.1 mL of 1 M glycine
ethyl ester (pH 8) per mL gel bead slurry and gently agitate the
mixture for 1 h at RT.
6. When beads have settled in the 15 mL test tube, determine the
protein concentration (by Bradford assay) in the supernatant to
calculate the amount of coupled protein. The density should be
10–12 mg galectin protein per mL packed volume.
7. Fill the slurries into Econo-columns® and wash the obtained
beads with approximately 10 to 15 bed volumes of cold MOPS
buffer until the gel is free of reactants.
8. Perform sequential (serial) affinity chromatography to remove
cross-reactivity from the IgG fraction raised against the galectin
of interest. Sequentially apply ca. 4–6 mL PBS containing the
IgG preparation (2 mg/mL) to columns containing each of the
cross-reactive galectins coupled to beads (ca. 1 mL) and equili-
brated with PBS. The eluate of the first column is hereby
applied to the next column(s) in a stepwise manner.
9. Always recirculate the solution with the IgG fraction using a
peristaltic pump (e.g., EP-1 Econo Pump) at a flow rate of
0.1–0.2 mL/min for about 8 h in each step of the serial affinity
chromatography (see Note 12).
10. Ascertain complete absence of contaminating cross-reactivity
by Western blot and ELISA analysis.
3.2 Measurement of 1. Determine the galectin quantity required to perform

Galectin Activity (See n experiments at concentration X in ITC loading volume V,
Note 13) where V ¼ 100–200 μL greater than the instrument cell vol-
ume (200 μL for Malvern PEAQ-ITC). Design the experiment
3.2.1 Preparation of considering the sigmoidicity factor: c ¼ n KA [M]t, where
Solutions of Galectins c ¼ the ratio of protein concentration in the cell to the KD
value, n ¼ stoichiometry of binding, KA ¼ binding constant,
[M]t ¼ macromolecule concentration, and 10 c 500 (see
Notes 13 and 14, also [48]).
2. Reconstitute galectin in HPLC grade water (see Note 3).

3. Quantitate galectin concentration in solution by absorbance at
280 nm, applying a predictive ε (molar absorption coefficient)
based on the protein sequence [49]. Spot 2–3 μL aliquots of
galectin solution and 2–3 μL of buffer solution (as blank) on a
Take3 plate (compatible with BioTek EPOCH Microplate
Spectrophotometer) and read the plate using protein A280
manufacturer protocol. Other means of protein quantitation
can be applied (see Note 15).
4. Prepare 10 L of dialysis buffer (see Subheading 2.2).
5. Transfer up to 12 mL (volume dependent on cassette) of dis-
solved galectin at 1 mg/mL into the dialysis cassette and
place the cassette into 2 L of dialysis buffer, stirring at 4 C.
6. Perform five subsequent 2 L buffer changes every 8–14 h, with
a total dialysis time of 72–84 h (see Note 16).
7. Remove galectin solution from the dialysis cassette and deter-
mine concentration, as performed in step 3 (see Note 17).
8. To avoid heat changes due to buffer mismatch, both protein
and ligand must be matched in the same buffer. Set aside
50 mL of final dialysis buffer for use in preparation of solution
with carbohydrate ligand, pre-run ITC cell equilibration, and
blank titrations.
9. Concentrate galectin-containing solution using Pall ultrafiltra-
tion centrifugation device at 4000 g and 4 C, to predeter-
mined concentration for use in the ITC experiment.
3.2.2 Preparation of 1. Prepare a stock solution of a carbohydrate ligand by adding a

Carbohydrate Ligand known volume of HPLC water to a defined quantity of carbo-
hydrate ligand (LacNAc, Lac, TF), vortex to dissolve, and store
stock solution at 4 C. In general, ligand stock concentrations
should be 2–10 times greater than experimental concentration
needed. Stock concentrations of 40 mM worked well for the
concentration range of galectins used.
2. Prepare the desired dilution of the carbohydrate-containing
solution (total volume of at least 30 μL greater than the total
injection volume needed for titration) by adding a known
volume of the ligand stock solution to a PCR tube followed
by a known volume of the buffer used in the last dialysis step.
Aim for a higher buffer-to-water ratio to avoid warming upon
ligand injection (see Note 18).
3.2.3 Isothermal Titration 1. Fill the reference cell with Milli-Q water or buffer using an
Calorimetry (ITC) instrument-specific Hamilton syringe.
2. Rinse or soak the sample cell with dialysis buffer to equilibrate/
passivate the cell prior to loading the sample.
3. Slowly load 350–400 μL of the galectin solution containing

BME (active protein concentration should be at least tenfold
lower than the titrant concentration with optimum protein
concentration being 10 higher than the KD) into the sample
cell, avoiding the creation of air bubbles. Remove the excess
volume not inside the cell.
4. Allow the galectin-containing solution to reach RT (or sample
cell temperature) prior to starting the experiment, to avoid
having to degas the solution.
5. Load the titrant syringe with the carbohydrate ligand-
containing solution mixed with buffer from the PCR tube.
Ensure that no air bubbles are present in the glass syringe
after loading.
6. Set modifiable experimental parameters: (a) temperature ( C),
(b) injection spacing (s), (c) injection duration (s), (d) stir
speed (rpm), (e) initial delay (s), (f) number of injections,
(g) injection volume in μL, (h) feedback, and (i) reference
power (μW). A reference power of 41.9 μW and high feedback
can remain unchanged for most ITC experiments that are
expected to produce high heat. A typical binding experiment
between galectin and carbohydrate ligand would entail:
(a) 25 C, (b) 180 s, (c) 4 s, (d) 750 rpm, (e) 60 s, (f) 19
injections, and (g) 2 μL (see Note 19).
7. Start ITC experiment. Make note of any acceptable abnormal-
ities (see Note 20) during initial instrument equilibration
period or abort the experiment if unacceptable abnormalities
occur.
3.2.4 Data Analysis Calorimetric data are typically analyzed with a single-site binding
model using Malvern software (version 1.22, MicroCal PEAQ-
ITC). Open source SEDPHAT [50, 51] and Affinimeter [52]
software may also be used to analyze data but will not be discussed
in detail here.
1. The processed ITC data is automatically categorized as “bind-
ing,” “no binding,” or “check data.” This allows for
non-subjective data analysis; however, it is suggested that the
autocategorization of each titration be independently, and
rationally, evaluated.
2. The raw ITC data set is presented in a thermogram, in which
the heat change (ΔQ) detected during each successive injection
is demarcated by “peaks” of changing differential power (DP,
μcal/s) measured between the sample and reference cell. As the
number of binding sites in the sample cell reaches saturation,
less ΔQ is detected and the peak areas regress towards smaller,
more constant measures associated with the heat of titrant
dilution.
3. The areas of the injection peaks are automatically integrated by

the software to measure the amount of heat released by each
injection of the ligand. A nonlinear regression is fitted to the
injection points in the form of a Wiseman plot (normalized
heat per mole of titrant injected (ΔH kcal/mol) versus the
molar ratio of the lectin).
4. The isotherm is generally fitted to a single-site “one-set-of-
sites” binding model. The true global molecular binding events
may be far more complex and require fitting in the two-sets-of-
sites or sequential models, in order to take into account multi-
valency and possible cooperativity. An analysis of cooperativity
of these systems can alternatively be made using data derived
from the one-set-of-sites analysis [38].
5. The isotherm yields the schematic representation of the
enthalpy change (ΔH), the inverse slope of the fitted curve
provides the binding constant (KD), and the center of the
binding curves’ alignment with the molar ratio (x-axis) eluci-
dates the stoichiometry (n), all while including fitting errors for
these parameters. Calculation of Gibbs free energy (ΔG) and
change in entropy (ΔS) is determined by the software from the
isotherm curve.
6. Malvern software automatically sets the integrated peak area;
however, data analysis requires correct assessment of the inter-
polated heat changes. In some cases, manual adjustment of the
integrated peak area and baseline generated by the fitting
model is required. Anomalous peaks can be excluded, fitted
offset control parameters can be replaced by a ligand-buffer
control, and variations in ligand/cell protein concentrations,
n, KD, and ΔH can be selected (see Notes 21–24).
7. Signature plots along with multiple in-built tools allow for
visualization of the thermodynamic parameters, making it eas-
ier to assess how the enthalpic and entropic components con-
tribute to the overall affinity. It is imperative that replicate
experiments are performed to ensure reproducibility of data
before conclusions are made about binding mechanisms.
3.3 Preparation of 1. Dissolve lyophilized iodoacetamide-treated recombinant galec-

Biotinylated Galectins tin in an appropriate volume (0.25–1 mL) of reconstitution
buffer containing 20 mM lactose for the protection of the
carbohydrate recognition site (see Subheading 2.3), resulting
in a final protein concentration of 2 mg/mL (see Note 25).
2. Prepare a 10 mM solution of the biotin NHS derivative in
DMF (1 mg in 50 μL) immediately prior to use.
3. Add the appropriate volume of solution containing the NHS
derivative of biotin to the galectin-containing solution (0.3 mg
per mg protein). This solution contains lactose to protect the

contact region for sugar from chemical modification.
4. Incubate mixture for 2 h at RT and separate the biotinylated
protein from free reagent using Sephadex G25 in PD-10
desalting columns (GE Healthcare).
5. Divide the fractions containing biotinylated proteins into ali-
quots and store the aliquots at 20 C for lyophilization.
3.4 Preparation of 1. Dissolve lyophilized iodoacetamide-treated galectin in recon-

Fluorescent Galectins stitution buffer for fluorescent labeling (see Subheading 2.4)
containing 20 mM lactose at a final concentration of 2 mg/mL
(see Note 25).
2. Dissolve 1 mg of Alexa Fluor® (488, 555 or 647) NHS ester
derivative in 120 μL DMF and carefully protect them from
light.
3. In a darkened room, slowly add 10 μL of the dye solution
(in DMF) per mg protein to the galectin solution, while mixing
gently. Protect the reaction mixtures from light, while
wrapping the reaction tubes in aluminum foil.
4. Incubate the mixture at RT for 2 h to complete the reaction.
5. Separate the conjugate from the residual reagent using Sepha-
dex G25 in PD-10 desalting columns (GE Healthcare). Con-
tinue to protect the labeled protein in solution from light.
6. Calculate the degree of labeling (see Note 26) using a
nanophotometer.
7. Divide the fractions containing dye-protein conjugate into
aliquots and store the aliquots in the dark at 20 C for
lyophilization.
3.5 Fixation, 1. Fix the tissue specimen in 4% (w/v) buffered PFA or in Bouin’s
Embedding, and solution. Penetration of tissue by fixative at rate of approxi-
Tissue Sectioning mately 1 mm/h (see Note 27). For safety reasons, place sam-
ples under a fume hood during fixation.
2. After fixation is completed (see Note 27), pour off fixative and
wash Bouin-fixed tissue specimens three times in 70% (v/v)
ethanol solution, PFA-fixed tissue three times in 20 mM PBS
for 1 h each.
3. Dehydrate serially using a panel of graded ethanol solutions:
70% (v/v) twice for 75 min each, 80% (v/v) twice for 75 min
each, and 99% (v/v) twice for 135 min each, followed by
isopropanol twice for 150 min each and xylene once for
150 min, twice for 60 min each. These steps should be per-
formed under a fume hood. Finally, immerse tissue pieces in
paraffin wax (60 C) three times for 120 min each.
4. Subsequently, place the fixed and paraffin-wax-soaked tissue

specimen in paraffin blocks using embedding forms and cas-
settes as well as forceps to adjust position of the specimen at a
temperature not higher than 60 C (see Note 28). Usually, for
this procedure an embedding workstation with heated work-
space and cold plate should be available.
5. Cut sections of appropriate thickness with a microtome
(3–7 μm) and mount sections onto slides.
6. Dry sections at 42 C in a drying oven (see Note 29).
7. For further histochemical processing (see Subheadings 3.6 or
3.7), place slides in a glass slide rack. Dewax and rehydrate
sections under a fume hood using glass staining dishes contain-
ing xylene (twice for 15 min each), then immerse in isopropa-
nol (twice for 10 min each) and in a series of graded ethanol
solutions (99%, 80%, and 70% (v/v); for 10 min).
8. Rinse thoroughly first in distilled water (twice for 5 min each)
and then once in PBS or PBS-T (protocols using fluorescence-
labeled probes) for 5 min.
3.6 Light Microscopy 1. After performing step 8 in Subheading 3.5, carefully cover
sections with PBS-T containing 1% BSA and 5% goat serum
3.6.1
(blocking solution). Incubate in a humid incubation box at RT
Immunohistochemistry:
for 1 h (see Notes 30 and 31).
Detection of Endogenous
Galectins 2. Remove and discard blocking solution. Do not allow to dry
and do not wash.
3. Cover sections with a freshly prepared solution of PBS contain-
ing 1% BSA and the appropriate quantity of non-cross-reactive
IgG directed against the specific galectin, whose presence
should be visualized, and incubate at 4 C overnight (see
Notes 31–33).
4. Place slides in a glass slide rack and wash sections three times in
a glass staining dish containing PBS for 10 min each.
5. Incubate sections with freshly prepared solution of PBS/1%
BSA containing an appropriate quantity of goat anti-rabbit IgG
antibody, alkaline phosphatase conjugate in a humid incuba-
tion box at RT for 1 h (see Notes 31 and 32).
6. Place slides in the glass slide rack and wash sections three times
in a glass staining dish containing PBS for 10 min each.
7. Wash slides twice in TRIS buffer for 5 min each.
8. Visualize staining with freshly prepared Vector®Red working
solution (see instructions of manufacturer) applied to sections
in a light-tight humid incubation box (see Notes 31, 34, and
35) and wash once in TRIS buffer for 5 min.
9. Discard Vector®Red solution, place slides in a glass slide rack in

a staining dish, and rinse thoroughly with distilled water.
Counterstain in hemalaun for 3–15 min (depending on desired
degree of counterstaining) and perform bluing of the hema-
laun stain by positioning slides in a glass slide rack in a staining
dish under running tap water for 10 min.
10. Dehydrate sections using a series of solutions with increasing
concentration of ethanol (70%, 80%, and 96% (v/v); for 5 min
each), then immerse twice in isopropanol and three times in
xylene for 5 min each. Perform dehydration steps under a
fume hood.
11. Place cover slips firmly on slides using forceps onto sections
covered with 1–2 drops of quick-hardening mounting medium
to cover sections under a fume hood.
12. Record photomicrographs of stained sections with a micro-
scope equipped with a digital color camera and an appropriate
analysis software (see Subheading 2.6, item 17).
3.6.2 Galectin 1. Subsequent to step 8 in Subheading 3.5, carefully cover and

Histochemistry: Detection incubate deparaffinized and rehydrated sections in PBS con-
of Accessible Binding Sites taining 2% BSA (w/v; blocking solution) in a humid incubation
box at RT for 1 h (see Notes 30 and 31).
2. Remove and discard blocking solution. Prevent sections from
drying and do not wash.
3. Cover sections overnight with PBS containing an appropriate
concentration of the biotinylated galectin and incubate in a
humid incubation box at 4 C overnight (see Notes 31–33
and 36).
4. Wash slides placed in a glass rack three times in a glass staining
dish containing PBS for 10 min each.
5. Cover and incubate sections with freshly prepared Vectastain®
ABC solution in a humid incubation box at RT for 1 h (see
Note 31).
6. Place slides in a glass slide rack and wash sections three times in
a glass dish with PBS for 10 min each.
7. Wash slides in TRIS buffer as in step 7 of Subheading 3.6.1.
8. Apply Vector®Red staining solution as described in step 8 of
Subheading 3.6.1.
9. Perform counterstaining in hemalaun as outlined in step 9 of
Subheading 3.6.1.
10. Dehydrate sections and mount cover slips (see steps 10 and 11
in Subheading 3.6.1).
11. Document signal profiles of stained sections as described in
Subheading 3.6.1, step 12.
3.7 Fluorescence 1. After step 8 in Subheading 3.5, carefully cover sections with
Microscopy (See PBS-T containing 1% BSA (w/v) and 5% (v/v) goat serum
Note 6) (blocking solution) and incubate in a humid incubation box
at RT for 1 h (see Notes 30 and 31).
3.7.1
Immunohistochemistry: 2. Remove and discard blocking solution. Do not wash and do
Detection of Endogenous not allow to dry.
Galectins 3. Cover sections with a solution of 1% BSA (w/v) in PBS-T
containing appropriate quantity of non-cross-reactive IgG
directed against the galectin. Incubate in a humid incubation
box at 4 C overnight (see Notes 31–33 and 37).
4. Place slides in a glass slide rack. Wash sections three times in a
glass staining dish with PBS-T for 10 min each.
5. Incubate sections in a solution of PBS-T containing 1% BSA
(w/v) and an appropriate concentration of fluorescent goat
anti-rabbit IgG (see Notes 31 and 32) and DAPI (1 μg/mL)
in a humid incubation box at RT for 1 h. Take necessary
handling precautions: wear gloves and protection glasses
while working with DAPI-containing solutions.
6. Remove and discard DAPI-containing solution. Place slides in
a glass slide rack and wash sections three times in a glass stain-
ing dish containing PBS-T for 10 min each.
7. Rinse slides thoroughly with water, cover tissue sections with
fluorescence mounting medium, and seal edges of the cover
slips with nail polish. Slides may be examined immediately but
it is advantageous to leave them at 4 C for 24 h in order for the
mounting medium to solidify (see Note 38).
8. Document photomicrographs of fluorescent sections with a
microscope equipped with a digital black and white camera
and an appropriate analysis software with a module for multi-
dimensional image acquisition and color assignment (see Sub-
heading 2.7, item 9).
3.7.2 Galectin 1. After step 8 in Subheading 3.5, carefully cover sections with
Histochemistry: Detection PBS-T containing 2% BSA (w/v; blocking solution) and incu-
of Accessible Binding Sites bate in a humid incubation box at RT for 1 h (see Notes 30
and 31).
2. Remove and discard blocking solution. Do not wash and do
not allow sections to dry.
3. Incubate sections with a solution of PBS 2% BSA (w/v) con-
taining an appropriate quantity of fluorescent galectin in a
humid incubation box at 4 C overnight (see Notes 31–33,
36 and 37).
4. Wash slides three times in PBS-T for 10 min each as described
in step 4 of Subheading 3.7.1.
5. Cover sections with DAPI solution (1 μg/mL) and incubate in

a humid incubation box at RT for 15 min. Follow safety
instructions as outlined in step 5 of Subheading 3.7.1.
7. Proceed as described in Subheading 3.7.1, step 7.
8. Document photomicrographs as outlined in step 8 of
Subheading 3.7.1.
3.7.3 Detection of 1. After step 8 in Subheading 3.5, block sites for non-specific
Endogenous Galectins and protein binding in sections with PBS-T containing 2% BSA
Accessible Binding Sites by (w/v) and 5% (v/v) goat serum at RT in a humid incubation
Fluorescence Microscopy box at RT for 1 h (see Notes 30 and 31).
2. Remove and discard blocking solution. Avoid drying of sec-
tions and do not wash.
3. Apply the solution of PBS-T containing 2% BSA (w/v) and the
appropriate quantity of non-cross-reactive primary antibody
against a distinct galectin and incubate in a humid incubation
box at 4 C overnight (see Notes 31–33 and 37).
5. Incubate sections with a solution of 2% BSA (w/v) in PBS-T
containing an appropriate quantity of fluorescent goat anti-
rabbit IgG (second-step antibody) at RT for 1 h (see Notes
31 and 32).
7. Perform successive detection of accessible sites for a galectin on
the slides pre-incubated with solutions of primary and second-
step antibody using a solution of PBS-T containing BSA and a
suitable quantity of fluorescent galectin (see Subheadings 2.4
and 3.4), incubate sections with this solution in a humid incu-
bation box at 4 C overnight (see Notes 31–33, 36, and 37).
8. Wash slides three times in PBS-T for 10 min each as outlined in
step 4 of Subheading 3.7.1.
9. Stain nuclei by applying a DAPI solution (1 μg/mL) to the
sections at RT for 15 min. Please follow safety instructions as
outlined in step 5 of Subheading 3.7.1.
11. Proceed as described in step 7 of Subheading 3.7.1.
12. Document photomicrographs as described in Subheading
3.7.1, step 8.
Scheme 1 Click Chemistry reaction
3.8 Synthesis of 1. Add 10 mL of THF to a round bottomed flask with a

Glycocompounds rubber seal.
3.8.1 Preparation of 2. Add 10 mL of water (H2O) to a round bottomed flask with a
Intermediate 3 by Click rubber seal.
Chemistry (Scheme 1) 3. Degas the water and THF using a gentle flow of nitrogen flow
for 15 min.
4. Add the dialkyne 2 [47] (108 mg, 0.42 mmol) to a 10 mL
microwave vessel (CEM, product code 909050) and dissolve in
a mixture of degassed THF-H2O (1:1, 6 mL).
5. Add azide 1 [47] (561 mg, 0.92 mmol), sodium ascorbate
(49 mg, 0.25 mmol), and Cu (II)SO4·5H2O (62 mg,
0.25 mmol) to the reaction vessel.
6. Stir under an inert atmosphere in a microwave reactor (Dis-
cover® SP; CEM) at 50 C, 120 W for 30 min in dynamic mode
with 15 s of premixing time [53] (see Note 39).
7. Monitor the reaction by TLC analysis using CH2Cl2-MeOH
(95:5) as mobile phase.
8. When complete, transfer the contents of the vessel to a 50 mL
round bottomed flask.
9. Remove volatile components under diminished pressure using
a rotary evaporator. This will leave a slurry.
10. Dilute with CH2Cl2 (15 mL) and H2O (10 mL) and transfer to
a separating funnel.
11. Drain the lower organic layer, which contains the desired prod-
uct into a conical flask and keep this.
12. Extract (2) the remaining aqueous layer with two further
15 mL portions of CH2Cl2.
13. Combine all organic layers and dry over Na2SO4.
14. Filter and remove the solvent under diminished pressure using
a rotary evaporator.
15. Place the residue at the top of a silica gel column (30 mm
diameter, 15 cm length) prepacked with CH2Cl2-MeOH (95:
5) and elute with the CH2Cl2-MeOH (95:5), collecting frac-

tions with test tubes.
16. Use TLC analysis to identify fractions with product, which will
be detectable using a UV lamp or by staining the plate with 5%
sulfuric acid in ethanol.
17. Combine all the fractions containing 3 and then remove the
solvent using a rotary evaporator (see Note 40).
3.8.2 Preparation of the 1. Transfer 10 mL of dry MeOH to a 50 mL flame-dried round

NaOMe Solution in MeOH bottomed flask. Cool in ice for 20 min.
for Deacetylation 2. Using a clean and dry tweezer transfer the metallic sodium
(stored in oil; see Note 41) into a beaker with cyclohexane,
swirl gently for few seconds. Repeat twice with fresh cyclohex-
ane portions.
3. Transfer the sodium cube onto a clean and dry glass surface
(a clock glass or Petri dish) and cut it into smaller pieces with a
clean and dry knife (see Note 42).
4. Slowly add the clean sodium pieces (ca 2.5 g) into the cold
MeOH and leave until fizzing stops.
3.8.3 Preparation of 1. Into a clean round bottomed flask, add compound 3 (400 mg,
Glycocluster 4 Via Zemplén 0.25 mmol) to anhydrous MeOH (10 mL) and cool to 0 C
Deacetylation (Scheme 2) over ice. Add the NaOMe solution in MeOH dropwise until
the solution reaches ~ pH 10.
2. Allow the solution to attain RT.
3. Monitor reaction progress by TLC analysis using CH2Cl2-
MeOH 95:5 as the mobile phase (see Note 43).
4. When complete, add glacial acetic acid dropwise to neutralize
(pH 7) the solution.
5. Remove the solvent under diminished pressure using a rotary
evaporator.
6. Dissolve the residue in the minimal amount of water with a few
drops of MeCN added.
7. If precipitates forms, add one drop of glacial acetic acid and
heat it slightly to dissolve.
8. When dissolved, load the solution to a reverse phase silica
column (C18, 20 mm diameter, 5 cm length), which has been
prepacked with water.
9. Elute with water (three column volumes) to remove any salts
present.
10. Elute with MeCN-H2O (6:4).
Scheme 2 Zemplén Deacetylation for the formation of 4
11. Locate the product using TLC analysis of collected tubes using
a UV lamp or by staining with 5% sulfuric acid in ethanol
followed by heating.
12. Remove the MeCN, from the combined fractions containing
the product, on a rotary evaporator, under diminished
pressure.
13. Remove any residual water by lyophilization (see Note 44).
This should give a white solid.
3.9 Evaluation of the 1. Follow instructions in steps 1 and 2 in Subheading 3.6.2 (for
Inhibitory Capacity of light microscopy) or in Subheading 3.7.2 (for fluorescence
Glycocompounds in microscopy).
Galectin 2. Prepare twofold serial dilutions starting at 200 mM to reach
Histochemistry (See 25 μM (for free lactose) or from 10 mM to 2.5 μM of lactose
Note 13) present in the glycocluster in PBS containing 1% BSA.
3. Mix solutions from step 2 with solutions of labeled galectin or
variant proteins adjusted to optimized concentration (strong
signal/minimal background, see Note 45) and incubate the

mixtures at RT for 1 h.
4. Cover serially cut tissue sections with solutions from step 3 and
incubate overnight at 4 C.
5. Proceed as described in steps 4–11 in Subheading 3.6.2 (for
light microscopy) or in steps 4–8 in Subheading 3.7.2 (for
fluorescence microscopy).
6. Examine processed slides for semi-quantitative grading (cate-
gories: ¼ no staining; (+) ¼ very weak but significant stain-
ing; + ¼ weak; ++ ¼ medium; +++ ¼ strong; ++++ ¼ very
strong staining) to determine the degree of inhibition (Fig. 4).
4 Notes
1. NaCl concentration can be varied depending on the solubility

of galectin.
2. Fresh BME should be added at each buffer exchange.
3. Please note that the nature of the protein determines the
optimal mode of processing for best solubility. For example,
turning wild-type proteins into variants by engineering often
has an impact on solubility. Also, freeze-drying affects this
property, becoming a source for differences between batches.
4. Cases can arise, where endogenous alkaline phosphatase activ-
ity leads to reagent-independent staining and inhibition of this
enzyme (e.g., intestinal activity) by levamisole is not possible.
Here, DAB/horseradish peroxidase-based detection kits,
which yield a dark brown product, can be recommended as an
alternative. When using this detection system, it is necessary to
block endogenous peroxidase activity. After rehydration, prior
to histochemical processing, slides with deparaffinized sections
are immersed in a glass dish containing a hydrogen peroxide
(1%) solution under gentle agitation for 10–30 min.
5. Microscopes by any other manufacturer can be used; we rec-
ommend equipping it with a motorized focus drive and module
for the acquisition of Z-stacks. This enables the generation of a
focused image of all planes throughout the thickness of the
tissue section.
6. Bleaching of the fluorophore by direct light must be avoided. A
darkened room with a red-light lamp is adequate for processing
of the slides. Incubation steps of slides are performed in a light-
tight box.
7. Serum contains immunoglobulins that block binding sites,
preventing non-specific binding of second-step antibodies via
Fc part. Serum must be taken from the species used for the
Fig. 4 Exemplary illustration of the correspondence between staining intensity and the semi-quantitative
scoring system using galectin staining in sections of fixed murine jejunum for documentation (for further
details, please see [53]). Effect of increasing concentrations of galectin inhibitor (lactose, Lac). v, Intestinal villi
(asterisks, cells of the lamina propria; white arrowheads, epithelial cells); c, crypts of Lieberkuhn (black
arrowheads)
production of the second-step antibody. We therefore chose

goat serum in this protocol.
8. If detection of endogenous galectin and of its accessible bind-
ing sites should be performed in one specimen, it is essential to
choose a fluorescent second-step goat anti-rabbit IgG with an
emission spectrum, which does not overlap with the spectrum
of the fluorophore used to label the galectin.
9. When biotinylated probes are available, it is possible to use
avidin-fluorophore conjugates. The solution of conjugate at
an appropriate concentration (usually in the range of
1–30 μg/mL) in PBS-T should be added to the slides after
overnight incubation with the biotinylated probes and

subsequent washing in PBS-T to remove labeled galectin.
After a 1 h incubation period, slides should be washed in
PBS-T and further processed as described (steps 5–8 in Sub-
heading 3.7.2 and steps 9–12 in Subheading 3.7.3).
10. To obtain optimal coupling yields, washing steps should be
done and the beads combined with the galectin solution within
20 min.
11. Addition of lactose to the coupling buffer is recommended to
preserve the activity of the galectin which should be coupled to
the beads.
12. Alternatively, mix the ligand solution with the pre-equilibrated
Affi-Gel® 10 or 15 resin and batch-bind by shaking the slurry
gently at 4 C for at minimum 4 h. As an alternative to recircu-
lation driven by a peristaltic pump, use Econo-columns
(Bio-Rad) or glass columns (with porous bed support at the
bottom to retain fine particles) and gravity flow in steps 7–9 in
Subheading 3.1.
13. For an example of protein activity measurements and evalua-
tion of inhibitory capacity of glycocompounds in galectin his-
tochemistry for wild-type and variant galectin, please see
[38, 53]. Description of calorimetric measurements on lectins
are found for example in [54, 55], Table 1 in [56] presents an
overview on methods towards this aim.
14. A c-value of 10 c 500 is necessary for ΔH, KA, and
n simultaneous determination. Low c-value (<1) titrations
essentially depend on the ratio of ligand concentration to KD
due to receptor occupancy becoming independent of receptor
concentration. Hence, it is necessary to either fix the binding
stoichiometry and/or use a ligand concentration at least
10 times the lectin concentration.
15. UV absorbance measurements can be similarly obtained using
different instrumentation (i.e., Thermo Fisher NanoDrop).
Galectin concentration can also be quantitated using a Brad-
ford assay or a BCA assay. Using an alternative method for
protein quantitation is a means to ensure the validity of the
predictive molar absorption coefficient applied.
16. This protocol has been optimized to remove bound lactose
from the purification steps. Insufficient dialysis will result in
lowered “active” protein concentrations.
17. Note any changes in concentration; it is not uncommon for
protein loss to occur due to precipitation or adherence to the
dialysis membrane.
18. It is imperative that the fixed titrant concentration is known
with high accuracy, whereas variability in the active lectin
concentration may be tolerated by letting n-value (stoichiome-

try) “float.”
19. The first injection volume and duration should be tenfold less
than experimental injection volume and duration. This first
injection should be discarded in the analysis for reasons high-
lighted elsewhere.
20. The differential power (DP) should match the pre-set reference
power input. If the system equilibrates to a DP above or below
the reference power input, the error may be the result of a
variety of factors, e.g., air bubbles in the sample or reference
cell, dirty sample cell or protein precipitation.
21. If the DP is not near the expected reference power during
equilibration or sporadic peaks develop in the baseline, it is
usually due to air bubbles in the cell. Bubbles are generated on
the cell walls upon injection of the protein or the reference
solution into the cell. These bubbles dislodge during titration
from heat and stirring resulting in showing as anomalous peaks
in the run. When this happens, their contribution is heavily
noticed in DP levels.
22. Extent of protein aggregation upon raising protein concentra-
tion may be lessened with optimal preparation that includes the
addition of reducing agents and pH, concentration, and ionic
strength adjustments. Additional attempts using higher or
lower temperatures may help subdue aggregation long enough
to afford a favorable thermogram.
23. Besides aggregation of the protein or bubble formation, dirty
equipment can lead to erroneous data. A common problem
experienced when working with multiple proteins (by different
people) or with aggregating macromolecules can be a dirty cell.
A simple cleaning of the cell will not likely resolve the residue
and a more extended heated cleaning and detergent soak may
be required. Additionally, thorough scrubbing of the cell and
the interior of the syringe may be necessary. To avoid any
buildup in the cell and the syringe, when aqueous solutions
evaporate, each should be rinsed with methanol prior the
cleaning, after the reactants have been removed.
24. Increased syringe spinning noise, causing trouble when aiming
to obtain optimum DP levels, noticeable baseline drift and
oscillating peaks are all indications of a possibly bent syringe.
25. Treatment with iodoacetamide is a standard procedure for
stable alkylation of cysteine residues to prevent oxidative inac-
tivation, which makes the addition of reducing agents such as
BME in subsequent steps unnecessary.
26. Quantitation of protein:dye conjugation (dye:conjugation or
F/P molar ratio) may be necessary to predict the amount of
probe for an experiment or for controlling fluorescence inten-

sity between experiments:
(a) Measure absorbance of protein:dye conjugate at 280 nm
(A280) and at the λmax (provided by the manufacturer) of
the dye (Amax).
(b) Then, the concentration of the protein (M) is calculated
by (A280 (Amax CF))/ε dilution factor (ε ¼ protein
molecular extinction factor, M1 cm1; CF ¼ Correction
factor, provided by the manufacturer; adjusts for the
amount of absorbance at 280 nm caused by the dye).
Then, moles of dye per moles of protein are calculated
by (Amax of the labeled protein)/(ε´ protein concentra-
tion) with ε´ ¼ molar extinction coefficient of the fluores-
cent dye (provided by the manufacturer).
27. Fixation time depends on the thickness and type of tissue
(typically 1 h fixation per mm of tissue). In some cases, fixation
and tissue processing can negatively affect binding of antibo-
dies and/or galectins. Use of organic solvents, for example,
typically extracts glycosphingolipids. In those cases, cryosec-
tioning can be an alternative. Snap-frozen tissue (by immersion
in liquid nitrogen) is cut on a cryostat, transferred to slides and
then dried to preserve morphology.
28. The type of paraffin blend can influence tissue sectioning and
processing. Generally speaking, paraffin blends with low con-
centrations of polymer offer rapid infiltration, while blends
with higher polymer concentration result in harder wax and
improve the cutting of thin sections (<5 μm).
29. Drying of slides (paraffinized tissue sections) should be per-
formed at 42 C and last for a minimum time period of 48 h.
30. If more than one section is mounted onto the same slide, a
hydrophobic barrier (drawn by a special pen such as the
ImmEdge® Hydrophobic Barrier PAP Pen from Vector
Laboratories) encircling each section is necessary in order to
avoid mixing of solutions on one slide. This can also be recom-
mended when just one section is present, as it can prevent
unintended loss of incubation solution and drying out of the
tissue section.
31. Once dewaxed, drying of sections must be avoided in order to
preserve integrity of tissue. For this reason, be sure to minimize
the time sections are not immersed in or covered with liquid.
32. Concentration of samples generating an optimal signal-to-
background ratio must be determined by systematic titrations.
Usually, concentrations for primary antibodies are in a range
from 0.5 to 4 μg/mL and for labeled galectins from 0.5 to
15 μg/mL. Optimal concentrations of second-step antibodies
are provided by the manufacturer but can be adjusted when

necessary to increase or decrease signal intensity.
33. To exclude any probe-independent signal generation caused by
any component of the detection system, a control reaction by
omitting the first step should be performed. In a parallel exper-
iment, tissue sections with known staining profile should be
used as a positive control.
34. Development of the signal (which should be performed in a
light-tight box) is usually complete after 20–30 min. In any
case, optimal development time should be visually determined
by the investigator.
35. To avoid precipitates in reagent 3 (signal-generating solution)
of the Vector®Red substrate kit and to maintain the strength of
the signal, we recommend to aliquot solutions of the three
reagents of the Vector®Red substrate kit and store them at
20 C.
36. To ascertain carbohydrate-inhibitable binding of probes, spec-
ificity controls using the canonical ligand lactose (Lac) should
be performed. In detail, solution containing galectin and 2%
BSA (w/v) in PBS (or PBS-T if performing fluorescence
microscopy) supplemented with 200 mM Lac is kept at RT
for 1 h. This mixture is then added to the tissue sections and
incubated at 4 C overnight prior to further processing (steps
4–11 in Subheading 3.6.2, steps 4–8 in Subheading 3.7.2, and
steps 8–12 in Subheading 3.7.3).
37. In fluorescence microscopy, we recommend including at least
one slide exclusively exposed to DAPI-containing solution, to
measure autofluorescence of processed tissue sections.
38. Storage at 4 C in the dark will preserve fluorescent signals of
sections for at least 1 month.
39. The reaction can also be carried out by heating the mixture at
reflux, without the need for a microwave reactor, although it
will take longer.
1
40. The product can be analyzed by H-NMR spectroscopic
analysis.
41. Metallic sodium reacts violently with water, keep the sodium in
oil. Do not scale up this procedure. Do not dispose of any
excess sodium metal in water.
42. It is also possible to clean the sodium surface by placing it in
MeOH for a short period before use and then transfer it to the
reaction mixture.
43. The reaction is typically complete after 12 h but may also be
much faster.
44. The product may be analyzed by 1H-NMR spectroscopy.
45. Systematic titrations with labeled galectins are recommended

to identify a concentration yielding a very strong signal with
minimal background to be tested in inhibition assays. This
allows most-practical quantitative detection of signal reduction
by the inhibitory compound in comparison.
References
1. Schrével J, Gros D, Monsigny M (1981) Cyto- 36(4):241–257. https://doi.org/10.1007/
chemistry of cell glycoconjugates. Prog Histo- s10719-019-09876-0
chem Cytochem 14(2):1–269. https://doi. 12. Kaltner H, Abad-Rodrı́guez J, Corfield AP,
org/10.1016/s0079-6336(81)80005-8 Kopitz J, Gabius H-J (2019) The sugar code:
2. Damjanov I (1987) Lectin cytochemistry and letters and vocabulary, writers, editors and
histochemistry. Lab Investig 57(1):5–20 readers and biosignificance of functional
3. Spicer SS, Schulte BA (1992) Diversity of cell glycan-lectin pairing. Biochem J
glycoconjugates shown histochemically: a per- 476(18):2623–2655. https://doi.org/10.
spective. J Histochem Cytochem 40(1):1–38. 1042/BCJ20170853
https://doi.org/10.1177/40.1.1370305 13. Amano M, Eriksson H, Manning JC, Detjen
4. Roth J (1993) Cellular sialoglycoconjugates: a KM, André S, Nishimura S-I, Lehtiö J, Gabius
histochemical perspective. Histochem J H-J (2012) Tumour suppressor p16INK4a:
25(10):687–710. https://doi.org/10.1007/ anoikis-favouring decrease in N/O-glycan/
BF00211765 cell surface sialylation by down-regulation of
5. Stowell SR, Ju T, Cummings RD (2015) Pro- enzymes in sialic acid biosynthesis in tandem
tein glycosylation in cancer. Annu Rev Pathol in a pancreatic carcinoma model. FEBS J
10:473–510. https://doi.org/10.1146/ 279(21):4062–4080. https://doi.org/10.
annurev-pathol-012414-040438 1111/febs.12001
6. Manning JC, Romero A, Habermann FA, Gar- 14. Bhide GP, Colley KJ (2017) Sialylation of
cı́a Caballero G, Kaltner H, Gabius H-J (2017) N-glycans: mechanism, cellular compartmen-
Lectins: a primer for histochemists and cell talization and function. Histochem Cell Biol
biologists. Histochem Cell Biol 147(2):149–174. https://doi.org/10.1007/
147(2):199–222. https://doi.org/10.1007/ s00418-016-1520-x
s00418-016-1524-6 15. Gao C, Hanes MS, Byrd-Leotis LA, Wei M,
7. Kaltner H, Garcı́a Caballero G, Ludwig A-K, Jia N, Kardish RJ, McKitrick TR, Steinhauer
Manning JC, Gabius H-J (2018) From glyco- DA, Cummings RD (2019) Unique binding
phenotyping by (plant) lectin histochemistry to specificities of proteins toward isomeric
defining functionality of glycans by pairing asparagine-linked glycans. Cell. Chem Biol
with endogenous lectins. Histochem Cell Biol 26(4):535–547 e534. https://doi.org/10.
149(6):547–568. https://doi.org/10.1007/ 1016/j.chembiol.2019.01.002
s00418-018-1676-7 16. Teichberg VI, Silman I, Beitsch DD, Resheff G
8. Cook GMW (1986) Cell surface carbohy- (1975) A β-D-galactoside-binding protein from
drates: molecules in search of a function? J electric organ tissue of Electrophorus electricus.
Cell Sci Suppl 4:45–70. https://doi.org/10. Proc Natl Acad Sci U S A 72(w4):1383–1387.
1242/jcs.1986.Supplement_4.5 https://doi.org/10.1073/pnas.72.4.1383
9. Reuter G, Gabius H-J (1999) Eukaryotic gly- 17. Harrison FL, Chesterton CJ (1980) Factors
cosylation: whim of nature or mediating cell-cell recognition and adhesion.
multipurpose tool? Cell Mol Life Sci Galaptins, a recently discovered class of bridg-
55(3):368–422. https://doi.org/10.1007/ ing molecules. FEBS Lett 122(2):157–165.
s000180050298 https://doi.org/10.1016/0014-5793(80)
80428-5
10. Gabius H-J, Roth J (2017) An introduction to
the sugar code. Histochem Cell Biol 18. Barondes SH (1984) Soluble lectins: a new
147(2):111–117. https://doi.org/10.1007/ class of extracellular proteins. Science
s00418-016-1521-9 223(4642):1259–1264. https://doi.org/10.
1126/science.6367039
11. Cummings RD (2019) Stuck on sugars: how
carbohydrates regulate cell adhesion, recogni- 19. Barondes SH (1997) Galectins: a personal
tion, and signaling. Glycoconj J overview. Trends Glycosci Glycotechnol
9(45):1–7. https://doi.org/10.4052/tigg.9.1
20. Kaltner H, Toegel S, Garcı́a Caballero G, Man- 29. Weinmann D, Kenn M, Schmidt S, Schmidt K,
ning JC, Ledeen RW, Gabius H-J (2017) Walzer SM, Kubista B, Windhager R,
Galectins: their network and roles in immu- Schreiner W, Toegel S, Gabius H-J (2018)
nity/tumor growth control. Histochem Cell Galectin-8 induces functional disease markers
Biol 147(2):239–256. https://doi.org/10. in human osteoarthritis and cooperates with
1007/s00418-016-1522-8 galectins-1 and -3. Cell Mol Life Sci
21. Hirabayashi J (2018) Special issue on galectins. 75(22):4187–4205. https://doi.org/10.
Trends Glycosci Glycotechnol 30(172): 1007/s00018-018-2856-2
SE1–SE223. https://doi.org/10.4052/tigg. 30. Manning JC, Garcı́a Caballero G, Knospe C,
1726.1SE Kaltner H, Gabius H-J (2017) Network analy-
22. Garcı́a Caballero G, Kaltner H, Kutzner TJ, sis of adhesion/growth-regulatory galectins
Ludwig A-K, Manning JC, Schmidt S, and their binding sites in adult chicken retina
Sinowatz F, Gabius H-J (2020) How galectins and choroid. J Anat 231(1):23–37. https://
have become multifunctional proteins. Histol doi.org/10.1111/joa.12612
Histopathol 35(6):509–539. https://doi.org/ 31. Manning JC, Garcı́a Caballero G, Knospe C,
10.14670/HH-18-199 Kaltner H, Gabius H-J (2018) Three-step
23. Gabius H-J, Brehler R, Schauer A, Cramer F monitoring of glycan and galectin profiles in
(1986) Localization of endogenous lectins in the anterior segment of the adult chicken eye.
normal human breast, benign breast lesions Ann Anat 217:66–81. https://doi.org/10.
and mammary carcinomas. Virchows Arch B 1016/j.aanat.2018.02.002
Cell Pathol Incl Mol Pathol 52(2):107–115. 32. Nio-Kobayashi J (2018) Histological mapping
https://doi.org/10.1007/BF02889955 and sub-type-specific functions of galectins in
24. Lahm H, André S, Höflich A, Fischer JR, health and disease. Trends Glycosci Glycotech-
Sordat B, Kaltner H, Wolf E, Gabius H-J nol 30(172):SE89–SE96. https://doi.org/10.
(2001) Comprehensive galectin fingerprinting 4052/tigg.1737.1SE
in a panel of 61 human tumor cell lines by 33. Garcı́a Caballero G, Schmidt S, Manning JC,
RT-PCR and its implications for diagnostic Michalak M, Schlötzer-Schrehardt U, Ludwig
and therapeutic procedures. J Cancer Res Clin A-K, Kaltner H, Sinowatz F, Schnölzer M,
Oncol 127(6):375–386. https://doi.org/10. Kopitz J, Gabius H-J (2020) Chicken lens
1007/s004320000207 development: complete signature of expression
25. Cooper DNW (2002) Galectinomics: finding of galectins during embryogenesis and evi-
themes in complexity. Biochim Biophys Acta dence for their complex formation with α-, β-,
1572(2–3):209–231. https://doi.org/10. δ- and τ-crystallins, N-CAM, and N-cadherin
1016/s0304-4165(02)00310-0 obtained by affinity chromatography. Cell Tis-
26. Kopitz J, von Reitzenstein C, André S, sue Res 379(1):13–35. https://doi.org/10.
Kaltner H, Uhl J, Ehemann V, Cantz M, 1007/s00441-019-03129-0
Gabius H-J (2001) Negative regulation of neu- 34. Gabius H-J, Wosgien B, Hendrys M, Bardosi A
roblastoma cell growth by carbohydrate- (1991) Lectin localization in human nerve by
dependent surface binding of galectin-1 and biochemically defined lectin-binding glycopro-
functional divergence from galectin-3. J Biol teins, neoglycoprotein and lectin-specific anti-
Chem 276(38):35917–35923. https://doi. body. Histochemistry 95:269–277. https://
org/10.1074/jbc.M105135200 doi.org/10.1007/BF00266777
27. Sturm A, Lensch M, André S, Kaltner H, 35. Hamburger V, Hamilton HL (1951) A series of
Wiedenmann B, Rosewicz S, Dignass AU, normal stages in the development of the chick
Gabius H-J (2004) Human galectin-2: novel embryo. J Morphol 88:49–92
inducer of T cell apoptosis with distinct profile 36. Hamburger V, Hamilton HL (1992) A series of
of caspase activation. J Immunol normal stages in the development of the chick
173(6):3825–3837. https://doi.org/10. embryo. Dev Dyn 195:231–272. https://doi.
4049/jimmunol.173.6.3825 org/10.1002/aja.1001950404
28. Stowell SR, Qian Y, Karmakar S, Koyama NS, 37. Ludwig A-K, Kaltner H, Kopitz J, Gabius H-J
Dias-Baruffi M, Leffler H, McEver RP, Cum- (2019) Lectinology 4.0: altering modular (ga)-
mings RD (2008) Differential roles of galectin- lectin display for functional analysis and bio-
1 and galectin-3 in regulating leukocyte viabil- medical applications. Biochim Biophys Acta
ity and cytokine secretion. J Immunol 1863(5):935–940. https://doi.org/10.1016/
180(5):3091–3102. https://doi.org/10. j.bbagen.2019.03.005
4049/jimmunol.180.5.3091 38. Kutzner TJ, Gabba A, FitzGerald FG, Shilova
NV, Garcı́a Caballero G, Ludwig A-K,
Manning JC, Knospe C, Kaltner H, localization. Cell Tissue Res 307(1):35–46.

Sinowatz F, Murphy PV, Cudic M, Bovin NV, https://doi.org/10.1007/s004410100457
Gabius H-J (2019) How altering the modular 46. Meldal MP, Meldal M, Tornøe CW (2017)
architecture affects aspects of lectin activity: Cu-catalyzed azide alkyne cycloaddition
case study on human galectin-1. Glycobiology Cu-catalyzed azide - alkyne cycloaddition.
29(8):593–607. https://doi.org/10.1093/ Chem Rev 108(8):2952–3015. https://doi.
glycob/cwz034 org/10.1021/cr0783479
39. Cerri DG, Arthur CM, Rodrigues LC, Fer- 47. Appukkuttan P, Mehta VP, van der Eycken EV
mino ML, Rocha LB, Stowell SR, Baruffi MD (2010) Microwave-assisted cycloaddition reac-
(2015) Examination of galectin localization tions. Chem Soc Rev 39(5):1467–1477.
using confocal microscopy. Methods Mol Biol https://doi.org/10.1039/b815717k
1207:343–354. https://doi.org/10.1007/ 48. Turnbull W, Daranas A (2004) On the value
978-1-4939-1396-1_23 of c: can low affinity systems be studied by
40. Lepp€anen A, Arthur CM, Stowell SR, Cum- isothermal titration calorimetry? J Am Chem
mings RD (2015) Examination of whole cell Soc 125:14859–14866. https://doi.org/10.
galectin binding by solid phase and flow cyto- 1021/ja036166s
metric analysis. Methods Mol Biol 1207: 49. Gasteiger E, Hoogland C, Gattiker A,
91–104. https://doi.org/10.1007/978-1- Duvaud S, Wilkins MR, Appel RD, Bairoch A
4939-1396-1_6 (2005) Protein identification and analysis tools
41. Lohr M, Kaltner H, Schwartz-Albiez R, on the ExPASy server. In: Walker JM (ed) The
Sinowatz F, Gabius H-J (2010) Towards func- proteomics protocols handbook. Humana
tional glycomics by lectin histochemistry: stra- Press, Totowa, NJ, pp 571–607. https://doi.
tegic probe selection to monitor core and org/10.1385/1-59259-890-0:571
branch-end substitutions and detection of 50. Garcı́a Caballero G, Beckwith D, Shilova NV,
cell-type and regional selectivity in adult Gabba A, Kutzner TJ, Ludwig A-K, Manning
mouse testis and epididymis. Anat Histol JC, Kaltner H, Sinowatz F, Cudic M, Bovin
Embryol 39(6):481–493. https://doi.org/10. NV, Murphy PV, Gabius H-J (2020) Influence
1111/j.1439-0264.2010.01019.x of protein (human galectin-3) design on
42. Ju T, Otto VI, Cummings RD (2011) The Tn aspects of lectin activity. Histochem Cell Biol
antigen: structural simplicity and biological 154(2):135–153. https://doi.org/10.1007/
complexity. Angew Chem Int Ed s00418-020-01859-9
50(8):1770–1791. https://doi.org/10.1002/ 51. Dam TK, Brewer CF (2003) Carbohydrate-
anie.201002313 lectin cross-linking interactions: structural,
43. Cervoni GE, Cheng JJ, Stackhouse KA, thermodynamic, and biological studies. Meth-
Heimburg-Molinaro J, Cummings RD ods Enzymol 362:455–486. https://doi.org/
(2020) O-glycan recognition and function in 10.1016/S0076-6879(03)01031-0
mice and human cancers. Biochem J 52. Dumas P, Ennifar E, Da Veiga C, Bec G,
477(8):1541–1564. https://doi.org/10. Palau W, Di Primo C, Piñeiro A, Sabin J,
1042/BCJ20180103 Muñoz E, Rial J (2016) Chapter seven -
44. Kaltner H, Manning JC, Garcı́a Caballero G, extending ITC to kinetics with kinITC. In:
Di Salvo C, Gabba A, Romero-Hernández LL, Feig AL (ed) Methods in enzymology, vol
Knospe C, Wu D, Daly HC, O’Shea DF, 567. Academic Press, pp 157–180. https://
Gabius H-J, Murphy PV (2018) Revealing bio- doi.org/10.1016/bs.mie.2015.08.026
medically relevant cell and lectin type- 53. Solı́s D, Bovin NV, Davis AP, Jiménez-
dependent structure-activity profiles for gly- Barbero J, Romero A, Roy R, Smetana K Jr,
coclusters by using tissue sections as assay plat- Gabius H-J (2015) A guide into glycosciences:
form. RSC Adv 8:28716–28735. https://doi. how chemistry, biochemistry and biology
org/10.1039/C8RA05382K cooperate to crack the sugar code. Biochim
45. Kaltner H, Seyrek K, Heck A, Sinowatz F, Biophys Acta 1850(1):186–235. https://doi.
Gabius H-J (2002) Galectin-1 and galectin-3 org/10.1016/j.bbagen.2014.03.016
in fetal development of bovine respiratory and 54. Schwarz A, Futerman AH (1997) Determina-
digestive tracts. Comparison of cell type- tion of the localization of gangliosides using
specific expression profiles and subcellular
anti-ganglioside antibodies: comparison of fix- Mol Biol 1207:51–62. https://doi.org/10.

ation methods. J Histochem Cytochem 1007/978-1-4939-1396-1_3
45(4):611–618. https://doi.org/10.1177/ 56. Mizoue LS, Tellinghuisen J (2004) The role of
002215549704500413 backlash in the “first injection anomaly” in iso-
55. Stowell SR, Arthur CM, Cummings RD, Feas- thermal titration calorimetry. Anal Biochem
ley CL (2015) Alkylation of galectin-1 with 326(1):125–127. https://doi.org/10.1016/j.
iodoacetamide and mass spectrometric ab.2003.10.048
mapping of the sites of incorporation. Methods
Chapter 18
Molecular Imaging for In Vivo Tracking and Detection

of Galectin Binding Partners
Thais Canassa De Leo, Sofia Nascimento dos Santos,
Emerson Soares Bernardes, Richard D. Cummings, Sean R. Stowell,
Abstract
Molecular imaging (MI) is a non-invasive growing technology that allows the investigation of cellular and
molecular processes in basic and clinical research and medicine. Luminescent proteins and radionuclides can
be associated to target molecules providing high-definition and real-time image of whole body in few
minutes or hours. Several MI studies have enabled the determination of molecular partners, in vivo
tracking, and fate of compounds in different disorders. Considering that galectins are multifaceted proteins
with great impact in many biological events, here we describe methods and strategies to generate labeled
galectins for in vivo non-invasive imaging studies.
Key words Molecular imaging, Bioluminescence, Radiolabeling, Galectin probes, Galectin-3
1 Introduction
Galectins are present at different levels of intra- and extra-cellular

compartments, playing important roles in a vast repertory of phys-
iological and pathophysiological processes. They have the intrinsic
ability to bind and crosslink glycan ligands and form dynamic net-
works that are continuously under discovery. Proliferative and
cytotoxic function of lymphocytes [1], tumor susceptibility [2],
autoimmune diseases [3], DNA splicing [4], and other biological
events are some of the described occurrences in which galectins are
involved. Besides, they have emerged as potential biomarkers of
several diseases, such as cardiovascular and inflammatory illness
[5, 6]. Hence, the galectins and their ligands represent molecular
targets of clinical relevance that continues to be clarified by tradi-
tional and new tools.
Molecular imaging (MI) has gained increasing space in the
research and diagnostic fields investigating specific in vivo cellular
339
340 Thais Canassa De Leo et al.
and molecular process in a non-invasive manner. It becomes possi-

ble to image whole body with advanced and high-definition imag-
ing in real time, dismissing the need to collect and treat tissues for
staining steps. Imaging modalities are mainly based on radio- and
isotope-labeling (PET/SPECT/MRI) and genetically engineered
targets using bioluminescent and fluorescent probes (optical
imaging) [7].
Radiolabel of ligands and drugs are considered promising tools
for the upgrading of personalized medicine [8]. The most common
radionuclides used are carbon, nitrogen, oxygen, fluorine, copper,
gallium, bromine, rubidium, yttrium, zirconium, technetium,
indium, and iodine [9]. Choosing the optimal radionuclide needs
to take into consideration whether its half-life is compatible with
the time for the drug to achieve the desired target and do not alter
the chemical properties and molecule structure [8]. Inhibitors,
drugs [10], monoclonal antibodies [11], ligands and tracers for
hormone receptors [12] have been radiolabeled for research, diag-
nosis, and therapy of several diseases, especially in oncology,
showing promising results. Nuclear medicine, employing the use
of radiopharmaceuticals, is a branch of MI that allows the measure-
ment of functional, metabolic, chemical, and biological processes
in vivo. Among the single-photon emission computed tomography
(SPECT) isotopes, 99mTc-based radiopharmaceuticals are the most
used radionuclides in the clinical practice. 99mTc can be easily
obtained through a 99Mo/99mTc generator and coupled to a pro-
tein, peptide, or other molecules. The injection of a 99mTc-labeled
molecule into a patient can be then detected by the radioactive
decay of 99mTc that leads to the production of γ-rays that are
captured by the SPECT cameras. SPECT imaging provides a
rapid, non-invasive evaluation of the biodistribution of 99mTc radi-
olabeled molecules in vivo [13].
On the other hand, bioluminescent reporter genes arises from
living organisms, such as sea pansy and firefly, which produce
luciferase enzymes able to catalyze luciferin substrates into photon
signal [14]. They are strategically engineered in fusion with targets
of interest, expressed in viruses, bacteria, plant, animal cells, or
whole animals, and administered along with its respective substrate
in order to monitor tumor growth [15], microorganisms’ spread
[16], track molecules in vivo, and identify targets and binding
partners [17]. To date, some authors have reported the develop-
ment of galectin probes as biotechnological tools to track their fate
and biodistribution in vivo [17–19] as well as galectin-labeled
inhibitors [20, 21].
The multifaceted biological features of galectins and the advan-
tages of modern MI encourage the development of galectin related
probes to be applied in basic and clinical non-invasive investiga-
tions. This chapter outlines the methods for preparing biolumines-
cent galectins using Renilla luciferase, a luciferase enzyme that
Molecular Imaging for In Vivo Tracking and Detection of Galectin Binding. . . 341
produces a blue light of 480 nm in the presence of coelenterazine

and oxygen, and radiolabeled galectins using the metastable tech-
netium form (99mTc) as powerful tools for in vivo tracking and
detection of galectin binding partners.
2 Materials
2.1 Bioluminescent 1. pGL4.75[hRluc/CMV/Hygro] plasmid (Promega) encoding

Galectins the Renilla luciferase (Rluc) gene.
2.1.1 Cloning Strategy: 2. pGEM-T cloning vector (Promega) cloned with mouse Galec-
Expression Vector for tin-3 (mGal-3) coding gene (see Note 1).
Recombinant 3. pET-28a(+) expression vector (see Note 2).
Bioluminescent Galectin 4. PCR primers for Rluc: we suggest Nde I and Bam HI restric-
tion sites to be included in the forward and reverse primers,
respectively.
5. PCR primers for mGal-3: we suggest Bam HI and Xho I
restriction sites to be included in the forward and reverse
primers, respectively. Include a stop codon in reverse primer
to stop transcription before His-tag on C-terminal.
6. Restriction enzymes: Nde I, Bam HI, Xho I (New England
Biolabs—NEB).
7. Taq DNA Polymerase (New England Biolabs—NEB).
8. Tris-acetate-EDTA Buffer (TAE) (Tris 40 mM, acetic acid
20 mM, EDTA 1 mM pH 7.6).
9. Agarose powder for DNA electrophoresis (Sigma-Aldrich).
10. Gel Extraction Kit (QIAquick—QIAGEN).
11. T4 DNA Ligase (New England Biolabs—NEB).
12. E. coli strain DH5-α (ThermoFisher).
13. QIAprep Spin Miniprep Kit (QIAGEN).
14. Luria Bertani (LB) broth (Sigma-Aldrich).
15. LB agar (Sigma-Aldrich).
16. Kanamycin (Sigma-Aldrich).
17. Electroporator (Bio-Rad).
2.1.2 Expression and 1. E. coli strain BL21-DE3 (Sigma-Aldrich) (see Note 3).
Purification of 2. Luria Bertani (LB) broth (Sigma-Aldrich).
Bioluminescent Galectin
3. LB agar (Sigma-Aldrich).
4. Kanamycin (Sigma-Aldrich) (see Note 4).
5. Isopropyl β-D-1-thiogalactopyranoside (IPTG) (Sigma-
Aldrich).
6. PBS (137 mM NaCl, 2.7 mM KCl, 10 mM Na2HPO4, 1.8 mM

KH2PO4 pH 7.3).
7. Lysis buffer: DNase 10 μg/mL, RNase 10 μg/mL, lysozyme
1 μg/mL, protease inhibitor cocktail (Roche), PBS 1 for
10 mL.
8. Sonicator.
9. Centrifuge.
10. α-Lactose-Agarose (Sigma).
11. HisTrap™ HP column (GE Healthcare).
12. Washing buffer (20 mM Na3PO4, 0.5 M NaCl pH 7.4).
13. Amicon® Ultra-15 Centrifugal Filter Device 30K (Millipore)
or Dialysis tubing membrane (Sigma).
14. Pierce™ BCA Protein Assay Kit (ThermoFisher Scientific).
2.1.3 Mouse Injection of 1. Desired mouse model with grown tumor (see Note 5).
Bioluminescent Galectin for 2. Isoflurane anesthetic (Cristália).
In Vivo Tracking
3. Pure and sterile bioluminescent mouse Galectin-3 (Rluc-
mGal3) preparation.
4. ViviRen™ substrate (Promega) (see Note 6).
5. PBS (137 mM NaCl, 2.7 mM KCl, 10 mM Na2HPO4, 1.8 mM
KH2PO4 pH 7.3).
6. Bovine serum albumin (Sigma-Aldrich).
7. MSFX-Pro equipment and Bruker Molecular Imaging software
(Bruker BioSpin Corporation).
2.2 Radiolabeling 1. Soluble and recombinantly expressed human Galectin-3 (hGal-

Galectins 3) (see Note 7).
2.2.1 Galectin 2. Hydrazinonicotinamide (HYNIC) (Future-Chem).
Conjugation to NHS-HYNIC 3. Sodium carbonate (0.1 M pH 9).
4. Dimethylformamide (DMF) (Sigma-Aldrich).
5. Magnetic agitator.
2.2.2 Galectin-HYNIC 1. PBS (137 mM NaCl, 2.7 mM KCl, 10 mM Na2HPO4, 1.8 mM

Purification KH2PO4 pH 7.3).
2. Sephadex G15 resin (GE Healthcare). Resuspend 3 g of Sepha-
dex G-15 resin per 9 mL of PBS, incubate overnight before use.
3. Glass wool.
4. Ammonium acetate (0.1 M pH 6).
5. Centrifuge.
2.2.3 Radiolabeling 1. Previously conjugated Galectin-HYNIC.

Galectin-HYNIC Using 2. Nitrogenated hydrochloric acid (HCl) (2 M) (Sigma) (see
99m
Tc Note 8).
3. Nitrogenated deionized water (see Note 8).
4. Stannous chloride (SnCl2) (Sigma-Aldrich).
5. Tricine (Sigma-Aldrich).
6. Ethylenediamine-N,N0 -diacetic acid (EDDA) (Sigma-Aldrich).
7. Technetium (99mTc) (99mTc generator, IPEN).
2.2.4 Galectin- 1. PBS (137 mM NaCl, 2.7 mM KCl, 10 mM Na2HPO4, 1.8 mM

HYNIC-99mTc Purification KH2PO4 pH 7.4).
2. Glass wool.
3. Sephadex G15 resin (GE Healthcare). Resuspend 3 g of Sepha-
dex G-15 resin per 9 mL of PBS, incubate overnight before use.
4. Microcentrifuge.
2.2.5 Mouse Injection of 1. Desired mouse model with grown tumor (see Note 5).
Galectin-HYNIC-99mTc for 2. Isoflurane anesthetic (Cristália).
In Vivo Tracking
3. Galectin-HYNIC-99mTc.
4. Albira microPET/SPECT/CT imaging system (Bruker BioS-
pin Corporation).
5. Albira and PMOD software.
3 Methods
3.1 Bioluminescent 1. Amplify Rluc gene from pGL4.75[hRluc/CMV/Hygro] plas-

Galectins mid (100 ng) to include Nde I (50 ) and Bam HI (30 ) restriction
sites by PCR using the respective primers and Taq DNA Poly-
3.1.1 Cloning Strategy:
merase as described by the manufacturer (see Note 9).
Expression Vector for
Recombinant 2. Resolve the reaction in 1% agarose gel (TAE buffer) and purify
Bioluminescent Galectin Rluc PCR amplicon (approximately 950 base pairs) using Gel
Extraction Kit as described by the manufacturer. Store at
20 C.
3. Amplify the mGal-3 gene from pGEM-T plasmid (100 ng) to
include Bam HI (50 ) and Xho I (30 ) restriction sites by PCR
using the respective primers and Taq DNA Polymerase as
described by the manufacturer (see Note 9).
4. Resolve the reaction in 1% agarose gel (TAE buffer) and purify
mGal-3 PCR amplicon (approximately 800 base pairs) using
Gel Extraction Kit as described by the manufacturer. Store at
20 C.
Fig. 1 pET-28-Rluc-mGal3 expression vector. Renilla luciferase and Galectin

coding sequences were inserted between Nde I/Bam HI and Bam HI/Xho I cloning
sites from a pET-28a(+) expression vector. This construction possesses a 6xHis-
tag at Rluc N-terminal, Gal C-terminal is free by adding a stop codon, and the
recombinant protein is under the control of lac operator and T7 promoter
5. Digest 1 μg of pET-28a(+) expression vector using Nde I and

Bam HI restriction enzymes as described by the manufacturer,
resolve in 1% agarose gel (TAE buffer), and purify the linear-
ized vector using Gel Extraction Kit as described by the manu-
facturer. In parallel, digest Rluc and mGal-3 amplicons using
Nde I/Bam HI and Bam HI/Xho I enzymes, respectively, to
open the cohesive cloning ends.
6. Clone Rluc purified amplicon into Nde I and Bam HI sites
from linearized pET-28a(+) using T4 DNA Ligase as described
by the manufacturer, transform into bacteria DH5-α by elec-
troporation and plate in LB agar plates supplemented with
kanamycin (50 μg/mL). Incubate overnight at 37 C.
7. Pick up colonies using a sterile p10 pipette tip, culture in 3 mL
of LB media supplemented with kanamycin (50 μg/mL), and
culture overnight at 37 C, 200 rpm.
8. Extract plasmids using QIAprep Spin Miniprep Kit and identify
positive colonies containing pET-28a(+) plasmids with Rluc
(pET-28-Rluc) gene by restriction enzyme digestion (Nde I
and Bam HI) using approximately 400 ng of plasmid. Store
positive clones at 20 C (see Note 10).
9. Digest 1 μg of pET-28-Rluc expression vector using Bam HI
and Xho I restriction enzymes as described by the
manufacturer, resolve in 1% agarose gel (TAE buffer), and

purify the linearized vector using Gel Extraction Kit as
described by the manufacturer.
10. Clone mGal-3 purified amplicon into Bam HI and Xho I sites
from linearized pET-28-Rluc using T4 DNA Ligase as
described by the manufacturer, transform into bacteria
DH5-α by electroporation, plate in LB agar plates supplemen-
ted with kanamycin (50 μg/mL) and incubate overnight at
37 C. Analyze positive clones by restriction enzyme digestion
(Bam HI and Xho I). The final vector pET-28-Rluc-mGal3 is
illustrated on Fig. 1. Store positive clones at 80 C (see
Note 11).
3.1.2 Expression and 1. Transform BL21-DE3 by heat shock method using 100 ng of
Purification of Recombinant pET-28-Rluc-mGal3 and plate into LB agar plates supplemen-
Bioluminescent Galectin ted with kanamycin (50 μg/mL).
2. Pick up a single colony using a sterile p10 pipette tip and
prepare a 5 mL starter culture overnight in LB media supple-
mented with kanamycin (50 μg/mL).
3. Inoculate the 5 mL starter culture in 1 L of sterile LB media
supplemented with kanamycin (50 μg/mL), culture at 37 C
and 200 rpm until optical density reaches 0.6 (600 nm), and
induce protein expression by adding 0.5 mM of IPTG. Then,
culture this suspension for 6 h at 25 C and 200 rpm (see
Note 12).
4. Harvest cells at 4 C, 4000 g, for 15 min.
5. Resuspend cell pellet using 10 mL of lysis buffer and keep on
ice for 30 min.
6. Divide the volume into 50 mL Falcon tubes and complete cell
lysis by sonication. Perform 12 short burst of 20 s and intervals
of 40 s (40% of maximum power) keeping tubes on ice bath.
Sonicator tip should be close to the tube bottom, but not
touching it. Avoid foaming.
7. Centrifuge at 4 C, 8000 g, for 40 min.
8. Collect supernatant (soluble fraction) and filter with 0.22 μm
filter.
9. Purify the supernatant containing Rluc-mGal3 protein by affin-
ity chromatography using Lactosyl-Sepharose resin, according
to manufacturer instructions.
10. Elute Rluc-mGal3 using 100 mM lactose and pool samples
according to OD 280 reading.
11. High levels of purity can be achieved by a second step purifica-
tion using nickel resin. Prepare a HisTrap™ HP resin
according to manufacturer instructions and apply Rluc-mGal3

lactose-purified sample.
12. Wash the resin with 10 column volumes of washing buffer.
13. Elute Rluc-mGal3 in 4 volumes of washing buffer supplemen-
ted with imidazole 150 mM and more 4 volumes of washing
buffer supplemented with imidazole 250 mM (see Note 13).
14. Read eluted protein samples at OD 280 nm, pool them, and
remove imidazole and/or lactose towards PBS using an Ami-
con® Ultra-15 Centrifugal Filter Device or a Dialysis tubing
membrane.
15. Calculate protein concentration using Pierce™ BCA Protein
Assay Kit according to manufacturer instructions.
16. Sterilize the purified protein with 0.22 μm filter and store at
4 C.
3.1.3 Mouse Injection of 1. Dilute ViviRen™ substrate in PBS (0.295 mM) and supple-
Rluc-mGal3 for In Vivo ment with bovine serum albumin 0.1% (w/v), protect from
Tracking light.
2. Inject 100 μg of Rluc-mGal3 intravenously/intratumorally
into a mouse tumor model (see Note 5). Negative control can
be performed administering Rluc-mGal3 in the presence of
lactose 100 Mm (see Note 14).
3. After 8 h, anesthetize animals using isoflurane 2% and inject
200 μL of ViviRen™ prepared on step 1.
4. After 15 min, acquire grayscale and luminescent color as well as
the overlay pseudocolor image using MSFX-Pro equipment
and process images (Fig. 2—Reproduced from De Leo 2020
with permission from Elsevier).
3.2 Radiolabeling 1. Tube A: Dilute 200 μg of hGal-3 in sodium carbonate buffer

Galectins (0.1 M pH 9) and reserve.
3.2.1 Galectin 2. Considering 200 μg of hGal-3 and its MW, calculate the respec-
Conjugation to NHS-HYNIC tive amount of HYNIC (MW ¼ 287.67 g/mol) using 20:1
molar ratio proportion (hGal-3:HYNIC).
3. Tube B: Prepare a fresh solution of succinimidyl-6-hydrazino-
pyridine-3-carboxylate (NHS-hynic) in dry DMF and add it
dropwise with agitation to a final galectin: HYNIC molar ratio
of 20:1 into tube A. DMF concentration in the final reaction
mixture should be <5%.
4. Complete the reaction mixture volume to 200 μL with sodium
carbonate buffer (0.1 M pH 9).
Fig. 2 Luminescent imaging of recombinant mGal-3 fused with Renilla luciferase located in the tumor region.
100 μg of Rluc-mGal3 was intratumorally injected in Balb/c nude mouse bearing MKN45 derived tumor. After
8 h, mouse was anesthetized with isoflurane, Renilla substrate (ViviRen) was administered, and images were
acquired for 15 min using MSFX-Pro equipment and processed using Bruker Molecular Imaging software.
Reproduced from Biochemical and Biophysical Research Communications: “Engineering of galectin-3 for
glycan-binding optical imaging” (2020), with permission from Elsevier
5. Incubate this reaction at room temperature under gentle agita-

tion for 60 min.
3.2.2 Galectin-HYNIC 1. With the aid of a needle, drill a little hole at the bottom of a
Purification 1.5 mL tube and fill the bottom with a small amount of
glass wool.
2. Apply 1 mL of Sephadex G-15 resin into the 1.5 mL tube from
step 1.
3. Place the 1.5 mL tube into a Falcon tube and centrifuge at
3075 g for 1 min at room temperature.
4. Wash Sephadex G-15 resin using 1 mL of ammonium acetate
for 4 times, in a total volume of 4 mL. Discard the eluted
volume.
5. Place the 1.5 mL tube into a new Falcon.
6. Gently apply hGal-3-HYNIC in the top of resin, avoiding

touching it.
7. Centrifuge at 5000 rpm for 1 min.
8. Collect and store the purified hGal-3-HYNIC for further
radiolabeling.
3.2.3 Radiolabeling 1. Tube A: Dilute 2 mg of SnCl2 in 100 μL of nitrogenated HCl

Galectin-HYNIC Using and protect from light.
99m
Tc 2. Tube B: Dilute 10 mg of tricine in 500 μL of nitrogenated
water.
3. Tube C: Dilute 10 mg of EDDA in 250 μL of nitrogenated
water.
4. Using a glass flask, mixture add: (1) 200 μg of purified hGal-3-
HYNIC, (2) 5 μL of SnCl2 (tube A), (3) 100 μL of tricine (tube
B), (4) 50 μL EDDA (tube C), and (5) 5 mCi (millicurie) or
185 MBq (megabecquerel) of 99mTc freshly eluted.
5. Seal the flask, nitrogenate the solution for 2 min, and incubate
at room temperature for 45 min (see Note 8).
3.2.4 Galectin- 1. With the aid of a needle, drill a little hole at the bottom of a
HYNIC-99mTc Purification 0.6 mL tube and compact a small amount of glass wool.
2. Apply 0.5 mL of Sephadex G-15 resin into the 0.6 mL tube
from step 1.
3. Place the 0.6 mL tube into a 2 mL tube and centrifuge at
5000 rpm for 1 min.
4. Wash Sephadex G-15 resin using 0.5 mL of PBS for 4 times.
Discard the eluted volume.
5. Place the 0.6 mL tube into a new 2 mL tube.
6. Gently apply the nitrogenated solution in the top of resin,
avoiding touching it.
7. Centrifuge at 5000 rpm for 1 min.
8. Collect and store the purified hGal-3-HYNIC-99mTc and verify
activity (mCi).
3.2.5 Mouse Injection of 1. Inject hGal-3-HYNIC-99mTc (1 mCi or 37 MBq) intrave-

Galectin-HYNIC-99mTc for nously/intratumorally into a mouse tumor model (see Note
In Vivo Tracking 5). Negative control can be performed administering hGal-3-
HYNIC-99mTc in the presence of lactose 100 mM.
2. After 2 h, anesthetize animals using isoflurane 2% and record
SPECT/CT data scan using Albira microPET/SPECT/CT
imaging system.
3. Images can be reconstructed using the Albira software and
analyzed with the PMOD software (Fig. 3).
Fig. 3 SPECT/CT imaging of recombinant hGal-3 radiolabeled with 99mTc located

in the tumor region. Recombinant hGal-3 was radiolabeled with 99mTc and
intravenously injected (37 MBq) in a Balb/c nude mouse bearing a
glioblastoma U87-derived tumor. 1-h post injection, mouse was anesthetized
with 2% isoflurane and SPECT (45 s/projection) and CT (35 keV, 400 μA) image
was acquired using Albira microPET/SPECT/CT imaging system (Brucker).
Images were reconstructed using the Albira System and analyzed with PMOD
software. A representative coronal static small animal SPECT/CT image is shown
4 Notes
1. Any other desired Galectin coding gene can be used depending

on the experiment proposal. Galectin coding sequences can be
obtained from the National Center for Biotechnology Infor-
mation (NCBI) website and cloned onto alternative cloning
vectors different from pGEM-T system. Usually, protein genes
are preserved cloned into cloning vectors at 20 C
and/or 80 C.
2. Alternatively, any other desired expression vector can be used
instead of pET-28a(+). This may change restriction enzymes
and a new cloning strategy should be designed.
3. BL21-DE3 is suggested as the classic and well-known E. coli

strain to express recombinant proteins, but other variants can
be tested and may require new expression tests.
4. Kanamycin is the antibiotic resistance marker from pET-28a
(+). If an alternative vector is used, check the resistance
marker gene.
5. Tumor bearing mouse model can be generated in approxi-
mately 3 weeks by inoculating 1 106 tumor cells subcutane-
ously in Balb/c mice. Some cell lineages may be more
successful to grow tumors than others, in this sense, we recom-
mend MKN45 cell line, which has glycan-binding partners
capable of being recognized by galectin-3 [17], and the aggres-
sive U87 glioblastoma cell line [22].
6. ViviRen™ has been showing promising results as substrate of
Renilla luciferase, but other options such as EnduRen™ and
novel synthetic photoactivatable substrates are available
[23, 24].
7. Any other desired soluble and recombinant galectin can be
used depending on the experiment proposal.
8. To remove oxygen, solutions should be nitrogenated, with the
inert gas N2 (99.9%). Briefly, solutions are kept in a glass flask
and capped with a rubber septum. A disposable needle attached
to a N2 line is used to purge N2 inside the flask and a second
needle is used to vent excess gas throughout the degassing. The
solution should be bubbled for 20 min at room temperature.
9. Tm Calculator is an online tool offered by ThermoFisher that
can be accessed to calculate temperature parameters for PCR
cycles.
10. Restriction digestion to identify positive clones can be analyzed
by resolving the reaction in 1% agarose gel (TAE buffer). Thus,
it is possible to visualize bands at approximately 5000 and
950 base pairs referent to pET-28a(+) vector and Rluc gene,
respectively, for positive clones.
11. Restriction digestion to identify positive clones can be analyzed
by resolving the reaction in 1% agarose gel (TAE buffer). Thus,
it is possible to visualize bands at approximately 6000 and
800 base pairs referent to pET-28-Rluc vector and mGal-3
gene, respectively for positive clones.
12. Different expression parameters may be required to produce
other bioluminescent galectins rather than mGal-3 to optimize
soluble protein yielding. Thus, new expression tests can be
performed according to the following steps: prepare approxi-
mately 24 Falcon tubes containing 10 mL of sterile LB media
culture supplemented with kanamycin (50 μg/mL). Each tube
will represent a different combination of parameters such as
IPTG (0.1–1 mM), time of induction (4–18 h), and tempera-

ture (18–37 C). Use a “zero” IPTG tube as a negative control.
Inoculate 50 μL of a starter culture in each test tube and
incubate under 200 rpm, 37 C until optical density reaches
0.4 (600 nm). Add IPTG and incubate according to time and
temperature test. Centrifuge at 4000 g, 10 min, 4 C,
remove supernatant, resupend cell pellet in 500 μL of deio-
nized water and mix with 1 Sample Buffer, Laemmli. Boil for
5 min and load into 12% or 4–20% SDS-PAGE gel using
denaturing and reducing conditions. Bands intensity can be
observed at predicted molecular weight by coomassie staining.
13. The purification parameters using HisTrap™ HP resin
described here were optimized for Rluc-Gal3 purification.
Other cloning strategies using different vectors and/or galec-
tins may require new optimizations of imidazole supply and
other additives, as observed in manufacturer instructions
manual.
14. Rluc-mGal3 should be pre-incubated at room temperature for
15 min with lactose 100 mM before administration.
Acknowledgments
We are thankful to the Brazilian National Council for Scientific

and Technological Development (CNPq—#312606/2019-2 for
M.D.B.).
References
1. Corapi E, Carrizo G, Compagno D, Laderach 5. Hernández-Romero D, Vı́lchez JA, Lahoz Á

D (2018) Endogenous galectin-1 in T lympho- et al (2017) Galectin-3 as a marker of intersti-
cytes regulates anti-prostate cancer immunity. tial atrial remodelling involved in atrial fibrilla-
Front Immunol 9:2190. https://doi.org/10. tion. Sci Rep 7:40378. https://doi.org/10.
3389/fimmu.2018.02190 1038/srep40378
2. Dos Santos SN, Sheldon H, Pereira JX et al 6. Yu TB, Dodd S, Yu L-G, Subramanian S
(2017) Galectin-3 acts as an angiogenic switch (2020) Serum galectins as potential biomarkers
to induce tumor angiogenesis via Jagged-1/ of inflammatory bowel diseases. PLoS One 15:
Notch activation. Oncotarget 8: e0227306. https://doi.org/10.1371/journal.
49484–49501. https://doi.org/10.18632/ pone.0227306
oncotarget.17718 7. Baker M (2010) The whole picture. Nature
3. Mendez-Huergo SP, Hockl PF, Stupirski JC 463:977–979. https://doi.org/10.1038/
et al (2019) Clinical relevance of galectin-1 463977a
and galectin-3 in rheumatoid arthritis patients: 8. Mammatas LH, Verheul HMW, Hendrikse
differential regulation and correlation with dis- NH et al (2015) Molecular imaging of targeted
ease activity. Front Immunol 9:3057. https:// therapies with positron emission tomography:
doi.org/10.3389/fimmu.2018.03057 the visualization of personalized cancer care.
4. Vyakarnam A, Dagher SF, Wang JL, Patterson Cell Oncol 38:49–64. https://doi.org/10.
RJ (1997) Evidence for a role for galectin-1 in 1007/s13402-014-0194-4
pre-mRNA splicing. Mol Cell Biol 17: 9. Vallabhajosula S (2009) Chemistry: basic
4730–4737. https://doi.org/10.1128/mcb. principles. In: Vallabhajosula S (ed) Molecular
17.8.4730
imaging—radiopharmaceuticals for PET and galectin-3 for glycan-binding optical imaging.

SPECT, 1st edn. Springer-Verlag, Heidelberg Biochem Biophys Res Commun 521:674–680.
10. Isin EM, Elmore CS, Nilsson GN et al (2012) https://doi.org/10.1016/j.bbrc.2019.
Use of radiolabeled compounds in drug metab- 10.161
olism and pharmacokinetic studies. Chem Res 18. Nakahara S, Oka N, Wang Y et al (2006) Char-
Toxicol 25:532–542. https://doi.org/10. acterization of the nuclear import pathways of
1021/tx2005212 galectin-3. Cancer Res 66:9995–10006.
11. Dammes N, Peer D (2020) Monoclonal https://doi.org/10.1158/0008-5472.CAN-
antibody-based molecular imaging strategies 06-1772
and theranostic opportunities. Theranostics 19. Farhadi SA, Bracho-Sanchez E, Fettis MM et al
10:938–955. https://doi.org/10.7150/thno. (2018) Locally anchoring enzymes to tissues
37443 via extracellular glycan recognition. Nat Com-
12. van Kruchten M, de Vries EGE, Brown M et al mun 9:4943. https://doi.org/10.1038/
(2013) PET imaging of oestrogen receptors in s41467-018-07129-6
patients with breast cancer. Lancet Oncol 14: 20. Kumar SR, Deutscher SL (2008) 111In-
e465–e475. https://doi.org/10.1016/ labeled galectin-3—targeting peptide as a
S1470-2045(13)70292-4 SPECT agent for imaging breast tumors. J
13. Boschi A, Uccelli L, Martini P (2019) A picture Nucl Med 49:796–803. https://doi.org/10.
of modern Tc-99m radiopharmaceuticals: pro- 2967/jnumed.107.048751
duction, chemistry, and applications in molec- 21. Bratteby K, Torkelsson E, L’Estrade ET et al
ular imaging. Appl Sci 9:2526. https://doi. (2020) In vivo veritas: 18F-radiolabeled glyco-
org/10.3390/app9122526 mimetics allow insights into the pharmacologi-
14. Mezzanotte L, van‘t Root M, Karatas H et al cal fate of galectin-3 inhibitors. J Med Chem
(2017) In vivo molecular bioluminescence 63:747–755. https://doi.org/10.1021/acs.
imaging: new tools and applications. Trends jmedchem.9b01692
Biotechnol 35:640–652. https://doi.org/10. 22. Diao W, Tong X, Yang C et al (2019) Behaviors
1016/j.tibtech.2017.03.012 of glioblastoma cells in in vitro microenviron-
15. Souza TKF, Nucci MP, Mamani JB et al (2018) ments. Sci Rep 9:85. https://doi.org/10.
Image and motor behavior for monitoring 1038/s41598-018-36347-7
tumor growth in C6 glioma model. PLoS 23. Otto-Duessel M, Khankaldyyan V, Gonzalez-
One 13:e0201453. https://doi.org/10. Gomez I et al (2006) In vivo testing of Renilla
1371/journal.pone.0201453 luciferase substrate analogs in an orthotopic
16. Ventura JD, Beloor J, Allen E et al (2019) murine model of human glioblastoma. Mol
Longitudinal bioluminescent imaging of Imaging 5:57–64. https://doi.org/10.2310/
HIV-1 infection during antiretroviral therapy 7290.2006.00006
and treatment interruption in humanized mice. 24. Zhang C, Cheng L, Dong G et al (2018) Novel
PLoS Pathog 15:e1008161. https://doi.org/ photoactivatable substrates for Renilla lucifer-
10.1371/journal.ppat.1008161 ase imaging in vitro and in vivo. Org Biomol
17. De Leo TC, Nascimento dos Santos S, Del Chem 16:4789–4792. https://doi.org/10.
Cistia AC et al (2020) Engineering of 1039/c8ob01192c
Chapter 19
Visualization of Cytosolic Galectin Accumulation Around

Damaged Vesicles and Organelles
Ming-Hsiang Hong, I-Chun Weng, Fang-Yen Li, and Fu-Tong Liu
Abstract
Galectins are animal lectins that recognize β-galactoside and bind glycans. Recent studies have indicated
that cytosolic galectins recognize cytosolically exposed glycans and accumulate around endocytic vesicles or
organelles damaged by various disruptive substances. Accumulated galectins engage other cytosolic pro-
teins toward damaged vesicles, leading to cellular responses, such as autophagy. Disruptive substances
include bacteria, viruses, particulate matters, and protein aggregates; thus, this process is implicated in the
pathogenesis of various diseases. In this chapter, we describe methods for studying three disruptive
substances: photosensitizers, Listeria monocytogenes, and Helicobacter pylori. We summarize the tools
used for the detection of cytosolic galectin accumulation around damaged vesicles.
Key words Galectins, Glycans, Endosomes, Phagosomes, Lysosomes, Photosensitizers, Listeria

monocytogenes, Helicobacter pylori, Fluorescence microscopy, APEX2-enhanced electron microscopy,
Immunofluorescence microscopy, Immunogold electron microscopy
1 Introduction
Galectins are synthesized without classical signal peptides and are

present in the cytosolic compartment, sometimes abundantly
[1]. Cytosolic galectins are reportedly associated with cytosolic
binding partners and modulate cell fate in a glycan-independent
manner [2]. Glycans displayed on glycoproteins and glycolipids are
galectin ligands. These glycans are present on the cell surface and
can be endocytosed and subsequently presented on the luminal side
of endocytic vesicles, including endosomes, phagosomes, and lyso-
somes. Moreover, lysosomes contain glycoproteins and glycolipids
transported from the endoplasmic reticulum or Golgi. Recent
research has indicated that cytosolic galectins can also control
cellular responses in a glycan-dependent manner by recognizing
glycans present on damaged endosomes or cellular organelles, con-
sequently accumulating around these vesicles [3]. The accumula-
tion of cytosolic galectins has been observed in response to various
353
354 Ming-Hsiang Hong et al.
cellular insults, including bacterial infection [4, 5], viral infection

[6, 7], and pathogenic protein aggregate deposition [8–10]. Here,
we summarize methods for detecting cytosolic galectin accumula-
tion in response to different stimuli, including light-illuminated
photosensitizers within endosomes, Listeria monocytogenes infec-
tion, and Helicobacter pylori infection, using fluorescence micros-
copy, immunofluorescence microscopy, enhanced ascorbate
peroxidase 2 (APEX2)-enhanced electron microscopy, and immu-
nogold electron microscopy [3, 11, 12].
Disulfonated aluminum phthalocyanine (AlPcS2a) is a far-red
fluorescent dye and an amphiphilic photosensitizer, which localizes
within cytosolic endosomes or lysosomes when endocytosed by
cells. The illumination of AlPcS2a-containing cells with red light
generates reactive oxygen species, leading to endosomal or lyso-
somal damage, which efficiently triggers cytosolic galectin accumu-
lation around these damaged vesicles. Accumulated galectins
recruit cytosolic autophagy adapter proteins or marker proteins to
damaged endocytic vesicles and target them for autophagy [3, 13,
14]. In this chapter, we summarize the use of fluorescence and
APEX2-enhanced electron microscopy to detect cytosolic galectin
accumulation around endosomes damaged by light-illuminated
photosensitizers.
Listeria monocytogenes (L. monocytogenes) is an intracellular
rod-shaped Gram-positive bacterium. L. monocytogenes infection
causes listeriosis, which results in sepsis, meningitis, encephalitis,
and pneumonia [15]. L. monocytogenes can invade phagocytic cells,
such as macrophages, and the bacteria can be internalized into
phagosomes. L. monocytogenes releases the virulence factor listerio-
lysin O (LLO) and two phospholipase C enzymes, which compro-
mise the vacuolar membrane of phagosomes and help the
bacterium to escape into the cytoplasm [16]. Cytosolic galectins,
including galectin-3 and galectin-8, sense and recognize cytosolic-
exposed glycans on damaged phagosomes, which are originally
displayed on the luminal side of phagosomes. These molecules
accumulate around the damaged organelles and modulate autop-
hagy targeting to them [11]. In this chapter, we summarize the use
of immunofluorescence and immunogold electron microscopy to
monitor the cytosolic galectin distribution in mouse macrophages
infected with L. monocytogenes.
Helicobacter pylori (H. pylori) is an extracellular, helix-shaped,
gram-negative bacterium. Its infection is associated with human
diseases, such as gastritis and gastric ulcers [17]. It has been
shown to trigger cytosolic galectin-8 aggregation in AGS gastric
epithelial cells. This aggregation results from the recognition of
exposed host glycans on the membrane of injured lysosomes.
Moreover, vacuolating cytotoxin A was found to be a critical bacte-
rial factor in driving H. pylori-associated lysosomal damage
[12]. Here, we describe a method of immunofluorescence to
Visualization of Cytosolic Galectin Accumulation Around Damaged Vesicles. . . 355
monitor cytosolic galectin distribution in gastric epithelial cells

infected with H. pylori.
2 Materials
2.1 Examination of 1. Chinese Hamster Ovary CHO-K1 cell line (ATCC: CCL-61).
Cytosolic Galectin 2. Minimum essential medium alpha (MEM-alpha; Life Technol-
Accumulation Around ogies Limited, Paisley, UK).
Endosomes Damaged
3. Phenol red-free MEM-alpha (Life Technologies Limited,
by a Light-Illuminated Paisley, UK).
Photosensitizer Using
4. Fetal bovine serum (FBS).
Fluorescence
Microscopy 5. Dulbecco’s phosphate-buffered saline (PBS; Life Technologies
Limited, Paisley, UK).
2.1.1 Cell Culture
6. Penicillin-streptomycin (Life Technologies Limited,
Paisley, UK).
2.1.2 Plasmids 1. TagRFP-galectin-3-, Galectin-3-EGFP-, and galectin-8-

EGFP-expressing plasmids [3].
2.1.3 Reagents 1. Lipofectamine 2000 transfection reagent (Invitrogen,

Paisley, UK).
2. Disulfonated aluminum phthalocyanine (AlPcS2a) (Frontier
Scientific, Logan, USA) (see Note 1).
3. Paraformaldehyde (Sigma, St. Louis, USA) (see Note 2).
4. Hoechst 33342 (Invitrogen, Paisley, UK).
2.1.4 Equipment 1. CO2 incubator.

2. Collimated LED red light source (Thorlabs, Newton, USA).
3. ImageXpress Micro system fluorescence microscope (Molecu-
lar Devices, San Jose, USA).
4. MetaXpress software (Molecular Devices, San Jose, USA).
2.2 Examination of 1. Galectin-3-Flag-APEX2-, and galectin-8-Flag-APEX2-expres-

Cytosolic Galectin sing plasmids [3].
Accumulation Around
Endosomes Damaged
by a Light-Illuminated
Photosensitizer Using
APEX2-Enhanced
Electron Microscopy
2.2.1 Plasmids
2.2.2 Reagents 1. Lipofectamine 2000 transfection reagent (Invitrogen,

Paisley, UK).
2. Disulfonated aluminum phthalocyanine (AlPcS2a) (Frontier
Scientific, Logan, USA).
3. Glutaraldehyde (Electron Microscopy Sciences,
Hatfield, USA).
4. Sodium cacodylate, trihydrate (Electron Microscopy Sciences,
Hatfield, USA).
5. Calcium chloride (Sigma, St. Louis, USA).
6. Glycine (Sigma, St. Louis, USA).
7. 3,3-Diaminobenzidine (DAB; Sigma, St. Louis, USA).
8. Hydrogen peroxide (H2O2; Sigma, St. Louis, USA).
9. Osmium tetroxide (OsO4; Electron Microscopy Sciences,
Hatfield, USA).
10. Uranyl acetate (Electron Microscopy Sciences, Hatfield, USA).
11. Ethanol (Fisher Scientific, Pittsburgh, USA).
12. Epon resin (Sigma-Aldrich, St. Louis, USA).
2.3 Examination of 1. Bone marrow-derived macrophages were isolated and prepared

Cytosolic Galectin as previously described [11].
Accumulation Around 2. Roswell Park Memorial Institute (RPMI)-1640 medium (Life
L. monocytogenes- Technologies Limited, Paisley, UK).
Containing 3. HEPES (Life Technologies Limited, Paisley, UK).
Phagosomes in Mouse
4. Non-essential amino acids (Life Technologies Limited,
Bone Marrow-Derived
Paisley, UK).
Macrophages Using
Immunofluorescence 5. FBS (Life Technologies Limited, Paisley, UK).
Microscopy 6. Gentamicin (Sigma, St. Louis, USA).
2.3.1 Cell Culture 7. Dulbecco’s PBS (Life Technologies Limited, Paisley, UK).
2.3.2 Bacterial Culture 1. Listeria monocytogenes (L. monocytogenes) strain 10403S (see
Note 3).
2. Brain Heart Infusion (BHI) broth (Becton, Dickinson and
Company, Franklin Lakes, USA).
2.3.3 Reagents 1. Paraformaldehyde (Sigma, St. Louis, USA).

2. Goat anti-human galectin-3 antibody (R&D Systems,
Minneapolis, USA).
Minneapolis, USA).
4. Alexa Fluor-conjugated donkey anti-goat IgG (Invitrogen,
Paisley, UK).
5. Saponin (Sigma, St. Louis, USA).
6. Casein (Sigma, St. Louis, USA).

7. Bovine serum albumin (BSA; Sigma, St. Louis, USA).
2.3.4 Equipment 1. Incubator (37 C) with orbital shaker.

2. CO2 incubator.
3. Centrifuge.
4. Spectrophotometer.
2.4 Examination of 1. Paraformaldehyde (Sigma, St. Louis, USA).

Cytosolic Galectin 2. Potassium phosphate (Sigma, St. Louis, USA).
Accumulation Around
3. Sodium phosphate (Sigma, St. Louis, USA).
L. monocytogenes-
Containing 4. Gelatin (Sigma, St. Louis, USA).
Phagosomes in Mouse 5. Goat anti-human galectin-3 antibody (R&D Systems,
Bone Marrow-Derived Minneapolis, USA).
Macrophages Using 6. 10-nm gold-conjugated anti-goat IgG (Sigma,
Immunogold Electron St. Louis, USA).
Microscopy 7. Uranyl acetate (Electron Microscopy Sciences, Hatfield, USA).
2.4.1 Reagents 8. Ethanol (Fisher Scientific, Pittsburgh, USA).
9. Propylene oxide (Sigma, St. Louis, USA).
10. Lead citrate (Sigma, St. Louis, USA).
11. Sodium hydroxide (NaOH; Sigma, St. Louis, USA).
12. Osmium tetroxide (OsO4; Electron Microscopy Sciences,
Hatfield, USA).
2.5 Examination of 1. Human gastric AGS cell line (ATCC: CRL-1739).

Cytosolic Galectin 2. RPMI-1640 medium (Life Technologies Limited,
Accumulation Around Paisley, UK).
Damaged Lysosomes
3. FBS (Life Technologies Limited, Paisley, UK).
in H. pylori-Infected
Gastric Epithelial Cells
4. Dulbecco’s PBS (Life Technologies Limited, Paisley, UK).
Using
Immunofluorescence
Microscopy
2.5.1 Cell Culture
2.5.2 Bacteria Culture 1. Helicobacter pylori (H. pylori) strain 26695 (ATCC: 700392).
2. Brucella broth (Becton, Dickinson and Company, Franklin
Lakes, USA).
2.5.3 Reagents 1. Paraformaldehyde (Sigma, St. Louis, USA).

Minneapolis, USA).
3. Goat anti-human galectin-3 antibody (R&D Systems, Minnea-
polis, USA).
4. Alexa Fluor-conjugated donkey anti-goat IgG (Invitrogen,
Paisley, UK).
5. Triton X-100 (Sigma, St. Louis, USA).
6. BSA (Sigma, St. Louis, USA).
2.5.4 Equipment 1. Anaerobic jar (Mitsubishi Gas Chemical, Tokyo, Japan).

2. Microaerophilic gas pack (Mitsubishi Gas Chemical, Tokyo,
Japan).
3. 37 C incubator with orbital shaker.
4. CO2 incubator.
5. Centrifuge.
7. T-25 flask.
3 Methods
3.1 Examination of 1. CHO cells were grown in MEM-alpha containing 10% (v/v)
Cytosolic Galectin FBS, 100 units/mL of penicillin, and 100 μg/mL of strepto-
Accumulation Around mycin at 37 C in 5% CO2.
Endosomes Damaged 2. CHO cells were transfected with TagRFP-galectin-3-, galectin-
by a Light-Illuminated 3-EGFP-, or galectin-8-EGFP-expressing plasmids using Lipo-
Photosensitizer Using fectamine 2000 reagent (see Notes 5 and 6).
Fluorescence 3. Cells were washed three times with PBS and incubated with
Microscopy phenol red-free MEM-alpha containing 10% (v/v) FBS and
(See Note 4) 1 μM AlPcS2a for 15 min.
4. Cells were washed three times with PBS and loaded with phe-
nol red-free MEM-alpha containing 10% (v/v) FBS.
5. Cells were illuminated with 635 nm red light.
6. Cells were washed three times with PBS and fixed using 4%
paraformaldehyde solution for 10 min at 25 C.
7. Cells were washed with PBS, and the cell nuclei were stained
with Hoechst 33342 for 15 min at 25 C. Cells were then
washed with PBS.
8. Images were automatically acquired using the ImageXpress

Micro system fluorescence microscope, and the distribution
of TagRFP-galectin-3, galectin-3-EGFP, or galectin-8-EGFP
was analyzed using MetaXpress software.
3.2 Examination of 1. CHO cells were grown in MEM-alpha containing 10% (v/v)
Cytosolic Galectin FBS and penicillin-streptomycin at 37 C in 5% CO2.
Accumulation Around 2. CHO cells were transfected with galectin-3-Flag-APEX2- or
Endosomes Damaged galectin-8-Flag-APEX2-expressing plasmids using Lipofecta-
by a Light-Illuminated mine 2000 reagent.
Photosensitizer Using 3. Cells were washed three times with PBS and incubated with
APEX2-Enhanced phenol red-free MEM-alpha containing 10% (v/v) FBS and
Electron Microscopy 1 μM AlPcS2a for 15 min.
(See Note 7)
4. Cells were washed three times with PBS and loaded with phe-
nol red-free MEM-alpha containing 10% (v/v) FBS.
5. Cells were illuminated with 635 nm red light.
6. Cells were washed three times with PBS and fixed with 2%
(v/v) glutaraldehyde in water at 25 C. Samples were then
placed on ice for 60 min.
7. Samples were washed using cold buffer containing 100 mM
sodium cacodylate and 2 mM calcium chloride.
8. Samples were incubated in cold buffer containing 20 mM gly-
cine and placed on ice for 5 min to quench the unreacted
aldehyde functional groups.
9. Samples were incubated in a solution containing 0.5 mg/mL
DAB and 10 mM H2O2 for 30 min. Galectin-APEX2 converts
DAB into an insoluble DAB polymer.
10. Samples were washed with cold buffer containing 100 mM
sodium cacodylate and 2 mM calcium chloride.
11. Samples were incubated in cold buffer containing 100 mM
sodium cacodylate, 2 mM calcium chloride, and 2% (w/v)
OsO4 for 30 min. The DAB polymer reacts with OsO4 and
deposits electron-dense osmium resulting in electron micros-
copy contrast.
12. Samples were washed three times using cold water.
13. Samples were incubated in a solution containing 2% (w/v)
uranyl acetate in the dark at 4 C for 1 h.
14. Samples were washed three times using cold water.
15. Samples were dehydrated by placing in ice-cold buffer for
2 min, as follows: 20, 50, 70, 90, and 100% (v/v) ethanol.
16. Ethanol was removed and the samples were submerged in an
EPON/ethanol mixture at 25 C for 30 min to infiltrate the
resin, as follows: 25, 50, 67, and 100% (v/v) EPON in ethanol.
17. Samples were embedded and polymerized in EPON embed-

ding medium for 48 h at 60 C.
18. Samples were ultramicrotomized and subjected to transmission
electron microscopy (TEM). Galectin-APEX2 distribution
presented as a dark color under electron microscopy.
3.3 Examination of 1. L. monocytogenes were inoculated into sterile BHI broth

Cytosolic Galectin L. monocytogenes culture and incubated at 37 C in an incubator
Accumulation Around with orbital shaking at 200 rpm.
L. monocytogenes- 2. After overnight (16–20 h) culture, the OD600 of the culture
Containing was measured using a spectrophotometer, and diluted with
Phagosomes in Mouse fresh sterile BHI broth to an OD600 of 0.1.
Bone Marrow-Derived 3. L. monocytogenes culture was incubated at 37 C with orbital
Macrophages Using shaking at 200 rpm for 4–6 h. This provides a L. monocytogenes
Immunofluorescence culture at the exponential mid-log growth phase with an
Microscopy OD600 of approximately 0.5.
(See Note 8)
4. Bone marrow-derived macrophages were prepared as previ-
ously described [11].
5. Mid-log-phase L. monocytogenes were washed with PBS and
resuspended in antibiotic-free RPMI-1640 medium supple-
mented with 10% FBS (v/v). The medium was then added to
the mouse bone marrow-derived macrophages at a multiplicity
of infection of 0.25.
6. Plates were then centrifuged at 500 g for 5 min to promote
contact between bacteria and bone marrow-derived macro-
phages. The cells were incubated at 37 C for 30 min and
washed twice with PBS. Fresh RPMI-1640 medium supple-
mented with 10% FBS (v/v) containing 10 μg/mL gentamicin
was then added to kill extracellular bacteria.
7. Cells were washed three times with PBS and fixed using 4%
paraformaldehyde solution in PBS for 10 min at 25 C.
8. Cells were washed three times with PBS, permeabilized with
0.05% saponin, and then blocked with 2.5% casein in PBS
containing 1% (w/v) BSA. Cells were stained with primary
antibodies against galectins overnight at 4 C. After washing
three times with PBS, cells were incubated with Alexa Fluor-
conjugated secondary antibodies for 1 h at 25 C.
9. Cells were washed three times with PBS, and cell nuclei were
stained with Hoechst 33342 for 15 min at 25 C. Cells were
washed three times with PBS.
of galectin-3 and galectin-8 was analyzed using MetaXpress
software.
3.4 Examination of 1. L. monocytogenes-infected bone marrow-derived macrophages

Cytosolic Galectin were washed three times with PBS and fixed with 4% parafor-
Accumulation Around maldehyde in 0.1 M Sorensen’s phosphate buffer (0.133 M
L. monocytogenes- potassium phosphate and 0.133 M sodium phosphate) for
Containing 30 min at 25 C. Samples were then washed three times with
Phagosomes in Mouse Sorensen’s phosphate buffer.
Bone Marrow-Derived 2. Samples were post-fixed with 1% OsO4 in Sorensen’s phos-
Macrophages Using phate buffer for 1–2 h at 4 C, and then washed 5 times in
Immunogold Electron distilled water.
Microscopy 3. Samples were dehydrated by placing in ice-cold buffer for
10 min, as follows: 30%, 50%, 70%, 80%, 90%, 100%, 100%,
and 100% (v/v) ethanol.
4. Samples were treated with cold ethanol containing 50% (v/v)
propylene oxide for 10 min. Samples were then washed for
10 min with cold propylene oxide for 3 times. Eventually
samples were slowly brought to 25 C.
5. Cells were then embedded in epoxy resin followed by ultrathin
cryosection.
6. Samples were blocked on gold grids with 1% gelatin in distilled
water for 30 min at 25 C.
7. Cells were incubated overnight at 4 C with 0.2, 0.5, or 1 μg/
mL goat anti-human galectin-3 antibody in PBS.
8. Samples were washed seven times with PBS. Samples were
incubated with 10 nm gold-conjugated anti-goat IgG diluted
1:50 in PBS, and incubated for 1 h at 25 C.
9. Samples were washed three times with PBS and then stained
with 4% uranyl acetate in 70% ethanol for 15 min.
10. Samples were washed with double-distilled water for 3 times,
and samples were stained with lead citrate solution (5 mg of
lead citrate in 1 mL of 4.5 mM NaOH) for 15 min. Then
samples were washed with double-distilled water for 3 times,
and dried on filter paper.
11. Samples were analyzed by TEM.
3.5 Examination of 1. H. pylori was cultured in a T-25 flask containing 10 mL sterile

Cytosolic Galectin Brucella broth supplemented with 10% FBS.
Accumulation Around 2. The T-25 flask was placed in a 2.5 L anaerobic jar containing a
Damaged Lysosomes microaerophilic gas pack.
in H. pylori-Infected 3. The anaerobic jar was incubated at 37 C on an orbital shaker
Human Gastric (100 rpm) overnight (16–18 h).
Epithelial Cells Using
4. After overnight culture, the OD600 of the overnight culture
Immunofluorescence
(OD600 of 1.0 ¼ 109 CFU/mL) was measured using a spec-
Microscopy
trophotometer (see Note 9).
5. H. pylori culture was centrifuged at 12,000 g at 25 C for

3 min.
6. The supernatant was discarded and the bacterial pellet was
washed with 1 mL sterile PBS.
7. The samples were then centrifuged, the supernatant was dis-
carded, and the bacteria were resuspended in 50 μL sterile PBS.
8. The bacterial suspension was diluted with RPMI-1640
medium supplemented with 10% FBS.
9. Human gastric AGS cells were grown at 70–80% confluence in
RPMI 1640 medium supplemented with 25 mM HEPES and
10% FBS, and then incubated in a 5% CO2 incubator at 37 C.
10. H. pylori suspension was added to AGS cells, and H. pylori-
infected cells were centrifuged at 400 g for 10 min at 25 C
to facilitate host-bacterial contact. H. pylori-infected cells were
incubated in a 5% CO2 incubator at 37 C for 8 h.
11. Cells were washed with PBS and fixed with 4% paraformalde-
hyde for 20 min at 25 C.
12. Cells were permeabilized by adding 0.2% Triton X-100 in PBS
at 25 C for 30 min.
13. Cells were washed with PBS and incubated with blocking
buffer (2% BSA in PBS) at 37 C for 30 min.
14. Cells were stained with primary antibodies against galectins
overnight at 4 C. After three washes with PBS, cells were
stained with Alexa Fluor-conjugated secondary antibodies for
1 h at 25 C.
15. Cells were washed three times with PBS and the cell nucleus
was stained with Hoechst 33342 for 15 min at 25 C. Cells
were washed three times with PBS.
of galectins was analyzed using MetaXpress software.
4 Notes
1. AlPcs2a powder was reconstituted using a small volume of

0.1 M NaOH and diluted using PBS to generate a stock
solution. The solution was passed through a 0.2 μm filter and
stored at 20 C in small aliquots [14]. To avoid any damage
induced by light to the photosensitizer, and to protect cells
from undesirable photochemical damage, tubes with
photosensitizer-containing solutions and dishes with
photosensitizer-loaded cells can be packed in aluminum foil.
2. Paraformaldehyde, glutaraldehyde, sodium cacodylate, DAB,

OsO4, and uranyl acetate solution are toxic and should be used
in a fume hood. Personal protective equipment should be worn
while manipulating these reagents, and the solution containing
these chemicals should be treated as hazardous waste.
3. L. monocytogenes and H. pylori are biosafety level 2 (BSL-2)
pathogens. The culture and infection of these bacteria should
be conducted in a BSL-2 laboratory with appropriate biosafety
practices and procedures.
4. Endosomal damage is efficiently induced in galectin-tag-
expressing cells when these cells are loaded with photosensiti-
zers followed by red light illumination. The accumulation of
fluorescent galectin-tag can be easily observed using fluores-
cence microscopy. This method was used to investigate the
contribution of glycan changes to cytosolic galectin accumula-
tion and galectin-mediated cellular responses. Additionally, we
observed that treatment with TD139, a small cell-permeable
galectin-3 inhibitor that targets carbohydrate binding domain
of galectin-3, could efficiently diminish the accumulation of
cytosolic galectin-3 [3]. This technique may be applied for the
development and screening of galectin inhibitors.
5. Galectin-EGFP- or galectin-TagRFP can rapidly accumulate
around damaged endocytic vesicles. Therefore, live tracing of
cytosolic galectin distribution is feasible using fluorescence
microscopy [3].
6. Transfection reagents have been reported to induce endosomal
damage and galectin accumulation [18, 19]. Therefore, we
aimed to establish stable galectin-tag-expressing cells in the
presence of selection antibiotics to determine whether chal-
lenges induce endocytic vesicle damage and galectin
accumulation.
7. Our research group previously reported that galectin-APEX2
converts DAB into an insoluble DAB polymer that reacts with
OsO4 and deposits electron-dense osmium to generate a con-
trast, which can be visualized using electron microscopy,
around damaged endosomes [3]. Therefore, galectin asso-
ciated with damaged membrane structures of endocytic vesicles
under different stimuli can be easily and directly observed using
APEX2-enhanced electron microscopy.
8. Using immunofluorescence microscopy and immunogold elec-
tron microscopy, we found that infection with L. monocytogenes
efficiently induces galectin accumulation in mouse macro-
phages. We have shown that galectin-3 accumulation inhibits
autophagy and promotes the survival of L. monocytogenes.
Using immunofluorescence microscopy, we also observed that
galectin-8 accumulated around damaged lysosomes when
gastric cells were infected with H. pylori. These methods have

been used to identify galectin accumulation under different
stimuli, such as Shigella flexneri and Salmonella Typhimurium
infection [4, 5].
9. Normally, H. pylori has a spiral shape. However, when H. pylori
culture is overgrown, the bacterium may become coccoid
instead of spiral-shaped. Although there is controversial infor-
mation regarding the infectivity of coccoid H. pylori, the coc-
coid form is non-culturable in vitro. To confirm the shape of
the bacteria, we added 10 μL of overnight culture to a glass
slide, covered with a coverslip, and viewed under a light
microscope.
References
1. Cooper DN, Barondes SH (1990) Evidence selective autophagy protects against seeded
for export of a muscle lectin from cytosol to tau aggregation. J Biol Chem
extracellular matrix and for a novel secretory 293(7):2438–2451
mechanism. J Cell Biol 110(5):1681–1691 9. Freeman D, Cedillos R, Choyke S, Lukic Z,
2. Yang RY, Hsu DK, Liu FT (1996) Expression McGuire K, Marvin S, Burrage AM,
of galectin-3 modulates T-cell growth and apo- Sudholt S, Rana A, O’Connor C, Wiethoff
ptosis. Proc Natl Acad Sci U S A CM, Campbell EM (2013) Alpha-synuclein
93(13):6737–6742 induces lysosomal rupture and cathepsin
3. Hong MH, Lin WH, Weng IC, Hung YH, dependent reactive oxygen species following
Chen HL, Chen HY, Chen P, Lin CH, Yang endocytosis. PLoS One 8(4):e62143
WY, Liu FT (2019) Intracellular galectins con- 10. Siew JJ, Chen HM, Chen HY, Chen HL, Chen
trol cellular responses commensurate with cell CM, Soong BW, Wu YR, Chang CP, Chan YC,
surface carbohydrate composition. Glycobiol- Lin CH, Liu FT, Chern Y (2019) Galectin-3 is
ogy 30(1):49–57 required for the microglia-mediated brain
4. Thurston TL, Wandel MP, von Muhlinen N, inflammation in a model of Huntington’s dis-
Foeglein A, Randow F (2012) Galectin 8 tar- ease. Nat Commun 10(1):3473
gets damaged vesicles for autophagy to defend 11. Weng IC, Chen HL, Lo TH, Lin WH, Chen
cells against bacterial invasion. Nature HY, Hsu DK, Liu FT (2018) Cytosolic
482(7385):414–418 galectin-3 and -8 regulate antibacterial autop-
5. Paz I, Sachse M, Dupont N, Mounier J, hagy through differential recognition of host
Cederfur C, Enninga J, Leffler H, Poirier F, glycans on damaged phagosomes. Glycobiol-
Prevost MC, Lafont F, Sansonetti P (2010) ogy 28(6):392–405
Galectin-3, a marker for vacuole lysis by inva- 12. Li FY, Weng IC, Lin CH, Kao MC, Wu MS,
sive pathogens. Cell Microbiol 12(4):530–544 Chen HY, Liu FT (2019) Helicobacter pylori
6. Maier O, Marvin SA, Wodrich H, Campbell induces intracellular galectin-8 aggregation
EM, Wiethoff CM (2012) Spatiotemporal around damaged lysosomes within gastric epi-
dynamics of adenovirus membrane rupture thelial cells in a host O-glycan-dependent man-
and endosomal escape. J Virol ner. Glycobiology 29(2):151–162
86(19):10821–10828 13. Chu YP, Hung YH, Chang HY, Yang WY
7. Staring J, von Castelmur E, Blomen VA, van (2017) Assays to monitor lysophagy. Methods
den Hengel LG, Brockmann M, Baggen J, Thi- Enzymol 588:231–244
baut HJ, Nieuwenhuis J, Janssen H, van Kup- 14. Hung YH, Chen LM, Yang JY, Yang WY
peveld FJ, Perrakis A, Carette JE, (2013) Spatiotemporally controlled induction
Brummelkamp TR (2017) PLA2G16 repre- of autophagy-mediated lysosome turnover. Nat
sents a switch between entry and clearance of Commun 4:2111
Picornaviridae. Nature 541(7637):412–416 15. Schlech WF (2019) Epidemiology and clinical
8. Falcon B, Noad J, McMahon H, Randow F, manifestations of listeria monocytogenes infec-
Goedert M (2018) Galectin-8-mediated tion. Microbiol Spectr 7(3):GPP3-0014-2018
16. Portnoy DA, Auerbuch V, Glomski IJ (2002) Autophagy induced by calcium phosphate pre-
The cell biology of Listeria monocytogenes cipitates targets damaged endosomes. J Biol
infection: the intersection of bacterial patho- Chem 289(16):11162–11174
genesis and cell-mediated immunity. J Cell 19. Wittrup A, Ai A, Liu X, Hamar P, Trifonova R,
Biol 158(3):409–414 Charisse K, Manoharan M, Kirchhausen T, Lie-
17. Atherton JC (2006) The pathogenesis of Heli- berman J (2015) Visualizing lipid-formulated
cobacter pylori-induced gastro-duodenal dis- siRNA release from endosomes and target gene
eases. Annu Rev Pathol 1:63–96 knockdown. Nat Biotechnol 33(8):870–876
18. Chen X, Khambu B, Zhang H, Gao W, Li M,
Chen X, Yoshimori T, Yin XM (2014)
Chapter 20
Transcytosis of Galectin-3 in Mouse Intestine

Alena Ivashenka, Christian Wunder, Valerie Chambon, Estelle Dransart,
Ludger Johannes, and Massiullah Shafaq-Zadah
Abstract
The GlycoLipid-Lectin (GL-Lect) hypothesis provides a conceptual framework to explain how endocytic
pits are built in processes of clathrin-independent endocytosis. According to this hypothesis, oligomeric
cellular or pathogenic lectins interact with glycosylated plasma membrane lipids in a way such as to drive the
formation of tubular endocytic pits that then detach to generate clathrin-independent endocytic carriers for
the cellular uptake of cellular or pathogenic products. This process operates in a complementary manner to
the conventional clathrin pathway for biological function linked to cell polarity. Up to date, the premises of
the GL-Lect hypothesis have been based on model membrane and cell culture experiments. It has therefore
become urgent to extend its exploration to complex organisms. In the current protocol, we describe
methods to study the endocytosis and transcytosis of a key driver of the GL-Lect mechanism, the cellular
galectin-3, and of one of its cargoes, lactotransferrin, in enterocytes of the intact jejunum of mice. In a step-
by-step manner, we present the generation of fluorescent endocytic ligands, tissue preparation for cellular
uptake measurements, binding and internalization assays, tissue fixation and preparation for sectioning,
light and electron microscopical observations, and quantification of data by image processing. Pitfalls are
discussed to optimize the chances of success with the described methods.
Key words Galectin-3, Lactotransferrin, Intestine, Transcytosis, Enterocytes, Mucus
1 Introduction
1.1 General Galectins are widely expressed small soluble lectins that undergo
Introduction on unconventional secretion, following their synthesis in the cytosol
Galectins [1]. Therefore, galectins may have a dual localization, extracellular
and intracellular. As a common feature, galectins contain a carbo-
hydrate recognition domain (CRD), specialized in high-affinity
binding to β-galactoside containing N-acetyllactosamine glycans
that are found on glycoproteins and glycolipids [2, 3]. To date,
15 galectins have been identified in mammals where they have been
implicated in a wide range of physiological and pathological func-
tions including cell adhesion, migration, apoptosis, inflammation,
367
368 Alena Ivashenka et al.
fibrosis, cell-host defense as well as immune responses [3–9]. Here,

we focus on the chimera type galectin-3 (Gal3), which has the
particularity to have, in addition to the C-terminal CRD domain,
a N-terminal oligomerization domain, responsible for the forma-
tion of higher-order complexes at the cell surface [10–13]. Gal3
thereby controls the plasma membrane dynamics of glycoproteins
that are implicated in differentiation, migration, or apoptosis, by
positively or negatively regulating their endocytotic uptake into
cells [13–16].
In some cases, galectins have been described to promote the
formation of galectin lattices at the plasma membrane, thereby
preventing the internalization of cargo proteins. Remarkably, this
process operates in breast cancer where Gal3 is associated with
EGFR and increases the membrane residence time of the latter
[17, 18]. A similar association and retention have been observed
for the VEGFR during angiogenesis [19]. For the GPI-anchored
protein CD59, internalization in HeLa cells was increased upon
Gal3 depletion, again suggesting lattice formation with Gal3 could
stabilize CD59 at the plasma membrane [20]. In contrast, the
impairment of all galectins with lactose as a competitive inhibitor
decreased CD59 internalization, suggesting that other galectins
could actually drive the endocytic uptake process.
Galectins have indeed also been described to stimulate the
internalization of glycoproteins [21]. Similar to what had been
reported before for glycosphingolipid (GSL)-binding pathogenic
lectins [22], it has recently been shown that Gal3 also drives the
formation of narrow membrane bending in interaction with GSLs
[23]: Monomeric Gal3 binds to glycosylated proteins such as
CD44 or α5β1 integrin, oligomerizes, then interacts with GSLs
to in fine induce the biogenesis of tubular endocytic pits from
which defined endocytic structures, the clathrin-independent car-
riers, are formed [23]. This mechanistic proposal has been termed
the GL-Lect hypothesis, where GL stands for glycolipids and Lect
for lectin [23, 24]. Recently, a similar GL-Lect process has been
reported for galectin-8 and GSL-dependent cellular uptake of
CD166 [25].
1.2 Endogenous The group of Kobayashi elegantly mapped galectin expression pat-
Galectin Expression in terns in the mouse digestive tract in a region- and cell type-specific
the Digestive Tract manner [26–28] (Fig. 1a). Initially, RNA expression was analyzed,
which led to the identification of galectin-2, -3, -4/6, -7, and -9
within the gut epithelium of mice [26]. Specific antibodies were
then used for the immunohistochemistry-based localization of the
following galectins along the mouse digestive tract:
1.2.1 Galectin- Gal2 is mainly expressed in the upper 2/3 of the crypts (prolifera-
2 (Glandular Stomach and tive zone), the base part of the villi, as well as in goblet cells
Small Intestine) (Fig. 1b). Gal2 is diffusely immune-reactive within the cytoplasm.
Transcytosis of Galectin-3 in Mouse Intestine 369
Fig. 1 Expression profile of key galectins in the digestive tract of mice (modified from Ref. 26). (a) Distribution
of multiple galectin subtypes within the differentiated epithelial tissues of the murine digestive tract. Of note,
Gal3 is broadly expressed, whereas Gal2 is mainly present in the small intestine, and Gal7 only in stratified
squamous epithelium, notably of esophagus and anus. (b) Sub-tissular expression profile of Gal2, Gal3, and
Gal4 within the small intestine: Gal3 is mainly located at the tip of the villus, Gal4 preferentially at the lower
part of the villus, and Gal2 even lower and also at the upper part of the crypt region. (c) Endogenous Gal3
expression pattern as analyzed at sub-cellular resolution on fixed tissue. Gal3 signal is found at the subapical
part of enterocytes (white arrow), and even more strongly at the basal side (white arrowhead). (d) Endogenous
expression of Gal3 using Carnoy’s tissue fixation to preserve the mucus layer. Immunofluorescence shows
strong Gal3 signal overlapping with the mucus marker UEA-1. Cells are systematically oriented as follows:
apical side (Ap) on top, and basolateral side (Bl) at the bottom. Scale bars ¼ 10 μm
Of note, no Gal2 immunoreactivity was observed within the large

intestine, including its goblet cells.
1.2.2 Galectin-3 (All Gal3 is widely expressed. In the gut, it is present from the esopha-
Along the Gut) gus, stomach, and small intestine to colon and anus (Fig. 1a). In the
small intestine, Gal3 immunoreactivity is restricted to the tips of
villi (Fig. 1b), where it shows a subapical cytoplasmic staining
pattern (Fig. 1c) which corresponds to the myosin-rich terminal
web. In contrast, in large intestine Gal3 is homogenously
distributed within the epithelium. In addition, Gal3 is immunore-

active at the surface and in the cytoplasm of mucous cells, and in the
mucus layer itself (Fig. 1d).
1.2.3 Galectin-4/6 Gal4/6 are mainly found in the lower part of villi and the upper half
(Glandular Stomach, Small of the crypts of absorptive intestinal cells (Fig. 1b). In villi, Gal4/6
Intestine, and Lower are diffusely located at the brush border and at the basolateral
Digestive Tract) membrane, whereas at the mature upper enterocyte zone, these
proteins are only found at the basolateral membrane. In the pro-
liferative zone of crypts, both galectins are found in the cytoplasm.
1.2.4 Galectin-7 Gal7 is preferentially found in keratinocytes and expressed in all

(Stratified Squamous stages of epidermal maturation. In these cells, Gal7 is found around
Epithelium) lateral intercellular junctions, and in the basal extracellular matrix,
hereby regulating cell-cell adhesion and cell migration, respectively.
In the digestive tract, Gal7 is highly immunoreactive within the
nonkeratinized cells of the stratified squamous epithelium, includ-
ing the tongue, esophagus, forestomach, and anus [29] (Fig. 1a).
1.3 Transcytosis Epithelial cells such as intestinal enterocytes are defined by apical
and basolateral membranes that are separated by tight junctions. In
order to maintain this apico-basal polarity, sorting of specific pro-
teins and lipids into the correct membranes is required. Such polar-
ized transport is in part ensured by the biosynthetic/secretory
pathway where neo-synthetized proteins transit from the ER via
the Golgi to the apical or the basolateral membrane [30, 31]. Some
of these proteins undergo subsequent trafficking across the cell and
reach the opposite membrane, a process that has been termed
transcytosis [32]. The polymeric immunoglobulin receptor
(pIgR) associated to its ligand (pIgA) was extensively used to
study this transcytotic route: pIgR binds pIgA at the basolateral
membrane, and the complex is then transcytosed to the apical
surface in a tyrosine kinase Src/calcium-dependent manner [33–
35]. The retromer complex has been implicated in the transcytosis
of pIgA/pIgR complex within hepatic tissue [35], in which neo-
synthesized proteins are exclusively transported from the Golgi to
the basolateral membrane, from which apical proteins must
undergo transcytosis to reach the apical membrane [36]. In the
same cellular context, the raftophilic GPI-anchored protein CD59
also undergoes this type of transcytosis to reach the apical mem-
brane [37, 38]. In kidney tissue, interleukin 11 (IL-11) is actively
signaling in association with its receptors IL-11R1 and IL-11R2,
both at apical and basolateral membranes [39]. However, the apical
pool of IL-11 is initially localized at the basolateral membrane,
from where it is then transcytosed and released apically in an
IL-11R1 receptor-dependent manner [39].
Proteins and lipids are also transcytosed from the apical to the
basolateral membranes, ensuring key physiological functions within
the absorptive intestinal tissue. For example, newly born mice
initially have an immature immune system and need to take up
immunoglobulins from their mother’s milk. In this context, the
IgG contained in the milk is transcytosed from the apical side and
delivered to the basolateral membrane, a process regulated by the
Arp2/3 complex, an actin nucleation factor [34, 40]. The other
major function of the small intestine is to absorb nutrients includ-
ing lipids. In neonates, triglycerides contained in the milk are
further digested into monoglycerides and fatty acids and are subse-
quently taken up from the apical membrane. These are then trans-
cytosed to the basolateral side for entry into the blood stream
[41, 42]. The lipid transporter CD36 is implicated in this process,
which again involves the actin nucleator Arp2/3, suggesting a
general requirement of the F-actin cytoskeleton in apical-to-baso-
lateral transcytosis [40].
Importantly, this nutrient uptake is regulated by several hor-
mones including leptin, which is secreted by the gastric mucosa and
is itself transcytosed in an apical-to-basolateral manner to be
released into the bloodstream. Bypassing the blood-brain barrier,
this hormone reaches the hypothalamus to either control hunger or
satiety sensing [43].
In the endothelial tissue, low-density lipoprotein (LDL) from
the plasma undergoes apical transcytosis before being released into
the subendothelial space. Interestingly, the retention of LDL in this
region is associated with the physio-pathological atherosclerosis
process, which is further stimulated by angiotensin II, in a LDLR-
mediated and caveolae-dependent manner [44].
LDLR is well expressed at the blood-brain barrier, where it
likely contributes to the receptor-mediated transcytosis of ligand
molecules from the circulating blood into the brain. This mecha-
nism represents a delivery route for many drugs that need to pass
the blood-brain barrier to reach the brain for the treatment of
central nervous system (CNS) diseases [45].
Transcytotic trafficking is also used by a range of microbial
pathogens to infect the digestive tract. Studies on reconstituted
intestinal M-like cells of the lymphoid tissue of the gastrointestinal
tract revealed that Salmonella was transcytosed across these cells in
a caveolae-dependent manner, as part of its invasion program
[46]. Shiga toxin 1, which is produced by enterohemorrhagic
E. coli strains is responsible for severe physiological alteration of
enterocyte functions, including hemorrhagic colitis. It has been
reported that Stx-1 was macropinocytosed in a myosin II and
CDC42-dependent manner, transcytosed across the enterocyte,
and released at the basolateral membrane to spread throughout
the infected organism [47]. Other pathogens such as Listeria
monocytogenes (Lm) use apically accessible E-cadherin on intesti-

nal mucus-expelling goblet cells as a host-receptor to achieve their
pathogenic activity. After rapid uptake, Lm is transcytosed through
the intestinal epithelium and basolaterally delivered for systemic
dissemination [48].
Taken together, these reports illustrate that the transcytosis
pathway within specialized intestinal cells is hijacked by several
bacterial pathogens to induce systemic infection. The molecular
characterization of this process may help to identify new targets
for therapeutic intervention.
1.4 Gal3 Dynamics in We have recently set up and optimized specific experimental pro-
the Small Intestine tocols to study the GL-Lect hypothesis, notably the endocytic
uptake of Gal3 and its interacting partners, in the physiological
context of the small intestine in mice [49]. Of note, the intestinal
epithelium is covered by a protective layer, the mucus, which is
essentially composed of heavily glycosylated proteins such as
mucins. Mucins bind to endogenous Gal3, which is thereby highly
enriched within the mucus (Fig. 1d). It was therefore a prerequisite
to permeabilize this mucosa to allow exogenous Gal3 and other
endocytic cargoes to have access to the apical membrane of enter-
ocytes within our experimental settings. This permeabilization was
achieved without perturbing the overall epithelial integrity by incu-
bation of the intestinal tissue with dithiothreitol (DTT) (Fig. 2).
How molecules are able to bypass the mucus layer for subsequent
endocytosis under physiological conditions is not known.
For these experiments, the small intestine was acutely removed
from mice, DTT-permeabilized, and then incubated with purified
recombinant Gal3 and/or interacting partners. Our pioneering
study [49] revealed that Gal3 first bound to the apical membrane
when incubated with the DTT-permeabilized tissue at 4 C
(Fig. 3a). Upon incubation at 37 C, the protein was transcytosed
to the basolateral membrane region (Fig. 3a), which was further
confirmed by electron microscopy (Fig. 3b). A similar trafficking
pattern was also observed for lactotransferrin (LTF), an interacting
partner of Gal3, whose transcytosis was in fact dependent on Gal3
and GSL expression [49] (Fig. 3c).
In this methods review, we provide step-by-step protocols to
analyze the transcytotic behavior of Gal3 and LTF within the
mouse small intestine. We describe a wide range of techniques,
including animal handling, live tissue manipulation for mucus per-
meabilization, protein purification and conjugation, endocytic
uptake assays, tissue fixation, cryo-sectioning, imaging using
photonic and electron microscopy.
Fig. 2 DTT treatment of the small intestine does not cause any major morphological alteration. (a) ZO-1, a
widely used apical tight junction marker, did not reveal any detectable mislocalization along the lateral
membrane upon DTT treatment. Scale bars ¼ 10 μm. (b) Tight junctions, visible as electron-dense structures,
were properly located at the subapical area of DTT-treated cells (red square). In addition, joint lateral
membranes were clearly observed below these junctions (yellow arrows), suggesting an intact cell-cell
cohesiveness. The apical microvilli are highlighted in blue. Black dashed squares indicate the area of
magnification. Scale bars ¼ 1 μm. (c) Untreated and DTT-treated tissues were incubated at 4 C with dextran
(10 kDa) to assess the sealing capacity of the apical tight junctions. Interestingly, in both conditions, the
dextran signal exclusively remained at the brush border, and no lateral leakage was detected. Scale
bars ¼ 10 μm
Fig. 3 Both Gal3 and its binding partner lactotransferrin (LTF) undergo transcytotic trafficking from the apical
side to the basolateral membrane. (a) DTT-treated jejunum was incubated with purified Gal3 for 30 min at 4 C
or at 37 C. Upon incubation at 37 C, Gal3 distinctly labels the basolateral membrane of enterocytes in a
lactose resistant manner, indicating that Gal3 is indeed internalized and no more accessible to the extracel-
lular milieu. Insets show one enterocyte in each condition. Scale bar ¼ 10 μm. (b) Electron microscopy
imaging was used to visualize Gal3-HRP transcytosis at the ultrastructural level. An individual enterocyte is
highlighted in pink. DAB-positive signal was visible after 15 min or 30 min incubation at 37 C within vesicular
structures (red squares) only in the presence of Gal3-HRP, notably close to the basal side. Of note, no
DAB-positive signal was detected on these structures in the absence of HRP. Scale bars ¼ 1 μm. (c)
DTT-treated tissues were incubated for 30 min at 4 C or 37 C with purified lactotransferrin (LTF). Similar
to Gal3, LTF also showed transcytotic trafficking with clear basolateral localization. Insets show one
enterocyte in each condition. Scale bars ¼ 10 μm
2 Materials
2.1 Mucus Fixation 1. Jejunum part of intestine (C57BL/6 mice).

with Carnoy’s Solution 2. Phosphate buffered saline (PBS), pH 7.4.
and Whole-Mount
3. Syringe 5 mL.
Immunofluorescence
4. Needle 26G.
5. Carnoy’s solution: 60% ethanol, 30% chloroform, and 10%
glacial acetic acid.
6. Scalpel.
7. 1% Triton buffer: 1% Triton X-100 prepared in PBS.
8. Wash buffer: 0.1% Triton X-100 prepared in PBS.
9. Triton-BSA buffer: 0.1% Triton X-100, 3% BSA buffer (w/v),
prepared in PBS.
10. Antibodies: Galectin-3 (from Fu-Tong Liu, UC Davis, USA),
UEA-1 (mucus marker, Vector Laboratories).
11. 40 ,6-diamidino-2-phenylindole (DAPI, Thermo Fisher
Scientific).
12. Glass slides.
13. Glass coverslips.
14. Aqua-Poly/Mount (aqueous mounting medium, Thermo
Fisher Scientific).
15. Scissors.
16. Petri dish or Eppendorf tube.
17. Tweezers.
2.2 Preparation of 1. LB agar plates (containing 100 μg/mL ampicillin, 34 μg/mL

Fluorescent or HRP chloramphenicol).
Coupled Gal3 and LTF 2. IPTG (Sigma-Aldrich).
Conjugates
3. Imidazole (Sigma-Aldrich).
4. Benzonase (Merck KGaA).
5. EDTA-free protease inhibitor cocktail (Sigma-Aldrich).
6. Glycerol (Sigma-Aldrich).
7. Sonicator (Branson Digital Sonifier).
8. HisPur-cobalt resin (Thermo Fisher Scientific).
9. Lactose powder (Sigma-Aldrich).
10. Superdex75 16/60 (GE Healthcare).
11. Alexa 488-NHS-ester (Thermo Scientific).
12. Cy3-NHS-ester mono-reactive (GE Healthcare).
13. PD-10 desalting columns (GE Healthcare).

14. Maleimide-activated horseradish peroxidase (HRP) (Thermo
Fisher Scientific).
15. FPLC (Äkta purifier).
16. 80 C freezer.
17. Purified recombinant LTF (SinoBiological) and Gal3-His or
Cys-Gal3-His in pHisParallel2 plasmid.
2.3 DTT Treatment of 1. Jejunum part of the intestine (C57BL/6 mice).

Intestine 2. Phosphate buffered saline (PBS), pH 7.4.
3. Syringe 5 mL.
4. Needle 26G.
5. DTT solution: 10 mM DTT prepared in PBS.
6. Tubing clamp forceps or as an alternative, small size binder
clips.
7. Crystallizing dish.
8. Scissors.
9. Tweezers.
2.4 Binding and 1. Jejunum part of the intestine (C57BL/6 mice).

Internalization Assay 2. Phosphate buffered saline (PBS), pH 7.4.
3. Syringe 5 mL.
4. Needle 26G.
5. 10 mM DTT, prepared in PBS.
6. Tubing clamp forceps or as an alternative, small size binder
clips.
7. Crystallizing dish.
8. Scissors.
9. Tweezers.
11. 20 μg/mL Gal3-Alexa 488 (as generated according to Sub-
headings 2.2 and 3.2).
12. 20 μg/mL LTF-Cy3 (as generated according to Subheadings
2.2 and 3.2).
13. 20 μg/mL dextran (10 kDa, Thermo Fisher Scientific).
14. Ice.
15. Icebox/plastic or metal plate.
16. Iso-osmolar lactose buffer: 200 mM lactose, 20 mM HEPES
pH 7.3, 50 mM NaCl, in water.
2.5 Tissue Fixation, 1. 4% PFA (w/v), Electron Microscopy Sciences), prepared

Generation of Frozen in PBS.
Blocks, and Sections 2. Phosphate buffered saline (PBS), pH 7.4.
for Imaging
3. 30% sucrose (w/v), prepared in PBS.
4. Optimal cutting temperature solution (OCT, Tissue-Tek).
5. Plastic molds (Tissue-Tek).
6. Box for dry ice.
7. Dry ice.
8. Cryostat (Leica).
9. Glass slides.
10. 80 C freezer.
2.6 Immuno- 1. Jejunum part of the intestine (C57BL/6 mice).

fluorescence 2. Phosphate buffered saline (PBS), pH 7.4.
3. NH4Cl buffer: 50 mM NH4Cl (Sigma-Aldrich), prepared
in PBS.
4. Permeabilization buffer: 0.2% Triton X-100, prepared in PBS.
5. Blocking buffer: 2% BSA (w/v), 0.2% Triton X-100, prepared
in PBS.
6. PBS/BSA buffer: 2% BSA (w/v), prepared in PBS.
7. Anti-ZO-1 antibody (tight junction marker, Thermo Fisher
Scientific).
8. 40 ,6-diamidino-2-phenylindole (DAPI, nuclear staining,
9. Glass coverslips.
10. Aqua-Poly/Mount (aqueous mounting medium, Thermo
Fisher Scientific).
2.7 Electron 1. Jejunum part of the intestine (C57BL/6 mice).

Microscopy 2. Phosphate buffered saline (PBS), pH 7.4.
3. Syringe 5 mL.
4. Needle 26G.
5. Diaminobenzidine (DAB) tablets SIGMAFAST (Sigma-
Aldrich) [50].
6. H2O2.
7. DAB solution: 2 DAB tablets in 10 mL of PBS, passed through
a 0.22 μm filter.
8. DAB/H2O2 solution: 1 tablet of DAB in 5 mL of PBS, 2 μL of
30% H2O2.
9. 25% Glutaraldehyde (w/v, Electron Microscopy Sciences).
10. Sodium cacodylate 0.2 M, pH 7.2 (Electron Microscopy

Sciences).
11. 2% Osmium tetroxide (w/v, Electron Microscopy Sciences).
12. Grids # Ni/Formvar/carbon square 100 mesh (Electron
Microscopy Sciences).
13. Scalpel.
15. Tecnai Spirit electron microscope (e.g., FEI, Eindhoven, The
Netherlands) equipped with a 4k CCD camera (EMSIS
GmbH, Münster, Germany).
16. Reichert Leica UCT microtome.
17. Minisart filters 0.22 μm.
18. Scissors.
19. Tweezers.
3 Methods
3.1 Mucus Fixation The small intestine has only one layer of mucus, which can be
with Carnoy’s Solution removed relatively easily [51]. Carnoy’s fixative solution is used in
and Whole-Mount order to preserve this mucus layer for imaging [51] (see Note 1).
Immunofluorescence 1. Sacrifice C57BL/6 mice (see Notes 2 and 3).
2. Open abdominal cavity.
3. Separate stomach from esophagus with scissor.
4. Gently pull up the stomach and detach at the same time the
mesenchyme, which is holding the intestine inside the abdom-
inal cavity.
5. Extract jejunum part of the intestine, which will start approxi-
mately 6–7 cm after the pylorus and is approximately 10 cm in
length in 3- to 4-month-old male mice (see Notes 4 and 5).
6. With a 26G needle and syringe filled with PBS, wash intestine
until the discarded fluid becomes visually clean (see Note 6).
7. Transfer intestine into Carnoy’s solution for overnight fixation
(see Note 7).
8. Remove intestine from Carnoy’s solution and place it on a glass
slide.
9. With a scalpel, cut the intestinal tube into ~1 mm sections, and
place it into an empty Eppendorf tube for further steps.
10. Add 1% Triton buffer to intestinal slices, and incubate for 2 h at
room temperature for permeabilization.
11. Exchange buffer to Triton-BSA buffer and incubate for 2 h at
room temperature (to reduce unspecific signal).
12. Exchange buffer to primary antibody solution (anti-UEA-1

antibody: 1:50 diluted in Triton-BSA buffer) (see Note 8).
14. Exchange buffer to 600 μL wash buffer.
15. Incubate for 30 min at room temperature.
16. Repeat washing procedure 4 times.
17. Exchange buffer to secondary antibody solution (antibody 1:
100 diluted in Triton-BSA buffer) and incubate overnight at
4 C (see Note 8).
18. Exchange buffer to 600 μL wash buffer.
19. Incubate for 30 min at room temperature.
20. Repeat washing procedure 4 times.
21. With tweezers, transfer and arrange intestinal cuts onto a
microscopical glass slide.
22. Remove excess of the buffer with Kimwipe paper.
23. Cover samples with approximately 10–15 μL Aqua-Poly/
Mount (aqueous mounting medium, Thermo Fisher Scien-
tific) solution, containing 0.5 μg/mL DAPI (Thermo Fisher
Scientific).
24. Place a cover glass on the samples.
25. Dry samples overnight at 37 C (protected from light).
26. Store samples at 4 C until imaging (see Note 9).
3.2 Preparation of His-tagged Gal3 or a His-tagged cysteine variant of Gal3 is purified

Fluorescent or HRP- with HisPur-cobalt-resin (see Note 10) and labeled with fluoro-
Labeled Gal3 and LTF phores (see Note 11) or horseradish peroxidase, respectively, as
Conjugates described below:
1. Inoculate a LB agar plate (100 μg/mL ampicillin, 34 μg/mL
chloramphenicol) with Rosetta2 pLys-transformed bacteria
(Gal3-His or Cys-Gal3-His in pHisParallel2 plasmid) (see
Note 10).
2. Leave overnight at 37 C.
3. Inoculate 50 mL pre-culture LB media (100 μg/mL ampicillin,
34 μg/mL chloramphenicol) by picking a single colony from
the LB plate.
4. Incubate overnight at 37 C on a rotary shaker, 170 rpm.
5. Measure optical density (OD) at 600 nm.
6. Inoculate fresh LB media (3 L, ampicillin/chloramphenicol)
by diluting pre-culture to an OD600 of 0.1.
7. Incubate at 37 C on a rotary shaker, 170 rpm, until the

bacterial culture reaches OD600 of 0.4 (approximately
1.5–2 h).
8. Cool down bacterial culture to 21 C.
9. Add IPTG to a final concentration of 60 μM.
10. Incubate for 14 h at 21 C on a rotary shaker, 170 rpm
(OD600 will be approximately 2.5).
11. Place all bacterial media bottles at 4 C.
12. Pellet bacteria by centrifugation for 20 min at 4500 g, 4 C.
13. Remove supernatant.
14. Resuspend pellet in 30 mL PBS, supplemented with 10 mM
imidazole +3000 units benzonase/L, 70 μL/L culture prote-
ase inhibitor (EDTA free), and 10% glycerol.
15. Break cells by sonication, use a big probe (5 mm diameter),
7 min, settings: 0.5 s ON/0.5 s OFF, power 40%, and keep
tube during sonication on ice water (see Note 12).
16. Centrifuge bacterial lysate at 4 C for 60 min, 76,000 g using
JA 25.5 rotor.
17. Equilibrate HisPur-cobalt resin (1 mL per L culture) by wash-
ing twice with 40 mL PBS-glycerol buffer (see above), and
centrifuge for 2 min at 700 g.
18. Incubate HisPur-cobalt resin with supernatant from step 16
for 1 h at 4 C on an end-over-end rotator.
19. Pour resin solution in an empty gravity column and wash the
resin with 250 mL ice-cold PBS-glycerol buffer.
20. Elute with 3 mL PBS, supplemented with 500 mM imidazole.
21. Pass eluate on a PBS equilibrated Superdex75 16/60 size-
exclusion chromatography column after clearing with a
0.2 μm filter.
22. Gal3 will elute in PBS at 68–75 mL.
23. Measure protein concentration of each fraction on a Nano-
Drop spectrophotometer, using 280 nm absorption (use PBS
as blank, the molar extinction coefficient of Gal3 is 1.287
(mg/mL)1 cm1).
24. Snap freeze Gal3 solution in liquid nitrogen.
25. For Gal3-Alexa 488 fluorescent labeling: Gal3 (2 mg/mL),
10 mM lactose in PBS is incubated under agitation for 1 h at
21 C with amine-reactive (NHS-ester) Alexa 488 at a molar
ratio of 1:4 (see Note 13).
26. Purify by using a PD-10 desalting column.
27. Measure protein concentration and labeling efficiency using

the “Proteins & labels” option on a NanoDrop
spectrophotometer.
28. For maleimide-activated horseradish peroxidase (maleimide-
HRP) coupling: His-tagged Cys-Gal3 (2 mg/mL) is incubated
at a molar ratio of 1:2 with maleimide-HRP for 12 h at 4 C in
PBS supplemented with 10 mM lactose.
29. Gel filtration chromatography with Superdex75 10/300 col-
umn is used to separate HRP, HRP-Gal3-His, and Cys-Gal3-
His (see Note 10).
30. Snap freeze Gal3-HRP solution in 50% glycerol, and store at
80 C for further use.
LTF is fluorescently labeled as described below:
1. LTF is solubilized in 50 μL water at a concentration of 0.4 mg/
mL and mixed with amine-reactive (NHS-ester) Cy3 at a molar
ratio of 1:3.
2. Incubate this mixture under agitation for 2 h at room
temperature.
3. Purify by using PD-10 desalting columns.
4. Measure protein concentration and labeling efficiency using
the “Proteins & Labels” option on a NanoDrop spectropho-
tometer at 280 nm (the molar extinction coefficient of LTF is
(1.148 mg/mL)1.cm1.
5. Store at 4 C for further use.
3.3 DTT Treatment of To permeabilize the mucus layer, disulfide bonds between mucins
Intestine are reduced with 10 mM DTT (see Note 14).
1. Sacrifice mice (see Notes 2 and 3).
2. Open abdominal cavity.
inal cavity.
5. Extract jejunum part of the intestine, which starts approxi-
mately 6–7 cm after the pylorus, and is approximately 10 cm
in length in 3- to 4-month-old male mice (see Notes 4 and 5).
7. Close intestinal tube with a tubing clamp from one side (see
Note 15), and inject the DTT solution into the other side (see
Note 7); close the second opening with a tubing clamp.
8. Incubate the intestine for 15 min at room temperature in a

crystallizing dish filled with PBS.
9. Wash 3 times each with 1 mL DTT solution.
10. Incubate for 10 min at room temperature with DTT solution
in a crystallizing dish filled with PBS.
11. Wash 3 times each with 1 mL DTT solution.
12. Wash once with 3 mL PBS solution.
13. Use intestine for further experiments.
3.4 Binding and The following assay has been developed to analyze the binding of
Internalization Assay Gal3 or LTF to the apical surface of enterocytes of the murine
jejunum, and their internalization into these cells. To test for tissue
integrity, dextran (10 kDa) was incubated in the intestinal tube
using experimental procedure that is similar to the binding step
(Fig. 2c).
3.4.1 Binding (See Note 1. Sacrifice mice (see Notes 2 and 3).
16) 2. Open abdominal cavity.
inal cavity.
5. Extract jejunum part of the intestine, which starts approxi-
mately 6–7 cm after the pylorus, and is approximately 10 cm
in length in 3- to 4-month-old male mice (see Notes 4 and 5).
7. Perform incubation of the intestinal tube with DTT
(as described in Subheading 3.3).
8. Cut intestine into two equal-sized pieces (one control sample
without ligand, and second sample with endocytic ligand).
9. Place intestinal samples for 30 min on ice.
10. Close intestinal samples with a tubing clamp (or small size
binder clips) from one side, and inject PBS solution containing:
20 μg/mL of Gal3-Alexa 488, 20 μg/mL of LTF-Cy3, or pure
PBS as a negative control (see Note 17), and close the second
opening with a tubing clamp.
11. For cell-cell leakage, incubate with 1 mg/mL dextran (10 kDa)
prepared in PBS using the same experimental procedure as
described above.
12. Incubate for 30 min at 4 C in a crystallizing dish filled
with PBS.
13. Remove clamps and open intestinal samples lengthwise (still

on ice).
14. Wash 3 times for 10 min at 4 C in ice-cold PBS to remove
unbound Gal3-Alexa 488, unbound LTF-Cy3, or unbound
dextran (10 kDa).
15. Process for cryo-embedding, as described in Subheading 3.5.
3.4.2 Internalization (See 1. Repeat procedure as above (Binding) until step 8.

Note 18) 2. Close intestinal samples with a tubing clamp (or small size
20 μg/mL of Gal3-Alexa 488, 20 μg/mL of LTF-Cy3, 20 μg/
mL of Gal3-HRP, or pure PBS as a negative control (see Note
6), and close the second opening with a tubing clamp.
3. Incubate for 30 min at 37 C in a crystallizing dish filled
with PBS.
4. After 30 min incubation, place intestinal samples on ice.
5. Remove clamps and open intestinal samples lengthwise with
scissors (still on ice).
6. Incubate 3 times in ice-cold iso-osmolar lactose solution
(200 mM) for 10 min at 4 C to remove surface-exposed
Gal3 (see Note 7).
7. Incubate 3 times in ice-cold PBS for 10 min at 4 C to partially
remove surface-exposed LTF.
8. Wash intestine with ice-cold PBS for 5 min.
9. Process for cryo-embedding as described below.
3.5 Fixation, To visualize and conserve the intestinal tissue from experiments
Generation of Frozen that were described above, the following protocol is used (except
Tissue Blocks, and for electron microscopy):
Sections for Imaging 1. Incubate intestine for 1 h at 4 C in freshly prepared 4% PFA in
PBS, then shift to room temperature for additional 4 h, pro-
tected from light (see Note 7).
2. Wash intestine 3 times for 5 min each with 25 mL PBS (mild
shaking).
3. Transfer to 30% sucrose in PBS and incubate at 4 C until the
tissue blocks drop down to the bottom of the tube (overnight
to 24 h incubation).
4. Transfer tissue into plastic molds.
5. Fill the plastic mold with OCT solution.
6. Orient tissue gently with plastic tips; microvilli should face the
bottom of plastic mold.
7. Place mold on dry ice and wait until the solution is solid (turns
from transparent to white).
8. Keep molds at 80 C for storage, or prepare cryo-sections as
described below.
9. For mounting tissue blocks on the cryostat specimen holder,
add OCT on top of the metal holder, and wait until it started to
freeze (see Note 19).
10. Pop the frozen intestinal tissue samples out of the plastic mold
and place it flatly onto the prepared cryostat specimen holder.
11. Cut sections of 25 μm and immediately mount them on slides.
12. Slides can be stored at the 80 C, mounted directly on glass
slides for imaging (see steps 13–16, below), or used for immu-
nofluorescence labeling (Subheading 3.6) (see Note 9).
13. Cover samples with approximately 6 μL Aqua-Poly/Mount
solution, containing DAPI.
14. Place a cover glass on the samples.
16. Store samples at 4 C until microscopical analysis.
3.6 Immuno- The integrity of the intestine under DTT treatment conditions is
fluorescence validated by immunofluorescence analysis using antibodies against
the tight junction marker ZO-1 (Fig. 2a).
1. Thaw slides for 7 min after cutting at room temperature.
2. Rehydrate slides in PBS for 7 min.
3. Incubate slides in NH4Cl buffer for 15 min.
4. Incubate slides with the permeabilization buffer for 10 min.
5. Incubate the tissue with PBS/BSA blocking buffer for 3 h at
room temperature to reduce any unspecific binding.
6. Apply primary antibodies (anti-Gal3 or anti- ZO-1) diluted
(1/100) in PBS/BSA blocking buffer (see Note 8).
7. Incubate samples overnight at 4 C.
8. Wash samples 4 times with PBS/BSA blocking buffer,
10 min each.
9. Incubate slides with the secondary antibody diluted (1/400) in
PBS/BSA blocking buffer for 90 min at room temperature (see
Note 8).
10. Wash samples 4 times, 10 min each, in PBS/BSA blocking
buffer.
11. Remove excess of the buffer with Kimwipe paper.
12. Cover samples with approximately 10–15 μL Aqua-Poly/
Mount solution containing DAPI.
13. Place glass coverslips on the samples.

15. Store samples at 4 C until imaging (see Note 9).
3.7 Electron The integrity of the intestine under DTT treatment conditions is
Microscopy validated by electron microscopy (Fig. 2b). The internalization of
Gal3-HRP is analyzed as shown in Fig. 3b.
1. Repeat procedure as described in Subheading 3.4 until step 8.
2. Close intestinal samples with a tubing clamp (or small size
20 μg/mL of Gal3-HRP, or pure PBS as a negative control (see
Note 17), and close the second opening with a tubing clamp.
3. Incubate for 15 min or 30 min at 37 C in a crystallizing dish
filled with 37 C pre-warmed PBS.
4. After incubation, place intestinal samples on ice.
5. Remove clamps and open the intestine lengthwise and cut into
3 mm 2 mm slices.
6. Wash intestine with ice-cold PBS for 5 min.
7. Perform immediately the DAB reaction in a petri dish or
Eppendorf tube as described below.
8. Incubate intestine for 15 min at 4 C in freshly prepared DAB
solution.
9. Replace the DAB solution with fresh DAB/H2O2 solution.
10. Incubate for 1 h at 4 C.
11. Wash intestine 3 times with PBS buffer at 4 C, 5 min each.
12. Fix intestine for 3 days at 4 C with 2.5% glutaraldehyde in
0.1 M sodium cacodylate (solution needs to be changed
every day).
13. Wash intestine 3 times 10 min at room temperature with 0.1 M
sodium cacodylate (pH 7.2).
14. Post fixation of intestine for 2 h at room temperature with 1%
osmium tetroxide in 0.1 M sodium cacodylate (pH 7.2).
15. Wash intestine 3 times 10 min at room temperature with 0.1 M
sodium cacodylate (pH 7.2).
16. Wash intestine for 5 min with water.
17. Dehydrate tissue at room temperature using the following
incubation scheme:
(a) 50% ethanol for 10 min
(b) 70% ethanol for 10 min
(c) 90% ethanol (2 times) for 15 min each
(d) 100% ethanol (3 times) for 20 min each.
18. Infiltration of Epon resin LX112 at room temperature using

the following incubation scheme:
(a) LX112/ethanol 100% (v/v) 30 min.
(b) LX112/ethanol 100% (2 v/v) 45 min.
(c) LX112 pure 1 h.
(d) LX112 pure overnight.
19. Pour samples in pure LX112 into embedding mold, and leave
for additional 3 days at 60 C to allow polymerization (for
2 mm tissue blocks).
20. Cut LX112 with Reichert Leica UCT microtome (pyramid top
0.5 0.8 mm) to obtain 65 nm slices for subsequent deposit
on Ni/formvar/carbon-coated grids.
21. Contrast for 10 min with 4% uranyl acetate prepared in water.
22. Rinse the grid 3 times with water.
23. Observe tissue with Tecnai Spirit transmission electron
microscope.
4 Notes
1. For mucus fixation with Carnoy’s solution, the indicated incu-

bation time was given for 1 mm intestinal cuts. This might be
extended up to 1 h for larger intestinal cuts.
2. It is preferred to perform experiments with animals in the
morning, due to the long experimental procedure and mouse
manipulation convenience.
3. 3- to 4-month-old mice were used for experiments.
4. In order to avoid intestine damage, always make sure that the
intestinal segment is being pulled on as gently as possible
during extraction.
5. To avoid tissue dehydration, always keep intestine covered with
buffer solution.
6. While performing PBS or DTT/PBS washes of intestine,
always make sure that the needle is gently inserted into the
intestinal tube, and gradually press syringe to avoid disruption
of microvilli. Always control the liquid flow.
7. DTT, lactose, PFA, and Carnoy’s solutions should be always
freshly prepared.
8. In order to avoid antibody solution spreading on the glass slide
and ensure uniform immunostaining, hydrophobic ink from a
Dako pen is used to delimit the sample contour prior to adding
the antibody solution in a dome shaped manner.
9. Imaging of the samples with immunofluorescence labeling

should be done within 3 weeks to make sure that the quality
of the signal is still optimal.
10. Wild-type Gal3-His can be purified the same way as His-tagged
Cys-Gal3.
11. Labeling efficiency should be between 1 and 3 fluorophores
per Gal3 molecule.
12. Avoid foam formation during sonication.
13. Use of 10 mM lactose during fluorophore coupling to Gal3 to
protect the carbohydrate recognition domain of Gal3
(active site).
14. Do not exceed the DTT treatment incubation time to avoid
tissue damage.
15. To close the intestinal tube during incubation as an alternative
to tubing clamp forceps, small size binder clips can be used.
16. For the binding assay, it is required to pre-cool all solutions
on ice.
17. In order to avoid loss of protein solutions (Gal3, LTF, or
dextran), the use of insulin syringes without dead-space is
advised.
18. For the internalization assay, pre-heat all solutions to 37 C.
19. In order to have flat OCT embedded intestine samples, the
following trick can be used: Put a flat metal plate inside the dry
ice box until the plate’s temperature has reached an equilib-
rium. Place OCT intestinal samples onto it and close the dry ice
box. Wait until intestine in OCT solution has turned solid
(switch from transparent into white).
References
1. Honig E, Ringer K, Dewes J, von Mach T, Oka T, Futai M, Muller WE, Yagi F, Kasai K
Kamm N, Kreitzer G, Jacob R (2018) (2002) Oligosaccharide specificity of galectins:
Galectin-3 modulates the polarized surface a search by frontal affinity chromatography.
delivery of beta1-integrin in epithelial cells. J Biochim Biophys Acta 1572(2–3):232–254.
Cell Sci 131(11):jcs213199. https://doi.org/ https://doi.org/10.1016/s0304-4165(02)
10.1242/jcs.213199 00311-2
2. Barondes SH, Cooper DN, Gitt MA, Leffler H 5. Liu FT, Patterson RJ, Wang JL (2002) Intra-
(1994) Galectins. Structure and function of a cellular functions of galectins. Biochim Bio-
large family of animal lectins. J Biol Chem phys Acta 1572(2–3):263–273. https://doi.
269(33):20807–20810 org/10.1016/s0304-4165(02)00313-6
3. Di Lella S, Sundblad V, Cerliani JP, Guardia 6. Liu FT, Rabinovich GA (2005) Galectins as
CM, Estrin DA, Vasta GR, Rabinovich GA modulators of tumour progression. Nat Rev
(2011) When galectins recognize glycans: Cancer 5(1):29–41. https://doi.org/10.
from biochemistry to physiology and back 1038/nrc1527
again. Biochemistry 50(37):7842–7857. 7. Ochieng J, Platt D, Tait L, Hogan V, Raz T,
https://doi.org/10.1021/bi201121m Carmi P, Raz A (1993) Structure-function rela-
4. Hirabayashi J, Hashidate T, Arata Y, Nishi N, tionship of a recombinant human galactoside-
Nakamura T, Hirashima M, Urashima T, binding protein. Biochemistry
32(16):4455–4460. https://doi.org/10. proliferation and differentiation. Cell

1021/bi00067a038 129(1):123–134. https://doi.org/10.1016/j.
8. Rabinovich GA, Gruppi A (2005) Galectins as cell.2007.01.049
immunoregulators during infectious processes: 18. Ramasamy S, Duraisamy S, Barbashov S,
from microbial invasion to the resolution of the Kawano T, Kharbanda S, Kufe D (2007) The
disease. Parasite Immunol 27(4):103–114. MUC1 and galectin-3 oncoproteins function
https://doi.org/10.1111/j.1365-3024.2005. in a microRNA-dependent regulatory loop.
00749.x Mol Cell 27(6):992–1004. https://doi.org/
9. Yang RY, Liu FT (2003) Galectins in cell 10.1016/j.molcel.2007.07.031
growth and apoptosis. Cell Mol Life Sci 19. Markowska AI, Jefferies KC, Panjwani N
60(2):267–276. https://doi.org/10.1007/ (2011) Galectin-3 protein modulates cell sur-
s000180300022 face expression and activation of vascular endo-
10. Ahmad N, Gabius HJ, Andre S, Kaltner H, thelial growth factor receptor 2 in human
Sabesan S, Roy R, Liu B, Macaluso F, Brewer endothelial cells. J Biol Chem
CF (2004) Galectin-3 precipitates as a penta- 286(34):29913–29921. https://doi.org/10.
mer with synthetic multivalent carbohydrates 1074/jbc.M111.226423
and forms heterogeneous cross-linked com- 20. Mathew MP, Donaldson JG (2018) Distinct
plexes. J Biol Chem 279(12):10841–10847. cargo-specific response landscapes underpin
https://doi.org/10.1074/jbc.M312834200 the complex and nuanced role of galectin-
11. Halimi H, Rigato A, Byrne D, Ferracci G, glycan interactions in clathrin-independent
Sebban-Kreuzer C, ElAntak L, Guerlesquin F endocytosis. J Biol Chem
(2014) Glycan dependence of Galectin-3 self- 293(19):7222–7237. https://doi.org/10.
association properties. PLoS One 9(11): 1074/jbc.RA118.001802
e111836. https://doi.org/10.1371/journal. 21. Furtak V, Hatcher F, Ochieng J (2001)
pone.0111836 Galectin-3 mediates the endocytosis of beta-1
12. Nieminen J, Kuno A, Hirabayashi J, Sato S integrins by breast carcinoma cells. Biochem
(2007) Visualization of galectin-3 oligomeriza- Biophys Res Commun 289(4):845–850.
tion on the surface of neutrophils and endothe- https://doi.org/10.1006/bbrc.2001.6064
lial cells using fluorescence resonance energy 22. Johannes L (2017) Shiga toxin-A model for
transfer. J Biol Chem 282(2):1374–1383. glycolipid-dependent and lectin-driven endo-
https://doi.org/10.1074/jbc.M604506200 cytosis. Toxins 9(11):340. https://doi.org/
13. Rabinovich GA, Toscano MA, Jackson SS, 10.3390/toxins9110340
Vasta GR (2007) Functions of cell surface 23. Lakshminarayan R, Wunder C, Becken U,
galectin-glycoprotein lattices. Curr Opin Howes MT, Benzing C, Arumugam S,
Struct Biol 17(5):513–520. https://doi.org/ Sales S, Ariotti N, Chambon V, Lamaze C,
10.1016/j.sbi.2007.09.002 Loew D, Shevchenko A, Gaus K, Parton RG,
14. Boscher C, Dennis JW, Nabi IR (2011) Glyco- Johannes L (2014) Galectin-3 drives
sylation, galectins and cellular signaling. Curr glycosphingolipid-dependent biogenesis of
Opin Cell Biol 23(4):383–392. https://doi. clathrin-independent carriers. Nat Cell Biol
org/10.1016/j.ceb.2011.05.001 16(6):595–606. https://doi.org/10.1038/
15. Brewer CF, Miceli MC, Baum LG (2002) ncb2970
Clusters, bundles, arrays and lattices: novel 24. Johannes L, Wunder C, Shafaq-Zadah M
mechanisms for lectin-saccharide-mediated cel- (2016) Glycolipids and lectins in endocytic
lular interactions. Curr Opin Struct Biol uptake processes. J Mol Biol S0022-2836
12(5):616–623. https://doi.org/10.1016/ (16):30453–30453. https://doi.org/10.
s0959-440x(02)00364-0 1016/j.jmb.2016.10.027
16. Partridge EA, Le Roy C, Di Guglielmo GM, 25. HFT R, Tyckaert F, Lo Guidice C, Hirsch T,
Pawling J, Cheung P, Granovsky M, Nabi IR, Velades Cruz CA, Lemaigre C, Shafaq-Zadah-
Wrana JL, Dennis JW (2004) Regulation of M, Wunder C, Wattiez R, Johannes L, van der
cytokine receptors by Golgi N-glycan proces- Bruggen P, Alsteens D, Morsomme P (2021)
sing and endocytosis. Science Endophilin-A3 and galectin-8 control the cla-
306(5693):120–124. https://doi.org/10. thrin -independent endocytosis of CD166. Nat
1126/science.1102109 Commun 11:1457
17. Lau KS, Partridge EA, Grigorian A, Silvescu 26. Nio J, Kon Y, Iwanaga T (2005) Differential
CI, Reinhold VN, Demetriou M, Dennis JW cellular expression of galectin family mRNAs in
(2007) Complex N-glycan number and degree the epithelial cells of the mouse digestive tract.
of branching cooperate to regulate cell
J Histochem Cytochem 53(11):1323–1334. 105(3):1241–1251. https://doi.org/10.

https://doi.org/10.1369/jhc.5A6685.2005 1083/jcb.105.3.1241
27. Nio-Kobayashi J (2017) Tissue- and cell- 37. Ait-Slimane T, Galmes R, Trugnan G, Maurice
specific localization of galectins, beta- M (2009) Basolateral internalization of
galactose-binding animal lectins, and their GPI-anchored proteins occurs via a clathrin-
potential functions in health and disease. Anat independent flotillin-dependent pathway in
Sci Int 92(1):25–36. https://doi.org/10. polarized hepatic cells. Mol Biol Cell
1007/s12565-016-0366-6 20(17):3792–3800. https://doi.org/10.
28. Nio-Kobayashi J, Takahashi-Iwanaga H, Iwa- 1091/mbc.E09-04-0275
naga T (2009) Immunohistochemical localiza- 38. Polishchuk R, Di Pentima A, Lippincott-
tion of six galectin subtypes in the mouse Schwartz J (2004) Delivery of raft-associated,
digestive tract. J Histochem Cytochem GPI-anchored proteins to the apical surface of
57(1):41–50. https://doi.org/10.1369/jhc. polarized MDCK cells by a transcytotic path-
2008.952317 way. Nat Cell Biol 6(4):297–307. https://doi.
29. Magnaldo T, Fowlis D, Darmon M (1998) org/10.1038/ncb1109
Galectin-7, a marker of all types of stratified 39. Monhasery N, Moll J, Cuman C, Franke M,
epithelia. Differentiation 63(3):159–168. Lamertz L, Nitz R, Gorg B, Haussinger D,
https://doi.org/10.1046/j.1432-0436.1998. Lokau J, Floss DM, Piekorz R, Dimitriadis E,
6330159.x Garbers C, Scheller J (2016) Transcytosis of
30. Jacob R, Naim HY (2001) Apical membrane IL-11 and apical redirection of gp130 is
proteins are transported in distinct vesicular mediated by IL-11alpha receptor. Cell Rep
carriers. Curr Biol 11(18):1444–1450. 16(4):1067–1081. https://doi.org/10.1016/
https://doi.org/10.1016/s0960-9822(01) j.celrep.2016.06.062
00446-8 40. Zhou K, Sumigray KD, Lechler T (2015) The
31. Kreitzer G, Schmoranzer J, Low SH, Li X, Arp2/3 complex has essential roles in vesicle
Gan Y, Weimbs T, Simon SM, Rodriguez- trafficking and transcytosis in the mammalian
Boulan E (2003) Three-dimensional analysis small intestine. Mol Biol Cell
of post-Golgi carrier exocytosis in epithelial 26(11):1995–2004. https://doi.org/10.
cells. Nat Cell Biol 5(2):126–136. https:// 1091/mbc.E14-10-1481
doi.org/10.1038/ncb917 41. Lynes M, Narisawa S, Millan JL, Widmaier EP
32. Mostov KE, Verges M, Altschuler Y (2000) (2011) Interactions between CD36 and global
Membrane traffic in polarized epithelial cells. intestinal alkaline phosphatase in mouse small
Curr Opin Cell Biol 12(4):483–490. https:// intestine and effects of high-fat diet. Am J
doi.org/10.1016/s0955-0674(00)00120-4 Physiol Regul Integr Comp Physiol 301(6):
33. Luton F, Verges M, Vaerman JP, Sudol M, R1738–R1747. https://doi.org/10.1152/
Mostov KE (1999) The SRC family protein ajpregu.00235.2011
tyrosine kinase p62yes controls polymeric IgA 42. Pepino MY, Kuda O, Samovski D, Abumrad
transcytosis in vivo. Mol Cell 4(4):627–632. NA (2014) Structure-function of CD36 and
https://doi.org/10.1016/s1097-2765(00) importance of fatty acid signal transduction in
80213-0 fat metabolism. Annu Rev Nutr 34:281–303.
34. Rojas R, Apodaca G (2002) Immunoglobulin https://doi.org/10.1146/annurev-nutr-
transport across polarized epithelial cells. Nat 071812-161220
Rev Mol Cell Biol 3(12):944–955. https:// 43. Cammisotto PG, Levy E, Bukowiecki LJ,
doi.org/10.1038/nrm972 Bendayan M (2010) Cross-talk between adi-
35. Verges M, Luton F, Gruber C, Tiemann F, pose and gastric leptins for the control of
Reinders LG, Huang L, Burlingame AL, Haft food intake and energy metabolism. Prog His-
CR, Mostov KE (2004) The mammalian retro- tochem Cytochem 45(3):143–200. https://
mer regulates transcytosis of the polymeric doi.org/10.1016/j.proghi.2010.06.001
immunoglobulin receptor. Nat Cell Biol 44. Bian F, Cui J, Zheng T, Jin S (2017) Reactive
6(8):763–769. https://doi.org/10.1038/ oxygen species mediate angiotensin II-induced
ncb1153 transcytosis of low-density lipoprotein across
36. Bartles JR, Feracci HM, Stieger B, Hubbard endothelial cells. Int J Mol Med
AL (1987) Biogenesis of the rat hepatocyte 39(3):629–635. https://doi.org/10.3892/
plasma membrane in vivo: comparison of the ijmm.2017.2887
pathways taken by apical and basolateral pro- 45. Molino Y, David M, Varini K, Jabes F,
teins using subcellular fractionation. J Cell Biol Gaudin N, Fortoul A, Bakloul K, Masse M,
Bernard A, Drobecq L, Lecorche P, 48. Nikitas G, Deschamps C, Disson O, Niault T,

Temsamani J, Jacquot G, Khrestchatisky M Cossart P, Lecuit M (2011) Transcytosis of
(2017) Use of LDL receptor-targeting peptide Listeria monocytogenes across the intestinal
vectors for in vitro and in vivo cargo transport barrier upon specific targeting of goblet cell
across the blood-brain barrier. FASEB J accessible E-cadherin. J Exp Med
fj.201600827R 1084/jem.20110560
46. Lim JS, Na HS, Lee HC, Choy HE, Park SC, 49. Ivashenka A, Wunder C, Chambon V,
Han JM, Cho KA (2009) Caveolae-mediated Sandhoff R, Jennemann R, Estelle D,
entry of Salmonella typhimurium in a human Podsypanina K, Lombard B, Loew D,
M-cell model. Biochem Biophys Res Commun Lamaze C, Poirier F, Gröne HJ, Johannes L,
390(4):1322–1327. https://doi.org/10. Shafaq Zadah M (2021) Glycolipid-dependent
1016/j.bbrc.2009.10.145 and lectin-driven transcytosis in mouse entero-
47. Lukyanenko V, Malyukova I, Hubbard A, cytes. Commun Biol 4:173
Delannoy M, Boedeker E, Zhu C, 50. Ravn V, Dabelsteen E (2000) Tissue distribu-
Cebotaru L, Kovbasnjuk O (2011) Enterohe- tion of histo-blood group antigens. APMIS
morrhagic Escherichia coli infection stimulates 108(1):1–28. https://doi.org/10.1034/j.
Shiga toxin 1 macropinocytosis and transcyto- 1600-0463.2000.d01-1.x
sis across intestinal epithelial cells. Am J Physiol 51. Hansson GC (2012) Role of mucus layers in
Cell Physiol 301(5):C1140–C1149. https:// gut infection and inflammation. Curr Opin
doi.org/10.1152/ajpcell.00036.2011 Microbiol 15(1):57–62. https://doi.org/10.
1016/j.mib.2011.11.002
Chapter 21
Evaluating the Role of Galectins in Clathrin-Independent

Endocytosis
Mohit P. Mathew, Julie G. Donaldson, and John A. Hanover
Abstract
Galectin-3 is a chimeric galectin involved in diverse intracellular and extracellular functions. Galectin-3 is
synthesized in the cytoplasm and then released extracellularly by a poorly understood non-canonical
secretion mechanism. As a result, it can play important roles both inside and outside the cell. One important
extracellular role of galectin-3 is in modulating clathrin-independent endocytosis (CIE), a form of cellular
internalization that is still not well understood. CIE, unlike clathrin-mediated endocytosis, has neither
defined signaling sequences nor cytoplasmic machinery. As a result, extracellular interactions like the
galectin-glycan interactions are thought to directly drive changes in CIE. This chapter discusses the
methods designed to study the role of galectin-glycan interactions in CIE, which have provided us with
insight into the functions of galectin-3 and cell surface glycans during CIE cargo internalization. These
methods include media supplementation for metabolic glycoengineering, antibody internalization assays,
lectin panels to assay changes in glycan patterns, exogenous galectin-3 supplementation, galectin-3 secre-
tion assays, and in vitro assays to monitor the effect of galectins on CIE.
Key words Galectin-3, Antibody internalization assay, Clathrin-independent endocytosis, Galectin-3

secretion assay
1 Introduction
Galectins are proteins that bind to terminal galactose residues

[1, 2]. They are known to play important physiological roles both
intracellularly [3] and extracellularly [4]. There are 3 types of
galectins based on the number of carbohydrate recognizing
domains: monomeric, dimeric, and chimeric [2]. Galectin-3 is the
only known chimeric galectin; it is synthesized in the monomeric
form but can multimerize to produce a pentameric form [1, 2].
Due to its ability to form multimeric structures, galectin-3 is
thought to play an important role in organizing large networks or
lattices of galectin-glycan interactions on the surface of cells termed
the “galectin lattice” [5]. These galectins have also been found to
be dysregulated in numerous diseases from cancer to diabetes and
391
392 Mohit P. Mathew et al.
cardiac hypertension [6–19]. One important emerging cellular role

for galectins has been in the modulation of clathrin-independent
endocytosis.
Clathrin-Independent endocytosis (CIE) is an essential cell
function in all nucleated cells [20, 21]. Unlike clathrin-mediated
endocytosis which is well characterized with extensive intracellular
machinery identified, CIE is still poorly understood with little
known about what cytoplasmic components regulate it [20, 22–
24]. CIE cargo, like most cell membrane bound proteins are gly-
cosylated. Since attempts to identify intracellular regulators of CIE
have been unfruitful, we decided to study whether CIE could be
modulated on the extracellular side of the membrane via well-
established galectin-glycan interactions.
The effect of galectin-glycan interactions had been previously
observed to have either a stimulatory role or inhibitory role on
CIE. The stimulatory role was shown by the Johannes group to be
due to initiation of membrane bending and inducing pit formation
[25, 26]. Galectins can also play an inhibitory role on CIE due to
increased cell surface sequestration via the galectin lattice as shown
by the Yarema and Dennis groups [27–29]. However, these obser-
vations are not mutually exclusive. Our study demonstrated that
these two disparate effects existed at either end of a spectrum of
behavior dependent on the extent of galectin-glycan interactions
[30, 31]. Based on the complexity of the behavioral landscape that
needed to be explored, a number of different tools and approaches
needed to move our cell lines in different directions along these
landscapes.
In order to properly explore these complex landscapes, several
tools need to be used to manipulate these galectin-glycan interac-
tions in different directions. These tools include metabolic oligo-
saccharide engineering where cell culture media is supplemented
with sugars or sugar analogs in order to drive shifts in glycan
patterns. Lactose and LacNAc can be used to competitively inhibit
all galectin interactions at the cell surface. Addition of exogenous
galectin-3 can be used to drive increased galectin-glycan interac-
tions. Galectin-3 overexpression can also be used to increase
galectin-glycan interactions. SiRNA knockdown of galectin-3 also
allows disruption of galectin-glycan interactions. Using these vari-
ous tools in different combinations it has been possible to compre-
hensively explore the galectin-glycan interaction landscapes that
appear to be important in modulating CIE [27, 28, 30, 31].
2 Materials
1. Dulbecco’s modified Eagle’s medium (Lonza).

2. Heat-inactivated fetal bovine serum (Atlanta Biologicals).
Evaluating the Role of Galectins in Clathrin-Independent Endocytosis 393
2.1 Cell Culture, 3. 100 glutamine solution (Lonza).

Reagents, and 4. 100 penicillin/streptomycin antibiotic solution (Lonza).
Antibodies
5. Eagle’s minimum essential medium (Lonza).
6. T75 tissue culture flasks.
7. 0.05% Trypsin (Gibco).
8. 15 ml Eppendorf tubes.
9. CO2 humidified cell incubator.
10. N-Acetyl glucosamine (GlcNAc) (Sigma-Aldrich).
11. Coverslips.
12. 60-mm culture plates.
13. Glucose (J.T. Baker).
14. Lactose (Sigma-Aldrich).
15. Sucrose (Sigma-Aldrich).
16. Recombinant human galectin-3 (R&D Systems, catalog
no. 8259-GA-050).
17. PBS (Sigma-Aldrich).
18. BSA (Sigma-Aldrich).
19. Benchtop centrifuge.
20. Monoclonal antibodies directed toward MHC class I (clone
w6/32), CD59 (clone p282/H19), and CD98 (clone
MEM-108) (Biolegend).
21. Monoclonal antibody directed toward galectin-3 (clone
A3A12) (Thermo Fisher Scientific).
22. Polyclonal antibody that detects Ezrin (catalog no. 3145S)
(Cell Signaling).
23. Alexa 594-conjugated transferrin (Molecular Probes).
24. Secondary antibodies conjugated to Alexa Fluor 594, 488,
647, 680, and 800 (Molecular Probes).
2.2 Lectin- 1. Dulbecco’s modified Eagle’s medium (Lonza).

Binding Assay 2. Heat-inactivated fetal bovine serum (Atlanta Biologicals).
3. 100 glutamine solution (Lonza).
4. 100 penicillin/streptomycin antibiotic solution (Lonza).
9. Fluorescein labeled lectins (Vector Laboratories): WGA, RCA,
UEA1, concanavalin A, or PHA-L.
10. 0.25% trypsin (Gibco).

11. 1.5 ml Eppendorf tubes.
13. 37.5% formaldehyde.
14. Flow cytometer—BD Fortessa LSR2.
2.3 Antibody 1. Dulbecco’s modified Eagle’s medium (Lonza).

Internalization Assay 2. Heat-inactivated fetal bovine serum (Atlanta Biologicals).
11. Coverslips.
no. 8259-GA-050).
19. Parafilm.
20. Monoclonal antibody directed toward galectin-3 (clone
A3A12) (Thermo Fisher Scientific).
21. Polyclonal antibody that detects Ezrin (catalog no. 3145S)
(Cell Signaling).
23. Secondary antibodies conjugated to Alexa Fluor 594, 488,
647, 680, and 800 (Molecular Probes).
24. Formaldehyde (Sigma-Aldrich).
25. Acetic acid (Sigma-Aldrich).
26. NaCl (Sigma-Aldrich).
27. Ice.
28. Sodium azide (Sigma-Aldrich).
29. HCS cell mask deep red stain (Molecular Probes).
30. Glass slides.

33. Confocal microscope (LSM 780 FCS, Carl Zeiss) with a 40
PlanApo oil immersion objective and 488- and 633-nm laser
excitation.
34. MetaMorph application (Molecular Devices).
2.4 Transferrin 1. Dulbecco’s modified Eagle’s medium (Lonza).

Internalization Assay 2. Heat-inactivated fetal bovine serum (Atlanta Biologicals).
11. Coverslips.
19. Acetic acid (Sigma-Aldrich).
20. NaCl (Sigma-Aldrich).
21. Ice.
25. Flow cytometer—BD Fortessa LSR2.
2.5 Galectin-3 ELISA 1. Dulbecco’s modified Eagle’s medium (Lonza).

Secretion Assay 2. Heat-inactivated fetal bovine serum (Atlanta Biologicals).

11. Coverslips.
no. 8259-GA-050).
18. BSA (Sigma-Aldrich) 6-well plate.
19. 1.5 ml Eppendorf tube.
21. Human galectin-3 ELISA kit (Abcam, ab 188394).
2.6 siRNA 1. Galectin-3 siRNA (Dharmacon, ON-TARGET plus human

Transfection LGALS2 (3958) siRNA-smartpool).
2. Nontargeting siRNA (Dharmacon, ON-TARGET plus
nontargeting pool).
3. Lipofectamine RNAiMAX (Invitrogen).
4. OptiMEM (Invitrogen).
2.7 Plasmids and 1. GFP-tagged galectin-3 (pEGFP-hGal3 (Addgene, plasmid

Transient Transfection no. 73080)).
2. pEGFP-N3 (Clontech, catalog no. 6080-1).
3. Xtremegene9 (Roche Diagnostics).
2.8 FRAP Assays 1. Dulbecco’s modified Eagle’s medium (Lonza).

Heat-inactivated fetal bovine serum (Atlanta Biologicals).

13. 4-well Lab-Tek chambered coverglass (Thermo Fisher Scientific).
14. anti-CD98 antibody (BioLegend).
15. Alexa Fluor 488 Zenon-labeling kit (Molecular Probes).
16. HEPES buffer.
17. Confocal Microscope incubation chamber (Zeiss).
18. Zeiss 780 FCS confocal microscope together with a 488-nm
argon ion laser.
19. GraphPad Prism version 6 software (GraphPad Software, Inc.,
La Jolla, CA).
2.9 Cell- 1. Dulbecco’s modified Eagle’s medium (Lonza).

Spreading Assay 2. Heat-inactivated fetal bovine serum (Atlanta Biologicals).
7. 0.05% Trypsin.
8. 0.25% Trypsin.
13. Coverslips.
no. 8259-GA-050).
22. Cell counter.
23. Hemocytometer slides.
24. Slides.
27. Alexa Fluor 594-linked phalloidin (Molecular Probes).
28. 10% Saponin (Sigma-Aldrich).
29. Mounting solution (Thermo Fisher Scientific).
excitation.
31. MetaMorph application (Molecular Devices) to quantify the
amount of cell spreading.
2.10 GFP-Tagged 1. Dulbecco’s modified Eagle’s medium (Lonza).

Galectin-3 2. Heat-inactivated fetal bovine serum (Atlanta Biologicals).
Secretion Assay
7. 0.05% Trypsin.
8. 0.25% Trypsin.
13. 12-well plate.
14. FluoroBrite Dulbecco’s modified Eagle’s medium (Life
Technologies).
15. GFP-tagged galectin-3 (pEGFP-hGal3 (Addgene, plasmid
no. 73080).
16. pEGFP-N3 (Clontech, catalog no. 6080-1).
17. Xtremegene9 (Roche Diagnostics).
21. Black 96-well plate.
22. PBS.
23. POLARstar Omega Microplate reader (BMG Labtech).
2.11 Macropinosome 1. Heat-inactivated fetal bovine serum (Atlanta Biologicals).

Analysis 2. 100 glutamine solution (Lonza).
6. 0.05% Trypsin.
7. 0.25% Trypsin.
12. Coverslips.
no. 8259-GA-050).
21. Cell counter.
22. Hemocytometer slides.
23. Slides.
24. 37.5% Formaldehyde (Sigma-Aldrich).
25. 5% Sodium azide (Sigma-Aldrich).
26. 10% Saponin (Sigma-Aldrich).
27. Mounting solution.
excitation.
29. MetaMorph application (Molecular Devices to quantify the
amount of cell spreading).
30. Primary antibody for CD59 (Biolegend), CD98 (BioLegend),
and MHC class I (Biolegend).
31. Glacial acetic acid.
32. NaCl.
33. Alexa Fluor 488-conjugated secondary antibody (Molecular
Probes).
34. Glass slides.

35. ImageJ software.
36. Epifluorescent upright microscope (Carl Zeiss).
3 Methods
3.1 Cell Culture, 1. HeLa and Beas2b cells are grown in Dulbecco’s modified
Reagents, and Eagle’s medium (Lonza) supplemented with 10% heat-
Antibodies inactivated fetal bovine serum (Atlanta Biologicals), 1.0%
100 glutamine solution (Lonza), and 1.0% 100 penicillin/
streptomycin antibiotic solution (Lonza).
2. HT1080 cells are grown in Eagle’s minimum essential medium
(Lonza) supplemented with 10% heat-inactivated fetal bovine
serum (Atlanta Biologicals), 1.0% 100 glutamine solution
(Lonza), and 1.0% 100 penicillin/streptomycin antibiotic
solution (Lonza).
3. Cells are maintained at 37 C in a humidified, 5% CO2-contain-
ing atmosphere.
4. For GlcNAc treatment, prepare a 100 mM stock solution of
GlcNAc (Sigma-Aldrich) in sterile complete culture medium,
seed cells on coverslips or on tissue culture plastic in 60-mm
culture plates in 5.0 ml of culture medium at a density of
3 105 cells/plate, add the appropriate volume of GlcNAc
stock solution to each well to achieve the desired sugar con-
centrations. Cells are typically incubated for 48 h with the
sugar. Similarly, for glucose treatment controls, a 100 mM
stock solution of glucose (J.T. Baker) in sterile complete cul-
ture medium is prepared and used.
5. For lactose treatments, prepare a 100 mM solution of lactose
(Sigma-Aldrich) in sterile complete culture medium, incubate
the cells in the lactose solution for 1 h before the start of the
experiment and during the experiment. Similarly, for sucrose
treatment controls, a 100 mM solution of sucrose (Sigma-
Aldrich) in sterile complete culture medium is prepared
and used.
6. For internalization assays with exogenous recombinant human
galectin-3, prepare a 50 μg/ml stock solution of galectin-3
(R&D Systems, catalog no. 8259-GA-050) in sterile PBS.
Add stock solution to conditioned complete medium to make
up the desired concentrations before the addition of primary
antibodies, and controls are made up to the highest concentra-
tion (10 μg/ml) with a similar stock solution of 50 μg/ml BSA
(Sigma-Aldrich).
3.2 Lectin- 1. Incubate cells for 48 h in recommended media with or without

Binding Assay GlcNAc.
2. Rinse cells with 500 μl PBS.
3. Split the cells (at least 50,000 cells each) and incubate with one
of the following fluorescein labeled lectins (Vector Labora-
tories): WGA, RCA, UEA1, concanavalin A, or PHA-L at a
concentration of 20 μg/ml in culture medium on ice for 1 h.
4. Rinse the cells three times in 500 μl media.
5. Detach the cells by incubation with 500 μl 0.25% trypsin for
1 min at 37 C.
6. Collect the cells in 1.5 ml Eppendorf tubes.
7. Spin down the cells in a centrifuge at 1000 rpm for 5 min.
8. Resuspend the cells in 200 μl of PBS and fix by adding 2 μl of
37.5% formaldehyde for a final concentration of 2%
formaldehyde.
9. After 10 min of fixation, rinse the cells three times in
500 μl PBS.
10. Analyze by flow cytometry using a BD Fortessa LSR2.
11. The cell population of interest is selected by gating appropri-
ately (see Note 1).
12. Set the gated cell count to 104 cells and determine mean
fluorescein fluorescence with an excitation maximum at
495 nm and an emission maximum at 515 nm.
3.3 Antibody 1. Grow cells on coverslips with relevant incubations, transfec-

Internalization Assay tions, and pre-treatments. Ideally 3 105 cells are seeded in
each well of a six-well plate. Incubations include supplementa-
tion with sugars like N-acetylglucosamine, transfections can
focus on siRNA knockdowns or plasmid overexpression of
various galectins and pre-treatments include lactose
inhibitions.
2. Place cells on coverslips face up on parafilm.
3. Add 50 μl of conditioned complete medium containing pri-
mary antibody at concentrations of 5 μg/ml for CD59 (Biole-
gend) or 10 μg/ml for MHCI (Biolegend) or 10 μg/ml Alexa
Fluor 488-linked transferrin (Molecular Probes). Schematic of
method presented in Fig. 1.
4. Incubate coverslips either on ice or at 37 C for 30 min.
Transferrin-labeled coverslips are incubated for 10 min.
5. Fix one set of coverslips that is incubated at 37 C in 2%
formaldehyde (representing total antibody bound both inter-
nally and to the surface, Ttot).
Fig. 1 Schematic representation of the workflow of the antibody internalization assay

6. Acid-wash for 40 s using a solution of 0.5% acetic acid, 0.5 M

NaCl, pH 3.0. The other set of coverslips that is held at 37 C
(representing antibody internalized, Tint) as well as the set that
is kept on ice (null control, T0) is fixed in 2% formaldehyde.
7. Surface block the coverslips for 30 min using a blocking solu-
tion of 10% FBS, 0.02% sodium azide in 50 μl PBS.
8. Label the coverslips at room temperature for 1 h with 1:500
Alexa Fluor 488-conjugated goat anti-mouse secondary anti-
body (Molecular Probes), 0.2% saponin, and 2 μg/ml HCS cell
mask deep red stain (Molecular Probes) in 50 μl blocking
solution.
9. Rinse the coverslips three times in 500 μl blocking solution and
a final time with PBS.
10. Mount the coverslips on glass slides (see Note 2).
11. Image at room temperature using a confocal microscope (LSM
780 FCS, Carl Zeiss) with a 40 PlanApo oil immersion
objective and 488- and 633-nm laser excitation.
12. For each coverslip, image six positions with 2 2 tiling with
the pinhole kept completely open. For each condition across
the six tiled images, ensure that at least 100 cells are imaged. All
images for each antibody must be taken with identical acquisi-
tion parameters, which are set based on the control’s Ttot
coverslip and tuned such that the signal is within the dynamic
range (see Note 3).
13. Use the MetaMorph application (Molecular Devices) to quan-
tify the percentage of antibody internalized.
14. Separate confocal images into the two different channel colors.
15. Set an antibody channel threshold for each condition using its
T0 coverslip images (see Note 4).
16. Threshold the Tint and Ttot conditions, and measure the
integrated signal intensity for the antibody channel.
17. Use autothresholding for the cell mask channel, and measure
the threshold area (see Note 5).
18. The integrated signal intensity for the antibody is then normal-
ized to the threshold area for each image.
19. The percentage internalized is then calculated with the equa-
tion, %internal ¼ 100 Ttot/area/Tint/area.
20. Finally, the internal percentages are normalized to the
untreated control.
3.4 Transferrin 1. Seed 1.5 105 cells in a 24-well plate with relevant incuba-
Internalization Assay tions, transfections, and pre-treatments.
2. Incubate cells with 10 μg/ml Alexa Fluor 546-linked transfer-

rin (Molecular Probes) in 500 μl complete medium for 10 min
either on ice (TransferrinSurface) or at 37 C (TransferrinTotal).
3. Wash the cells three times in 500 μl of cold PBS and detach
using 100 μl of 0.05% trypsin on ice for 10 min.
4. Collect the cells in 1.5 ml Eppendorf tubes.
5. Spin down the cells in a centrifuge at 1000 rpm for 5 min.
6. Resuspend the cells in 200 μl of PBS and fix by adding 2 μl of
37.5% formaldehyde for a final concentration of 2%
formaldehyde.
7. After 10 min of fixation, rinse the cells three times in
500 μl PBS.
8. Analyze by flow cytometry using a BD Fortessa LSR2.
9. The cell population of interest is selected by gating appropri-
ately (see Note 1).
10. Set the gated cell count to 104 cells and determine mean
fluorescence.
11. Calculate the percentage internalized using the equation, %
internal ¼ 100 (TransferrinTotal TransferrinSurface)/
TransferrinTotal.
12. Normalize the internal percentages to the untreated control.
3.5 Galectin-3 ELISA 1. Seed cells on a 6-well plate at a density of 3 105 cells/well in
Secretion Assay 1 ml of complete medium in quadruplicate.
2. After 48 h in culture, collect the supernatants (i.e., conditioned
medium) in a 1.5 ml Eppendorf tube.
3. Spin down supernatants at 3000 g for 10 min to remove any
cell debris.
4. For lactose and sucrose extraction conditions, after the super-
natant is removed, rinse the cells on the plate with complete
medium and then incubate for 1 h with 1 ml of 100 mM lactose
(Sigma-Aldrich) or sucrose (Sigma-Aldrich) in serum-free
medium.
5. After the incubation, collect the lactose or sucrose extract and
spin down at 3000 g for 10 min to remove any cell debris.
6. Analyze samples of complete medium, serum-free medium,
conditioned medium, and the lactose and sucrose extracts for
Galectin-3 content using a human Galectin-3 ELISA kit
(Abcam, ab 188394) as recommended by the manufacturer.
3.6 siRNA 1. Seed 3 10 cells per well in a 6-well plate in antibiotic-free

Transfection complete medium and transfect with a 50 nM final concentra-
tion of Galectin-3 siRNA (Dharmacon, ON-TARGET plus
human LGALS2 (3958) siRNA-smartpool) or nontargeting

siRNA (Dharmacon, ON-TARGETplus nontargeting pool)
using Lipofectamine RNAiMAX (Invitrogen) as recommended
by the manufacturer.
2. Cells should be used in experiments 48 h after transfection.
3. SiRNA transfection efficiency is confirmed by western blotting.
3.7 Plasmids and 1. Seed 3 10 cells per well in a 6-well plate in complete medium.
Transient Transfection 2. After a 24-h incubation, transfect cells with GFP-tagged Galec-
tin-3 (pEGFP-hGal3 (Addgene, plasmid no. 73080); pEGFP-
N3 (Clontech, catalog no. 6080-1)) at 2.5 μg of DNA/60-mm
well using Xtremegene9 (Roche Diagnostics) as recommended
by the manufacturer.
3. Experiments should be performed 48 h after transfection.
4. Transfection is confirmed by western blotting.
3.8 FRAP Assays 1. Culture 5 104 cells in 4-well Lab-Tek chambered coverglass
(Thermo Fisher Scientific).
2. Label cells with 1:100 anti-CD98 (BioLegend), which was
Zenon-labeled with Alexa Fluor 488 (Molecular Probes) fol-
lowing the manufacturer’s instructions.
3. After transfections, incubations, and lactose pre-treatments,
cells in complete medium with 25 mM HEPES buffer are
then analyzed for fluorescence recovery after photobleaching
(explained step by step below).
4. Cells are incubated at 37 C for the duration of the FRAP
experiments using an incubation chamber.
5. Using a Zeiss 780 FCS confocal microscope together with a
488-nm argon ion laser for excitation of Alexa Fluor 488 or
GFP, monitor emissions at 525 nm.
6. Adjust the bleaching laser intensity and number of bleaching
scan iterations to obtain a 75% loss in fluorescence in a circular
2-μm diameter photobleached region on the apical, medial, or
basal focal planes of the cell membrane.
7. Image multiple regions pre- and post-photobleaching using
low laser intensities, and recovery fluorescence in the selected
regions is tracked over time.
8. Fluorescence in unbleached regions on a cell (Iref) and on the
coverglass (Ibackground) are also monitored over time.
9. Subtract Ibackground(t) from the measured fluorescence intensity
at each time point.
10. Iref(t) is used to account for photobleaching over time as
follows.

I norm ðt Þ ¼ I refprebleach =I ref ðt Þ I ðt Þ=I prebleach ð1Þ
11. Convert the fluorescence intensities (Inorm(t)) to normalized

fluorescence intensities (NFI(t)) using the following
equation.

NFIðt Þ ¼ I norm ðt Þ I postbleach = I prebleach I postbleach ð2Þ
12. Plot the NFI against time and fit to a one-phase exponential
association curve using the GraphPad Prism version 6 software
(GraphPad Software, Inc., La Jolla, CA).
13. From the fit of the curves, time constants for half-recovery are
derived (t1/2).
14. Calculate the diffusion coefficients using the Soumpasis equa-
tion as follows: D ¼ 0.224 (r2/t1/2), where D is the
diffusion coefficient, r is the radius of the region, and t1/2 is
the time constant for half-recovery.
15. For each condition, the FRAP is measured for at least
20 regions on separate cells in each focal plane, and the entire
experiment should be done in biological triplicate.
3.9 Cell- 1. Grow 1 106 cells in 60 mm culture plates with required

Spreading Assay incubations and transfections.
2. Trypsinize cells with 1 ml of 0.25% trypsin. At 37 C.
3. Collect cells in a 1.5 ml Eppendorf tube.
4. Pellet cells by centrifuging at 300 g for 5 min.
5. Resuspend cells in 0.5 ml complete medium.
6. Count the cells and split equal numbers of cells into four 1.5 ml
Eppendorf tubes, pellet, and resuspend in complete medium
with or without 100 mM lactose or 100 mM sucrose.
7. Incubate the cells in suspension at 37 C for 1 h and then seed
at equal density on coverslips.
8. Allow the cells to attach and spread for 2 or 4 h at 37 C.
9. Rinse the coverslips with 0.5 ml PBS.
10. Fix in 0.5 ml of 2% formaldehyde for 10 min.
11. Block the coverslips for 30 min using 0.5 ml of a blocking
solution of 10% FBS, 0.02% sodium azide in PBS.
12. Label the coverslips at room temperature for 1 h with
0.67 units/ml Alexa Fluor 594-linked phalloidin (Molecular
Probes) and 0.2% saponin in 50 μl blocking solution.
13. Rinse the coverslips three times in blocking solution and a final
time with 0.5 ml PBS.
14. Mount the coverslips on glass slides (see Note 2).
15. Image the coverslips at room temperature using a confocal
microscope (LSM 780 FCS, Carl Zeiss) with a 40 PlanApo
oil immersion objective and 488- and 594-nm laser excitation.
16. For each coverslip, 10 positions are imaged. For each condition
across the 10 images, at least 100 cells should be imaged.
17. Ensure that all images are imaged with identical acquisition
parameters.
18. Use the MetaMorph application (Molecular Devices) to quan-
tify the amount of cell spreading.
19. Draw a region of interest around each individual cell, and set an
automatic threshold for light objects.
20. Measure the threshold area in each region (which corresponds
to the individual cell area). In the case of transient plasmid
expression, only cells that are expressing GFP should be out-
lined and measured.
3.10 GFP-Tagged 1. Seed cells on a 12-well plate at a density of 5 104 cells/well in

Galectin-3 1 ml of complete medium in quadruplicate.
Secretion Assay 2. After 24 h, replace the medium with 500 μl of FluoroBrite
Dulbecco’s modified Eagle’s medium (Life Technologies) sup-
plemented with 10% heat-inactivated fetal bovine serum
(Atlanta Biologicals), 1.0% 100 glutamine solution (Lonza),
and 1.0% 100 penicillin/streptomycin antibiotic solution
(Lonza).
3. Transfect cells with GFP-tagged Galectin-3 (Addgene) or GFP
(Clontech) using Xtremegene9 (Roche Diagnostics) as recom-
mended by the manufacturer.
4. 48 h post-transfection, collect the supernatants in 1.5 ml
Eppendorf tubes and spin down at 3000 g for 10 min to
remove any cell debris.
5. Transfer 200 μl of the supernatant to a black 96-well plate in
duplicate.
6. Detach the cells by incubating with 200 μl 0.25% trypsin for
1 min at 37 C.
7. Cells are collected, spun down, and resuspended in 500 μl
of PBS.
8. Transfer 200 μl of the cell suspensions to the black 96-well
plate in duplicate.
9. Measure the bulk fluorescence in each well using a Synergy H1
multimode microplate reader (BioTek).
10. After background-subtracting any fluorescence from the

medium, PBS, and cell autofluorescence, normalize the fluo-
rescence from the supernatant wells to the fluorescence in the
cell suspension wells.
3.11 Macropinosome 1. Seed 3 105 HT1080 cells on coverslips in a six-well plate.

Analysis 2. After required incubations and pre-treatments, place the cells
on coverslips face-up on parafilm, and add 50 μl of complete
medium with primary antibody at concentrations of 5 μg/ml
for CD59 (Biolegend) or CD98 (BioLegend), or 10 μg/ml for
MHC class I (Biolegend).
3. Incubate the coverslips at 37 C for 30 min.
4. Acid-wash the coverslips for 40 s using 200 μl of a solution of
0.5% acetic acid, 0.5 M NaCl, pH 3.0.
5. Fix the coverslips in 50 μl 2% formaldehyde for 10 min.
6. Block the coverslips for 30 min using 200 μl of a blocking
solution of 10% FBS, 0.02% sodium azide in PBS.
7. Label the coverslips at room temperature for 1 h with Alexa
Fluor 488-conjugated secondary antibody (Molecular Probes)
and 0.2% saponin in 50 μl blocking solution.
8. Rinse the coverslips three times in 200 μl blocking solution and
a final time with PBS.
9. Mount coverslips on glass slides.
10. Image coverslips at room temperature using a confocal micro-
scope (LSM 780 FCS, Carl Zeiss) with a 40 PlanApo oil
immersion objective and 488- or 594-nm laser excitation.
11. For each coverslip, image five positions. For each condition
across the five images, at least 100 cells should be imaged.
12. Ensure that all images for each antibody are taken with identi-
cal acquisition parameters.
13. Use the ImageJ software to measure the size of the individual
macropinosomes.
14. Each coverslip is also studied using an epifluorescent upright
microscope (Carl Zeiss).
15. Record the number of cells, the number of cells performing
macropinocytosis, and the number of macropinosomes for
numerous fields of view, with a minimum of 100 cells counted
for each coverslip (see Note 6).
16. Calculate the fraction of cells performing micropinocytosis and
the number of macropinosomes/cell for each condition.
4 Notes
1. Gates for flow cytometry are set to gate for intact cells. The
gates are set up to count as many of the main cell population as
possible while excluding broken cells and cell fragments. Gates
are set up based on the forward and side scatter and the primary
intact cell population is gated for.
2. To mount coverslips, a drop of mounting solution is placed on
a slide, the coverslip is blotted to remove excess water and then
placed face down on the mounting solution droplet. After the
mounting solution is allowed to fry, the coverslips are sealed
using a thin ring of clear nail polish around the edge of the
coverslip.
3. In order to identify a dynamic range for imaging the control
conditions, coverslip is viewed under the confocal and the gain
and laser power are adjusted so that the image is clear with no
saturation in the field of view. These settings are fixed and used
for all imaging for that antibody.
4. When using the metamorph software to analyze the images, the
first step is to set thresholds to subtract any background fluo-
rescence. To do this the images for each condition, T0 coverslip
are viewed and a threshold is set such that minimal fluorescence
is picked up (as close to 0 as possible without increasing the
threshold too high). The average of the thresholds for this set
of images is then used to threshold the Tint and Ttot images for
each condition to subtract background fluorescence.
5. For each of the Tint and Ttot images, the cell area is determined
by using the software to do an automatic threshold for the cell
mask channel and using the thresholded area measured to
determine total cell area per field of view.
6. Cells performing macropinocytosis are easily identifiable by the
presence of micron size vesicles with clearly visible lumens
unlike clathrin-independent and clathrin-mediated endocytosis
which produce small 50–200 μm diameter vesicles that appear
as dots under confocal imaging. Macropinosomes are notice-
ably large and easily countable.
References
1. Johannes L, Jacob R, Leffler H (2018) Galec- 3. Liu F-T, Patterson RJ, Wang JL (2002) Intra-
tins at a glance. J Cell Sci 131(9). https://doi. cellular functions of galectins. Biochim Bio-
org/10.1242/jcs.208884 phys Acta 1572(2):263–273. https://doi.
2. Leffler H, Carlsson S, Hedlund M, Qian Y, org/10.1016/S0304-4165(02)00313-6
Poirier F (2002) Introduction to galectins. 4. Ochieng J, Furtak V, Lukyanov P (2002)
Glycoconj J 19(7):433–440. https://doi.org/ Extracellular functions of galectin-3.
10.1023/B:GLYC.0000014072.34840.04
Glycoconj J 19(7):527–535. https://doi.org/ 14. O’Seaghdha CM, Hwang S-J, Ho JE, Vasan
10.1023/B:GLYC.0000014082.99675.2f RS, Levy D, Fox CS (2013) Elevated galectin-
5. Nabi IR, Shankar J, Dennis JW (2015) The 3 precedes the development of CKD. J Am Soc
galectin lattice at a glance. J Cell Sci Nephrol 24(9):1470–1477. https://doi.org/
128(13):2213–2219. https://doi.org/10. 10.1681/ASN.2012090909
1242/jcs.151159 15. Pasquini LA, Millet V, Hoyos HC, Giannoni
6. Anand IS, Rector TS, Kuskowski M, JP, Croci DO, Marder M, Liu FT, Rabinovich
Adourian A, Muntendam P, Cohn JN (2013) GA, Pasquini JM (2011) Galectin-3 drives oli-
Baseline and serial measurements of galectin-3 godendrocyte differentiation to control myelin
in patients with heart failure: relationship to integrity and function. Cell Death Differ
prognosis and effect of treatment with valsar- 18(11):1746–1756. https://doi.org/10.
tan in the Val-HeFT. Eur J Heart Fail 1038/cdd.2011.40
15(5):511–518. https://doi.org/10.1093/ 16. Piper SE, de Courcey J, Sherwood RA, Amin-
eurjhf/hfs205 Youssef GF, McDonagh TA (2016) Serial
7. Bresalier RS, Mazurek N, Sternberg LR, Byrd galectin-3 for the monitoring of optimally trea-
JC, Yunker CK, Nangia-Makker P, Raz A ted stable chronic heart failure: a pilot study.
(1998) Metastasis of human colon cancer is Int J Cardiol 207:279–281. https://doi.org/
altered by modifying expression of the 10.1016/j.ijcard.2016.01.179
β-galactoside-binding protein galectin 3. Gas- 17. Takemoto Y, Ramirez RJ, Yokokawa M,
troenterology 115(2):287–296. https://doi. Kaur K, Ponce-Balbuena D, Sinno MC, Willis
org/10.1016/S0016-5085(98)70195-7 BC, Ghanbari H, Ennis SR, Guerrero-Serna G,
8. Furtak V, Hatcher F, Ochieng J (2001) Henzi BC, Latchamsetty R, Ramos-
Galectin-3 mediates the endocytosis of β-1 Mondragon R, Musa H, Martins RP, Pandit
integrins by breast carcinoma cells. Biochem SV, Noujaim SF, Crawford T,
Biophys Res Commun 289(4):845–850. Jongnarangsin K, Pelosi F, Bogun F,
https://doi.org/10.1006/bbrc.2001.6064 Chugh A, Berenfeld O, Morady F, Oral H,
9. Iacobini C, Blasetti Fantauzzi C, Bedini R, Jalife J (2016) Galectin-3 regulates atrial fibril-
Pecci R, Bartolazzi A, Amadio B, Pesce C, lation remodeling and predicts catheter abla-
Pugliese G, Menini S (2018) Galectin-3 is tion outcomes. JACC Basic Transl Sci
essential for proper bone cell differentiation 1(3):143–154. https://doi.org/10.1016/j.
and activity, bone remodeling and biomechan- jacbts.2016.03.003
ical competence in mice. Metabolism 83: 18. Takenaka Y, Fukumori T, Raz A (2002)
149–158. https://doi.org/10.1016/j. Galectin-3 and metastasis. Glycoconj J
metabol.2018.02.001 19(7):543–549. https://doi.org/10.1023/B:
10. Iurisci I, Tinari N, Natoli C, Angelucci D, GLYC.0000014084.01324.15
Cianchetti E, Iacobelli S (2000) Concentra- 19. Thurston TLM, Wandel MP, von Muhlinen N,
tions of galectin-3 in the sera of normal con- Foeglein Á, Randow F (2012) Galectin-8-
trols and cancer patients. Clin Cancer Res targets damaged vesicles for autophagy to
6(4):1389–1393 defend cells against bacterial invasion. Nature
11. Kajitani K, Yanagimoto K, Nakabeppu Y 482(7385):414–418. https://doi.org/10.
(2017) Serum galectin-3, but not galectin-1, 1038/nature10744
levels are elevated in schizophrenia: implica- 20. Mayor S, Pagano RE (2007) Pathways of
tions for the role of inflammation. Psychophar- clathrin-independent endocytosis. Nat Rev
macology 234(19):2919–2927 Mol Cell Biol 8(8):603–612. http://www.
12. Kim H-J, Do I-G, Jeon H-K, Cho YJ, Park YA, nature.com/nrm/journal/v8/n8/suppinfo/
Choi J-J, Sung CO, Lee Y-Y, Choi CH, Kim nrm2216_S1.html
T-J, Kim B-G, Lee J-W, Bae D-S (2013) Galec- 21. McMahon HT, Boucrot E (2011) Molecular
tin 1 expression is associated with tumor inva- mechanism and physiological functions of
sion and metastasis in stage IB to IIA cervical clathrin-mediated endocytosis. Nat Rev Mol
cancer. Hum Pathol 44(1):62–68. https://doi. Cell Biol 12(8):517–533. http://www.nature.
org/10.1016/j.humpath.2012.04.010 com/nrm/journal/v12/n8/suppinfo/nrm31
13. Melin EO, Dereke J, Thunander M, Hillman 51_S1.html
M (2018) Depression in type 1 diabetes was 22. Howes MT, Mayor S, Parton RG (2010)
associated with high levels of circulating Molecules, mechanisms, and cellular roles of
galectin-3. Endocr Connect 7(6):819. clathrin-independent endocytosis. Curr Opin
https://doi.org/10.1530/ec-18-0108
Cell Biol 22(4):519–527. https://doi.org/10. Med Chem Lett 25(6):1223–1227. https://

1016/j.ceb.2010.04.001 doi.org/10.1016/j.bmcl.2015.01.060
23. Mayor S, Parton RG, Donaldson JG (2014) 28. Mathew MP, Tan E, Saeui CT, Bovonratwet P,
Clathrin-independent pathways of endocytosis. Sklar S, Bhattacharya R, Yarema KJ (2016)
Cold Spring Harb Perspect Biol 6(6):a016758. Metabolic flux-driven sialylation alters internal-
https://doi.org/10.1101/cshperspect. ization, recycling, and drug sensitivity of the
a016758 epidermal growth factor receptor (EGFR) in
24. Sandvig K, Kavaliauskiene S, Skotland T SW1990 pancreatic cancer cells. Oncotarget
(2018) Clathrin-independent endocytosis: an 7(41):66491–66511. https://doi.org/10.
increasing degree of complexity. Histochem 18632/oncotarget.11582
Cell Biol 150(2):107–118. https://doi.org/ 29. Lajoie P, Partridge EA, Guay G, Goetz JG,
10.1007/s00418-018-1678-5 Pawling J, Lagana A, Joshi B, Dennis JW,
25. Johannes L, Parton RG, Bassereau P, Mayor S Nabi IR (2007) Plasma membrane domain
(2015) Building endocytic pits without cla- organization regulates EGFR signaling in
thrin. Nat Rev Mol Cell Biol 16(5):311–321. tumor cells. J Cell Biol 179(2):341–356.
https://doi.org/10.1038/nrm3968. http:// https://doi.org/10.1083/jcb.200611106
www.nature.com/nrm/journal/v16/n5/ 30. Mathew MP, Donaldson JG (2018) Distinct
abs/nrm3968.html#supplementar y- cargo-specific response landscapes underpin
information the complex and nuanced role of galectin-
26. Lakshminarayan R, Wunder C, Becken U, glycan interactions in clathrin-independent
Howes MT, Benzing C, Arumugam S, endocytosis. J Biol Chem
Sales S, Ariotti N, Chambon V, Lamaze C 293(19):7222–7237. https://doi.org/10.
(2014) Galectin-3 drives glycosphingolipid- 1074/jbc.RA118.001802
dependent biogenesis of clathrin-independent 31. Mathew MP, Donaldson JG (2019) Glycosyla-
carriers. Nat Cell Biol 16(6):595 tion and glycan interactions can serve as extra-
27. Mathew MP, Tan E, Saeui CT, Bovonratwet P, cellular machinery facilitating clathrin-
Liu L, Bhattacharya R, Yarema KJ (2015) Met- independent endocytosis. Traffic
abolic glycoengineering sensitizes drug- 20(4):295–300. https://doi.org/10.1111/
resistant pancreatic cancer cells to tyrosine tra.12636
kinase inhibitors erlotinib and gefitinib. Bioorg
Chapter 22
Examination of Galectin-3 Recruitment into Multivesicular

Bodies for Exosomal Secretion
Sebastian B€anfer, Sophie Kutscher, and Ralf Jacob
Abstract
Cells use unconventional secretion to deliver the β-galactoside binding lectin galectin-3 from the cell
interior into the extracellular milieu. This process starts with galectin-3 recruitment into intraluminal
vesicles (ILVs), which are later released at the plasma membrane as exosomes. Electron microscopy is
utilized to determine the location of GFP-tagged galectin-3 in pelleted exosomes. We also describe how
these vesicles are harvested from cell culture media to determine their composition. The fluorescent protein
GFP was fused with the exosomal sorting motif of galectin-3 to direct GFP into exosomes. Recruitment of
this fusion construct into the lumen of exosomes can be assessed by proteinase K accessibility analysis.
Key words Unconventional secretion, Galectins, Exosomes, ESCRT, P(S/T)AP motif
1 Introduction
Almost all secretory proteins and transmembrane proteins that are

transported to the plasma membrane and into the extracellular
space follow the conventional route via endoplasmic reticulum
(ER) and Golgi apparatus. The starting point of this targeted
transport is based on the intrinsic signal sequence that directs the
nascent polypeptide chain to the ER. However, more and more
extracellular proteins were identified which do not have a (classic)
signal sequence or which have a signal sequence but are neverthe-
less secreted independently of the conventional route. This trans-
port route is therefore summarized as unconventional protein
secretion (UPS), but comprises a whole series of different routes
(UPS Type-I to -IV) [1].
So how do such unconventionally secreted proteins reach their
target in the extracellular space? In particular, what are the under-
lying molecular mechanisms, i.e., the sorting signals and receptors,
which could guarantee such a targeted process? These fundamental
questions in the growing field of exocytosis have recently been
elucidated for a small subset of proteins. These include galectin-3,
413
414 €nfer et al.
Sebastian Ba
the only chimeric member of the galectin family [2]. In addition to

a C-terminal carbohydrate recognition domain (CRD), this galec-
tin has a unique N-terminus, which enables galectin-3 to form
sugar-independent oligomers and high-molecular clusters with
other ligands [3] and also seems to be responsible for unconven-
tional secretion of the lectin [4].
Galectin-3 plays multiple roles intracellularly as well as extra-
cellularly. For example, extracellular galectin-3 influences the cell
cycle, apoptosis, immunomodulation, as well as tumor progression
and metastasis [5, 6]. Within the lumen of endosomes galectin-3
acts as a sorting receptor in post-Golgi compartments [7, 8]. How
can this lectin perceive these tasks in the extracellular milieu with-
out any ER signal sequence? This question came up very early.
Various groups were able to show that galectin-3 is secreted regard-
less of the classic secretion pathway [9–12]. In this regard, it has
been shown that inhibitors of classic secretion had no influence on
the secretion of galectin-3. Only the destruction of the microtubule
cytoskeleton or incubation at 20 C prevents secretion.
Functional roles for bypassing the classical secretion pathway
could lie in prohibition of galectin-3 cross-linking with ligands in
early compartments of the secretory pathway and/or a putative
inactivation of the lectin when passing through early secretory
compartments. It is worth taking a look at the few known excep-
tions of the galectin family that have a signal sequence. One exam-
ple is the galectin GCLT1 [13] from the sponge Geodia cydonium.
GCLT1 has a signal sequence, is glycosylated in the course of the
secretory pathway and can homooligomerize [14]. Another exam-
ple is Lec-5 from Caenorhabditis elegans, which also has a signal
sequence [15]. Lec-5 is N-glycosylated in direct vicinity of an
amino acid that is essential for the carbohydrate recognition
domain [16, 17]. Interestingly, Lec-5 does not bind to asialofetuin
[18], a known and preferred binding partner of galectin-3. This
would suggest that galectins with a certain sugar binding affinity are
blocked if they follow the classical secretory pathway, while others
remain intact.
Regarding the mechanism for non-classical secretion of galec-
tin-3, there is very little evidence for a direct translocation process
across the plasma membrane [12], either by induced pore forma-
tion (UPS Type-I) or by a transporter (UPS Type-II) that would
explain the large proportion of extracellular galectin-3 [19, 20].
In contrast, there is a lot of evidence that galectin-3 is secreted
using vesicular structures, which strongly suggests a sorting mech-
anism by a UPS type-III transport route. However, the nature of
secreted galectin-3 containing vesicles has to be clarified.
Non-classically secreted extracellular vesicles are generated at the
plasma membrane (microvesicles) or are released by fusion of multi-
vesicular bodies (MVBs) with the plasma membrane (exosomes).
Most commonly, these vesicle populations are specifically isolated
Examination of Galectin-3 Recruitment into Multivesicular Bodies for. . . 415
by differential centrifugation, which involves several centrifugation

and ultracentrifugation steps. Since microvesicles are larger than
exosomes, they pellet at 10,000 g, whereas the roughly 100 nm
small exosomes pellet at 100,000 g. Thus, microvesicles and
exosomes can be clearly separated from each other based on their
sedimentation profile.
A very large number of scientific studies that could detect
galectin-3 in exosomal vesicles of very different origins confirms
this type of secretion for galectin-3 [21–24]. Protocols used in
these studies slightly vary amongst users. Especially the viscosity
of different body fluids as primary material alters the efficiency of
vesicle recovery.
In our own work, a molecular mechanism of galectin-3 recruit-
ment and sorting into exosomes has been elucidated [4]. Using
epithelial Madin Darby canine kidney cells (MDCK) in culture, we
showed that galectin-3 is found in the lumen of apically secreted
exosomes (Fig. 1), but not in microvesicles. Using electron micros-
copy and superresolution fluorescence microscopy, intracellular
galectin-3 recruitment onto the membranes of MVBs was visua-
lized. This results in packaging of the lectin into the budding
intraluminal vesicles (ILVs), which are then secreted as exosomes.
The process is induced by the highly conserved tetrapeptide motif
PSAP in the N-terminus of galectin-3, which enables the lectin to
interact directly with the tsg101 polypeptide. Tsg101 is a compo-
nent of the ESCRT-I complex and is involved in cargo recruitment
into exosomes [25]. In accordance, inhibition of exosomal secre-
tion by dimethylamiloride (DMA; Fig. 2), tsg101-knockdown, and
the expression of the dominant-negative mutant Vps4aE228Q
blocked the release of galectin-3 into the medium. The use of
DMA in particular allows it to draw conclusions from the specific
inhibition of exosome secretion as to whether the respective pro-
tein of interest is secreted via exosomes and whether this may also
represent the exclusive route of secretion. In this way it was possible
to show that exosomal secretion is thus most likely the exclusive
secretion pathway for galectin-3 in MDCK cells.
The methods that our lab has used to describe the unconven-
tional secretion of galectin-3 primarily aim to purify and analyze
exosomes and to investigate the process of secretion by specific
inhibitors. Labeling of exosomal galectin-3 for electron microscopy
also plays essential roles to visualize distinct steps in exosomal
galectin-recruitment.
2 Materials
2.1 Cell Lines Madin Darby canine kidney (MDCK) strain II cells were used [26].
416 €nfer et al.
Sebastian Ba
exosomal fraction
galectin-3-eGFP
Fig. 1 Electron microscopy analysis of galectin-3-eGFP localization in exosomes.

The pellet from exosomal fractionation was embedded, cut and finally stained
against galectin-3-eGFP. As a result, 15 nm gold-labeled GFP-nanobodies
(arrowheads) were detected in exosomes. The lipid bilayer of the exosomes is
also visible (arrow). Scale bars: 100 nm; inset 50 nm
MW lysate exosomes
(kDa)
- + - + DMA
55
tsg101
36
galectin-3
Fig. 2 Dimethylamiloride-inhibition of galectin-3 secretion. Exosomes were

collected from MDCK cells incubated in the presence or absence of the exosome
release blocker dimethylamiloride (DMA). Cell lysates and isolated exosomes
were separated by SDS-PAGE and analyzed by immunoblot with antibodies
directed against the ESCRT I-component tsg101 or galectin-3. Take note of a
dramatic reduction of both components in the exosomal pellet prepared from the
supernatants of DMA-treated MDCK cells
2.2 Plasmids The plasmids pGal3-eGFP [27] and peGFP-PSAP were used. To
generate peGFP-PSAP, the late domain like-motif PSAP of eGFP-
PSAP was inserted by DpnI-mediated site-directed mutagenesis of
pEGFP-N1 plasmid (Clontech) with primers 50 -C ATG GAC GAG
CTG ccc ggg cca agt gct cct ggt cca TAC AAG TAA AGG-30 and
50 -CCT TTA CTT GTA tgg acc agg agc act tgg ccc ggg-CAG CTC
GTC CAT G-30 (inserted nucleotides are depicted in small letters).
2.3 Antibodies
Antibodies: GFP (Clontech, JL-8).

GFP-nanobody (Chromotek; labeled with 15 nm gold,
Cytodiagnostics).
tsg101 (Abcam 4A10).
Gp80 (Santa Cruz, Clusterin-α, C-18).
Galectin-3 (Abcam, EPR19244).
2.4 Buffers Phosphate buffered saline (PBS), pH 8.0.

3 SDS-PAGE sample buffer: 6% (w/v) Sodium dodecyl sulfate,
30% (v/v) Glycerol, 0.02% (w/v) Bromophenol blue, 150 mM
Tris-HCl pH 6.8, 150 mM Dithiothreitol.
Electrophoresis buffer: 25 mM Tris, 190 mM Glycine, 0.1% (w/v)
Sodium dodecyl sulfate.
Transfer buffer: 48 mM Tris, 39 mM Glycine, 0.04% Sodium
dodecyl sulfate, 20% Ethanol.
Cell lysis buffer: 25 mM Tris-HCl pH 7.6, 150 mM NaCl, 5 mM
EDTA, 1% NP-40, 1% Sodium dodecyl sulfate.
2.5 Cell Cell culture medium: MEM high glucose supplemented with
Culture Media 2 mM glutamine, 100 U/ml penicillin, 100 mg/ml strepto-
mycin, and 10% FCS.
Cell culture medium for transfection: Opti-MEM (Gibco).
3 Methods
3.1 Cell Culture MDCKII cells were cultured in MEM medium at 37 C in humi-
dified atmosphere containing 5% CO2.
3.2 Generation of 1. Prepare a 10 cm-cell culture dish with 80% confluence of

Stable MDCK Cell Lines MDCK cells.
2. Prepare two tubes each with 1.5 ml Opti-MEM.
3. Add 10 μl of Lipofectamine 2000 (Invitrogen) to one tube and
5 μg of the respective DNA into the other and incubate sepa-
rately for 5 min at room temperature.
4. Mix both mixtures and incubate for 30 min at room
temperature.
5. In the meantime, wash the cells 2 with 10 ml PBS to remove
all medium.
6. Then drop the prepared Opti-MEM solution with the DNA
onto the cells.
418 €nfer et al.
Sebastian Ba
7. After 6 h, replace the transfection mixture with 10 ml cell

culture medium without FCS.
8. If necessary, repeat the procedure the next day.
9. On the following day, replace the second transfection mixture
or the FCS-free medium with 10 ml with normal cell culture
medium.
10. Wait until individual areas of single cell clones are visible.
11. For generation of stable cell lines, split the transfected cells 1:
1000 2 days after transfection in MEM medium containing
0.4 mg/ml G418 or an equivalent antibiotic for selection.
12. Transfer single clones to 24-well plates with trypsin/EDTA-
soaked Whatman slices.
13. Analyze single clones for expression of the exogenous protein
by immunoblot and fluorescence microscopy.
14. In a second round of single cell clone isolation, dilute cells of
positive clones to a dilution of 0.5 cell per 100 μl. Disperse cell
suspension in 96-well plate (100 μl per well).
15. Use single cell clones for further analysis.
3.3 Inhibition of The inhibitor dimethylamiloride (DMA) can be used to investigate

Exosomal Secretion the exosomal secretion pathway [28]. The content of the protein in
with the medium as well as in the exosomal fraction can be measured by
Dimethylamiloride immunoblot.
1. Prepare cell culture medium containing a final concentration of
15 nM DMA. In order to exclude false positive results due to
contamination with bovine exosomes, the cell culture medium
was centrifuged at 100,000 g, the pellet removed, and the
supernatant sterile filtered.
2. Incubate up to 18 cell culture dishes (10 cm) with the prepared
DMA-cell culture medium overnight.
3. Collect cell culture medium and/or purify exosomal fraction.
3.4 Purification of Exosomes can be purified from the cell culture supernatant by
Exosomes from differential centrifugation [29] (see Note 1). For this purpose, up
MDCK Cells to 18 cell culture dishes (10 cm) were incubated with cell culture
medium at 37 C for 16 h (see Note 2). As above, to prevent false
positive results due to contamination with bovine exosomes, the
cell culture medium was centrifuged at 100,000 g, the pellet
removed and the supernatant sterile filtered.
1. Collect cell culture media (see Note 3).
2. Centrifuge at 500 g for 6 min (removal of remaining cells).
The pellet can be discarded. Use the supernatant for the next
centrifugation.
3. Centrifuge at 5000 g for 30 min (removal of bulky apoptotic

cell debris). The pellet can be discarded. Use the supernatant
for the next centrifugation.
4. Centrifuge at 20,000 g for 30 min (removal of microvesi-
cles). The pellet can be discarded. Use the supernatant for the
next centrifugation.
5. Centrifuge at 100,000 g for 1 h (pelleting of exosomes).
Discard the supernatant.
6. Wash the pellet with 1 ml ice-cold PBS.
7. Centrifuge at 100,000 g for 1 h (pelleting the exosomes).
Discard the supernatant.
8. Resuspend the exosome pellet in either SDS-PAGE sample
buffer or in PBS for further analysis (see Note 4). Figure 3
shows the collection of exosomal GFP-PSAP and soluble Gp80
(Clusterin) in distinct fractions collected during exosome
preparation.
3.5 Gold Labeling of Due to the very small size and the very high affinity, nanobodies are
Exosomal Galectin-3 particularly suitable for immune labeling in electron microscopy.
for Electron GFP-nanobodies (Chromotek) were decorated with
Microscopy NHS-activated gold nanoparticles (Cytodiagnostics) for direct
labeling of galectin-3-eGFP.
supernatant
exosomes
100,000 g
100,000 g
20,000 g
medium
5000 g
500 g
MW
(kDa)
55
gp80
36
eGFP-PSAP
Fig. 3 Purification of exosomes by differential centrifugation. Cell culture super-

natant (medium) was collected from stably eGFP-PSAP expressing MDCK cells.
From each subsequent centrifugation step the supernatant was applied to the
western blot in order to illustrate the enrichment of the exosomal proteins.
Ultimately, what remains is the exosome pellet, which contains the exosomal
component eGFP-PSAP. This consists of the late domain PSAP-motif from
galectin-3 fused to eGFP. Gp80 (Clusterin) a marker protein for exocytosis in
MDCK cells was used as positive control for a predominantly soluble secreted
polypeptide. Antibodies directed against eGFP and Gp80 were used for
immunoblot-detection as indicated
420 €nfer et al.
Sebastian Ba
1. All reagents should be used at room temperature. Due to the

very high stability of nanobodies, this is not a problem.
2. Dilute nanobodies to a final concentration of 0.5 mg/ml. The
supplied protein resuspension buffer can be used for this step.
It is important that it is not a primary amine buffer, such as Tris,
as these groups would otherwise compete with the conjugation
process. Since the hydrolysis of the NHS ester also plays a role
in the coupling efficiency, a pH of 7.2–8.5 should be adjusted.
3. Combine 24 μl of the diluted protein with 84 μl of the supplied
reaction buffer in a microcentrifuge tube.
4. Transfer 90 μl of the protein/reaction buffer mix directly to
one vial containing lyophilized NHS-activated gold nanoparti-
cles (15 nm) and immediately mix well by pipetting up and
down. In this context, it is important that the NHS-activated
gold particles must not be resuspended with buffer before the
reaction in order to avoid rapid hydrolysis in aqueous solution
and thus reduced conjugation efficiency.
5. Incubate at room temperature for 2 h.
6. Add 10 μl of the supplied quencher buffer to stop the reaction.
7. Centrifuge for 30 min at 17,000 g (15 nm gold particles; if
other particle sizes should be used, the centrifugation must be
adapted to the respective size according to the manufacturer’s
instructions).
8. Remove the supernatant, which contains unbound protein.
9. Store at 4 C until use.
10. Prepare either cells or the pellet from exosomal fractionation
according to standard immuno-electron microscopy protocols
(see Subheading 3.6). Antigen retrieval, the appropriate con-
centration and incubation period should be adapted to the
respective experiment.
3.6 Material 1. Fix pellets of the exosomal fraction in 2 ml of an ice-cold

Preparation for mixture of 2.5% glutaraldehyde, 2.5% paraformaldehyde, and
Electron Microscopy 0.05% picric acid in 67 mM cacodylate buffer (pH 7.4) at 4 C
for 45 min.
2. Wash with ice-cold 100 mM cacodylate buffer (pH 7.4).
3. Incubate pellets with 1% osmium tetroxide for 1 h at RT for
postfixation. The osmium tetroxide mixture is prepared from
equal parts of 2% osmium tetroxide and 3% potassium ferro-
cyanate-2-trihydrate in diH2O.
4. Incubate overnight in 0.3% uranyl acetate dissolved in 50 mM
maleate buffer (pH 5) at RT.
5. Dehydration at RT: 50% EtOH 2 10 min/70% EtOH
2 10 min/90% EtOH 2 10 min/100% EtOH
2 10 min, 3 15 min.
6. Incubate in propylene oxide (Serva, 33715) 2 15 min at RT.

7. Incubate for 2 h in a 1:1 mixture of propylene oxide and Epon.
Epon is a mixture of Epon “solution A” and “solution B.”
Solution A consists of 62 ml glycidether 100 (Roth) and
100 ml of dodecenylsuccinic anhydride (Serva), whereas solu-
tion B consists of 100 ml glycidether 100 and 89 ml methylna-
dic anhydride (Serva). The finished mixture is obtained by a
ratio of 40:60 (%) of solution A to B. In addition, 3% 2,4,6-tri
(dimethylaminoethyl)phenol (DMP-30, accelerator; Serva)
should be added.
8. Move samples to embedding mold filled with freshly
mixed Epon.
9. Leave samples to polymerize for 12 h at 60 C.
10. Cut ultrathin sections in an Ultracut FC4E (Reichert-Jung)
ultramicrotome.
11. Contrast thin sections with lead citrate. To do this, incubate
the ultrathin sections for 5–10 min in a drop of 2% uranyl
acetate (Merck).
12. Wash the sections in carefully boiled, CO2-free diH2O for
1 min in a large beaker.
13. In a second step, incubate the ultrathin sections in a drop of
Reynolds lead citrate [30] for 5 min.
14. Wash the sections again as described in step 12 and then let
them dry at RT.
15. Etch thin sections with 10% hydrogen peroxide for 10 min to
unveil eGFP-epitopes at RT.
16. Incubate with gold-labeled GFP-nanobodies for 2 h at RT.
17. Image acquisition on a Zeiss EM 109S electron microscope
equipped with a bottom-mounted-TEM-CCD camera (TRS
Systems).
3.7 Proteinase K A proteinase K sensitivity test can be used to determine whether a

Treatment of Purified protein is inside or outside of a membrane-enclosed vesicular struc-
Exosomes ture. Since Proteinase K is not membrane-permeable, exclusively
accessible proteins on the surface are non-specifically degraded.
Accordingly, intravesicular proteins located in the lumen are pro-
tected from degradation by the membrane of the vesicle. Mem-
brane permeabilization by a detergent act as a positive control to
make all proteins accessible to the protease.
1. Purify exosomes from cell culture medium.
2. Divide the preparation into several aliquots, e.g., 10 μl (see
Note 5).
422 €nfer et al.
Sebastian Ba
+ + + exosomes
MW - + + proteinase K
(kDa) - - + Triton X-100
55
tsg101
36
eGFP-PSAP
Fig. 4 Proteinase protection assay of purified exosomes. Exosomes isolated from

stably eGFP-PSAP expressing MDCK cells were treated or not treated with
proteinase K. The luminal exosome component tsg101 and eGFP-PSAP were
insensitive to the proteinase, indicating that both polypeptides reside in the
exosome lumen. Accessibility to the proteinase was achieved by addition of the
detergent Triton X-100, which disrupts exosomal membranes. Antibodies
against eGFP and tsg101 were used for immunoblot-detection as indicated
3. As a positive control, prepare one aliquot with 1 μl 1% Triton

X-100.
4. Start the reaction by adding 1 μl proteinase K (0.5 mg/ml) and
incubate for 10 min at RT (see Note 6).
5. Stop the reaction by adding 1 μl PMSF (0.4 M) and incubate
for 10 min at RT.
6. Add SDS sample buffer and incubate at 95 C for 10 min to
denature proteins for SDS-PAGE. Figure 4 shows results on
the insensitivity of luminal exosomal proteins to proteinase K
treatment.
4 Notes
1. To analyze the exosomal secretion of galectins, the purification

of exosomes is the most critical step. In this regard, differential
centrifugation is the most widely used method to selectively
isolate exosomes from the cell culture supernatant or other
fluids. However, it should be noted that this method is suscep-
tible to contamination with cellular and protein debris. This
also applies to various commercially available exosome purifica-
tion kits, which are less user-intensive than ultracentrifugation,
but can be contaminated with debris. In contrast, the method
of discontinuous sucrose or iodixanol gradient centrifugation
at 100,000 g is the method of choice in terms of purity, but
also very user-intensive and not suitable for high-throughput
applications [31].
2. Fetal calf serum (FCS) can contain exosomes. To exclude iso-
lation of bovine exosomes from FCS, centrifuge medium at
100,000 g before addition to cell culture [32].
3. Work at 4 C from this step on.

4. Avoid harsh resuspension not to destroy exosomes.
5. An input control should be retained at this stage.
6. Proteinase K-concentration, temperature, and time are critical
factors for this reaction. Adjustment to local lab conditions
could be necessary.
References
1. Rabouille C (2017) Pathways of unconven- endoplasmic reticulum-Golgi complex. Exp
tional protein secretion. Trends Cell Biol Cell Res 207(1):8–18. https://doi.org/10.
27(3):230–240. https://doi.org/10.1016/j. 1006/excr.1993.1157. S0014-4827(83)
tcb.2016.11.007 71157-2 [pii]
2. Leffler H, Carlsson S, Hedlund M, Qian Y, 11. Cleves AE, Cooper DN, Barondes SH, Kelly
Poirier F (2004) Introduction to galectins. RB (1996) A new pathway for protein export in
Glycoconj J 19(7–9):433–440 Saccharomyces cerevisiae. J Cell Biol
3. Lepur A, Salomonsson E, Nilsson UJ, Leffler 133(5):1017–1026. https://doi.org/10.
H (2012) Ligand induced galectin-3 protein 1083/jcb.133.5.1017
self-association. J Biol Chem 12. Hughes RC (1999) Secretion of the galectin
287(26):21751–21756. https://doi.org/10. family of mammalian carbohydrate-binding
1074/jbc.C112.358002. C112.358002 [pii] proteins. Biochim Biophys Acta
4. B€anfer S, Schneider D, Dewes J, Strauss MT, 1473(1):172–185
Freibert S-A, Heimerl T, Maier UG, Els€asser 13. Stalz H, Roth U, Schleuder D, Macht M,
H-P, Jungmann R, Jacob R (2018) Molecular Haebel S, Strupat K, Peter-Katalinic J, Hanisch
mechanism to recruit galectin-3 into multive- FG (2006) The Geodia cydonium galectin
sicular bodies for polarized exosomal secretion. exhibits prototype and chimera-type character-
Proc Natl Acad Sci 115(19):E4396–E4405. istics and a unique sequence polymorphism
https://doi.org/10.1073/pnas.1718921115 within its carbohydrate recognition domain.
5. Fortuna-Costa A, Gomes AM, Kozlowski EO, Glycobiology 16(5):402–414. https://doi.
Stelling MP, Pavão MSG (2014) Extracellular org/10.1093/glycob/cwj086
galectin-3 in tumor progression and metastasis. 14. Muller WE, Conrad J, Schroder C, Zahn RK,
Front Oncol 4:138–138. https://doi.org/10. Kurelec B, Dreesbach K, Uhlenbruck G (1983)
3389/fonc.2014.00138 Characterization of the trimeric, self-
6. Johannes L, Jacob R, Leffler H (2018) Galec- recognizing Geodia cydonium lectin I. Eur J
tins at a glance. J Cell Sci 131(9). https://doi. Biochem 133(2):263–267
org/10.1242/jcs.208884 15. Wagner-Hülsmann C, Bachinski N, Diehl-
7. Delacour D, Jacob R (2006) Apical protein Seifert B, Blumbach B, Steffen R, Pancer Z,
transport. Cell Mol Life Sci Müller WEG (1996) A galectin links the aggre-
63(21):2491–2505 gation factor to cells in the sponge (Geodia
8. Delacour D, Cramm-Behrens CI, Drobecq H, cydonium) system. Glycobiology
Le Bivic A, Naim HY, Jacob R (2006) Require- 6(8):785–793. https://doi.org/10.1093/
ment for galectin-3 in apical protein sorting. glycob/6.8.785-d
Curr Biol 16(4):408–414 16. Fan X, She Y-M, Bagshaw RD, Callahan JW,
9. Lindstedt R, Apodaca G, Barondes SH, Mos- Schachter H, Mahuran DJ (2005) Identifica-
tov KE, Leffler H (1993) Apical secretion of a tion of the hydrophobic glycoproteins of Cae-
cytosolic protein by Madin-Darby canine kid- norhabditis elegans. Glycobiology
ney cells. Evidence for polarized release of an 15(10):952–964. https://doi.org/10.1093/
endogenous lectin by a nonclassical secretory glycob/cwi075
pathway. J Biol Chem 268(16):11750–11757 17. Di Lella S, Sundblad V, Cerliani JP, Guardia
10. Sato S, Burdett I, Hughes RC (1993) Secretion CM, Estrin DA, Vasta GR, Rabinovich GA
of the baby hamster kidney 30-kDa galactose- (2011) When galectins recognize glycans:
binding lectin from polarized and nonpolarized from biochemistry to physiology and back
cells: a pathway independent of the again. Biochemistry 50(37):7842–7857.
https://doi.org/10.1021/bi201121m
424 €nfer et al.
Sebastian Ba
18. Ideo H, Fukushima K, Gengyo-Ando K, 25. Pornillos O, Alam SL, Rich RL, Myszka DG,
Mitani S, Dejima K, Nomura K, Yamashita K Davis DR, Sundquist WI (2002) Structure and
(2009) A Caenorhabditis elegans glycolipid- functional interactions of the Tsg101 UEV
binding galectin functions in host defense domain. EMBO J 21(10):2397–2406.
against bacterial infection. J Biol Chem https://doi.org/10.1093/emboj/21.10.
284(39):26493–26501. https://doi.org/10. 2397
1074/jbc.M109.038257 26. Richardson JC, Scalera V, Simmons NL (1981)
19. Rabouille C, Malhotra V, Nickel W (2012) Identification of two strains of MDCK cells
Diversity in unconventional protein secretion. which resemble separate nephron tubule seg-
J Cell Sci 125(22):5251–5255. https://doi. ments. Biochim Biophys Acta 673(1):26–36
org/10.1242/jcs.103630 27. Delacour D, Greb C, Koch A, Salomonsson E,
20. Nickel W, Rabouille C (2009) Mechanisms of Leffler H, Le Bivic A, Jacob R (2007) Apical
regulated unconventional protein secretion. sorting by galectin-3-dependent glycoprotein
Nat Rev Mol Cell Biol 10(2):148–155. clustering. Traffic 8(4):379–388
https://doi.org/10.1038/nrm2617 28. Savina A, Furlan M, Vidal M, Colombo MI
21. Madrigal-Matute J, Lindholt JS, Fernandez- (2003) Exosome release is regulated by a
Garcia CE, Benito-Martin A, Burillo E, calcium-dependent mechanism in K562 cells.
Zalba G, Beloqui O, Llamas-Granda P, J Biol Chem 278(22):20083–20090. https://
Ortiz A, Egido J, Blanco-Colio LM, Martin- doi.org/10.1074/jbc.M301642200
Ventura JL (2014) Galectin-3, a biomarker 29. Raposo G, Nijman HW, Stoorvogel W,
linking oxidative stress and inflammation with Liejendekker R, Harding CV, Melief CJ,
the clinical outcomes of patients with athero- Geuze HJ (1996) B lymphocytes secrete
thrombosis. J Am Heart Assoc 3(4):e000785. antigen-presenting vesicles. J Exp Med
https://doi.org/10.1161/jaha.114.000785 183(3):1161–1172. https://doi.org/10.
22. Gonzales PA, Pisitkun T, Hoffert JD, 1084/jem.183.3.1161
Tchapyjnikov D, Star RA, Kleta R, Wang NS, 30. Reynolds ES (1963) The use of lead citrate at
Knepper MA (2009) Large-scale proteomics high pH as an electron-opaque stain in electron
and phosphoproteomics of urinary exosomes. microscopy. J Cell Biol 17:208–212. https://
J Am Soc Nephrol 20(2):363–379. https:// doi.org/10.1083/jcb.17.1.208
doi.org/10.1681/ASN.2008040406 31. Kalra H, Adda CG, Liem M, Ang C-S,
23. Liang B, Peng P, Chen S, Li L, Zhang M, Mechler A, Simpson RJ, Hulett MD, Mathiva-
Cao D, Yang J, Li H, Gui T, Li X, Shen K nan S (2013) Comparative proteomics evalua-
(2013) Characterization and proteomic analy- tion of plasma exosome isolation techniques
sis of ovarian cancer-derived exosomes. J Pro- and assessment of the stability of exosomes in
teome 80:171–182. https://doi.org/10. normal human blood plasma. Proteomics
1016/j.jprot.2012.12.029 13(22):3354–3364. https://doi.org/10.
24. Welton JL, Khanna S, Giles PJ, Brennan P, 1002/pmic.201300282
Brewis IA, Staffurth J, Mason MD, Clayton A 32. Théry C, Amigorena S, Raposo G, Clayton A
(2010) Proteomics analysis of bladder cancer (2006) Isolation and characterization of exo-
exosomes. Mol Cell Proteomics somes from cell culture supernatants and
9(6):1324–1338. https://doi.org/10.1074/ biological fluids. Curr Protoc Cell Biol
mcp.M000063-MCP201 Chapter 3:Unit 3.22. https://doi.org/10.
1002/0471143030.cb0322s30
Chapter 23
Manipulating Galectin Expression in Zebrafish (Danio rerio)

Chiguang Feng, Mihai Nita-Lazar, Nuria González-Montalbán,
Jingyu Wang, Justin Mancini, Sheng Wang, Chinnarajan Ravindran,
Hafiz Ahmed, and Gerardo R. Vasta
Abstract
Techniques for disrupting gene expression are invaluable tools for the analysis of the biological role of a
gene product. Because of its genetic tractability and multiple advantages over conventional mammalian
models, the zebrafish (Danio rerio) is recognized as a powerful system for gaining new insight into diverse
aspects of human health and disease. Among the multiple mammalian gene families for which the zebrafish
has shown promise as an invaluable model for functional studies, the galectins have attracted great interest
due to their participation in early development, regulation of immune homeostasis, and recognition of
microbial pathogens. Galectins are β-galactosyl-binding lectins with a characteristic sequence motif in their
carbohydrate recognition domains (CRDs), that constitute an evolutionary conserved family ubiquitous in
eukaryotic taxa. Galectins are emerging as key players in the modulation of many important pathological
processes, which include acute and chronic inflammatory diseases, autoimmunity and cancer, thus making
them potential molecular targets for innovative drug discovery. Here, we provide a review of the current
methods available for the manipulation of gene expression in the zebrafish, with a focus on gene knock-
down [morpholino (MO)-derived antisense oligonucleotides] and knockout (CRISPR-Cas) technologies.
Key words Galectins, Zebrafish, Morpholino, CRISPR-Cas, Gene expression, Microinjection
1 Introduction
The zebrafish (Danio rerio) is a widely used model organism to

study a broad spectrum of human normal and pathological pro-
cesses, including development, autoimmune, neoplastic, and infec-
tious disease [1–7]. The growing interest in using zebrafish as a
genetically tractable model system is due to its multiple advantages
relative to the mammalian models, which include high fecundity,
external fertilization, rapid development, transparent embryo, low
maintenance cost, easy genetic manipulation, and an extensive
collection of mutants currently available [8–10]. Zebrafish and
Sheng Wang is a visiting student and Chinnarajan Ravindran is a visiting scientist.
425
426 Chiguang Feng et al.
human genomes have been shown to share highly conserved struc-

tural and functional features [11]. Hundreds of zebrafish genes
have been identified [12] and its full genome sequence is now
available online (http://www.ncbi.nlm.nih.gov/genome/guide/
zebrafish/). Moreover, D. rerio exhibits some higher-level beha-
viors previously observed only in mammals, such as memory,
conditioned responses, and schooling [13–17]. In recent years,
the zebrafish model has proven useful to gain new insight into the
functional aspects of protein-carbohydrate interactions, such as
those mediated by galectins [18–22]. Galectins are β-galactosyl-
binding lectins, which share primary structural homology in their
carbohydrate recognition domains (CRDs) [23–25]. They are clas-
sified into three major structural types: (1) proto-type galectins,
which contain one CRD and form homodimers; (2) chimera-type
galectins, which have a single CRD and can oligomerize forming
trimers and pentamers; (3) tandem-repeat-type galectins, which are
comprised of two CRDs joined by a linker peptide [26]. Galectins
participate in a multitude of biological processes, such as develop-
ment, apoptosis, tumor metastasis, and regulation of immune
responses [22, 25, 27–35]. All three structural types of galectins
have been identified and characterized in various tissues, plasma,
and mucus of teleost fish, such as zebrafish [20, 22, 36–
39]. Among the various methodologies for elucidating the
biological role(s) of a particular protein of interest, disruption of
gene expression represents a useful approach. Multiple strategies
have been developed to modulate gene expression at a genetic or
epigenetic level (Table 1). Several of these methodologies have
been recently applied to functional studies of galectins [18, 42,
45, 54–56].
Morpholinos (MOs) are antisense oligonucleotides derivatized
with morpholine rings to increase stability, and designed to anneal
close to the start codon of the selected gene and disrupt its transla-
tion, or to the splice acceptor sequence to induce incorrectly spliced
mRNA [41]. The MOs do not degrade their mRNA targets, but
(1) block mRNA translation by targeting the 50 -UTR through the
first 25 bases of coding sequence [18], or (2) alter the translation
modifying pre-mRNA processing by targeting splice junctions or
regulatory sites (Fig. 1) [57, 58]. Thus, the “morphant” phenotype
results from disrupted protein expression levels. Although their
effects are only transient, MOs are relatively long-lived inside the
cell upon delivery, with the effect on pre-mRNA splicing or trans-
lation lasting for up to 5 days following microinjection. This tech-
nique allows for the rapid manipulation and interrogation of
complex processes such as embryonic development, organ forma-
tion, innate immunity, and host-pathogen interactions [3, 18,
58]. However, since the effects of mRNA suppression by MO
oligos are temporary and not inheritable, and there is no suitable
way to deliver them systemically in adult fish, this system is not
Manipulating Galectin Expression in Zebrafish (Danio rerio) 427
Table 1
Reverse genetic tools and transgenic methodologies for disruption of gene expression
Procedure Mechanism References

Morpholinos Custom-designed MO-oligonucleotides block Summerton
Photomorpholinos translation or pre-mRNA processing of the [40]
targeted gene. Nasevicius
et al. [41]
Tallafuss
et al. [42]
TILLING Formation of DNA hetero-duplexes between multiple McCallum
(Targeting Induced Local alleles amplified by PCR and cleaved by single et al. [43]
Lesions in Genomes) stranded nucleases. Mutagenized products are Wienholds
screened by size. et al. [44]
Heat shock promoters Targeted gene is engineered to be under inducible Halloran
control of a heat shock promoter. Transgene et al. [45]
expression induced by heat shock. Hardy et al.
[46]
Targeted cell ablation Expression of bacterial nitro-reductase (NTR) under a Curado
promoter that controls the proliferation and survival et al. [47]
of a particular cell type.
ZNFs Engineered DNA-binding endonucleases to break in the Ekker [48]
(Zinc-Finger Nucleases) targeted double-stranded DNA for specific site
mutagenesis.
Gal4/UAS Targeted gene is engineered to be under inducible Scheer et al.
control of UAS promoter. Transgene expression [49]
induced by tissue-specific Gal4/UAS system. Scott [50]
TALEN Engineered TAL effector DNA domain fused to a DNA Clark et al.
(Transcription Activator-Like restriction site binds and cleaves targeted gene, [51]
Effector Nuclease) introducing a custom-designed mutagenesis site.
CRISPR-Cas Engineered CRISPR system with custom-designed Jinek et al.
(Clustered Regularly Cas9 nuclease directed by “guide RNA” to introduce [52]
Interspaced Short single break in the targeted double-stranded DNA Hwang et al.
Palindromic Repeats) sequence. [53]
suitable for more integrated approaches. For inheritable genetic

modulation, a set of other techniques has been introduced in the
last 15 years (Table 1).
CRISPR-Cas is a genome editing approach based on the pro-
karyotic immune system. By using a segment of virus-derived DNA
from its CRISPR array and processing to crRNA targeting the viral
genome, the system leads to the inactivation or degradation of
targeted DNA by the CAS-crRNA complex [52]. Cas9 nuclease
transgenically expressed in vertebrates is active and able to cleave
target DNA when directed by a short guide RNA (sgRNA), which
at its 50 -end contains 20 base pairs of complementary target DNA
[53, 59, 60]. This technology has recently been expanded rapidly
Fig. 1 Genetic modulation of galectins by morpholinos and CRISPR-Cas microinjection. MO-oligos may be
designed to target an exonic sequence, thus blocking mRNA translation. Alternatively, MO oligos may be
designed to target an exon-intron junction, preventing the maturation of the targeted pre-mRNAs. MO-oligos
and CRISPR transcripts are microinjected into the embryo yolk. In the case of morpholinos (orange arrows), the
microinjection will result in a mutant phenotype either/both in embryo and/or adult. When injecting CRISPR
transcripts, the screening of the microinjected embryos will result in the selection of the mutant genotypes
(usually heterozygous). After crossing individuals with mutant genotypes, fish with mutant phenotypes
(homozygous) will be selected from the resultant progeny
from human and mouse to other mammals [61], poultry [62],

invertebrates [63] and plants [64], as well as zebrafish [65–67].
Here we describe in detail the use of MOs and CRISPR-Cas to
modulate galectin expression in the zebrafish model to elucidate
their functions in development and immunity. The MOs Drgal1-
L1-MO and Drgal1-L2-MO were designed to block translation
initiation based on 50 -UTR sequences of Drgal1-L1 and Drgal1-
L2, respectively (Table 2). MOs are usually validated by in vitro
blockade of the corresponding protein expression using the TNT
SP6 Coupled Rabbit Reticulocyte Lysate System. Following
MO-oligo injection of zebrafish embryos, expression of the
corresponding protein is inspected by whole mount antibody stain-
ing. Potential phenotypes are inspected under microscope and by
whole mount antibody staining of a specific marker. Zebrafish
embryos injected with validated Drgal1-L2-MO (Fig. 2a) show
dramatically reduced Drgal1-L2 expression as observed by whole
mount antibody staining (Fig. 2c). Drgal1-L2 is strongly expressed
in the notochord during early embryogenesis and Drgal1-L2
knockdown results in a characteristic phenotype with a short and
bent tail (Fig. 2c). Microscopically, the phenotype exhibits dis-
rupted muscle fiber organization as observed by whole mount
immunostaining with the F59 antibody (monoclonal anti-myosin
antibody), a marker for slow muscle (Fig. 2e). On the other hand,
the CRISPR-Cas approach induces efficient non-homologous end
joining-mediated small insertions or deletions (indels) in genomic
DNA [53, 59, 60], which can be verified with Sanger sequencing or
resolvase digestion. Compared to genomic DNA extracted from
wild-type embryos, which display a uniform chromatogram, geno-
mic DNA extracted from treated embryos showed a distorted
profile around the gRNA targeting site (Fig. 3a), corresponding
to the superposition of the different alleles with mutated sites. In
addition, reannealing of the PCR amplicons covering the gRNA
targeted site form the mismatched double strand DNA, can be
cleaved by Guide-it Resolvase and visualized in agarose electropho-
resis (Fig. 3b).
Table 2
Verified morpholino-modified antisense oligonucleotides
Designation Sequence Reference

0 0
DrGal1-L1-MO 5 - TCTATAAGCACAGTCTCATGCA -3 Ahmed et al. [18]
0 0
DrGal1-L2-MO 5 - ATAAGCACACCGGCCATTTTGACGT -3
Fig. 2 Validation of Drgal1-L2-MO. (a) In vitro blocking of Drgal1-L2 protein in a rabbit reticulocyte system.
SDS-PAGE/autoradiography analysis of the in vitro labeled Drgal1-L2 translation products. Lane 1: translation
reaction in the absence of Drgal1-L2-MO (positive control); lanes 2 and 3: reactions in the presence of 170 ng
and 340 ng of Drgal1-L2-MO, respectively. (b) Whole mount immunostaining of an uninjected embryo probed
with anti-Drgal1-L2 antibodies (positive control). Lateral and dorsal view of a 27 hours post-fertilization (hpf)
embryo. Drgal1-L2 was seen expressed in the notochord. (c) In vivo blocking of Drgal1-L2 protein as shown by
whole mount immunostaining of an embryo injected with Drgal1-L2-MO and probed with anti-Drgal1-L2
antibodies. Lateral and dorsal view of a 27 hpf embryo. (d and e) Whole mount immunostaining of an
uninjected embryo (d) and an embryo injected with Drgal1-L2-MO (e) probed with F59 antibody. Lateral and
dorsal view of 48 hpf embryos. In e, muscle fibers were found disorganized
2 Materials
2.1 Preparation and 1. MO oligos (Gene tools) (see Note 1).

Validation of 2. pCS2+ vector (Addgene).
Morpholino (MO)
3. pCS-Drgal1-L2 plasmid DNA (prepared in Vasta’s lab [18]).
Oligos
4. Drgal1-L2-MO (custom-synthesized by Gene Tools LLC).
Fig. 3 Induction of genomic mutation by Cas9/gRNA. 1–2 cell stage embryos of wild-type Tübingen zebrafish
(Tu) were microinjected with Cas9/gRNA complex (gRNA1, gRNA2) targeting zebrafish GRIFIN (DrGRIFIN,
Galectin-related interfiber protein). Genomic DNA was extracted from 5 to 10 embryos at 48 hpf from each
group or normal embryos (Tu), and PCR was performed with primer covering the targeted sites. (a) The PCR
products were purified, and sequences were determined. Chromographs from normal embryos (Tu) and Cas9/
gRNA2 are shown, with the gRNA targeted site highlighted (in blue). (b) PCR product was subjected for
resolvase digestion (Treated) according to the Guide-it Mutation Detection Kit and analyzed in agarose
electrophoresis. Untreated samples (Untreated), and positive control (PC) from the kit are included. Digested
PCR products suggest successful mutation in the treated fish
5. TNT SP6 Coupled Rabbit Reticulocyte Lysate System

(Promega).
6. [35S] methionine (PerkinElmer).
7. 15% polyacrylamide gel.
2.2 Preparation 1. Single guide RNA (sgRNA) expression vector pDR274

of sgRNA (Addgene) (see Note 2).
2. Cas9 nuclease expression plasmid pMLM3613 (Addgene).
3. Gene specific oligos (Sigma) (see Notes 3 and 4).
4. pDR274 (Addgene).
5. 1% Agarose gel.
6. LB agar.
7. 42 C water bath.
8. Kanamycin.
9. Sterile 30% glycerol.
10. QIAprep Spin Miniprep Kit/QIAfilter Plasmid Midi Kit
(Qiagen).
11. MinElute Gel Extraction Kit (Qiagen).
12. Rapid DNA Ligation kit (Roche).
13. DH5α Competent cells (Invitrogen).
14. SOC medium (Sigma).
15. Restriction enzymes: BsaI, PmeI, DraI (New England
Biolabs).
16. MAXIscript T7 kit.
2.3 Embryo 1. Morpholino oligo (Drgal-1-L2, etc. See Subheading 3.1).

Preparation 2. RNase-free dH2O.
3. Danieau’s solution: 58 mM NaCl, 0.7 mM KCl, 0.4 mM
MgSO4, 0.6 mM Ca(NO3)2, 5.0 mM HEPES, pH 7.6.
4. 1% Phenol red dye (Sigma).
5. Glass microinjection needles (Tritech Research) (see Notes 5
and 6).
6. Borosilicate glass capillary tubes (Narishige).
7. Micropipette puller (Sutter Instrument CO., model P-97).
8. Agarose (Sigma).
9. 90 mm Petri dishes.
10. Adult male and female zebrafish (Purchased from Zebrafish
International Resource Center, maintained and bred in Aqua-
culture Research Center of IMET).
11. Crossing Tank Kit (Thoren Aquatics Housing Systems).
12. Tank divider.
13. Disposable plastic transfer pipette (RPI Corp).
2.4 Embryo 1. Dissecting forceps.

Microinjections 2. Micropipettes and disposable tips.
3. PBS (Bio-Rad).
4. Incubator at 28 C.
5. 15 ml and 50 ml Conical tube (BD Biosciences).
6. Eppendorf tube (Thermo Scientific).
7. ECL Plus detection kit (Amersham Biosciences).
8. Polyclonal antibodies to Drgal1, 3 and 9 were custom-
produced (Thermo Scientific).
2.5 Whole Mount 1. 4% paraformaldehyde in PBS (ThermoFisher Scientific).

Antibody Staining 2. PBS-Tween: Phosphate buffered saline with 0.1% Tween 20.
3. 0.1% BSA/1% dimethyl sulfoxide/phosphate buffered saline
(BDP).
4. Avidin (Vector Laboratories).
5. 10% goat serum (Invitrogen) in BDP.
6. Anti-Drgal1-L2 antibody (custom-produced by ThermoFisher
Scientific).
7. Biotin-labeled goat anti-rabbit IgG (Vector Laboratories).
8. 1:1 diluted avidin-biotin complex solution (Vector
Laboratories).
9. DAB substrate (Vector Laboratories).
10. Glass vials with closed-top cap (ThermoFisher Scientific).
2.6 Rescuing the 1. 4% paraformaldehyde in PBS (ThermoFisher Scientific).

Phenotypes by Co- 2. In vitro transcription kit (mMESSAGE mMACHINE SP6).
injection of Embryos
3. Formaldehyde 1.5% agarose gels (see Note 7).
with Drgal1-L2 mRNA
and Drgal1-L2-MO 4. Distilled water.
(Ectopic Expression of 5. Micropipettes and disposable tips.
Native Drgal1-L2 on
Drgal1-L2-MO Injected
Embryos)
2.7 Verification of 1. DNeasy Blood & Tissue Kit (Qiagen).

Induced Indels 2. DreamTaq Green PCR Master Mix (2) (Thermo Scientific).
3. MinElute PCR Purification Kit (Qiagen).
4. Guide-it Mutation Detection Kit (TakaraBio).
2.8 Special 1. Microinjection system: Pico-injector PLI-100 (Harvard Medi-

Instruments cal Systems Research Products). It is connected to compress
nitrogen gas and to adjust the air pressure for reagent delivery.
Modulator M152 (Narishige).
2. Sanger sequencer: ABI PRISM 3100xl Genetic Analyzer
(Applied Biosystems).
2.9 General 1. Microscope Stemi 2000 (ZEISS).

Instruments 2. Microcentrifuge (Beckman Coulter).
3. Incubator at 28 C (Fisher Scientific).
4. MyCycler Thermal Cycler (Bio-Rad).
3 Methods
3.1 Preparation and 1. Synthesis of MO oligos: (see Notes 1 and 8).

Validation of 2. Prepare an expression construct for a particular gene for in vitro
Morpholino (MO) protein expression (see Note 9).
Oligos
3. Perform in vitro direct translation of Drgal1-L2 from
pCS-Drgal1-L2 plasmid DNA (0.5 μg) in the presence or
absence of the Drgal1-L2-MO (170 ng and 340 ng) using
TNT SP6 Coupled Rabbit Reticulocyte Lysate System accord-
ing to manufacturer’s instructions.
4. To detect the translated product, use [35S] methionine with the
methionine free amino acid mixture.
5. After completion of the reaction, analyze the translated prod-
uct on 15% SDS-PAGE followed by autoradiography. The
translated protein was transferred to the PVDF membrane
and the radioactivity was captured into an X-ray-film. As shown

in Fig. 2a, Drgal1-L2-MO blocked translation of Drgal1-L2
protein.
3.2 Transformation 1. Digest the pDR274 with BsaI for 1 h at 37 C by following

and Preparation manufacturer’s protocol. The linearized fragment is separated
of sgRNA in a 1% agarose gel and recovered using Gel Extraction kit.
2. Mix the 50 ng of linearized vector with annealed targeted
sequences (1:3 molar ratio) in 20 μl of ligation solution from
Quick Ligation kit and incubate for 15 min at room tempera-
ture. The ligates can be used immediately or stored at 20 C
for future use.
3. Add 2 μl of the above ligate in 50 μl of DH5α competent cells
and incubate on ice for 20 min.
4. Heat-shock the mixture for 45 s in 42 C water bath, and
incubate on ice for at least 2 min.
5. Afterward, add 500 μl of SOC medium and incubate the
bacteria without shaking for 1 h at 37 C.
6. Spread the bacteria onto LB agar with 50 μg/ml of kanamycin
for overnight incubation at 37 C.
7. Inoculate each colony into one 15 ml tube with 3 ml of LB
broth supplemented with 50 μg/ml of kanamycin for over-
night incubation at 37 C with shaking at 220 rpm.
8. Extract plasmid using Mini-prep kit for sequence analysis from
2 ml of the culture bacteria.
9. Stock colonies in 30% glycerol with the remaining culture
bacteria for future use.
10. Select the colony with expected sequence for Midi-prep.
11. Prepare sgRNA using DraI-digested sgRNA expression vector
and the MAXIscript T7 kit as manufacturer’s recommendation;
prepare Cas9 mRNA using PmeI-digested Cas9 expression
vector and mMESSAGE mMACHINE T7 ULTRA kit by
following manufacturer’s suggested protocol.
12. Purify both RNA by LiCl precipitation by following manufac-
turer’s suggested protocol and use for microinjection.
3.3 Embryo 1. (a) Prepare morpholino oligo (MO) stocks: Aliquots of 5 mM

Preparation MO stock solution are prepared by adding RNAse-free dH2O
(or Danieau’s solution) and stored at 20 C. Before use,
mixed with Danieau’s solution and 1% phenol red dye (final
0.1%) to make 0.5–2 mM working solution (see Note 10).
Keep working solution at room temperature.
(b) Prepare Cas9/sgRNA working solution: Incubate

50 pg of sgRNA at 37 C for 30 min and mix with 300 pg of
Cas9 RNA and 1 μl of 1% phenol red dye to a final 10 μl of
working solution.
2. Prepare agarose injection bed: Prepare 1% agarose in water into
90 mm Petri dish, lay carefully several glass micropipettes par-
allel onto agarose surface to make the troughs. After cooling
down, discard micropipettes and the injection bed is ready for
use or store at 4 C.
3. The day before spawning, separate adult male and female zeb-
rafish into spawning tanks with divider. Set up two to four pairs
of animals for each experiment.
4. In the morning of the day for injection, remove the divider to
allow mating. Usually, the females will begin to lay eggs within
15–45 min.
5. Within 30 min of spawning, collect embryos into a fish tray,
wash thoroughly several times with fresh water.
6. Use a disposable plastic transfer pipette (RPI Corp) to transfer
the embryos onto the injection bed. Carefully arrange the eggs
on the troughs of injection bed.
7. Quick spin the working solution [e.g., morpholino solution for
30 s] at room temperature and use the supernatant for
microinjection.
3.4 Embryo 1. Set up the microinjector: Attach the microinjection needles to

Microinjections the needle holder of the microinjector (see Notes 5 and 6).
2. Place the injection bed on the stage and adjust the modulator
to appropriated position for injection.
3. Load the microinjection needle with 2 μl of working solution
on it. (Optional: Test injection in a few embryos. Adjust pres-
sure values to 90–95 psi if needed) (see Note 11).
4. Penetrate the needle to the yolk and deliver 1 nl of working
solution. The flow is visible due to the phenol red added to the
working solution (see Note 12).
5. Retract the pipette and repeat on remaining embryos.
6. After microinjection, wash the embryos into a Petri dish with
fresh water and place them in incubator at 28 C.
7. Evaluate the injected embryos under the microscope a few
hours after injections and remove the damaged embryos. If
the water in the dish is cloudy, transfer the embryos to a new
dish with fresh water.
8. After 24 h, dechorionate the embryos under microscope using
forceps if needed. Under the microscope (2–5), use fine
forceps in both hands to grasp the envelope, carefully tear it
open to release the embryos.
9. Visually inspect embryos for any potential phenotypes, such as

delayed developmental stage, bended tails. Further investiga-
tion to verify the effect of MO injection on protein expression
is done by Western blot and whole mount antibody staining
(see Subheading 3.5). Further investigation to verify the geno-
mic indel mutation of Cas9/sgRNA injection is done by either
Sanger sequencing or resolvase digestion, respectively (see
Subheading 3.7).
10. Following Drgal1-L2-MO injection, we performed whole
mount immunostaining with Drgal1-L2 antibody (1:1000)
(to validate the Drgal1-L2-MO by in vivo blocking of
Drgal1-L2 protein expression) and F59 antibody (1:15)
(to observe potential defects in muscle fiber organization).
11. Once we confirmed the phenotype, we then rescued the defect
by injection of embryos with 100 μg/ml of Drgal1-L2 mRNA
(for ectopic expression of Drgal1-L2, 100–200 pg per embryo)
along with Drgal1-L2-MO (see Subheading 3.6).
3.5 Whole Mount 1. Fix embryos with 4% paraformaldehyde for 1 h at RT,

Antibody Staining 10–15 embryos per ml in a glass vial with closed-top cap
(ThermoFisher Scientific).
2. Wash twice with 5 ml of PBS-Tween for 5 min each.
3. Replace with cold acetone and incubate for 10 min at 20 C
freezer.
4. Wash embryos twice with 5 ml of PBS-Tween for 5 min. Each
followed by washing once with 0.1% BSA/1% dimethyl sulfox-
ide/PBS (BDP) for 5 min.
5. Incubate with 1 ml of avidin (Vector) (4 drops/ml) in blocking
buffer (10% of goat serum in BDP) for 30 min at RT.
6. Wash the embryos twice with 5 ml of BDP for 5 min each and
incubate with 1 ml of diluted (1:1000) anti-Drgal1-L2 anti-
body in blocking buffer containing biotin (Vector) (4 drops/
ml) overnight at 4 C.
7. Wash the embryos three times for 30 min with 5 ml of BDP,
followed by incubation with 1 ml of diluted (1:1000) biotin-
labeled secondary antibody (goat anti-rabbit IgG) (Vector) in
BDP for 1 h at RT.
8. Wash the embryos three times for 30 min with 5 ml of BDP and
incubate with 1:1 diluted avidin-biotin complex solution (Vec-
tor) for 30 min at RT.
9. Wash the embryos three times for 30 min in 5 ml of BDP and
develop color with DAB substrate according to the manufac-
turer’s protocol.
3.6 Rescuing the 1. To determine if the ectopic expression of native Drgal1-L2

Phenotypes by Co- rescues the phenotypes observed, Drgal1-L2 mRNAs were
injection of Embryos synthesized from a pCS-Drgal1-L2 construct using an in vitro
with Drgal1-L2 mRNA transcription kit by following manufacturer’s guidelines (see
and Drgal1-L2-MO Note 13).
(Ectopic Expression of 2. The integrity of the transcribed RNA was examined on formal-
Native Drgal1-L2 on dehyde 1.5% agarose gels, with the sharp, clear 28S and 18S
Drgal1-L2-MO Injected rRNA bands.
Embryos) 3. For microinjection, mRNA was dissolved in distilled water to a
final concentration of 100 μg/ml. The transcribed RNA solu-
tion (approximately 2 nl) was microinjected into the cytoplasm
of zebrafish embryos at the one- or two-cell stage, and subse-
quently, the Drgal1-L2-MO was microinjected into the
yolk sac.
4. After 24 h, visually inspect embryos for any phenotypes, such as
delay development, bended tails, or mortality, and perform
whole mount immunostaining with Drgal1-L2 and F59 anti-
bodies as described before.
3.7 Verifying the 1. Extract genomic DNA from 5–10 of 24–48 hpf (hours post-
Induction of Genomic fertilization) embryos, using the DNeasy Blood & Tissue Kit
Indels by following manufacturer’s guidelines.
2. Perform PCR with primer pairs covering the gRNA targeted
sites (see Note 14).
3. Purify the PCR amplicons with MinElute PCR Purification Kit
(Qiagen).
4. Analyze the purified amplicons with Sanger sequencing (BioA-
nalytical Services Laboratory, IMET).
5. Transfer 10 μl of the PCR product and 5 μl of PCR-grade water
in a fresh PCR tube. Add 15 μl of positive control DNA from
the Guide-it Mutation Detection Kit to a fresh PCR tube.
Prepare duplicates for each sample.
6. Perform DNA hybridization with heating at 95 C for 5 min,
cooling from 95 C to 85 C at the rate of 2 C per second,
followed by cooling from 85 C to 25 C at the rate of 0.1 C
per second, and finally cooling to 4 C.
7. Add 1 μl of Guide-it Resolvase into one set of the samples, 1 μl
of water into another set.
9. Run entire reaction on a 1.5–2% agarose gel.
10. Verify the digestion of positive control DNA and the experi-
mental samples. The positive control containing the Resolvase
will generate bands of 470 bp, 290 bp, and 180 bp. The
control without Resolvase will generate a single band of
470 bp.
4 Notes
1. GeneTools, LLC (Philomath, OR; https://www.gene-tools.

com/oligo-design-form) provides a free design service that
does not require prior knowledge of the start codon of the
mRNA target. Despite being very user-friendly and not being
held to a mandatory order online from the same site, this
software provides very limited sequence design and analysis
option to the user. In that regard, Vector NTI® software (Life
Technologies: Grand Island, NY) offers an excellent alternative
to MO oligo design provided the user has previous knowledge
of the target sequence. Additionally, MOrpholino DataBase
(http://www.morpholinodatabase.org/) is a public
web-based database with more than 700 morpholinos to date
against zebrafish genomic sequences.
2. Selected 20 variable nucleotides to base pair with a target
genomic DNA sequence will be inserted at its 50 end of
pDR274 to produce a short 102 nucleotides guide RNA (see
Subheading 3.2).
3. One pair of oligos for each target genes are designed using
ZiFiT Targeter. The ZiFiT Targeter website (http://zifit.
partners.org/ZiFiT/) offers an online option to identify
potential target sites for the described CRISPR-Cas system.
By default, the sequences that meet the following criteria:
50 -GG- (N)18-NGG-30 will be identified. ZiFiT Targeter will
analyze the user-input sequences and returns a list of recom-
mended target sites and sequences of oligonucleotides that
need to be synthesized for cloning into the pDR274 vector.
4. The single strand oligos are synthesized, mixed at 95 C for
10 min, then cooled down gradually at room temperature to
allow annealing. The annealed oligos contain unique overhangs
on each site for directional cloning into BsaI-digested pDR274
(see Subheading 3.2).
5. Break off the tip of the microinjection needle with your forceps
under the microscope, so that they have an open tip.
6. Obtained the microinjection needles from commercial ven-
dors, such as Tritech Research Inc. Alternatively, prepare the
microinjection needles by heating at a Bunsen burner and pull-
ing borosilicate glass capillary tubes in a micropipette puller.
Store the needles in a Petri dish on top of small amount of clay
or adhesive tape.
7. Melt 1.5 g of agarose in 10 ml of 10 MOPs buffer and 72 ml
of water in microwave. Let it cool in 65 C water bath. Then in
fume hood, add 18 ml of 37% formaldehyde, mix well, and cast
the gel.
8. MOs for each gene are designed based on the gene sequence to
obtain translational blocker (designed to bind close to start
codon disrupting the translation) or splice blocker (designed to
bind to splice acceptor sequence to induce incorrectly spliced
mRNA). For example, Drgal1-L1-MO and Drgal1-L2-MO
shown in Table 2 are translational blockers of the Drgal1-L1
and Drgal1-2, respectively. MOs were custom synthesized by
Gene Tools (www.gene-tools.com).
9. For example, we cloned Drgal1-L2 with 50 -UTR sequences
into a pCS2+ vector (a gift from D. Turner, R. Rupp, J. Lee,
and H. Weintraub, Fred Hutchinson Cancer Research Center,
Seattle, WA) to obtain the pCS-Drgal1-L2 construct. In this
construct, a 27-nucleotide untranslated 50 leader (derived from
the Xenopus β-globin mRNA 50 -end) is introduced between the
SP6 promoter and the Drgal1-L2 insert. The pCS-Drgal1-L2
construct is expected to generate protein when added to a cell-
free protein synthesis system that is initiated by SP6 RNA
polymerase.
10. The first step in using a new MO is to determine the optimum
delivery dose. Thus, MOs can be initially injected at different
doses, such as 0.5–1 mM, and the dosages are increased or
decreased to optimize the phenotype to toxicity ratio. High
concentration of morpholino may cause non-specific toxicity,
such as increased mortality rate after microinjection. Working
stocks of MOs were prepared to use as near-isotonic solutions
for the zebrafish (i.e., Danieau’s solution).
11. Alternatively, “freehand” injection can be practiced without a
micromanipulator in a normal Petri dish. It is more robust but
time-consuming as proper embryo orientation and microinjec-
tion technique is required.
12. The phenol red color from the injected sample will not change
if sample delivered into the embryonic cells. A shift to pink in
phenol red color will indicate the microinjection was delivered
outside the embryo.
13. In this construct, 27 nucleotides derived from the Xenopus
β-globin 50 -UTR were used to replace the Drgal1-L2 50 -UTR
as the Drgal1-L2-MO was specifically targeted to the Drgal1-
L2 50 -UTR. Thus, the Drgal1-L2-MO would only inhibit
expression of the endogenous Drgal1-L2, but not of the
injected Drgal1-L2 mRNA.
14. PCR conditions may vary. We used 35 cycles of 95 C 30 s,
60 C 30 s, and 72 C 30 s, after 3 min of denaturing at 95 C,
followed by 5 min of elongation at 72 C and cool down to
4 C. Forward primer: ACC GCG ATG ACA GAG TAG CA;
reverse primer: TCT TCA GGT CTT CCA CAC GG. Each
20 μl of PCR reaction contains 10 μl of DreamTaq Green PCR
Master Mix (2), 0.5 μM of each primer, and 1 μl of
genomic DNA.
Acknowledgments
Experimental work described here was supported by grant

5R01GM070589-06 from the National Institutes of Health to G.
R.V.
References
1. Meeker ND, Trede NS (2008) Immunology Eeden FJ, Jiang YJ, Heisenberg CP, Kelsh RN,
and zebrafish: spawning new models of Furutani-Seiki M, Vogelsang E, Beuchle D,
human disease. Dev Comp Immunol Schach U, Fabian C, Nusslein-Volhard C
32(7):745–757. https://doi.org/10.1016/j. (1996) The identification of genes with unique
dci.2007.11.011 and essential functions in the development of
2. Feitsma H, Cuppen E (2008) Zebrafish as a the zebrafish, Danio rerio. Development 123:
cancer model. Mol Cancer Res 6(5):685–694. 1–36
https://doi.org/10.1158/1541-7786.MCR- 11. Barbazuk WB, Korf I, Kadavi C, Heyen J,
07-2167 Tate S, Wun E, Bedell JA, McPherson JD,
3. Mahmood F, Mozere M, Zdebik AA, Stanescu Johnson SL (2000) The syntenic relationship
HC, Tobin J, Beales PL, Kleta R, of the zebrafish and human genomes. Genome
Bockenhauer D, Russell C (2013) Generation Res 10(9):1351–1358
and validation of a zebrafish model of EAST 12. Howe DG, Ramachandran S, Bradford YM,
(epilepsy, ataxia, sensorineural deafness and Fashena D, Toro S, Eagle A, Frazer K,
tubulopathy) syndrome. Dis Model Mech Kalita P, Mani P, Martin R, Moxon ST,
6(3):652–660. https://doi.org/10.1242/ Paddock H, Pich C, Ruzicka L, Schaper K,
dmm.009480 Shao X, Singer A, Van Slyke CE, Westerfield
4. Weyand AC, Shavit JA (2014) Zebrafish as a M (2020) The Zebrafish information network:
model system for the study of hemostasis and major gene page and home page updates.
thrombosis. Curr Opin Hematol Nucleic Acids Res 49(D1):D1058–D1064.
21(5):418–422. https://doi.org/10.1097/ https://doi.org/10.1093/nar/gkaa1010
MOH.0000000000000075 13. Vital C, Martins EP (2013) Socially-central
5. Lohi O, Parikka M, Ramet M (2013) The zeb- zebrafish influence group behavior more than
rafish as a model for paediatric diseases. Acta those on the social periphery. PLoS One 8(1):
Paediatr 102(2):104–110. https://doi.org/ e55503. https://doi.org/10.1371/journal.
10.1111/j.1651-2227.2012.02835.x pone.0055503
6. Kanwal Z, Wiegertjes GF, Veneman WJ, Meijer 14. Manabe K, Dooling RJ, Takaku S (2013) Dif-
AH, Spaink HP (2014) Comparative studies of ferential reinforcement of an approach
toll-like receptor signalling using zebrafish. response in zebrafish (Danio rerio). Behav Pro-
Dev Comp Immunol 46(1):35–52. https:// cess 98:106–111. https://doi.org/10.1016/j.
doi.org/10.1016/j.dci.2014.02.003 beproc.2013.05.013
7. Saralahti A, Ramet M (2015) Zebrafish and 15. Fu CW, Horng JL, Tong SK, Cherng BW, Liao
Streptococcal infections. Scand J Immunol BK, Lin LY, Chou MY (2021) Exposure to
82(3):174–183. https://doi.org/10.1111/ silver impairs learning and social behaviors in
sji.12320 adult zebrafish. J Hazard Mater 403:124031.
8. Streisinger G, Walker C, Dower N, Knauber D, https://doi.org/10.1016/j.jhazmat.2020.
Singer F (1981) Production of clones of homo- 124031
zygous diploid zebra fish (Brachydanio rerio). 16. Perathoner S, Cordero-Maldonado ML, Craw-
Nature 291(5813):293–296 ford AD (2016) Potential of zebrafish as a
9. Driever W, Solnica-Krezel L, Schier AF, Neu- model for exploring the role of the amygdala
hauss SC, Malicki J, Stemple DL, Stainier DY, in emotional memory and motivational behav-
Zwartkruis F, Abdelilah S, Rangini Z, Belak J, ior. J Neurosci Res 94(6):445–462. https://
Boggs C (1996) A genetic screen for mutations doi.org/10.1002/jnr.23712
affecting embryogenesis in zebrafish. Develop- 17. Gerlai R (2017) Zebrafish and relational mem-
ment 123:37–46 ory: could a simple fish be useful for the analy-
10. Haffter P, Granato M, Brand M, Mullins MC, sis of biological mechanisms of complex
Hammerschmidt M, Kane DA, Odenthal J, van vertebrate learning? Behav Process 141
(Pt 2):242–250. https://doi.org/10.1016/j. Adv Exp Med Biol 598:389–406. https://doi.

beproc.2017.01.016 org/10.1007/978-0-387-71767-8_27
18. Ahmed H, Du SJ, Vasta GR (2009) Knock- 27. Karlsson A, Christenson K, Matlak M,
down of a galectin-1-like protein in zebrafish Bjorstad A, Brown KL, Telemo E,
(Danio rerio) causes defects in skeletal muscle Salomonsson E, Leffler H, Bylund J (2009)
development. Glycoconj J 26(3):277–283. Galectin-3 functions as an opsonin and
https://doi.org/10.1007/s10719-008- enhances the macrophage clearance of apopto-
9178-9 tic neutrophils. Glycobiology 19(1):16–20.
19. Ahmed H, Vasta GR (2008) Unlike mamma- https://doi.org/10.1093/glycob/cwn104
lian GRIFIN, the zebrafish homologue 28. Vasta GR (2009) Roles of galectins in infection.
(DrGRIFIN) represents a functional Nat Rev Microbiol 7(6):424–438. https://doi.
carbohydrate-binding galectin. Biochem Bio- org/10.1038/nrmicro2146
phys Res Commun 371(3):350–355. https:// 29. Davicino RC, Elicabe RJ, Di Genaro MS, Rabi-
doi.org/10.1016/j.bbrc.2008.04.078 novich GA (2011) Coupling pathogen recog-
20. Vasta GR, Ahmed H, Du S, Henrikson D nition to innate immunity through glycan-
(2004) Galectins in teleost fish: Zebrafish dependent mechanisms. Int Immunopharma-
(Danio rerio) as a model species to address col 11(10):1457–1463. https://doi.org/10.
their biological roles in development and 1016/j.intimp.2011.05.002. S1567-5769
innate immunity. Glycoconj J 21 (11)00213-X [pii]
(8–9):503–521. https://doi.org/10.1007/ 30. Rabinovich GA, Toscano MA, Jackson SS,
s10719-004-5541-7 Vasta GR (2007) Functions of cell surface
21. Ghosh A, Banerjee A, Amzel LM, Vasta GR, galectin-glycoprotein lattices. Curr Opin
Bianchet MA (2019) Structure of the zebrafish Struct Biol 17(5):513–520. https://doi.org/
galectin-1-L2 and model of its interaction with 10.1016/j.sbi.2007.09.002. S0959-440X(07)
the infectious hematopoietic necrosis virus 00130-3 [pii]
(IHNV) envelope glycoprotein. Glycobiology 31. Guha P, Kaptan E, Bandyopadhyaya G,
29(5):419–430. https://doi.org/10.1093/ Kaczanowska S, Davila E, Thompson K, Mar-
glycob/cwz015 tin SS, Kalvakolanu DV, Vasta GR, Ahmed H
22. Nita-Lazar M, Mancini J, Feng C, Gonzalez- (2013) Cod glycopeptide with picomolar affin-
Montalban N, Ravindran C, Jackson S, Heras- ity to galectin-3 suppresses T-cell apoptosis and
Sanchez Ade L, Giomarelli B, Ahmed H, prostate cancer metastasis. Proc Natl Acad Sci
Haslam SM, Wu G, Dell A, Ammayappan A, U S A 110(13):5052–5057. https://doi.org/
Vakharia VN, Vasta GR (2016) The zebrafish 10.1073/pnas.1202653110
galectins Drgal1-L2 and Drgal3-L1 bind 32. Feng C, Ghosh A, Amin MN, Bachvaroff TR,
in vitro to the infectious hematopoietic necro- Tasumi S, Pasek M, Banerjee A, Shridhar S,
sis virus (IHNV) glycoprotein and reduce viral Wang LX, Bianchet MA, Vasta GR (2015)
adhesion to fish epithelial cells. Dev Comp Galectin CvGal2 from the eastern oyster (Cras-
Immunol 55:241–252. https://doi.org/10. sostrea virginica) displays unique specificity for
1016/j.dci.2015.09.007 ABH blood group oligosaccharides and differ-
23. Taylor ME, Drickamer K (2003) Structure- entially recognizes sympatric perkinsus species.
function analysis of C-type animal lectins. Biochemistry 54(30):4711–4730. https://doi.
Methods Enzymol 363:3–16. https://doi. org/10.1021/acs.biochem.5b00362
org/10.1016/S0076-6879(03)01039-5 33. Shi X-Z, Wang L, Xu S, Zhang X-W, Zhao X-F,
24. Vasta GR, Feng C, González-Montalbán N, Vasta GR, Wang J-X (2014) A galectin from
Mancini J, Yang L, Abernathy K, Frost G, the kuruma shrimp (Marsupenaeus japonicus)
Palm C (2017) Functions of galectins as ’self/ functions as an opsonin and promotes bacterial
non-self’-recognition and effector factors. clearance from hemolymph. PLoS One 9(3):
Pathog Dis 75(5):ftx046. https://doi.org/10. e91794. https://doi.org/10.1371/journal.
1093/femspd/ftx046 pone.0091794
25. Vasta GR (2020) Galectins in host-pathogen 34. Nita-Lazar M, Banerjee A, Feng C, Vasta GR
interactions: structural, functional and evolu- (2015) Galectins regulate the inflammatory
tionary aspects. Adv Exp Med Biol 1204: response in airway epithelial cells exposed to
169–196. https://doi.org/10.1007/978- microbial neuraminidase by modulating the
981-15-1580-4_7 expression of SOCS1 and RIG1. Mol Immunol
26. Vasta GR, Ahmed H, Tasumi S, Odom EW, 68(2 Pt A):194–202. https://doi.org/10.
Saito K (2007) Biological roles of lectins in 1016/j.molimm.2015.08.005
innate immunity: molecular and structural 35. Nita-Lazar M, Banerjee A, Feng C, Amin MN,
basis for diversity in self/non-self recognition. Frieman MB, Chen WH, Cross AS, Wang LX,
Vasta GR (2015) Desialylation of airway epi- of transgenic zebrafish. Development

thelial cells during influenza virus infection 127(9):1953–1960
enhances pneumococcal adhesion via galectin 46. Hardy ME, Ross LV, Chien CB (2007) Focal
binding. Mol Immunol 65(1):1–16. https:// gene misexpression in zebrafish embryos
doi.org/10.1016/j.molimm.2014.12.010 induced by local heat shock using a modified
36. Ahmed H, Fink NE, Vasta GR (1994) Elasmo- soldering iron. Dev Dyn 236(11):3071–3076.
branch and teleost fish contain thiol-dependent https://doi.org/10.1002/dvdy.21318
beta-galactoside-binding lectins that are cross- 47. Curado S, Anderson RM, Jungblut B,
reactive with those identified and characterized Mumm J, Schroeter E, Stainier DY (2007)
in bovine spleen. Ann N Y Acad Sci 712: Conditional targeted cell ablation in zebrafish:
318–320 a new tool for regeneration studies. Dev Dyn
37. Muramoto K, Kagawa D, Sato T, Ogawa T, 236(4):1025–1035. https://doi.org/10.
Nishida Y, Kamiya H (1999) Functional and 1002/dvdy.21100
structural characterization of multiple galectins 48. Ekker SC (2008) Zinc finger-based knockout
from the skin mucus of conger eel, Conger punches for zebrafish genes. Zebrafish
myriaster. Comp Biochem Physiol B Biochem 5(2):121–123. https://doi.org/10.1089/
Mol Biol 123(1):33–45 zeb.2008.9988
38. Inagawa H, Kuroda A, Nishizawa T, Honda T, 49. Scheer N, Campos-Ortega JA (1999) Use of
Ototake M, Yokomizo U, Nakanishi T, Soma the Gal4-UAS technique for targeted gene
G (2001) Cloning and characterisation of expression in the zebrafish. Mech Dev
tandem-repeat type galectin in rainbow trout 80(2):153–158
(Oncorhynchus mykiss). Fish Shellfish Immu- 50. Scott EK (2009) The Gal4/UAS toolbox in
nol 11(3):217–231 zebrafish: new approaches for defining behav-
39. Ahmed H, Du SJ, O’Leary N, Vasta GR (2004) ioral circuits. J Neurochem 110(2):441–456.
Biochemical and molecular characterization of https://doi.org/10.1111/j.1471-4159.2009.
galectins from zebrafish (Danio rerio): 06161.x
notochord-specific expression of a prototype 51. Clark KJ, Voytas DF, Ekker SC (2011) A TALE
galectin during early embryogenesis. Glyco- of two nucleases: gene targeting for the masses?
biology 14(3):219–232. https://doi.org/10. Zebrafish 8(3):147–149. https://doi.org/10.
1093/glycob/cwh032 1089/zeb.2011.9993
40. Summerton J (1999) Morpholino antisense 52. Jinek M, Chylinski K, Fonfara I, Hauer M,
oligomers: the case for an RNase Doudna JA, Charpentier E (2012) A program-
H-independent structural type. Biochim Bio- mable dual-RNA-guided DNA endonuclease
phys Acta 1489(1):141–158 in adaptive bacterial immunity. Science
41. Nasevicius A, Ekker SC (2000) Effective tar- 337(6096):816–821. https://doi.org/10.
geted gene ‘knockdown’ in zebrafish. Nat 1126/science.1225829
Genet 26(2):216–220. https://doi.org/10. 53. Hwang WY, Fu Y, Reyon D, Maeder ML, Tsai
1038/79951 SQ, Sander JD, Peterson RT, Yeh JR, Joung JK
42. Tallafuss A, Gibson D, Morcos P, Li Y, (2013) Efficient genome editing in zebrafish
Seredick S, Eisen J, Washbourne P (2012) using a CRISPR-Cas system. Nat Biotechnol
Turning gene function ON and OFF using 31(3):227–229. https://doi.org/10.1038/
sense and antisense photo-morpholinos in zeb- nbt.2501
rafish. Development 139(9):1691–1699. 54. Oliveira-de-Abreu E, Silva-Dos-Santos D,
https://doi.org/10.1242/dev.072702 Lepletier A, Ramos TDP, Ferreira-Reis R,
43. McCallum CM, Comai L, Greene EA, Henik- Vasconcelos-Fontes L, Ramos MT, Torres
off S (2000) Targeted screening for induced RC, Cotta-de-Almeida V, Carvalho VF, Villa-
mutations. Nat Biotechnol 18(4):455–457. Verde DMS (2018) Lack of galectin-3 disrupts
https://doi.org/10.1038/74542 thymus homeostasis in association to increase
44. Wienholds E, van Eeden F, Kosters M, of local and systemic glucocorticoid levels and
Mudde J, Plasterk RH, Cuppen E (2003) Effi- steroidogenic machinery. Front Endocrinol 9:
cient target-selected mutagenesis in zebrafish. 365. https://doi.org/10.3389/fendo.2018.
Genome Res 13(12):2700–2707. https://doi. 00365
org/10.1101/gr.1725103 55. Ferragut F, Cagnoni AJ, Colombo LL, Sán-
45. Halloran MC, Sato-Maeda M, Warren JT, Su F, chez Terrero C, Wolfenstein-Todel C, Tron-
Lele Z, Krone PH, Kuwada JY, Shoji W (2000) coso MF, Vanzulli SI, Rabinovich GA, Mariño
Laser-induced gene expression in specific cells KV, Elola MT (2019) Dual knockdown of
Galectin-8 and its glycosylated ligand, the
activated leukocyte cell adhesion molecule 12(10). https://doi.org/10.1242/dmm.

(ALCAM/CD166), synergistically delays 039982
in vivo breast cancer growth. Biochim Biophys 62. Park JS, Lee KY, Han JY (2020) Precise
Acta, Mol Cell Res 1866(8):1338–1352. genome editing in poultry and its application
https://doi.org/10.1016/j.bbamcr.2019. to industries. Genes 11(10):1182. https://doi.
03.010 org/10.3390/genes11101182
56. Yang N, Zhang W, He T, Xing Y (2017) Silenc- 63. Douris V, Denecke S, Van Leeuwen T, Bass C,
ing of galectin-1 inhibits retinal neovasculariza- Nauen R, Vontas J (2020) Using CRISPR/
tion and ameliorates retinal hypoxia in a murine Cas9 genome modification to understand the
model of oxygen-induced ischemic retinopa- genetic basis of insecticide resistance: Drosoph-
thy. Exp Eye Res 159:1–15. https://doi.org/ ila and beyond. Pestic Biochem Physiol 167:
10.1016/j.exer.2017.02.015 104595. https://doi.org/10.1016/j.pestbp.
57. Morcos PA (2007) Achieving targeted and 2020.104595
quantifiable alteration of mRNA splicing with 64. Wada N, Ueta R, Osakabe Y, Osakabe K
Morpholino oligos. Biochem Biophys Res (2020) Precision genome editing in plants:
Commun 358(2):521–527. https://doi.org/ state-of-the-art in CRISPR/Cas9-based
10.1016/j.bbrc.2007.04.172 genome engineering. BMC Plant Biol
58. Benato F, Skobo T, Gioacchini G, Moro I, 20(1):234. https://doi.org/10.1186/
Ciccosanti F, Piacentini M, Fimia GM, s12870-020-02385-5
Carnevali O, Dalla Valle L (2013) Ambra1 65. Albadri S, De Santis F, Di Donato V, Del Bene
knockdown in zebrafish leads to incomplete F (2017) CRISPR/Cas9-mediated knockin
development due to severe defects in organo- and knockout in zebrafish. In: Jaenisch R,
genesis. Autophagy 9(4):476–495. https:// Zhang F, Gage F (eds) Genome editing in
doi.org/10.4161/auto.23278 neurosciences. Springer, Cham, pp 41–49.
59. Cong L, Ran FA, Cox D, Lin S, Barretto R, h t t p s : //d oi . o r g / 1 0. 1 0 07 / 9 78 - 3 - 3 1 9-
Habib N, Hsu PD, Wu X, Jiang W, Marraffini 60192-2_4
LA, Zhang F (2013) Multiplex genome engi- 66. Liu K, Petree C, Requena T, Varshney P, Varsh-
neering using CRISPR/Cas systems. Science ney GK (2019) Expanding the CRISPR tool-
339(6121):819–823. https://doi.org/10. box in Zebrafish for studying development and
1126/science.1231143 disease. Front Cell Dev Biol 7:13. https://doi.
60. Mali P, Yang L, Esvelt KM, Aach J, Guell M, org/10.3389/fcell.2019.00013
DiCarlo JE, Norville JE, Church GM (2013) 67. Wu N, Liu B, Du H, Zhao S, Li Y, Cheng X,
RNA-guided human genome engineering via Wang S, Lin J, Zhou J, Deciphering Disorders
Cas9. Science 339(6121):823–826. https:// Involving S, Study CO, Qiu G, Wu Z, Zhang J
doi.org/10.1126/science.1232033 (2019) The progress of CRISPR/Cas9-
61. Kang Y, Chu C, Wang F, Niu Y (2019) mediated gene editing in generating mouse/
CRISPR/Cas9-mediated genome editing in zebrafish models of human skeletal diseases.
nonhuman primates. Dis Model Mech Comput Struct Biotechnol J 17:954–962.
https://doi.org/10.1016/j.csbj.2019.06.006
Chapter 24
Examining Galectin Gene Regulation by Reporter Assays

Sebastian Schmidt, Herbert Kaltner, and Hans-Joachim Gabius
Abstract
Matching their role as potent and versatile effectors in cellular homeostasis and disease processes, galectins
are subject to a fine-tuned transcriptional regulation of gene expression. It can apparently even involve
coregulation with certain elements of the enzymatic machinery for glycan biosynthesis/remodeling and/or
functional carriers of galectin-binding glycans such as the α5β1-integrin. All this suggests not yet fully
known combinatorial processes to reach the desired outcome. Identification of transcription start point(s),
cloning of upstream promoter region, and the design of plasmids for luciferase-based reporter assays
establish the platform to initiate a systematic search of regulatory sequences. Their elucidation is also a
step toward rationally manipulating expression of galectin genes in pathogenesis.
Key words Crystallin, L-Maf, Luciferase, Pax6, Promoter, Transcription
1 Introduction
The sheer size of the glycome can be overwhelming. The concept of

the sugar code has given this structural complexity a physiological
meaning [1–5]. Accordingly, the glycan diversity is teeming with
biochemical signals, and—together with lectins—specific interac-
tions translate the meaning of glycan-encoded signals into cellular
activities. In fact, glycan-lectin recognition is a versatile mode
within the flow of biological information. It is increasingly deli-
neated to underlie a large variety of processes in cellular homeosta-
sis and in diseases.
On the side of the receptors (“readers” and “translators” of
glycan-encoded information), galectins have earned a special place
because of their multifunctionality, by presence within the cell, at
the cell surface and in the extracellular space and by interacting with
glycan and peptide ligands [6–12]. The discovery of individual
expression profiles, for example initially detected by immunocyto-
and histo-chemistry for galectin-1 (galaptin) and galectin-3 (Mac-2
antigen) [13–15], argue in favor of (mostly) non-redundant activity
profiles and intimate expression control. The paralogue pair of
445
446 Sebastian Schmidt et al.
chicken galectins-1A and -1B that originated by duplication from

an ancestral gene, probably not long after separation of lineages
leading to mammals and birds [16], provides a graphic example:
these two closely related proteins have markedly different expres-
sion profiles in embryonic and adult tissues [17, 18].
Of particular note for reaching optimal functionality, evidence
is emerging even for the presence of molecular switches that make
coregulation of a counter-receptor for the pairing possible. The
case studies of pancreatic carcinoma cell line (Capan-1) growth
regulation by a tumor suppressor (i.e., p16INK4a) [19] or of cross-
talk between effector/regulatory T cells [20] have disclosed differ-
ent routes toward the aim of tailoring binding and post-binding
signaling. Equally important, such coregulation also appears to be
operative for structure building by contact, as seen for an eye lens-
specific galectin, i.e., the galectin-related inter-fiber protein (GRI-
FIN) [21], and α-crystallins [22–25]. Moreover, coregulation
enables team building, the simultaneous upregulation of human
galectins-1, -3, and -8 in the pathogenesis of osteoarthritis
providing a clinically relevant example [26, 27]. Considering
these insights, “it is [still] rather unexpected that very few studies
have been performed on the molecular mechanisms governing the
activity of galectin genes” [28].
The identification of functional sequences in the upstream
promoter region by such studies encourages systematic efforts.
The first molecular switch for activation of transcription of the
human galectin-1 gene was the 50 -proximal Sp1 site at -57 bp
connected to butyrate-induced expression [29]. Assumed to be
involved in strong dexamethasone-induced upregulation of galec-
tin-1 in human T leukemic cells associated with cell death, respec-
tive steroid-responsive elements are present in the promoter, too
[30, 31]. So the puzzle-like complexity of occurrence of sequence
motifs with putative capacity to interact with transcription factors
poses a challenge to be addressed (for a computationally obtained
profile of putative sites of binding transcription factors in the
upstream promoter region of the human galectin-1 gene, please
see Fig. 1a; for respective examples in the cases of human and
chicken galectin-3 genes, please see [32, 33]; for extent of sharing
such sequence motifs between the three human galectin genes
mentioned above, please see Fig. 1b).
Toward this aim, that is to answer the big question on galectin
gene regulation by transcription factors, we here describe methods
to identify the transcription start point(s) (tsps) and active sequence
motifs in the upstream promoter region. The flow diagram in Fig. 2
presents a graphical overview. Engineering of vectors with respec-
tive sequence sections into a luciferase-based detection platform for
reporter assays and a transfection protocol enable functional
Examining Galectin Gene Regulation by Reporter Assays 447
Fig. 1 Illustrations of data from in silico analyses (based on applying the search algorithm MatInspector from
Genomatix (Munich, Germany); for details on settings and thresholds, please see ref. 24) for putative
mapping and deletion analysis. Application of these protocols will

not only clarify the nature of the switch on/off elements. The
resulting insights may then offer cues to guide experiments toward
the aim to rationally manipulate galectin expression at the level of
transcription.
Fig. 2 Flow diagram illustrating the experimental outline of promoter analyses for chicken galectins
Fig. 1 (continued) transcription factor-binding sites in the promoter region 2500 bp upstream from the tsps
of genes for human Gal-1, -3, and -8. (a) Positions of the sequence motifs unique for Gal-1 (obtained from
comparison with the corresponding sequence regions for the genes of Gal-3 and -8; the two sites mentioned
in the introduction [29, 30] were detected but not found to be unique, thus not presented here). (b) Compilation
of the motifs unique for Gal-1 (blue), Gal-3 (yellow), and Gal-8 (red) or shared by the three sequence regions
(central circle) for the three galectins (modified, from [12])
2 Materials
2.1 RACE (Rapid 1. Chicken cell lines (DT40, F6CC-PR9692, other suitable cell
Amplification of cDNA lines) or tissue (e.g., eye lens or kidney from 14-day- or
Ends): Determination 18-day-old chicken embryos; fresh or frozen at 80 C).
of Transcription Start 2. RNA preparation kit (RNeasy Mini Kit; Qiagen, Hilden,
Points (TSPs) at the 50 - Germany).
Ends of Chicken 3. LE Agarose (EEO 0.09–0.13; Biozym Scientific, Hessisch Old-
Galectin Gene endorf, Germany).
Sequences
4. TAE buffer (5 mM ethylenediaminetetraacetic acid (EDTA),
40 mM Tris, 20 mM acetic acid, pH 8.5).
5. Equipment for agarose gel electrophoresis: submerged hori-
zontal electrophoresis cell with power supply (Bio-Rad,
Munich, Germany).
6. 10 DNA loading buffer (Qiagen).
7. GelRed® (Biotium; VWR, Darmstadt, Germany).
tem; Bio-Rad).
9. GeneRacer™ kit containing GeneRacer™ 50 primer 5’- CGA
CTGGAGCACGAGGACACTGA -30 (Thermo Fisher Scien-
tific, Dreieich, Germany).
10. Chicken total RNA.
11. Thermomixer comfort (Eppendorf, Hamburg, Germany).
13. dNTPs (10 mM; Promega, Walldorf, Germany).
14. TopTaq™ DNA polymerase and 10 buffer (Qiagen).
15. Oligonucleotide primers specific for a fragment of the chicken
actin gene (Metabion, Planegg, Germany).
16. Galectin gene-specific reverse oligonucleotide primer
(Metabion).
Molecular, Berlin, Germany).
19. TA cloning vector (e.g., pGEM®-T Easy; Promega).
20. T4 DNA ligase and 10 buffer (Promega).
21. Chemically competent E. coli cells (One Shot™ TOP10; Invi-
trogen, Dreieich, Germany).
22. Luria Broth (LB) agar plates containing antibiotic(s) as well as
isopropyl-β-D-thiogalactopyranoside (IPTG) and 5-bromo-4-
chloro-3-indolyl-β-D-galactopyranoside (X-Gal) (add 1 μL of
100 mM IPTG and 1 μL of 40 mg/mL X-Gal solution directly
per 1 mL of liquid LB agar, kept at about 50 C; pour 15 mL

into a Petri dish) to perform blue-white screening of colonies.
23. LB medium (Roth, Karlsruhe, Germany).
Bühler, Bodelshausen, Germany).
25. Incubator (Type BM100; Memmert, Schwabach, Germany).
Invitek Molecular).
27. Restriction endonucleases, provided 10 buffer and 100
bovine serum albumin (BSA) (all reagents supplied by
Promega).
28. Photometer (NanoPhotometer® P-Class P 300; Implen,
Munich, Germany).
29. Sequence editing software (DNAStar Lasergene 15.0;
Madison, USA).
2.2 Generation of 1. Chicken cell lines or tissue (see Subheading 2.1, item 1).
Vectors for Analyzing 2. LE Agarose (EEO 0.09–0.13; Biozym Scientific).
Regulatory DNA
3. TAE buffer.
Sequences in
Eukaryotic Cells 4. Equipment for agarose gel electrophoresis: submerged hori-
zontal electrophoresis cell with power supply (Bio-Rad).
5. 10 DNA loading buffer (Qiagen).
6. GelRed® (VWR).
tem; Bio-Rad).
9. Phusion® High-Fidelity DNA polymerase and 5 buffer (New
England Biolabs, Frankfurt, Germany).
11. Gel extraction kit (Invitek Molecular).
12. Restriction endonucleases, provided 10 buffer and 100
BSA (all reagents supplied by Promega).
13. Alkaline phosphatase and buffer (rAPid Alkaline Phosphatase;
Roche Diagnostics, Mannheim, Germany).
14. T4 DNA ligase and 10 buffer (Promega).
15. Chemically competent E. coli cells (One Shot™ TOP10;
Invitrogen).
16. LB agar plates with antibiotic(s).
17. LB medium (Roth).
Invitek Molecular).
2.2.1 Cloning of DNA 1. Wizard® genomic DNA purification kit (Promega).

Segments of the (Putative) 2. TopTaq™ DNA polymerase (Qiagen).
Regulatory Region of
3. Oligonucleotide primers (for specific amplification of
Interest into Firefly
sequences; Metabion).
Reporter Vector pGL4.20
4. GC-rich PCR system (Roche Diagnostics).
5. Luciferase reporter vector (pGL4.20; Promega).
2.2.2 Cloning of 1. Software tool for detection of putative binding motifs for
Transcription Factor- transcription factors in DNA sequences (MatInspector; Geno-
Encoding cDNA into matix, Munich, Germany).
Eukaryotic Expression 2. RNA preparation kit (RNeasy Mini Kit; Qiagen).
Vector pcDNA™3.1(+)
3. GoScript™ reverse transcription system (Promega).
4. Oligonucleotide primers (for specific amplification of
sequences; Metabion).
5. Expression vector (pcDNA™3.1(+); Thermo Fisher
Scientific).
2.3 Transfection of 1. Suitable cell lines (e.g., F6CC-PR9692, DLD-1, PC-3, or any
Adherent Eukaryotic other adherent cell line).
Cells with Luciferase 2. Cell culture medium (for F6 cells: Iscove’s Modified Dulbec-
Reporter Vectors and co’s Medium (IMDM) supplemented with 2% (v/v) chicken
Transcription Factor- serum, 10% (v/v) fetal calf serum (FCS); 1% (w/v) penicillin/
Encoding Vector streptomycin/L-glutamine solution).
3. Phosphate-buffered saline (PBS: 20 mM KH2PO4/Na2HPO4,
150 mM NaCl, pH 7.2) with 2 mM EDTA (PBS/EDTA).
4. Culture flasks (Techno Plastic Products AG, Trasadingen,
Switzerland).
5. 24-Well cell culture plate (Techno Plastic Products AG).
6. Light microscope (Model BH-2; Olympus, Hamburg,
Germany).
7. Centrifuge (Model EBA 20; Hettich, Kirchlengern, Germany).
8. Neubauer chamber.
9. jetPEI® transfection reagent (Polyplus-transfection®, Illkirch,
France).
10. 150 mM NaCl.
11. Firefly reporter vector (pGL4 vector containing DNA seg-
ments of the putative promoter sequences).
12. Normalization vector (Renilla luciferase-expressing vector;
pGL4.74; Promega).
13. Expression vector pcDNA™ 3.1 (+) containing transcription
factor-encoding sequence.
14. pGEM®-T Easy (Promega).

15. Incubator set at 37 C with humid atmosphere with 5% CO2
(Eppendorf).
2.4 Luciferase Assay 1. Cells transfected with firefly reporter vector, normalization
of Galectin Promoter vector and, if needed, also the expression vector (see Subhead-
Activity ing 2.3, items 11–13).
2. PBS.
3. Dual-Luciferase® Reporter assay system (Promega), containing
Luciferase Assay Reagent II (LAR II), Stop & Glo Reagent and
Passive Lysis Buffer (PLB).
5. 96-Well microtiter plate, white (for luminescence detection;
6. Infinite F200 plate reader (TECAN, M€annedorf, Switzerland).
3 Methods
3.1 RACE (Rapid 1. Isolate total chicken RNA from chicken cell pellet (DT40 or
Amplification of cDNA F6CC-PR9692, each 107 cells) or suitable tissue specimen
Ends): Determination (~20 mg) using reagents of an RNA preparation kit.
of TSPs at the 50 -Ends 2. Measure the concentration and the purity of the RNA using a
of Chicken Galectin nanophotometer (see Note 1).
Gene Sequences 3. Use a small aliquot (about 1 μg) of the RNA to check the
quality by agarose gel electrophoresis: Mix the RNA sample
with loading buffer (containing 3.0 μL 10,000 GelRed® per
mL) to a final concentration of 1 and load the mixture to a
slot of a 1% (w/v) agarose gel in TAE buffer. Run the electro-
phoresis at 94 V for 10 min (see Note 1).
4. Prepare RACE-ready cDNA (RcDNA) according to the
instructions outlined in the GeneRacer™ kit’s protocol.
5. Examine the quality of the RcDNA by PCR, using a gene-
specific primer pair directed against a DNA segment of chicken
actin and TopTaq™ DNA polymerase. Use the following ther-
mocycler program: initial denaturation at 94 C for 3 min, then
30 cycles of denaturation at 94 C for 30 s, annealing at 55 C
for 30 s and extension at 72 C for 1 min, final extension step at
72 C for 10 min.
6. Mix the sample with DNA loading buffer (see step 3) and
perform electrophoresis at 100 V for 30 min. Visualize by
using gel documentation system.
7. Tsps determination starts with amplification of cDNA ends by
PCR using the GeneRacer™ 50 primer 50 -
CGACTGGAGCACGAGGACACTGA-30 and a gene-specific

reverse primer to generate a product upstream of the start ATG
of the coding sequence (see Notes 2 and 3).
8. Perform agarose gel electrophoresis (see steps 3 and 6) and
excise the amplified product from the gel.
9. Recover the DNA fragments from the gel slice using a gel
extraction kit.
10. Ligate the fragments into TA cloning vector using T4 DNA
ligase (at room temperature (RT) for 1 h or 4 C overnight).
Component Volume
®
pGEM -T Easy (50 ng/μL) 0.5 μL
Insert 5.0 μL
10 ligase buffer 1.0 μL
T4 DNA ligase (3 units/μL) 1.0 μL
Sterile distilled water to a final volume of 10.0 μL
11. Transform chemically competent E. coli cells in a 50 μL sus-

pension by first adding 5 μL of the ligation reaction followed by
incubation on ice for 45 min and 90 s at 42 C. After adding
1 mL LB medium to the suspension, shake the cells at 37 C at
350 rpm in a thermomixer. Then streak 100 μL of the suspen-
sion on LB plates containing the appropriate antibiotic(s),
IPTG and X-Gal for blue/white selection. Incubate them at
37 C overnight.
12. Inoculate 4 mL LB medium containing the appropriate antibi-
otic(s) with a single colony and agitate this setup in an orbital
shaker (225 rpm) at 37 C overnight.
13. Purify the plasmids from the bacterial suspensions by using a
plasmid preparation kit and determine the concentration using
a nanophotometer.
14. Examine the length of inserted fragments after digestion (2 h at
37 C) by appropriate restriction enzymes using agarose gel
electrophoresis (see steps 3 and 6), visualized by gel documen-
tation system.
Component Volume
Vector 100 ng
10 buffer 2.0 μL
100 BSA 0.2 μL
Restriction endonuclease(s) 0.1 units
Sterile distilled water to a final volume of 20.0 μL
15. Sequence the inserted fragments (as part of the plasmid) by a

commercial service, then analyze the sequences for determina-
tion of their tsps by sequence editing (software used: DNAStar
Lasergene 15.0).
3.2 Generation of 1. Prepare genomic DNA from chicken cell pellet or tissue using
Vectors for Analyzing the Wizard® Genomic DNA purification kit, following the
Regulatory DNA instructions of the manufacturer.
Sequences in 2. Check integrity of the genomic DNA (200 ng) by agarose gel
Eukaryotic Cells electrophoresis as described in Subheading 3.1, steps 3 and
6 (see Note 4).
3.2.1 Cloning of DNA
Segments of the (Putative) 3. Design sequences of oligonucleotide primer pairs to be specific
Regulatory Region of for a distinct genomic DNA sequence of interest by selecting
Interest into Firefly suitable flanking regions. Primers should be between 18 and
Reporter Vector pGL4.20 23 bp in length. Add restriction recognition sites at the 50 -ends
with an at least 4 bp overhang (see Note 5).
4. Perform a control amplification reaction with DNA oligonu-
cleotide primers specific for a segment of the chicken actin gene
using TopTaq™ DNA polymerase. Use the following thermo-
cycler program: initial denaturation at 94 C for 3 min, then
30 cycles of denaturation at 94 C for 30 s, annealing at 55 C
for 30 s, and extension at 72 C for 1 min, final extension step
at 72 C for 10 min.
5. Check the length of amplification product by agarose gel elec-
trophoresis (see Subheading 3.1, steps 3 and 6).
6. Amplify the promoter region (2.5 kbp or shortened fragments)
from the genomic DNA template (purified in step 1) with the
oligonucleotide primer pairs (described in step 3) and Phu-
sion® DNA polymerase. Use the following thermocycler pro-
gram: initial denaturation at 98 C for 3 min, 35 cycles:
denaturation at 98 C for 10 s, annealing at 50–60 C for
10 s, extension at 72 C for 2 min, final extension step at
72 C for 10 min (see Notes 3, 6, and 7).
7. Examine the length of the amplification products by agarose
gel electrophoresis (see Subheading 3.1, steps 3 and 6).
8. Excise the DNA fragment from the agarose gel and purify DNA
from slices with a gel extraction kit according to the manufac-
9. Digest the promoter fragment and the pGL4.20 vector in
separate reaction tubes with appropriate restriction enzymes
(see Subheading 3.1, step 14).
10. Reduce re-ligation of the linearized vector by transferring a
mixture of digestion solution and DNA loading buffer to a slot
of a 1% (w/v) agarose gel. Perform electrophoresis at 100 V for
30 min.
11. Purify linearized vector DNA as outlined in step 8 and pro-

moter fragments directly from solutions of enzymatic treat-
ment using spin columns of the gel extraction kit.
12. Dephosphorylate the 50 ends of the linearized vector DNA
using alkaline phosphatase. Mix 2 μL alkaline phosphatase
buffer (10), 17 μL of vector DNA, and 1 μL of alkaline
phosphatase (1 unit). Incubate the reaction mix for 20 min at
37 C and inactivate the enzyme at the end of the incubation
period for 10 min at 75 C.
13. Perform T4 DNA ligase reaction in a setup containing DNA
fragments and the enzymatically treated vector DNA at a molar
ratio of 3:1 to 5:1 at 4 C overnight:
Component Amount
Vector 15 fmol (~50 ng)
Insert 45–75 fmol
10 ligation buffer 1 μL
T4 DNA ligase (3 units/μL) 1 μL
With sterile water to a final volume of 10 μL
14. Transform chemically competent E. coli cells (see Subheading

3.1, step 11).
15. Inoculate 4 mL LB medium containing the appropriate antibi-
otic(s) with a single colony and agitate this setup in an orbital
shaker (225 rpm) at 37 C overnight.
16. Prepare DNA plasmids using a commercial plasmid preparation
kit, determine the concentration using a nanophotometer,
digest plasmids with the appropriate restriction enzymes (see
Subheading 3.1, step 14), and perform agarose gel electro-
phoresis (see Subheading 3.1, steps 3 and 6) to ascertain integ-
rity of the inserted fragment.
17. Verify the sequence and the orientation of the inserted DNA
fragment by commercial sequencing.
3.2.2 Cloning of 1. Perform in silico analyses of the promoter sequences using a

Transcription Factor- software tool to detect sequence motifs with putative affinity
Encoding cDNA into for transcription factors (e.g., MatInspector) (see Note 8).
Eukaryotic Expression 2. Choose transcription factors of interest (e.g., organ-specific
Vector pcDNA™3.1(+) factors).
3. Isolate total chicken RNA from chicken cell pellet (107 cells) or
suitable tissue specimen (~20 mg) using RNA preparation kit,
following the manufacturer’s manual.
4. Measure the concentration and the purity of the RNA using a
nanophotometer.
5. Test quality of isolated RNA with a small aliquot (about 1 μg)

transferred to an agarose gel electrophoresis (see Subheading
3.1, step 3).
6. For a first strand cDNA synthesis with the GoScript™ reverse
transcription system, add 1 μL of Oligo(dT) Primer
(100 pmol/μL) to a solution of 2 μg isolated RNA and fill up
to a final volume of 5 μL with nuclease-free water.
7. Incubate the solution 5 min at 70 C, followed by 5 min at
4 C. During that time, prepare the reverse transcription mix as
follows:
Component Volume
5 reaction buffer 4.0 μL
MgCl2 (25 mM) 1.5 μL
dNTPs (10 mM) 1.0 μL
GoScript™ reverse transcriptase 1.0 μL
®
RNAsin ribonuclease inhibitor 0.5 μL
Nuclease-free water 7.0 μL
8. Combine the reverse transcription setup and the

RNA-containing solution. Incubate the reaction mixture
5 min at 25 C, 60 min at 42 C, and 15 min at 72 C. The
cDNA can be stored at 20 C or used directly for PCR.
9. To obtain transcription factor-encoding DNA sequences,
design specific oligonucleotide primer pairs for full-length
cDNAs (see Subheading 3.2.1, step 3).
10. Run a control amplification reaction to test the quality of
synthesized cDNA (see Subheading 3.1, steps 5 and 6).
11. Apply Phusion® DNA polymerase and oligonucleotide primer
pairs for the amplification of full-length DNA sequences from
1 μL of cDNA containing solution (synthesized cDNA diluted
1:10 in water). Use the following thermocycler program: initial
denaturation at 98 C for 3 min, then 35 cycles of denaturation
at 98 C for 10 s, annealing at 50–60 C for 10 s, and extension
at 72 C for 1–2 min, final extension step at 72 C for 10 min
(see Note 6).
12. Purify the products (see Subheading 3.2.1, steps 7 and 8).
13. Treat amplified full-length DNA and linearize the
pcDNA™3.1(+) vector with appropriate restriction enzymes
at 37 C for 2 h (see Subheading 3.1, step 14).
14. Process restriction enzyme-treated full-length DNA and vector
as outlined in steps 10–17 in the previous section.
3.3 Transfection of 1. Remove culture medium and incubate the cells of the layer
Adherent Eukaryotic with 5 mL PBS/2 mM EDTA solution at 37 C for 10 min.
Cells with Luciferase 2. Transfer cell suspension into a 15 mL centrifuge tube.
Reporter Vectors and
3. Centrifugate at 1500 rpm (214 g) for 3 min and resuspend
Transcription Factor- the cell pellet in culture medium, count the cells using a Neu-
Encoding Vector bauer chamber and adjust the cell number to 2 105 cells/
mL.
4. Seed 0.5 mL of cell suspensions (contain 1 105 cells) per well
of a 24-well cell culture plate.
5. Grow cells at 37 C for approximately 24 h.
6. Prepare DNA-containing solution as follows (see Notes 9–11):
Vector Amount
pGL4 with promoter sequences 0.1 to 100 ng/well
(see Note 12)
pGL4.74 10 ng/well
(normalizing vector expressing Renilla
luciferase)
pcDNA™3.1(+) with transcription factor 0 to 100 ng/well
sequence
Any suitable empty DNA vector Add to final quantity of
(e.g., pGEM-T Easy) 500 ng
7. Calculate the volume necessary for three replicates and pipet

the DNA-containing solution into one tube.
8. Adjust the total volume to 30 μL with sterile, distilled water
and add 120 μL of 150 mM NaCl solution (solution A).
9. Use 6 μL jetPEI transfection reagent per three wells and dilute
this volume in 144 μL 150 mM NaCl solution (solution B).
10. Add solution B to the DNA-containing solution A and incu-
bate the mixture at RT for at least 15 min.
11. During the incubation step, substitute the culture medium in
the wells through 500 μL of fresh medium per well.
12. Add 100 μL per well of the DNA/jetPEI mix dropwise and
agitate the culture plate gently. Incubate the cells in an incuba-
tor at 37 C and 5% CO2.
13. After 2 h, add 500 μL medium to the cells per well and grow
them for a further period of 24 h (see Note 13).
3.4 Luciferase Assay 1. Dilute the 5 PLB to 1 PLB with sterile distilled water.
of Galectin Promoter 2. Remove the growth medium and wash the cells with
Activity 300 μL PBS.
3. Carefully pour off the PBS and lyse the cells in 100 μL 1 PLB
per well of a 24-well cell culture plate, shake at 800 rpm at RT
for 15 min using a thermomixer.
4. Prepare the plate reader by priming the injectors with LAR II
and Stop & Glo reagent. Adjust the injected volume of the
reagents to 50 μL per well each and set the measurement time
to 5 s.
5. Transfer 20 μL of each cell lysate to separate wells of a white
96-well microtiter plate (see Notes 14 and 15).
6. Measure the luciferase activity (see Note 16).
7. Calculate the ratio between the firefly and Renilla signal values
(see Note 17).
4 Notes
1. RNA preparations should be of high quality. The ratio of

absorbance of the sample at 260 nm to the absorbance at
280 nm should be between 1.9 and 2.1. Addition of formalde-
hyde is not necessary for gel electrophoresis. Ensure that the
ratio of intensity of the 28S rRNA to the 18S rRNA band is
approximately 2:1 (if possible, quantify the signals of both
bands and calculate the ratio by using a gel documentation
system and gel analysis software, e.g., Image Lab software
(Bio-Rad)).
2. If no or only a small quantity of amplification products can be
obtained, applying nested PCR technique is recommended.
3. Perform a touch-down protocol to increase specificity of the
amplification reaction. Apply the following thermocycler pro-
gram: initial denaturation at 98 C for 3 min, 5 cycles of
denaturation at 98 C for 10 s, annealing at 60 C for 10 s,
and extension at 72 C for 90 s, 5 cycles: denaturation at 98 C
for 10 s, annealing at 55 C for 10 s, extension at 72 C for 90 s,
25 cycles: denaturation at 98 C for 10 s, annealing at 50 C for
10 s, extension at 72 C for 90 s, final extension step at 72 C
for 10 min. This protocol may need adjustment by varying the
annealing temperatures.
4. Do not freeze to avoid shearing of the genomic DNA. Geno-
mic DNA should be stored at 4 C and checked by gel electro-
phoresis to form a sharp band of high molecular weight before
use. Even a slight smearing indicates presence of
degraded DNA.
5. When a higher level of specificity is needed, primer sequences
of increased length can be used. Analyze primers sequences in
silico for self-complementarity and hairpin formation using
either commercial DNA tools (e.g., SeqBuilder Pro 15, part of

DNAStar Lasergene 15.0) or open-access online tools (e.g.,
http://www.molbiotools.com/primerinspector.php).
Sequences should not exceed a GC content of 70%.
6. Phusion® DNA polymerase should be the first choice because
of its comparatively reduced error rate. A disadvantage,
though, is that production rates can be comparatively low in
some cases.
7. For GC-rich sequences, which are often difficult to amplify, the
GC-RICH PCR System (Roche Diagnostics, Sigma-Aldrich)
with its enzyme blend of thermostable Taq (Thermus aquati-
cus) and Tgo (Thermococcus gorgonarius) DNA polymerases is
recommended.
8. A sequence region of about 2500 bp upstream of the tsps is
commonly mapped for presence of putative sites of transcrip-
tion factors and sorted according to occurrence of
corresponding factors in chicken tissues. Only sequences with
a core similarity of 1 compared with the respective matrix were
considered as hits that have a perfect match to the respective
motif. Presence of more than a single site for a certain tran-
scription factor gives incentive for planning deletion analysis.
Of note, the relationship between such site(s) and expression
can be specific for a galectin-motif pair (please see refs. 24, 25
for the example of eye lens-specific C-GRIFIN). A case such as
upregulation of galectin expression for more than one member
of this family, as seen for butyrate as induces [34], points to
coordinated modulation and the possibility of sharing motifs.
9. For storage at 20 C, concentration of plasmids should be
adjusted to 100 ng/μL.
10. Always use fresh aliquots of plasmid DNA for transfection to
avoid degradation of DNA by repeated freeze/thaw cycles.
11. For practical reasons, it is suggested to dilute the DNA with
water to a concentration of 10 ng/μL or less, allowing to work
with volumes large enough for pipetting.
12. Titrations of the amount of transfected reporter plasmid DNA
are necessary. After finding the concentration for an optimal
signal (~20,000 relative light units, RLU) of Renilla activity, a
second titration with increasing concentrations of firefly
reporter plasmid DNA should be performed. Apply highest
possible concentrations of transfected firefly reporter plasmid
DNA, which do not lead to a reduction of the optimized signal
for Renilla activity.
13. Each assay series should consist of a minimum of six indepen-
dent experiments, each including three technical replicates.
14. Protect luciferase substrates from light.
15. Firefly luciferase and Renilla luciferase in PLB are stable at RT

for at least 6 h or on ice for 16 h. For long-term storage, the
lysates should be frozen at 20 C.
16. Counts for activity of Renilla luciferase should be approxi-
mately 20,000 RLU for normalization. Values for firefly lucif-
erase signal should be at least 200,000 RLU, which is then
sufficient to detect significant activation or downregulation of
reporter activity.
17. Correct the measured values with the blank sample (PLB). To
obtain a normalized signal value, calculate the ratio of firefly
RLU to Renilla RLU. This ratio is set into relation with a
control ratio (RLU of basic firefly vector without promoter
sequence/Renilla RLU or with a respective promotor
sequence), then the percentage of increase/decrease of pro-
moter activity can be calculated.
Acknowledgments
We are grateful to Dr. P. Muschler for his valuable advice and

inspiring discussions.
References
1. Winterburn PJ, Phelps CF (1972) The signifi- 8. Funasaka T, Raz A, Nangia-Makker P (2014)
cance of glycosylated proteins. Nature Nuclear transport of galectin-3 and its thera-
236(5343):147–151. https://doi.org/10. peutic implications. Semin Cancer Biol 27:
1038/236147a0 3 0 – 3 8 . h t t p s : // d o i . o r g / 1 0 . 1 0 1 6 / j .
2. Sharon N (1975) Complex carbohydrates. semcancer.2014.03.004
Their chemistry, biosynthesis, and functions. 9. Arthur CM, Baruffi MD, Cummings RD,
Addison-Wesley, Reading, MA Stowell SR (2015) Evolving mechanistic
3. Roseman S (2001) Reflections on glycobiol- insights into galectin functions. Methods Mol
ogy. J Biol Chem 276(45):41527–41542. Biol 1207:1–35. https://doi.org/10.1007/
https://doi.org/10.1074/jbc.R100053200 978-1-4939-1396-1_1
4. Gabius H-J, Roth J (2017) An introduction to 10. Kaltner H, Toegel S, Garcı́a Caballero G, Man-
the sugar code. Histochem Cell Biol ning JC, Ledeen RW, Gabius H-J (2017)
147(2):111–117. https://doi.org/10.1007/ Galectins: their network and roles in immu-
s00418-016-1521-9 nity/tumor growth control. Histochem Cell
5. Kaltner H, Abad-Rodrı́guez J, Corfield AP, Biol 147(2):239–256. https://doi.org/10.
Kopitz J, Gabius H-J (2019) The sugar code: 1007/s00418-016-1522-8
letters and vocabulary, writers, editors and 11. Kasai K-i (2018) Galectins: quadruple-faced
readers and biosignificance of functional proteins. Trends Glycosci Glycotechnol
glycan-lectin pairing. Biochem J 30(172):SE221–SE223. https://doi.org/10.
476(18):2623–2655. https://doi.org/10. 4052/tigg.1745.7SE
1042/BCJ20170853 12. Garcı́a Caballero G, Kaltner H, Kutzner TJ,
6. Hirabayashi J (1997) Recent topics on galec- Ludwig A-K, Manning JC, Schmidt S,
tins. Trends Glycosci Glycotechnol Sinowatz F, Gabius H-J (2020) How galectins
9(45):1–180 have become multifunctional proteins. Histol
7. Liu F-T, Patterson RJ, Wang JL (2002) Intra- Histopathol 35(6):509–539. https://doi.org/
cellular functions of galectins. Biochim Bio- 10.14670/HH-18-199
phys Acta 1572(2–3):263–273. https://doi. 13. Flotte TJ, Springer TA, Thorbecke GJ (1983)
org/10.1016/S0304-4165(02)00313-6 Dendritic cell and macrophage staining by
monoclonal antibodies in tissue sections and 23. Barton KA, Hsu CD, Petrash JM (2009) Inter-
epidermal sheets. Am J Pathol actions between small heat shock protein
111(1):112–124 α-crystallin and galectin-related interfiber pro-
14. Gabius H-J, Brehler R, Schauer A, Cramer F tein (GRIFIN) in the ocular lens. Biochemistry
(1986) Localization of endogenous lectins in 48(18):3956–3966. https://doi.org/10.
normal human breast, benign breast lesions 1021/bi802203a
and mammary carcinomas. Virch Arch [Cell 24. Garcı́a Caballero G, Schmidt S, Schnölzer M,
Pathol] 52(2):107–115. https://doi.org/10. Schlötzer-Schrehardt U, Knospe C, Ludwig
1007/BF02889955 A-K, Manning JC, Muschler P, Kaltner H,
15. Allen HJ, Sucato D, Gottstine S, Kisailus E, Kopitz J, Gabius H-J (2019) Chicken GRI-
Nava H, Petrelli N, Castillo N, Wilson D FIN: binding partners, developmental course
(1991) Localization of endogenous of localization and activation of its lens-specific
β-galactoside-binding lectin in human cells gene expression by L-Maf/Pax6. Cell Tissue
and tissues. Tumour Biol 12(1):52–60. Res 375(3):665–683. https://doi.org/10.
https://doi.org/10.1159/000217688 1007/s00441-018-2931-x
16. Sakakura Y, Hirabayashi J, Oda Y, Ohyama Y, 25. Garcı́a Caballero G, Schmidt S, Manning JC,
Kasai K-i (1990) Structure of chicken 16-kDa Michalak M, Schlötzer-Schrehardt U, Ludwig
β-galactoside-binding lectin. Complete amino A-K, Kaltner H, Sinowatz F, Schnölzer M,
acid sequence, cloning of cDNA and produc- Kopitz J, Gabius H-J (2020) Chicken lens
tion of recombinant lectin. J Biol Chem development: complete signature of expression
265(35):21573–21579 of galectins during embryogenesis and evi-
17. Beyer EC, Barondes SH (1982) Secretion of dence for their complex formation with α-, β-,
endogenous lectin by chicken intestinal goblet δ- and τ-crystallins, N-CAM, and N-cadherin
cells. J Cell Biol 92(1):28–33. https://doi. obtained by affinity chromatography. Cell Tis-
org/10.1083/jcb.92.1.23 sue Res 379(1):13–35. https://doi.org/10.
1007/s00441-019-03129-0
18. Kaltner H, Gabius H-J (2012) A toolbox of
lectins for translating the sugar code: the galec- 26. Toegel S, Bieder D, André S, Kayser K, Walzer
tin network in phylogenesis and tumors. Histol SM, Hobusch G, Windhager R, Gabius H-J
Histopathol 27(4):397–416. https://doi.org/ (2014) Human osteoarthritic knee cartilage:
10.14670/HH-27.397 fingerprinting of adhesion/growth-regulatory
galectins in vitro and in situ indicates differen-
19. Amano M, Eriksson H, Manning JC, Detjen tial upregulation in severe degeneration. His-
KM, André S, Nishimura S-I, Lehtiö J, Gabius tochem Cell Biol 142(4):373–388. https://
H-J (2012) Tumour suppressor p16INK4a: doi.org/10.1007/s00418-014-1234-x
anoikis-favouring decrease in N/O-glycan/
cell surface sialylation by down-regulation of 27. Weinmann D, Kenn M, Schmidt S, Schmidt K,
enzymes in sialic acid biosynthesis in tandem Walzer SM, Kubista B, Windhager R,
in a pancreatic carcinoma model. FEBS J Schreiner W, Toegel S, Gabius H-J (2018)
279(21):4062–4080. https://doi.org/10. Galectin-8 induces functional disease markers
1111/febs.12001 in human osteoarthritis and cooperates with
galectins-1 and -3. Cell Mol Life Sci
20. Ledeen RW, Kopitz J, Abad-Rodrı́guez J, 75(22):4187–4205. https://doi.org/10.
Gabius H-J (2018) Glycan chains of ganglio- 1007/s00018-018-2856-2
sides: functional ligands for tissue lectins
(siglecs/galectins). Progr Mol Biol Transl Sci 28. Chiariotti L, Salvatore P, Frunzio R, Bruni CB
156:289–324. https://doi.org/10.1016/bs. (2004) Galectin genes: regulation of expres-
pmbts.2017.12.004 sion. Glycoconj J 19(7–9):441–449. https://
doi.org/10.1023/B:GLYC.0000014073.
21. Garcı́a Caballero G, Manning JC, Ludwig A-K, 23096.3a
Ruiz FM, Romero A, Kaltner H, Gabius H-J
(2018) Members of the galectin network with 29. Lu Y, Lotan R (1999) Transcriptional regula-
deviations from the canonical sequence signa- tion by butyrate of mouse galectin-1 gene in
ture. 1. Galectin-related inter-fiber protein embryonal carcinoma cells. Biochim Biophys
(GRIFIN). Trends Glycosci Glycotechnol Acta 1444(1):85–91. https://doi.org/10.
30(172):SE1–SE9 1016/s0167-4781(98)00257-7
22. Ogden AT, Nunes I, Ko K, Wu S, Hines CS, 30. Gitt MA, Barondes SH (1991) Genomic
Wang AF, Hegde RS, Lang RA (1998) GRI- sequence and organization of two members of
FIN, a novel lens-specific protein related to the a human lectin gene family. Biochemistry
galectin family. J Biol Chem 30(1):82–89. https://doi.org/10.1021/
273(44):28889–28896. https://doi.org/10. bi00215a013
1074/jbc.273.44.28889
31. Goldstone SD, Lavin MF (1991) Isolation of a Lehmann WD, André S, Solı́s D, Gabius H-J
cDNA clone, encoding a human β-galactoside- (2011) Toward comprehensive analysis of the
binding protein, overexpressed during galectin network in chicken: unique diversity of
glucocorticoid-induced cell death. Biochem galectin-3 and comparison of its localization
Biophys Res Commun 178(2):746–750. profile in organs of adult animals to the other
https://doi.org/10.1016/0006-291x(91) four members of this lectin family. Anat Rec
90171-3 294(3):427–444. https://doi.org/10.1002/
32. Kadrofske MM, Openo KP, Wang JL (1998) ar.21341
The human LGALS3 (galectin-3) gene: deter- 34. Katzenmaier E-M, André S, Kopitz J, Gabius
mination of the gene structure and functional H-J (2014) Impact of sodium butyrate on the
characterization of the promoter. Arch Bio- network of adhesion/growth-regulatory galec-
chem Biophys 349(1):7–20. https://doi.org/ tins in human colon cancer in vitro. Anticancer
10.1006/ABBI.1997.0447 Res 34(10):5429–5438
33. Kaltner H, Kübler D, López-Merino L,
Lohr M, Manning JC, Lensch M, Seidler J,
Chapter 25
Examining the Impact of Galectin-9 on Latent HIV

Transcription
Opeyemi S. Adeniji, Leila B. Giron, and Mohamed Abdel-Mohsen
Abstract
The β-galactoside-binding protein Galectin-9 (Gal-9) functions as a double-edged sword during HIV
infection. On the one hand, Gal-9 can reactivate HIV latently infected cells, the main barrier to achieving
HIV eradication, making them visible to immune clearance. On the other hand, Gal-9 induces latent HIV
transcription by activating T cell Receptor (TCR) signaling pathways. These signaling pathways induce
undesirable pro-inflammatory responses. While these unwanted responses can be mitigated by rapamycin
without impacting Gal-9-mediated latent HIV reactivation, this effect raises the concern that Gal-9 may
play a role in the chronic immune activation/inflammation that persists in people living with HIV despite
antiretroviral therapy. Together, these data highlight the need to understand the positive and negative
impacts of galectin interactions on immunological functions during HIV infection. In this chapter, we
describe methods that can be used to investigate the effects of galectins, in particular Gal-9, on latent HIV
transcription in vitro and ex vivo.
Key words Galectin-9, Galectins, HIV persistence, HIV, Latent HIV transcription, J-Lat cell line
1 Introduction
Antiretroviral therapy (ART) does not eradicate HIV due to the

persistence of long-lived CD4+ T cells carrying integrated viral
genomes that do not express viral transcripts or proteins (latent
infection). These cells carrying latent proviruses are hidden from
the immune system leading to persistent infection [1, 2]. This
persistent infection leads to a state of chronic inflammation and
immune dysfunction that has been linked to several comorbidities
prevalent in people living with HIV despite long-term ART
[3, 4]. Several of these HIV-associated immune dysfunctions can
be linked to interactions between glycans and glycan-binding pro-
teins (lectins). Glycan-lectin interactions on cell surface can activate
Opeyemi S. Adeniji and Leila B. Giron contributed equally to this chapter and are listed alphabetically by last
name.
463
464 Opeyemi S. Adeniji et al.
or inhibit downstream signaling pathways that control a variety of

cellular processes and immunological functions [5–8]. For exam-
ple, β-galactoside glycans on cell surface binding to galectins pro-
mote immune evasion by inducing T cell exhaustion and apoptosis,
expanding regulatory T cells (Tregs), and inhibiting natural killer
(NK) cells [9–13].
Among galectins, galectin-9 (Gal-9) plays an important role in
regulating adaptive and innate defense mechanisms and has been
linked in HIV pathogenesis [14–16]. The secretion of endogenous
Gal-9 rapidly increases after HIV infection, and the elevated levels
of endogenous Gal-9 do not return to normal after ART
[15]. Galectin-9 has several effects on T cells; it induces apoptosis
[17]; activates cells through T cell Receptor (TCR) signaling [18],
renders CD4+ T cells less susceptible to HIV infection by inducing
the expression of anti-HIV host restriction factors [19], but it also
increases HIV entry by inducing the cell-surface concentration of
protein disulfide isomerase (PDI) [20].
We recently found that the endogenous plasma levels of Gal-9
are associated with HIV transcription levels during ART-suppressed
HIV infection [21]. We have also found that exogenous Gal-9
treatment induces latent HIV transcription in vitro and ex vivo
[21]. This ability of Gal-9 to induce latent HIV transcription
suggested that it can be considered within the “shock and kill”
HIV eradication framework [21]. The “shock and kill” strategy is
one of the approaches currently being considered to eliminate the
HIV reservoir. In this strategy, latency-reversing agents (LRAs) are
administered to force HIV latently infected cells to produce virus
and become visible to the immune system. The premise is that
immune cells would recognize these cells for immune-mediated
clearance, while uninfected cells will be protected from the reacti-
vated virus by ART [22, 23]. More recently, we identified the
mechanism underlying the Gal-9-mediated reactivation of latent
HIV [24]. We found that Gal-9 modulates HIV transcription by
activating ERK and CREB signaling pathways in an Lck-dependent
manner [24]. We also found that this signaling pathway not only
induces HIV transcription but also induces an undesirable
pro-inflammatory response that can be mitigated by rapamycin
(an mTOR inhibitor), without impacting Gal-9-mediated reactiva-
tion of latent HIV. Uncoupling Gal-9-mediated viral reactivation
from undesirable pro-inflammatory effects, using rapamycin, may
increase the potential utility of Gal-9 within the reversal of HIV
latency eradication framework. However, the Gal-9-mediated
induction of TCR signaling and the downstream ERK and CREB
pathways raises an important concern that Gal-9 may play a role in
maintaining chronic T cell activation during ART-suppressed HIV
infection. The elevated immune activation during ART-suppressed
HIV infection is associated with the development of several comor-
bidities in people living with HIV [25–28]. Together, these data
highlight the need to understand the positive and negative impacts
Examining the Impact of Galectin-9 on Latent HIV Transcription 465
of galectin interactions on immunological functions during HIV

infection. In this chapter, we describe methods that can be used to
investigate the effects of galectins, in particular Gal-9, on latent
HIV transcription in vitro and ex vivo.
2 Materials
2.1 Cell Lines In vitro experiments can be done using the established “J-Lat”
model of HIV latency. There are several clones of this cell line
including:
1. 5A8 (can be provided by Dr. Warner Greene at the Gladstone
Institute of Virology and Immunology).
2. 6.3 (NIH AIDS Reagent Program (Germantown, MD);
catalog# 9846).
catalog# 9847).
catalog# 9848).
catalog# 9849).
catalog# 9850).
2.2 General 1. Roswell Park Memorial Institute (RPMI) medium with L-glu-
Reagents tamine (Corning Cellgro, Tewksbury, MA; catalog#
10-040-CM).
2. Fetal Bovine Serum (FBS) (Gibco, Thermo Fisher Scientific,
Waltham, MA; catalog# 26140079).
3. Penicillin/Streptomycin (Thermo Fisher Scientific, Waltham,
MA; catalog# 15140148).
4. A stable form of recombinant galectin-9 (developed by partial
deletion of the linker peptide) [29], which can be obtained
through GalPharma Co., Ltd. (Kagawa, Japan). Alternatively, a
natural form of recombinant galectin-9 can be purchased from
R&D Systems (E. coli-derived human Galectin-9 protein; Min-
neapolis, MN; catalog # 2045-GA-050).
5. InSolution Rapamycin (Millipore Sigma (Burlington, MA; cat-
alog# 553211).
6. Phorbol myristate acetate (PMA) (Sigma-Aldrich, St. Louis,
MO; catalog# P8139).
7. Ionomycin (Sigma-Aldrich, St. Louis, MO; catalog# I0634).
8. Dimethyl Sulfoxide (DMSO) (Sigma-Aldrich, St. Louis, MO;
catalog# 20-139).
9. Cell-culture treated 24-well flat-bottom plates (Fisher Scien-

tific, Waltham, MA; catalog# 12-565-163).
10. Cell-culture treated 6-well flat-bottom plates (Fisher Scientific,
Waltham, MA; catalog# 14-832-11).
2.3 Reagents for 1. ImmunoCult Human CD3/CD28 T Cell Activator (STEM-

Testing the Impact of CELL, Vancouver, BC, Canada; catalog# 10971).
Gal-9 on Latent HIV 2. Human Recombinant TNF-alpha (STEMCELL, Vancouver,
Transcription In Vitro BC, Canada; catalog# 78068.1).
3. 5 ml FACS tubes (VWR, Radnor, PA; catalog# 60818-565).
4. LSR II flow cytometer and FACSDiva software (Becton Dick-
inson, Mountain View, CA) or equivalent flow cytometer.
2.4 Reagents for 1. SepMate™-50 (STEMCELL, Vancouver, BC, Canada; cata-

Testing the Impact of log# 85450).
Gal-9 on Latent HIV 2. EasySep Human CD4+ T Cell Enrichment Kit (STEMCELL,
Transcription Ex Vivo Vancouver, BC, Canada; catalog# 17952).
Using CD4+ T Cells 3. Dynabeads™ Human T-Activator CD3/CD28 (Thermo
Isolated from HIV- Fisher Scientific, Waltham, MA; catalog# 11132D).
Infected ART-
4. AllPrep DNA/RNA/miRNA Universal Kit (Qiagen, Hilden,
Suppressed
Germany; catalog# 80224).
Individuals
5. Nuclease-free water (Ambion, Austin, TX; catalog# AM9938).
6. TaqMan® RNA-to-CT™ 1-Step Kit Life (Thermo Fisher Sci-
entific, Waltham, MA; catalog# 4392938).
7. 1.5 ml LoBindtubes (Fisher Scientific, Waltham, MA; catalog#
13–698-791).
8. MicroAmp® Fast Optical 96-Well Reaction Plate (Thermo
Fisher Scientific, Waltham, MA; catalog# 4346906).
9. MicroAmp® Optical Adhesive Film (Applied Biosystems
Thermo Fisher Scientific, Waltham, MA; catalog# 4311971).
10. Pipette tips with an aerosol barrier (plugged, DNA/RNAse-
free).
11. Forward qPCR primer “F522–43” (50 GCC TCA ATA AAG
CTT GCC TTG A 30 ) [30].
12. Reverse qPCR primer “R626-43” (50 GGG CGC CAC TGC
TAG AGA 30 ) [30].
13. qPCR Taqman Probe “Kumar” (FAM/ZEN/BHQ-1) probe
(50 CCA GAG TCA CAC AAC AGA CGG GCA CA 30 ) [30].
14. RPLP0 Primer/probe mix (Thermo Fisher Scientific, Wal-
tham, MA; catalog# 4310879E).
15. Centrifuge which can hold 96-well reaction plate.
16. StepOnePlus™ (or equivalent instrument).
3 Methods
3.1 Testing the 1. In vitro experiments can be performed using the “J-Lat” model
Impact of Gal-9 on of HIV latency. J-Lat cells harbor latent, transcriptionally com-
Latent HIV petent HIV provirus that encodes green fluorescent protein
Transcription In Vitro (GFP) as an indicator of viral reactivation [31, 32]. Levels of
Using the J-Lat HIV latent HIV transcription after stimulation can be measured
Latency Model using flow cytometry.
2. Maintain the J-Lat cell line (different clones, see Note 1) in
Roswell Park Memorial Institute (RPMI) medium supplemen-
ted with L-glutamine, 10% Fetal Bovine Serum (FBS), and 1%
Penicillin/Streptomycin at 106 cells/ml in a 24-well flat-bot-
tom plate.
3. Treat the J-Lat cells with escalating concentrations of Gal-9
(in PBS buffer) (50–500 nM), 25 μl/ml of ImmunoCult
human CD3/CD28 T Cell Activator (positive control for the
5A8 clone), PMA/ionomycin (16 nM/500 nM; positive con-
trol for all clones), TNFα (10 ng/ml; positive control for all
clones), or 0.2–0.5% DMSO (negative control) for 24, 48, or
72 h.
4. These treatments can be done with and without a
pre-incubation step (for 1 h) with 5 μM for rapamycin.
5. Assess the percentage of GFP+ cells and mean fluorescence
intensity of GFP expression using LSR II flow cytometer and
FACSDiva software (Becton Dickinson, Mountain View, CA)
or equivalent flow cytometer.
6. Analyze data with FlowJo (TreeStar Inc., Ashland, OR). See an
example in Fig. 1.
3.2 Testing the 1. Collect fresh blood (100 ml) from HIV-infected ART-sup-
Impact of Gal-9 on pressed individuals.
Latent HIV 2. Isolate peripheral blood mononuclear cells (PBMCs) from
Transcription Ex Vivo whole blood using SepMate (STEMCELL Technologies) or
Using CD4+ T Cells Ficoll density gradient method.
Isolated from HIV- 3. Isolate CD4+ T cells from PBMCs using negative selection
Infected ART- (EasySep, STEMCELL Technologies) according to the manu-
Suppressed facturer’s instructions.
Individuals
4. Maintain primary CD4+ T cells in RPMI supplemented with L-
3.2.1 Isolation and glutamine, 20% FBS at 106 cells/ml in a 6-well or 24-well flat-
Treatment of Primary CD4+ bottom plate.
T Cells (See Note 2) 5. Primary CD4+ T cells can be pretreated (for 1 h) or not with
5 μM for rapamycin.
6. Treat cells with DMSO 0.2–0.5% as a negative control, PMA at
2 nM, and Ionomycin at 0.5 μM (as a positive control), one cell
J-Lat HIV Latency Model

DMSO control 500nM Gal-9
250K 250K
0.2% 13.3%
200K 200K
Clone 5A8
150K 150K
100K 100K
50K 50K
0 0
0 102 103 104 105 0 102 103 104 105
250K 250K
0.03% 7.6%
Clone 6.3
200K 200K
150K 150K
100K 100K
50K 50K
0 0
0 102 103 104 105 0 102 103 104 105
250K 250K
1.74% 40.8%
Clone 11.1
200K 200K
150K 150K
100K 100K
50K 50K
SSC-A
0 0
0 102 103 104 105 0 102 103 104 105
GFP
Fig. 1 An example of the reactivation of latent HIV transcription by Gal-9 in vitro

using the J-Lat cells. J-Lat 5A8 cells, 6.3 cells, and 11.1 cells were treated with
500 nM Gal-9 for 24 h. Cells were analyzed by flow cytometry to assess
HIV-encoded GFP expression. SSC side scatter
to one Dynabead Human T-Activator CD3/CD28 (as another

potential positive control), or 500–1000 nM of Gal-9 (in PBS
buffer) for 24 h (see Note 3).
3.2.2 Isolation of RNA Extract total RNA from more than 1 106 CD4+ T cells using the
from CD4+ T Cells AllPrep DNA/RNA/miRNA Universal Kit (Qiagen) with the
optional on-column DNase treatment step, following the manu-
facturer’s instructions. Other Qiagen column-based kits or Trizol-
based extraction methods may also be used.
3.2.3 Measuring Levels 1. Quantify RNA amount by spectrophotometry and normalize

of CD4+ T Cell-Associated RNA samples to 50–100 ng/μl. Samples can be concentrated
HIV RNA (See Notes 4 using a SpeedVac vacuum concentrator, if needed or diluted in
and 5) nuclease-free water.
2. RNA samples should be tested in triplicate for HIV-1 RNA
quantification and duplicate for RPLP0 housekeeping gene
quantification: 5 μl for each assay (250–500 ng/5 μl), with a
total of 25 μl needed for both assays.
3. Keep RNA on ice until further use.
4. In PCR template-free area, prepare the master mix for the HIV
RNA quantification plate and RPLP0 plate using cold block.
5. Calculate master mix volume requirement, enough for
unknown samples and negative control in triplicate.
6. Prepare master mix in sterile, RNAse-free low-binding 1.5 ml
tube according to Table 1 for HIV quantification and Table 2
for RPLP0.
7. Distribute 15 μl of the master mix to each designated well of
Fast Optical 96-well reaction plate using automatic pipettor.
8. Move plate to template-allowed area.
9. Add 5 μl of each sample and water to the negative control
designated well, mix by pipetting up and down.
10. Seal plate carefully, centrifuge for 2 min.
11. Perform RT PCR using StepOnePlus™ (or equivalent instru-
ment) under the conditions described in Tables 3 and 4.
12. Analyze data using the delta-delta Ct method, also known as
the 2–ΔΔCt method, to calculate the relative levels of CD4+ T
cell-associated HIV RNA and fold induction of HIV RNA (see
an example in Fig. 2). Higher levels of CD4+ T cell-associated
HIV RNA compared to negative controls is indicative of reac-
tivated HIV transcription.
4 Notes
1. There are several clones of this cell line. These clones respond
differently to stimulation, likely reflecting different molecular
mechanisms maintaining HIV latency in them. For example,
the 5A8 clone responds well to αCD3/αCD28 stimulation,
while other clones are not responsive to the same stimulation.
Another example, some clones, such as 11.1, exhibit a higher
HIV transcription background without stimulation when com-
pared to other clones, such as 6.3 or 5A8. It is recommended to
test several clones [21].
Table 1
HIV RNA quantification master mix
(1 plate)
Reagent 1 reaction 96-well plate Final conc.
Nuclease-free water 3.3 μl 330 μl –
RNA to CT master mix 10 μl 1000 μl 1
F522-43 primer 0.4 μl 40 μl 0.2 μM
(10 μM—working stock)
R626-43 primer 0.4 μl 40 μl 0.2 μM
Kumar probe 0.4 μl 40 μl 0.2 μM
Reverse transcriptase 0.5 μl 50 μl 1
Final volume 15 μl 1500 μl –
Table 2
RPLP0 quantification master mix
(1 plate)
Reagent 1 reaction 96-well plate Final conc.
Nuclease-free water 3.5 μl 350 μl –
RNA to CT Master Mix 10 μl 1000 μl 1
RPLP0 primers/probe mastermix 1 μl 100 μl 1
Reverse transcriptase 0.5 μl 50 μl 1
Final volume 15 μl 1500 μl –
Table 3
Cycling conditions for the HIV-1 RNA qPCR
Initiation Enzyme activation PCR

48 C, 95 C, 95 C,
20 min 10 min 15 s
59 C,
1 min
1 cycle 1 cycle 60 cycles
2. Ex vivo experiments were described to be performed using

total CD4+ T cells; however, they can also be performed
using resting CD4+ T cells (CD4+ T cells that do not express
CD69, CD25, or HLA-DR activation markers). Resting CD4+
T cells can be further enriched from total CD4+ T cells through
Table 4
Cycling conditions for the RPLP0 RNA qPCR
Initiation Enzyme activation PCR

48 C, 95 C, 95 C,
20 min 10 min 15 s
60 C,
1 min
1 cycle 1 cycle 40–60 cycles
100
Fold induction of HIV RNA
10
1
l
-9
I
tro
A/
tin
PM
on
ec
C
al
G
Fig. 2 An example of the reactivation of HIV transcription by Gal-9 ex vivo using

primary CD4+ T cells isolated from ART-suppressed HIV-infected individuals.
DMSO 0.5% was used as a control, PMA/ionomycin (2 nM/500 nM) was used as
a positive control, and 500 nM of Gal-9 was used. Fold increase in HIV RNA was
determined relative to the corresponding DMSO-treated control for each
individual. Statistical significance was calculated using Kruskal–Wallis test.
*** ¼ p < 0.001, **** ¼ p < 0.0001
depletion of cells expressing CD69, CD25, or HLA-DR sur-

face markers using the CD69 MicroBead Kit II (Miltenyi Bio-
tec; catalog number 130-092-355); CD25 MicroBeads II
(Miltenyi Biotec; catalog number 130-092-983); and anti-

HLA-DR MicroBeads (Miltenyi Biotec; catalog number
130-046-101).
3. The synergy between Gal-9 and other latency-reversing agents
can be calculated using the Bliss independence model [33] as
implemented by Jiang et al. and Laird et al. [34, 35].
4. The impact of Gal-9 on CD4+ T cell activation can be examined
by measuring the co-expression of T cell surface activation
markers (CD69 and CD25) using flow cytometry.
5. Workflow for qPCR assay described above to measure the
relative levels of CD4+ T cell-associated HIV RNA must pro-
ceed in a unidirectional flow beginning in the pre-amplification
(template-free), moving to the post-amplification/low copy
areas then post-amplification/high copy detection.
Pre-amplification activities must begin with reagent prepara-
tion and proceed to specimen preparation/amplification.
Supplies and equipment must be dedicated to each
pre-amplification activity and not used for other activities or
moved between areas. Gloves and lab coats must be worn in
each area and must be changed before leaving that area. Equip-
ment and supplies used for reagent preparation must not be
used for specimen preparation/amplification activities or for
pipetting or processing amplified DNA or other sources of
target DNA. Post-amplification supplies and equipment must
remain in the post-amplification area at all times.
6. The absolute, instead of relative, levels of CD4+ T cell-
associated HIV RNA can be calculated by including calibrated
standards of HIV RNA and human RPLP0 in the qPCR assay.
Acknowledgments
M.A.-M. is supported by NIH grants (R01 DK123733, R01

AG062383, R01 NS117458, R21 AI143385, R21 AI129636,
and R21 NS106970), The Foundation for AIDS Research
(amfAR) impact grant # 109840-65-RGRL, and W.W. Smith Char-
itable Trust grant # A1901.
References
1. Wong JK, Hezareh M, Gunthard HF, Havlir Gallant J, Markowitz M, Ho DD, Richman
DV, Ignacio CC, Spina CA, Richman DD DD, Siliciano RF (1997) Identification of a
(1997) Recovery of replication-competent reservoir for HIV-1 in patients on highly active
HIV despite prolonged suppression of plasma antiretroviral therapy. Science
viremia. Science 278(5341):1291–1295 278(5341):1295–1300
2. Finzi D, Hermankova M, Pierson T, Carruth 3. Wandeler G, Johnson LF, Egger M (2016)
LM, Buck C, Chaisson RE, Quinn TC, Trends in life expectancy of HIV-positive
Chadwick K, Margolick J, Brookmeyer R, adults on antiretroviral therapy across the
globe: comparisons with general population. 13. Smith LK, Boukhaled GM, Condotta SA,
Curr Opin HIV AIDS 11(5):492–500. Mazouz S, Guthmiller JJ, Vijay R, Butler NS,
h t t p s : // d o i . o r g / 1 0 . 1 0 9 7 / C O H . Bruneau J, Shoukry NH, Krawczyk CM,
0000000000000298 Richer MJ (2018) Interleukin-10 directly inhi-
4. Rodger AJ, Lodwick R, Schechter M, Deeks S, bits CD8(+) T cell function by enhancing
Amin J, Gilson R, Paredes R, Bakowska E, N-glycan branching to decrease antigen sensi-
Engsig FN, Phillips A, Insight Smart ESG tivity. Immunity 48(2):299–312.e5. https://
(2013) Mortality in well controlled HIV in doi.org/10.1016/j.immuni.2018.01.006
the continuous antiretroviral therapy arms of 14. Sehrawat S, Reddy PB, Rajasagi N,
the SMART and ESPRIT trials compared with Suryawanshi A, Hirashima M, Rouse BT
the general population. AIDS 27(6):973–979. (2010) Galectin-9/TIM-3 interaction regu-
h t t p s : // d o i . o r g / 1 0 . 1 0 9 7 / Q A D . lates virus-specific primary and memory CD8
0b013e32835cae9c T cell response. PLoS Pathog 6(5):e1000882.
5. Barrera C, Espejo R, Reyes VE (2002) Differ- https://doi.org/10.1371/journal.ppat.
ential glycosylation of MHC class II molecules 1000882
on gastric epithelial cells: implications in local 15. Tandon R, Chew GM, Byron MM, Borrow P,
immune responses. Hum Immunol Niki T, Hirashima M, Barbour JD, Norris PJ,
63(5):384–393 Lanteri MC, Martin JN, Deeks SG, Ndhlovu
6. de Freitas Junior JC, Silva Bdu R, de Souza LC (2014) Galectin-9 is rapidly released during
WF, de Araujo WM, Abdelhay ES, Morgado- acute HIV-1 infection and remains sustained at
Diaz JA (2011) Inhibition of N-linked glyco- high levels despite viral suppression even in
sylation by tunicamycin induces E-cadherin- elite controllers. AIDS Res Hum Retrovir
mediated cell-cell adhesion and inhibits cell 30(7):654–664. https://doi.org/10.1089/
proliferation in undifferentiated human colon AID.2014.0004
cancer cells. Cancer Chemother Pharmacol 16. Jost S, Moreno-Nieves UY, Garcia-Beltran WF,
68(1):227–238. https://doi.org/10.1007/ Rands K, Reardon J, Toth I, Piechocka-
s00280-010-1477-8 Trocha A, Altfeld M, Addo MM (2013) Dysre-
7. Dwek RA, Butters TD, Platt FM, Zitzmann N gulated Tim-3 expression on natural killer cells
(2002) Targeting glycosylation as a therapeutic is associated with increased Galectin-9 levels in
approach. Nat Rev Drug Discov 1(1):65–75. HIV-1 infection. Retrovirology 10:74.
https://doi.org/10.1038/nrd708 https://doi.org/10.1186/1742-4690-10-74
8. Walt D, Aoki-Kinoshita KF, Bendiak B, Ber- 17. Lanteri M, Giordanengo V, Hiraoka N, Fuzi-
tozzi CR, Boons GJ, Darvill A, Hart G, Kies- bet JG, Auberger P, Fukuda M, Baum LG,
sling LL, Lowe J, Moon R, Paulson J, Lefebvre JC (2003) Altered T cell surface gly-
Sasisekharan R, Varki A, Wong CH (2012) cosylation in HIV-1 infection results in
Transforming glycoscience: a roadmap for the increased susceptibility to galectin-1-induced
future. National Academies Press, Washington cell death. Glycobiology 13(12):909–918.
9. Mendez-Huergo SP, Blidner AG, Rabinovich https://doi.org/10.1093/glycob/cwg110
GA (2017) Galectins: emerging regulatory 18. Lhuillier C, Barjon C, Niki T, Gelin A, Praz F,
checkpoints linking tumor immunity and Morales O, Souquere S, Hirashima M, Wei M,
angiogenesis. Curr Opin Immunol 45:8–15. Dellis O, Busson P (2015) Impact of exoge-
https://doi.org/10.1016/j.coi.2016.12.003 nous galectin-9 on human T cells: contribution
10. Zhuo Y, Bellis SL (2011) Emerging role of of the T cell receptor complex to antigen-
alpha2,6-sialic acid as a negative regulator of independent activation but not to apoptosis
galectin binding and function. J Biol Chem induction. J Biol Chem
286(8):5935–5941. https://doi.org/10. 290(27):16797–16811. https://doi.org/10.
1074/jbc.R110.191429 1074/jbc.M115.661272
11. Barondes SH, Cooper DN, Gitt MA, Leffler H 19. Elahi S, Niki T, Hirashima M, Horton H
(1994) Galectins. Structure and function of a (2012) Galectin-9 binding to Tim-3 renders
large family of animal lectins. J Biol Chem activated human CD4+ T cells less susceptible
269(33):20807–20810 to HIV-1 infection. Blood
119(18):4192–4204. https://doi.org/10.
12. Gordon-Alonso M, Hirsch T, Wildmann C, 1182/blood-2011-11-389585
van der Bruggen P (2017) Galectin-3 captures
interferon-gamma in the tumor matrix reduc- 20. Schaefer K, Webb NE, Pang M, Hernandez-
ing chemokine gradient production and T-cell Davies JE, Lee KP, Gonzalez P, Douglass MV,
tumor infiltration. Nat Commun 8(1):793. Lee B, Baum LG (2017) Galectin-9 binds to
https://doi.org/10.1038/s41467-017- O-glycans on protein disulfide isomerase.
00925-6
Glycobiology 27(9):878–887. https://doi. 28. Kaplan RC, Sinclair E, Landay AL, Lurain N,
org/10.1093/glycob/cwx065 Sharrett AR, Gange SJ, Xue X, Parrinello CM,
21. Abdel-Mohsen M, Chavez L, Tandon R, Chew Hunt P, Deeks SG, Hodis HN (2011) T cell
GM, Deng X, Danesh A, Keating S, Lanteri M, activation predicts carotid artery stiffness
Samuels ML, Hoh R, Sacha JB, Norris PJ, among HIV-infected women. Atherosclerosis
Niki T, Shikuma CM, Hirashima M, Deeks 217(1):207–213. https://doi.org/10.1016/j.
SG, Ndhlovu LC, Pillai SK (2016) Human atherosclerosis.2011.03.011
galectin-9 is a potent mediator of HIV tran- 29. Niwa H, Satoh T, Matsushima Y, Hosoya K,
scription and reactivation. PLoS Pathog 12(6): Saeki K, Niki T, Hirashima M, Yokozeki H
e1005677. https://doi.org/10.1371/journal. (2009) Stable form of galectin-9, a Tim-3
ppat.1005677 ligand, inhibits contact hypersensitivity and
22. Siliciano JD, Siliciano RF (2013) HIV-1 eradi- psoriatic reactions: a potent therapeutic tool
cation strategies: design and assessment. Curr for Th1- and/or Th17-mediated skin inflam-
Opin HIV AIDS 8(4):318–325. https://doi. mation. Clin Immunol 132(2):184–194.
org/10.1097/COH.0b013e328361eaca https://doi.org/10.1016/j.clim.2009.04.012
23. Bullen CK, Laird GM, Durand CM, Siliciano 30. Kumar AM, Borodowsky I, Fernandez B,
JD, Siliciano RF (2014) New ex vivo Gonzalez L, Kumar M (2007) Human immu-
approaches distinguish effective and ineffective nodeficiency virus type 1 RNA levels in differ-
single agents for reversing HIV-1 latency ent regions of human brain: quantification
in vivo. Nat Med 20(4):425–429. https://doi. using real-time reverse transcriptase-
org/10.1038/nm.3489 polymerase chain reaction. J Neurovirol
24. Colomb F, Giron LB, Premeaux TA, Mitchell 13(3):210–224. https://doi.org/10.1080/
BI, Niki T, Papasavvas E, Montaner LJ, 13550280701327038
Ndhlovu LC, Abdel-Mohsen M (2019) 31. Chavez L, Kauder S, Verdin E (2011) In vivo,
Galectin-9 mediates HIV transcription by in vitro, and in silico analysis of methylation of
inducing TCR-dependent ERK signaling. the HIV-1 provirus. Methods 53(1):47–53.
Front Immunol 10:267. https://doi.org/10. https://doi.org/10.1016/j.ymeth.2010.
3389/fimmu.2019.00267 05.009
25. Hunt PW, Martin JN, Sinclair E, Bredt B, 32. Jordan A, Bisgrove D, Verdin E (2003) HIV
Hagos E, Lampiris H, Deeks SG (2003) T cell reproducibly establishes a latent infection after
activation is associated with lower CD4+ T cell acute infection of T cells in vitro. EMBO J
gains in human immunodeficiency virus- 22(8):1868–1877. https://doi.org/10.1093/
infected patients with sustained viral suppres- emboj/cdg188
sion during antiretroviral therapy. J Infect Dis 33. Bliss CI (1939) The toxicity of poisons applied
187(10):1534–1543. https://doi.org/10. jointly. Ann Appl Biol 26:585–615
1086/374786 34. Jiang G, Mendes EA, Kaiser P, Wong DP,
26. Hunt PW, Cao HL, Muzoora C, Ssewanyana I, Tang Y, Cai I, Fenton A, Melcher GP, Hildreth
Bennett J, Emenyonu N, Kembabazi A, Nei- JE, Thompson GR, Wong JK, Dandekar S
lands TB, Bangsberg DR, Deeks SG, Martin (2015) Synergistic reactivation of latent HIV
JN (2011) Impact of CD8+ T-cell activation on expression by ingenol-3-angelate, PEP005,
CD4+ T-cell recovery and mortality in targeted NF-kB signaling in combination with
HIV-infected Ugandans initiating antiretrovi- JQ1 induced p-TEFb activation. PLoS Pathog
ral therapy. AIDS 25(17):2123–2131. https:// 11(7):e1005066. https://doi.org/10.1371/
doi.org/10.1097/QAD.0b013e32834c4ac1 journal.ppat.1005066
27. Kaplan RC, Sinclair E, Landay AL, Lurain N, 35. Laird GM, Bullen CK, Rosenbloom DI, Mar-
Sharrett AR, Gange SJ, Xue X, Hunt P, tin AR, Hill AL, Durand CM, Siliciano JD,
Karim R, Kern DM, Hodis HN, Deeks SG Siliciano RF (2015) Ex vivo analysis identifies
(2011) T cell activation and senescence predict effective HIV-1 latency-reversing drug combi-
subclinical carotid artery disease in nations. J Clin Invest 125(5):1901–1912.
HIV-infected women. J Infect Dis https://doi.org/10.1172/JCI80142
203(4):452–463. https://doi.org/10.1093/
infdis/jiq071
Chapter 26
Evaluation of the Role of Galectins in Parasite Immunity

Jaclyn Swan, Dhanasekaran Sakthivel, Travis Beddoe, Michael Stear,
David Piedrafita, and Sarah Preston
Abstract
Galectin-11 (LGALS-11) and galectin-14 (LGALS-14) are ruminant specific galectins, first reported in
sheep. Although their roles in parasite immunity are still being elucidated, it appears that they influence
protection against parasites. In gastrointestinal infections with the nematode Haemonchus contortus, both
galectin-11 and galectin-14 appear to be protective. However, in a chronic infection of liver fluke, Fasciola
hepatica, these galectins may aid parasite survival. To unravel the structural, functional, and ligand profile of
galectin-11 and galectin-14, recombinant production of these proteins is vital. Here we present the
recombinant production of soluble galectin-11 and galectin-14 from domestic sheep for in vitro and
structural biology studies. These methods include parasite cultivation and infection, galectin staining of
host and parasite tissue, surface staining of parasites with recombinant galectins, pull-down assays to
identify endogenous galectin binding proteins, and in vitro assays to monitor the effect of galectins on
parasite development.
Key words Galectin-11, Galectin-14, Haemonchus contortus, Fasciola hepatica, Larval culture, Immu-
nohistochemistry, Recombinant proteins, N-terminal His-tag, Pull-down assays, Larval molting assay,
Larval growth assay, Larval feeding assay
1 Introduction
Galectins are produced by parasites and by hosts in response to

parasite infections, suggesting that they play a role in host–parasite
interactions [1]. Of the 17 mammalian galectins described to date,
at least five are upregulated during parasitic infection (galectin-1,
-3, -9, -11 and -14). Since the initial discovery of galectin-11
(LGALS-11) and galectin-14 (LGALS-14) in parasite-infected
sheep by our laboratory, we and our collaborators have focused
on elucidating their structural and functional role in parasitic
infections.
Jaclyn Swan and Dhanasekaran Sakthivel joint first authors.
475
476 Jaclyn Swan et al.
We have focused on the gastrointestinal nematode parasite,

Haemonchus contortus (H. contortus) and the trematode parasite,
Fasciola hepatica (F. hepatica) which have considerable economic
impact on ruminant animal production worldwide. Our laboratory
explores the immune response to these parasites to aid vaccine
design, selected breeding, and disease diagnostics. As it is believed
that the larval stages of parasitic helminths are the most vulnerable
to host immune attack, parasitic larvae are the focus of our studies.
H. contortus causes significant production losses due to reduced
growth rates. In addition, its blood feeding can kill heavily infected
hosts. Anthelminthic resistance is a major concern as resistance to
all drug classes has been reported in different areas of the world
[2, 3]. Sheep (and goats) are naturally infected while grazing on
pasture contaminated by the third-stage H. contortus larvae (L3).
This is replicated experimentally by oral infection with parasitic
larvae using a gelatin capsule delivered by a dosing gun. The L3
for infection can be initially obtained from veterinary departments
or companies, but once an infection is established the cultivated L3
can be used to reinfect, creating a continuous infection cycle. Once
ingested, the L3 will move through the rumen (first stomach of
ruminants) to the abomasum (fourth or true stomach) where they
develop through the fourth larval stage (L4), prior to developing
into an immature adult stage that grows and becomes sexually
mature. Under optimal conditions, 21 days is the time required
for eggs to appear in the feces. When the infection has reached this
stage, feces can be collected to cultivate the eggs which in nature
would be excreted to contaminate pasture and cause reinfection in
grazing small ruminants.
Intra-mammalian life stages are difficult to replicate in vitro,
particularly for adult stages; however, conditions can be mimicked
or induced artificially to cultivate some life stages. Exsheathment
(release of the outer protective cuticle) can be induced by two
methods, either by treatment with carbon dioxide (CO2) or incu-
bation with sodium hypochlorite. The CO2 method is gentler and
has little to no effect on the epicuticle. Whereas, the hypochlorite
method can be used if the intention is reinfection or studying later
larval stages, such as the L4. For L3-L4 cultivation, aseptic techni-
ques are required and bleach (hypochlorite) has the advantage of
reducing the chance of bacteria and fungi that would cause culture
contaminations. The L4 stage of H. contortus is an important stage
to study the interaction of galectin-11 and galectin-14 as this is the
stage where it actively interacts with the host, as a mouth has now
developed and it begins feeding from the host.
To study the natural immune response to H. contortus infec-
tions in vivo, naı̈ve sheep (young animals with no or little prior
exposure to parasites) are usually primed to the infection. Priming
involves a series of infections which aims to mimic natural kinetics
where animals would undergo constant reinfection from pasture
Role of Galectins in Parasite Immunity 477
grazing. Following repeated priming and a rest period, a challenge

infection is administered, which in resistant animals will provoke a
strong adaptive immune response resulting in rejection of the
parasites. Immediate challenge without a rest period may result in
rapid rejection of incoming larvae [4, 5]. The size of sheep is
advantageous as it allows for collection of large tissue samples and
in vivo cannulation studies during the development of parasite
rejection [6, 7]. Using this experimental model, the immune
response can be strategically investigated to study either the larval
or adult parasitic stage, as differential immune responses are known
to exist [4, 5].
A similar scenario is true for F. hepatica which infects rumi-
nants, humans, and many other mammals. Infection begins by the
ingestion of metacercariae that are encysted larvae on vegetation.
When experimentally infecting an animal, it is most common to
purchase metacercariae collected from laboratory snail colonies
which are orally delivered by a dosing gun using a gelatin capsule
containing the metacercariae, to study the immune response to
acute and chronic infection. Following ingestion excystment occurs
in the small intestines and the resulting newly excysted juvenile
(NEJ) migrates through the peritoneal cavity into the liver. For
several weeks, the NEJ migrates through the liver tissue causing
pathology and hence is the cause of the majority of the economic
losses and welfare issues in the ruminant. After 6–8 weeks, the
immature fluke will reach the biliary ducts where it matures into
an egg producing adult. When adult flukes are needed for experi-
ments, they can be collected from the livers of cattle or sheep
sourced from abattoirs as they cannot be grown artificially in the
laboratory. Eggs secreted by sexually mature adults are excreted in
the feces. Unlike gastrointestinal nematodes, liver fluke requires an
intermediate host, a semi aquatic snail, to complete their life cycle
to produce infective metacercariae [8, 9]. Currently the main drug
of choice to treat liver fluke is Triclabendazole as it targets both the
immature and adult life stages; however, drug resistance to Tricla-
bendazole and other active ingredients is now globally distributed
[10, 11] hence new preventative treatments such as vaccination is
highly desirable.
Both galectins have been implicated in host immune responses
against internal parasitic infection in ruminants [12–18]. Elevated
expression levels of galectin-11 were originally reported from a
sheep abomasum infected with a gastrointestinal parasite
(H. contortus) and confirmed by cDNA cloning and RNA hybridi-
zation [19]. The coding region of LGALS-11 is 468 bp and the
translated amino acid sequence is 137 amino acids with a molecular
weight of ~14 kDa. Although galectin-11 has some sequence iden-
tity (34%) to human galectin-10 (Fig. 1), galectin 10 exhibits weak
binding to lactose and high binding affinity to mannose. In con-
trast, galectin-11 has a strong preference for lactose and galactose.
Fig. 1 Sequence and structural similarities between sheep galectin-11 and human galectin-10. (a) Amino acid
sequence alignment of sheep galectin-11 and human galectin-10. Horizontal arrows indicate the positions of
β-strands (S-sheet (S1–S6) and F-sheets (F1–F5)) for galectin 10. Conserved residues between the two
galectins are highlighted in red. A column framed in blue if more than 70% of its residues are similar according
to physicochemical properties. (b). Structural overlay of sheep galectin-11 (blue; PDB code 6N3R) and human
galectin-10 (Brown; PDB code 6GKT) highlighting the structure similarities and region of carbohydrate binding
(glycan recognition domain)
The galectin-11 gene does not have a signaling peptide for its
localized expression or a sequence encoding a transmembrane
domain that allows its secretion into the gastrointestinal tract
[19, 20]. Increased secretion of galectin-11 protein was reported
in the abomasum and small intestinal epithelium of animals infected
with gastrointestinal parasites compared to uninfected animals
[17, 19]. Immunohistochemical studies detected galectin-11 in
the cytoplasm and nucleus of epithelial cells of the gastrointestinal
tract [19]. Galectin-11 (renamed for the publication as Galectin-
15) was also found to be secreted in the ovine uterus during
pregnancy where it is hypothesized to be involved in trophoblast
migration and adhesion [21, 22].
Increased mucus stickiness in the gastrointestinal tract corre-
sponds with the production of galectin-11. Galectin-11 may
impede the motility of H. contortus and other gastrointestinal
nematodes [7]. Binding and accumulation of epithelial specific
galectin-11 in the anterior region of H. contortus L4 has also been
associated with the inhibition of larval development [16]. A pull-
down assay to identify ligands for galectin-11 revealed that recom-
binant galectin-11 binds to at least 34 proteins from adult parasites,
but identified no potential L4 ligand [23]. This seems to correlate
with fluorescent staining using recombinant galectin-11 where
specific binding around the anterior region was noted for L4 but
large surface area binding was observed for the adult stage
[16]. The lack of L4 ligands could be due to limitations of the
pull-down procedure or that galectin-11 is binding to excreted/

secreted molecules rather than the L4 worm itself.
Intriguingly, in sheep there are at least two different isoforms of
galectin-11, differing by a substitution of 15 amino acids [15]. Iso-
form 2 contains hydrophobic and electrostatic differences in the
glycan binding domain, when compared to the crystal structure of
isoform 1, potentially resulting in altered glycan recognition
[15, 24]. Additionally, differences at the dimer interface have also
been identified and have been demonstrated to cause isoform 2 to
exist as a monomeric dimer, rather than a tetramer as is the case for
isoform 1. In vitro experiments with isoform 2 and mutated forms
of isoform 1 (CDR mutation and dimer interface) have demon-
strated that the lack of dimerization of isoform 2 impeded the
effectiveness of galectin-11 to inhibit L4 development [15]. This
highlights the importance of galectin-11 structure in mediating
anti-parasitic activity. As these two natural genetic variants of
galectin-11, with differential anti-parasitic activity, have been iden-
tified in sheep [15], what role these two isoforms of galectin-11
may play in gastrointestinal nematode resistance, warrants further
investigation.
Galectin-14 was identified in a cDNA library prepared from
eosinophil-rich mammary lavage of sheep [25]. The entire coding
region of galectin-14 is 489 bp, spans a single open reading frame
and the translated protein was found to have 162 amino acids. The
predicted protein sequence of galectin-14 was found to be
18.2 kDa; however, the actual size (under reducing and
non-reducing conditions) was found to be 17 kDa [25]. Galectin-
14 has the closest sequence identity to galectin-9/ecalectin of rats
(58% identity) and humans (57% identity) (Fig. 2). Although func-
tionally it appears closest to galectin 10 which is also found in
eosinophil granules. Comparative amino acid sequences of known
galectins from other species revealed that galectin-14 contains six of
the seven residues considered as important for sugar recognition
(His74, Asn76, Arg78, Asn88, Trp95, and Glu98) [26, 27]. North-
ern blot and flow cytometric analysis demonstrated that a high level
of galectin-14 expression is observed only in eosinophil-rich leuko-
cyte populations, with no expression seen in neutrophils and
macrophages [25]. Worm extracts and excretory-secretory pro-
ducts of H. contortus and other parasites have been found to have
eosinophil chemotactic activity and eosinophils can kill larvae of
H. contortus suggesting a possible role of host galectins [28–
30]. The finding that galectin-14 is localized within eosinophils is
intriguing as this is one of the main immune cells implicated in the
resistance to gastrointestinal worm infections [4, 31, 32].
Analysis of H. contortus-infected sheep demonstrated release of
galectin-14 into the gastrointestinal mucus, the interface of host
and parasite interaction [13]. In addition, kinetic studies of
galectin-14 showed that release into the mucus occurred soon
Fig. 2 Sequence and structural similarities between sheep galectin-14 and human galectin-9. (a) Amino acid
sequence alignment of sheep galectin-14 and human galectin-9. Horizontal arrows indicate the positions of
β-strands for galectin 14. Conserved residues between the two galectins are highlighted in red. A column
framed in blue if more than 70% of its residues are similar according to physicochemical properties. (b)
Structural overlays of sheep galectin-14 (predicted structure) and human galectin-9 (PDB code 2YY1) high-
lighting the similarities. The glycan recognition domain is highlighted
after challenge infection, correlated with a reduction in parasitic

burden [7] and modulation of eosinophils (the main repository
immune cell of galectin-14) decreased resistance to infection in
sheep [31]. How and what host–parasite interactions mediate this
resistance is unknown. Recombinant galectin-14 protein
(GST-tagged galectin expression; r-galectin-14) has been passed
over a commercial glycan plate array developed by the Consortium
of Functional Glycomics, demonstrating high affinity to a low
number of repeats of type 2 lactosamine (Galß1-4GlcNac)
[13]. Recently, using a pull-down assay, rgalectin-14 has been
shown to interact with several adult H. contortus proteins, including
Zinc metalloprotease (M13) and hematophagous stage specific
sperm coating protein, both previous vaccine candidates [23] and
the majority of the interacting proteins inferred to take part in
metabolic and regulatory processes.
The above work suggests that galectin-14 plays a role in resis-
tance to H. contortus, but its role in infection in F. hepatica may tell
a different story. Eosinophilia is also a hallmark of F. hepatica
infections, and the secretion of galectin-14 from eosinophils was
found to occur in the bile duct during chronic liver fluke infection
[12]. Additionally, transcriptomic studies of experimentally
infected Merino sheep have demonstrated that galectin-14 is upre-
gulated in peripheral blood mononuclear cells at 2- and 8-weeks
post infection (wpi) [33] and 8 wpi in liver tissue [34]. Although,
endogenous galectin-14 has not been found attached to the surface
of F. hepatica, rGST-galectin-14 was shown to bind to the fluke
surface in frozen tissue sections. This is most likely due to an
evasion strategy of helminths which involves shedding of adhered
host proteins such as antibodies [35–37]. This process can be a

hindrance when trying to stain live parasites but can be counter-
acted by snap-freezing the parasites prior to antibody incubation.
As resistance to chronic F. hepatica infections in sheep is not
believed to exist [38], it is likely that these parasites have “high-
jacked” eosinophilia and galectin-14 secretion for their own
benefit, unlike the probable protective role of eosinophils in
H. contortus infections.
Using a pull-down method with rgalectin-14 bound to
NHS-sepharose beads and F. hepatica lysate passed over, it was
found that rgalectin-14 interacted with a number of F. hepatica
glycoproteins with the most abundant being an uncharacterized
C-type lectin, as well as a number of other antigenic proteins,
including previous vaccine candidates [39]. This procedure was
repeated using rgalectin-11; however; it interacted with far fewer
proteins, which was reflective of histochemical staining, [12] where
rGST-galectin-11 did not show binding to the surface of adult fluke
although this was assumed to be due to the inactivity of the recom-
binant protein. However; native galectin-11 was detected in bile
and bile duct tissue of naturally infected sheep [12]. To date a
correlation between galectin-11 expression and F. hepatica viability
has not been concluded. Although it is likely that galectin-11 is
induced shortly following infection and is possibly having a host–
host interaction, perhaps like the affects seen to mucus viscosity
following H. contortus infection.
In summary, recombinant galectins are useful tools to gain
insights into the structural basis of protein–glycan interactions
and critical for studying its function in the host–parasite interac-
tion. In this chapter, we present an updated version of methods
involved in molecular cloning, recombinant protein production of
Ovis aries (sheep) soluble galectin-11 and galectin-14 for in vitro
functional and structural biology studies. These methods include
parasite cultivation and infection, galectin staining of host and
parasite tissue, surface staining of parasites with recombinant galec-
tins, pull-down assays to identify endogenous galectin binding
proteins, and in vitro assays to monitor the effect of galectins on
parasite development.
2 Materials
2.1 Cultivation of 1. 3–4 Merino wethers aged between 3 months and 1 year old (see
Parasites Note 1).
2.1.1 Hatching H. 2. 20,000 H. contortus L3 larvae per sheep for inoculation.
contortus Eggs from 3. Gelatin capsules.
Infected Sheep to the Third 4. Plastic dosing gun (Torpac, USA).
Larval Stage (L3)
5. 0.9% (w/v) sterile sodium chloride (saline).

6. Fecal collection bags (see Note 2).
7. Plastic trays.
8. Clear plastic sheets (long enough to cover plastic trays).
9. Room with morning sunlight.
10. Spray bottle containing tap water.
11. Large glass jar.
12. 75 cm2 tissue culture flask.
13. Light microscope.
14. Centrifuge.
2.1.2 Fecal Egg Count for 1. 2–5 g of feces from each sheep to be tested (collected fresh
Determining H. contortus from rectum).
Infection 2. Electronic balance.
3. Saturated salt solution (0.32 kg/l).
4. Whitlock McMaster counting chamber.
5. Plastic pipettes.
6. Glass jar.
7. Handheld blender.
8. 2 mm sieve.
9. Metal teaspoon.
2.1.3 Storage and 1. Cool room or refrigerator at 4 C.

Monitoring of H. contortus 2. Centrifuge.
L3 Larvae
3. Tap water.
4. 75 cm2 tissue culture flasks.
2.1.4 Exsheathment of H. 1. H. contortus L3 larvae.

contortus Third Larval 2. Pyrogen-free saline (PFS).
Stage (L3) Using Sodium
3. 0.15% (v/v) sodium hypochlorite in PFS.
Hypochlorite
4. Centrifuge.
5. 50 ml tubes.
6. Water bath set at 37 C.
8. Penicillin/streptomycin solution (Invitrogen; 5000 units peni-
cillin and 5000 μg streptomycin/ml).
2.1.5 Exsheathment of H. 1. H. contortus L3 larvae (1000/ml).

contortus L3 Using CO2 2. Phosphate-buffered saline (PBS; 140 mM NaCl, 2.7 mM KCl,
10 mM Na2HPO4, 1.8 mM KH2PO4, pH 7.3).
3. 100 ml 0.09% (w/v) glucose.
4. 250 ml glass bottle which can be sealed tightly with a lid.
5. Gas cylinder of 100% CO2.
6. Orbital incubator set to 40 C.
7. Centrifuge.
2.1.6 Culturing of H. 1. L3 H. contortus exsheathed by sodium hypochlorite (see Sub-

contortus L3 to the Fourth heading 3.1.4).
Larval Stage (L4) 2. Incubator set at 40 C, 10% CO2.
3. Centrifuge.
5. Tissue culture medium Dulbecco’s modified Eagle Medium
+GlutaMax (DMEM) supplemented with 1% penicillin strep-
tomycin and 1% fungisome.
6. 75 cm2 tissue culture flask.
2.1.7 Collection of Adult 1. H. contortus-infected sheep (at stage where parasitic eggs are
H. contortus being detected in the feces) (see Subheading 3.2 for infection
protocols).
2. Sodium pentobarbital.
3. Dissection equipment (including blunt nose tweezers).
4. Phosphate-buffered saline (PBS; 140 mM NaCl, 2.7 mM KCl,
10 mM Na2HPO4, 1.8 mM KH2PO4, pH 7.3) warmed to
40 C.
5. Plastic tray.
6. 1 l Beaker.
7. Incubator set to 40 C.
2.1.8 Infection of Sheep 1. F. hepatica metacercariae (see Note 3).

with Liver Fluke (F. 2. Filter paper.
hepatica)
3. Gelatin capsules.
4. Parasite-free adult sheep.
5. Plastic dosing gun (Torpac, USA).
7. Dissection equipment and tweezers.
8. Tissue homogenizer.
10. 2 2 mm sieves.
11. 250 μm sieve.
12. Tweezers.
13. Optimal cutting temperature (OCT) compound.
14. Liquid nitrogen and aluminum foil boats.
2.2 RNA Isolation, 1. Abomasal tissues of 3–4 Merino sheep aged between 3 months
cDNA Synthesis, and and 1 year old infected with gastrointestinal parasites.
Cloning of LGALS-11 2. Phosphate-buffered saline (PBS; 140 mM NaCl, 2.7 mM KCl,
and LGALS-14 10 mM Na2HPO4, 1.8 mM KH2PO4, pH 7.3).
3. RNeasy RNA isolation kit (Qiagen).
4. TissueLyser-II (Qiagen).
5. QuantiTect® cDNA synthesis kit (Qiagen).
6. PrimeSTAR® GXL DNA polymerase (Takara, Cat No#
R050Q).
7. PCR primers;
LGLAS-11 (468 bp);
Forward: 50 - CTTGCCATGCC ATGGACTCCTTGCC -30
(NCO);
Reverse: 50 CCGGAATTC GTGGTGGTGGTGGT -30
(EcoRI);
LGLAS-14 (489 bp);
Forward: 50 CCCCATGG GCATGCAAAGCGAGAGTGGT
CACGAAG-30 (NCO-1);
Reverse: 50 -CCGCTCGAG TATCTGGAAGCTGATATAG
GAC -30 (Xho-I).
8. Luria-Bertani (LB) medium.
9. Kanamycin (50 μg/ml).
10. Agar plates containing kanamycin (50 μg/ml).
11. Restriction enzymes (New England Biolabs).
12. pET28a vector.
13. T4 DNA ligase.
14. E. coli XL-1 blue cells and E. coli BL-21 (DE3) cells.
2.3 Recombinant 1. E. coli BL-21 (DE3) cells.

Protein Production 2. Agar plates containing kanamycin (50 μg/ml).
2.3.1 Protein Expression 3. Luria-Bertani (LB) medium.
4. Kanamycin (50 μg/ml).

5. Isopropyl β-D-1-thiogalatopyranoside (IPTG).
6. Lysis buffer: 20 mM Tris–HCl pH 8.0, 150 mM NaCl, 0.1%
(v/v) Triton X-100, 10 mM Imidazole pH 8.0, 1 mM PMSF
(Phenylmethylsulfonyl fluoride).
7. DNase (10 μg/ml).
8. MgCl2 (5 mM).
9. Centrifugal filters (3-kDa molecular-weight cutoff)
(Millipore).
10. Thermal cycler.
11. Incubator and orbital shaking incubator.
12. Centrifuge.
13. Sonicator.
14. 0.22 μm syringe filter.
2.3.2 Immobilized Metal 1. Nickel-charged resin (Agarose resin).

Affinity Chromatography 2. Gravity column.
(IMAC) Purification of
3. Immobilized nickel affinity purification buffer A (binding
LGALS-11 and LGALS-14
buffer): 20 mM Tris–HCl pH 8.0, 150 mM NaCl, 0.1%
(v/v) Triton X-100, 10 mM Imidazole, 1 mM Dithiothreitol,
1 mM PMSF.
4. Immobilized nickel affinity purification buffer B (wash buffer):
50 mM Tris–HCl pH 8.0, 500 mM NaCl, 30 mM Imidazole.
5. Imidazole Immobilized nickel affinity purification buffer C
(elution buffer): 50 mM Tris–HCl pH 8.0, 500 mM NaCl,
300 mM Imidazole.
6. Tris(2-carboxyethyl) phosphine hydrochloride (TCEP-HCL).
7. Amicon ultracentrifugal filters (5 kDa molecular-weight cutoff;
Millipore).
2.3.3 Removal of 1. Human rhinovirus-3C protease (Thermo Fisher

Hexahistidine from Scientific, USA).
Recombinant LGALS-11
and LGALS-14
2.3.4 Size-Exclusion 1. Superdex S75 16/60 gel filtration column.

Chromatography (SEC) 2. AKTA-basic Fast Protein Liquid Chromatography (FPLC)
system.
3. Buffer D: 20 mM Tris–HCl pH 8.0, 100 mM NaCl, 10 mM
TCEP [Tris (2-carboxyethyl) 2 phosphine].
4. Protein concentrator (MWCO3000, GE healthcare Life
Sciences).
2.4 Recombinant 1. Polyacrylamide gel (4–12% gel).

LGALS-11 and LGALS- 2. Staining solution (0.22% (w/v) Coomassie Brilliant Blue R, in
14 Characterization 40% (v/v) ethanol and 7% (v/v) glacial acetic acid).
2.4.1 SDS-PAGE Analysis 3. Destaining solution (20% (v/v) ethanol and 7% (v/v) glacial
acetic acid).
2.4.2 Western Blot 1. Nitrocellulose membrane (Trans-blot® mini nitrocellulose

Analysis membrane, Bio Rad).
2. 5% (w/v) Bovine Serum Albumin (BSA).
3. 1 Phosphate-buffered Saline (PBS) (137 mM NaCl, 2.7 mM
KCL, 10 mM Na2HPO4, 2 mM KH2PO4 pH 7.4).
4. Anti-LGALS-11 antibody and anti-LGALS-14 antibody
(we have produced rabbit polyclonal anti-galectin-11 and
mouse monoclonal anti-galectin-14 antibodies).
5. Anti-rabbit or rabbit anti-mouse IgG horseradish peroxidase
conjugate.
6. Pierce™ ECL plus western blotting substrate (Thermo Fisher).
2.5 Preparation of 1. Parasite-free sheep.

Parasite-Infected 2. H. contortus L3 larvae gelatin capsules (prepared in Subheading
Gastrointestinal 3.1.1, steps 1–3).
Tissues (H. contortus-
3. Plastic dosing gun (Torpac, USA).
Infected Abomasum)
from Primary and
4. Anthelmintic, levamisole.
Primed Infections 5. Commercial feed pellets (such as lucerne pellets).
7. Dissection equipment.
2.6 Detection of 1. Infected and control tissues (prepared in Subheading 3.5).

Endogenous Galectins 2. Optimal cutting temperature (OCT) compound.
in Tissues
3. Containers to hold OCT-embedded tissue (we use small plastic
(Immunohisto-
weigh boats).
chemistry in OCT-
Embedded Frozen
4. Liquid nitrogen.
Tissues) 5. Aluminum foil.
6. Cryostat.
7. Microscope slides.
8. 95% (v/v) ethanol.
9. Peroxidase blocking reagent (DAKO).
11. Equipment to create a humidity chamber (water-saturated

material such as wet tissues in the bottom of an enclosed
container helps to prevent sections from drying out).
12. Primary anti-galectin antibody (we have used rabbit polyclonal
anti-galectin-11 and mouse monoclonal anti-galectin-14 anti-
bodies produced in our lab).
13. HRP-conjugated secondary antibody (e.g., horseradish
peroxidase-conjugated anti-rabbit IgG).
14. 1.5 mM 3,30 -diaminobenzidine tetrahydrochloride (DAB).
15. Hematoxylin or Hematoxylin and Eosin Y (H & E).
2.7 Recombinant 1. 5-iodoacetamidofluorescein (Molecular Probes, Invitrogen).

Galectin Binding to 2. Recombinant GST-galectin-14 (rGST-galectin-14) and recom-
Frozen Tissue Sections binant GST (rGST).
2.7.1 Fluorescent 3. Rabbit erythrocytes—rabbit blood collected 1:1 with Alsevers
Labeling of Galectin solution.
Probes, Assessment of 4. 50 ml sterile tubes.
Activity by
5. Trypsin.
Hemagglutination Assay
7. 0.15 M sodium chloride (NaCl) in phosphate-buffered saline.
8. 1% (v/v) glutaraldehyde in phosphate-buffered saline (PBS;
140 mM NaCl, 2.7 mM KCl, 10 mM Na2HPO4, 1.8 mM
KH2PO4, pH 7.3).
9. 0.1 M glycine in phosphate-buffered saline (PBS; 140 mM
NaCl, 2.7 mM KCl, 10 mM Na2HPO4, 1.8 mM KH2PO4,
pH 7.3).
10. V-well microtiter plates.
11. DES (0.15 M NaCl, 2 mM EDTA, 2 mM DTT made fresh
daily).
12. 0.5% (w/v) bovine serum albumin (BSA) in 0.15 M NaCl.
2.7.2 Binding of 1. Blocking solution: 3% BSA (w/v) in phosphate-buffered saline

Fluorescent Galectin to (PBS; 140 mM NaCl, 2.7 mM KCl, 10 mM Na2HPO4,
Frozen Tissue Sections 1.8 mM KH2PO4, pH 7.3).
2. Fluorescence microscope.
3. 100 mM lactose.
4. Fluorescein-labeled recombinant GST-galectin-14 (rGST-
galectin-14) and recombinant GST (rGST) (See
Subheading 3.3).
2.7.3 Binding of Non- 1. rGST-galectin-14, inactive rGST-galectin-11, and rGST.

fluorescent Galectin to 2. Anti-GST horseradish peroxidase (HRP)-conjugated second-
Frozen Tissue Sections ary antibody.
3. 1.5 mM 3,30 -diaminobenzidine tetrahydrochloride (DAB).
4. Hematoxylin or Hematoxylin and Eosin Y (H & E).
5. Humidity chamber (see Subheading 2.6, step 11).
2.8 Galectin Binding 1. L3 (CO2 exsheathed), L4, and adult stage H. contortus
to Parasites (H. parasites.
contortus and F. 2. Liquid nitrogen.
hepatica) 3. Phosphate-buffered saline (PBS; 140 mM NaCl, 2.7 mM KCl,
2.8.1 Fluorescent 10 mM Na2HPO4, 1.8 mM KH2PO4, pH 7.3).
Staining of H. contortus L3, 4. Eppendorf tubes (1.5 ml).
L4, and Adult Stage
5. Eppendorf centrifuge.
Parasites
6. Water bath set at 37 C.
7. 100 mM D-galactose.
8. Recombinant galectin-11 (5 μg/ml–100 μg/ml).
9. Rabbit anti-ovine galectin-11 polyclonal antibody (1:1000).
10. Goat anti-rabbit conjugated to Alexa Fluor 647 (1:1000).
11. Aluminum foil.
12. Microscopic glass slide and coverslips.
13. Fluorescent microscope.
2.8.2 Binding of Non- 1. Liver fluke F. hepatica collected from infected sheep livers and
fluorescent Galectin Probes bile ducts (from the local abattoir, or collected from deliber-
to F. hepatica ately infected sheep, as described in Subheading 3.1.8).
2. Optimal cutting temperature compound (OCT).
3. Cryostat.
4. 95% (v/v) ethanol.
5. 1.5% (v/v) hydrogen peroxide diluted in phosphate-buffered
saline (PBS; 140 mM NaCl, 2.7 mM KCl, 10 mM Na2HPO4,
1.8 mM KH2PO4, pH 7.3).
6. 3% (w/v) Bovine serum albumin (BSA) in phosphate-buffered
saline (PBS; 140 mM NaCl, 2.7 mM KCl, 10 mM Na2HPO4,
1.8 mM KH2PO4, pH 7.3).
7. Galectin probe (rGST-galectin-14) and control probes (inac-
tive rGST-galectin-11, and rGST).
8. Anti-GST HRP-conjugated antibody.
9. DAB (3,30 -diaminobenzidine tetrahydrochloride).

10. Hematoxylin.
2.9 Identification of 1. PBS (140 mM NaCl, 2.7 mM KCl, 10 mM Na2HPO4, 1.8 mM

Galectin Ligands Using KH2PO4, pH 7.3).
a Pull-Down Assay 2. 50 mM galactose and 50 mM lactose.
2.9.1 Preparation of 3. Liquid nitrogen.
Parasite Lysate 4. Mortar and pestle.
5. RIPA buffer (20 mM Tris–HCl pH 8.0, 150 mM NaCl,
0.5 mM ethylenediaminetetraacetic acid (EDTA), 1% (v/v)
Nonidet P-40, 0.1% (w/v) sodium deoxycholate, 0.05%
(w/v) sodium dodecyl sulfate (SDS), 1% (v/v) Triton X-100).
6. Dialysis tube 3.5 kDa molecular-weight cutoff.
7. RIPA dialysis buffer (20 mM Tris–HCl pH 8.0, 150 mM NaCl,
0.5 mM EDTA, 0.05% (v/v) Nonidet P-40, 0.01% (w/v)
sodium deoxycholate, 1% (v/v) Triton X-100).
8. Pierce™ BCA protein assay kit (Thermo Fisher
Scientific, USA).
2.9.2 Conjugation of 1. Galectin coupling buffer (20 mM 4-(2-hydroxyethyl)-1-piper-

Galectin to NHS-Activated azinee thanesulfonic acid (HEPES) pH 8.8, 100 mM NaCl,
Sepharose and 10 mM TECP.
2. NHS-activated Sepharose (GE Healthcare Life Science).
3. Rotating wheel.
4. Ice cold 1 mM HCl.
5. 500 mM HEPES.
6. Blocking buffer (100 mM Tris–HCl pH 8.5, 500 mM NaCl,
and 10 mM TCEP).
7. Wash buffer 1 (100 mM Tris–HCl pH 8.8, 500 mM NaCl, and
10 mM TCEP).
8. Wash buffer 2 (100 mM HEPES pH 6.8, 500 mM NaCl, and
10 mM TCEP).
9. Storage buffer (100 mM Tris–HCl pH 8.0, 150 mM NaCl, and
0.05% (w/v) sodium azide).
10. Galectin ligand elution buffer (250 mM galactose, 20 mM
Tris–HCl pH 8.0, 100 mM NaCl, and 10 mM TCEP).
2.9.3 Mass Spectrometry 1. SpeedVac Concentrator and Savant Refrigerated Vapor trap
Preparation and Analysis (Thermo Fisher).
2. 8 M Urea, 100 mM Tris–HCl pH 8.3.
3. 200 mM TCEP.
4. ThermoMixer (Thermo Fisher).
5. 1 M iodoacetamide (IAA; in water).

6. 50 mM Tris–HCl (pH 8.3).
7. Sequencing-grade trypsin (Promega, USA).
8. Incubator 37 C.
9. Trifluoroacetic acid (TFA).
10. Poly(styrene-divinylbenzene) copolymer (SDB) StageTips
(3 M, USA).
11. 2% (v/v) acetonitrile (ACN), 0.1% (v/v) TFA.
12. In line HPLC fitted with Guard column (C18 PepMap 100 mm
inner diameter (ID) 2 cm trapping column, Thermo Fisher
Scientific) and analytical column (Vydac MS C18, 3 mm,
300 Å and 75 mm ID 25 cm).
13. Buffer A (2% (v/v) ACN, 0.1% (v/v) formic acid).
14. Buffer B (80% (v/v) ACN, 0.1% (v/v) formic acid).
15. Orbitrap Elite mass spectrometer (Thermo Fisher).
2.10 Functional 1. Tissue culture media (Dulbecco’s modified Eagle Medium

Assays with H. +GlutaMax (DMEM) supplemented with 1% penicillin strep-
contortus tomycin and 1% fungisome).
2.10.1 Larvae Moulting 2. Sodium hypochlorite exsheathed L3 H. contortus parasites (see
Assay Using H.
Subheading 3.1.4, steps 1–5).
contortus L3 3. 24-well plates.
4. Recombinant galectin-11 (5 μg/ml–100 μ/ml).
5. Tissue culture incubator set to 40 C and 10% CO2 flow.
6. 1.5 ml eppendorf tubes.
7. Iodine.
10. Light microscope with camera.
2.10.2 Larvae Growth 1. Tissue culture media (Dulbecco’s modified Eagle Medium
Assay Using H. +GlutaMax (DMEM) supplemented with 1% penicillin strep-
contortus L4 tomycin and 1% fungisome).
2. L4 H. contortus parasites (see Subheading 3.1.6).
3. 24-well plates.
7. Iodine.
8. Recombinant galectin-11 (5–100 μg/ml).

10. Light microscope with camera.
11. Imaging software such as Image J.
2.10.3 Larvae Feeding 1. Tissue culture media (Dulbecco’s modified Eagle Medium
Assay Using H. +GlutaMax (DMEM) supplemented with 1% penicillin strep-
contortus L4 tomycin and 1% fungisome).
2. L4 H. contortus parasites (see Subheading 3.1.6).
3. 24-well plates.
4. Recombinant galectin-11 (5–100 μg/ml).
8. Protein labeled to fluorescence [we used chicken ovalbumin-
fluorescein isothiocyanate (OVA-FITC) concentrated at 2 mg/
ml].
11. Fluorescent microscope.
3 Methods
3.1 Cultivation of the 1. Wash H. contortus L3 larvae three times in sterile saline by
Larval Stages of H. spinning at 350 g.
contortus 2. Resuspend L3 in sterile saline at concentration of 10,000/ml
3.1.1 Hatching H. (see Note 4).
contortus Eggs from 3. Transfer 1 ml into each gelatin capsule. (You will need to
Infected Sheep to the Third prepare two capsules per sheep—see Note 5).
Larval Stage (L3) 4. Infect sheep with two capsules each, using a dosing gun (see
Note 6).
5. After 4 weeks check if H. contortus infection has established by
performing fecal egg counts (FEC) (see Subheading 3.1.2).
6. Perform FEC weekly until infection level is approximately
1000 eggs per gram. Once infection level is 1000 eggs per
gram, collect feces.
7. To collect feces, attach fecal collection bags to the animals.
Feces should be collected twice daily.
8. Place collected feces into plastic trays, moisten with tap water
using a spray bottle, and cover with clear plastic sheets. Place in
a position where morning sunlight will be received.
9. After 10 days, check for L3 (see Note 7). To collect L3, rinse
plastic sheet with tap water into large glass jar. Transfer con-
tents into a 75 cm2 tissue culture flasks (100 ml/flask) and
store at 4 C. L3 can be visualized under a light microscope.
Replace system with fresh feces when L3 collection begins to
decline or feces become overgrown with fungi.
3.1.2 Fecal Egg Count for 1. Collect approximately 5 g of fresh feces from rectum of the
Determining H. contortus animal.
Infection 2. Weigh out 5 g of feces.
3. Place 5 g of feces in large glass jar with 50 ml of saturated salt
solution and create a homogenous mixture using handheld
blender.
4. Strain mixture through sieve to remove particulates. Push
down on particulates in sieve with teaspoon to collect remain-
ing liquid.
5. Using a plastic pipette transfer 5 ml of the solution into a
Whitlock McMaster counting chamber and count the number
of eggs in each chamber under light microscope at 20
magnification.
6. To calculate the number of eggs per gram (epg), use the
following formula:
Eggs per gram ðEPGÞ ¼ ðegg number solution volume ðmlÞÞ=ðfaeces ðgÞ volume in counting chamberÞ
¼ ðegg# 50 mlÞ=ð5 g 0:5 mlÞ
3.1.3 Storage and 1. Stored L3 will need to be washed every month to prevent
Monitoring of H. fungal/bacteria buildup and to remove dead parasite larvae.
contortus L3 2. Wash L3 by centrifuging at 350 g (with no break) and
resuspend in tap water with total volume of approximately
100 ml/flask.
3.1.4 Exsheathment of H. 1. Transfer ~10,000 H. contortus L3 larvae to a 50 ml tube con-

contortus Third Larval taining 10 ml of 0.15% sodium hypochlorite in PFS.
Stage (L3) Using Sodium 2. Incubate at 37 C for 15 min with occasional agitation.
Hypochlorite
3. Add more PFS up to a total volume of 50 ml.
4. Centrifuge at 500 g for 5 min (with no brake).
5. Wash larvae three times in PFS.
6. Count the percentage of exsheathed larvae (see Note 8 and
Fig. 3).
7. Resuspend ~1000 larvae/ml in saline containing 0.5% penicil-
lin/streptomycin solution.
Fig. 3 Haemonchus contortus larvae illustrating the morphological features distinguishing third-stage larvae
(L3), exsheathed L3 (xL3), and fourth-stage larvae (L4). (a) Sheathed L3 collected from fecal culture. Arrow
indicates sharp tail used to distinguish between L3 and xL3. (b) xL3, arrow indicates a rounder tail compared
to L3 which is a morphological feature used to estimate percentage of exsheathed larvae post sodium
hypochlorite or carbon dioxide treatment. (c) xL3 undergoing exsheathment to L4, arrow indicates moulting
sheath. (d) L4 H. contortus, arrows indicate characteristic mouth and tail morphological development used to
distinguish between xL3 and L4. (e) Arrow indicates L4 mouth piece. Magnification: (a, b, c, and d) 20; (e)
100
3.1.5 Exsheathment of H. 1. Wash L3 H. contortus 3 times in PBS by centrifuging at 350 g

contortus L3 Using CO2 (with no break) for 5 min.
2. Resuspend in 50 ml PBS.
3. Transfer into 250 ml glass bottle containing 100 ml of 0.09%
glucose warmed to 40 C.
4. Bubble solution with CO2 gas for 20 s, immediately seal
with lid.
5. Place in orbital incubator heated to 40 C at a speed of 100 rpm
for 15 min.
6. Re-bubble L3 solution with CO2 for 2 min and place back into
incubator for 90 min or leave overnight.
7. Examine exsheathment under light microscope (see Note 8 and
Fig. 3). A substantial exsheathment rate is 75%.
3.1.6 Culturing H. 1. Exsheath L3 with sodium hypochlorite (see Subheading 3.1.4).

contortus L3 to the Fourth 2. At step 5, wash exsheathed L3 (xL3) five times with saline by
Larval Stage (L4) spinning at 350 g (with no break) and resuspend at 1000
xL3/ml into tissue culture medium.
3. Transfer medium containing xL3 into a 75 cm2 tissue culture
flask and place in tissue culture incubator set to 40 C, 10%
CO2 flow.
4. Incubate for 5 days or until moulting to L4 greater than 80%.
5. L4 can be distinguished under light microscope by the devel-
opment of mouth parts and distinct markings on the tail (see
Fig. 3).
3.1.7 Collection of Adult 1. Sacrifice infected sheep with an overdose of sodium

H. contortus pentobarbital.
2. Open up sheep and remove abomasum.
3. Open abomasum by cutting along the greater curvature.
4. Place abomasum in plastic tray filled with warm PBS and rinse.
5. Use tweezers to collect dislodged adult worms from solution
and transfer into a beaker containing a warm PBS solution.
6. Use tweezers to remove adult worms still lodged inside the
tissue, check under abomasal fold.
7. Keep at 40 C if performing staining straight away or preserve
for later use by snap-freezing (see Note 9).
3.1.8 Infection of Sheep 1. Obtain F. hepatica metacercariae (see Note 3).

with Liver Fluke (F. 2. Using a pipette, load the metacercariae onto filter paper.
hepatica)
3. Insert loaded filter paper into gelatin capsules.
4. Infect sheep with 250 metacercariae of F. hepatica by orally
administering the metacercariae gelatin capsules with a
dosing gun.
5. 3 months post-infection with F. hepatica metacercariae, sacri-
fice sheep by lethal overdose with sodium pentobarbital to
obtain infected liver and bile duct (see Note 10).
6. Collect juvenile F. hepatica parasites from the infected liver by
slicing tissue, homogenizing the tissue slices in 4 C PBS, and
then washing this homogenized tissue consecutively through
two sieves of 2 mm and 250 μm mesh. Liver flukes (between
2 and 5 mm long) will be caught by the 250 μm screen. Adult
liver flukes in the bile duct can be visualized by eye and
removed with tweezers. Isolated liver fluke and infected tissue

samples can be embedded in OCT, frozen on the surface of
liquid nitrogen, and stored frozen for subsequent immunohis-
tology or galectin binding assays (see Note 11).
3.2 RNA Isolation The LGALS-11 and LGALS-14 encoding sequence from sheep
from Sheep (Ovis aries) was amplified using cDNA prepared from sheep aboma-
Abomasum, cDNA sum tissue by PCR with galectin-specific primers (Table 1; Fig. 4).
Synthesis, and Cloning Sheep LGALS-11 and LGALS-14 genes has polymorphism in the
of LGALS-11 and ORF region (see Note 12). The recombinant LGALS-11 and
LGALS-14 LGALS-14 proteins were produced for structural biology studies
and functional studies using an established method [24].
1. Thaw previously collected frozen tissue samples at room tem-
perature and wash three times with 1 PBS buffer.
2. After complete removal of PBS from tissue samples, weigh
0.5 mg in a sterile Qiagen TissueLyser-II tube and add 1 ml
of tissue lysis buffer and incubated at 22 C for 30 min. Lyse
the tissue cells by TissueLyser-II at 750 oscillations/min.
3. Isolate the total RNA by following RNeasy tissue RNA isola-
tion protocol (Qiagen).
4. Mix 1 μg of total RNA with 2 μl of genomic DNA wipeout
(7), 14 μl nuclease-free water and incubate at 42 C for 2 min
(as stated in the QuantiTect® cDNA synthesis kit (Qiagen)).
5. Following incubation, add 1 μl of quantiscript reverse-
transcriptase, 4 μl of quantiscript RT buffer (5), and 1 μl of
OligodT primer and incubate at 42 C for 45 min (as stated in
the QuantiTect® cDNA synthesis kit (Qiagen)).
Table 1
Details of the tools used to clone LGALS-11 and LGALS-14 gene
Source organism Ovis aries

Gene source Abomasum tissue
Cloning vector pET28
Expression vector pET28
Expression host E. coli BL21 (DE3)
Fig. 4 Illustration of recombinant LGALS-11 or LGLAS-14. Both the galectins were produced with an
N-terminal histidine tag and a 3C enzyme cleavable site using pET28a vector
Table 2
Details of PCR cycle conditions for LGALS-11 and LGALS-14 gene
PCR Cycle condition
Step LGALS-14 LGALS-11 Cycles

Initial denaturation 94 C—3 min 94 C—3 min 1

Denaturation 94 C—30 s 94 C—45 s 34 times
Annealing 56 C—30 s 54 C—45 s
Extension 74 C—45 s 74 C—1 min
Final extension 74 C—7 min 74 C—10 min 1
6. Prepare 50 μl of PCR reaction mixture containing LGALS-11

and LGALS-14 gene specific primers using PrimeSTAR® GXL
DNA polymerase (Takara, Cat No# R050Q).
7. Amplify the LGALS-11 or LGALS-14 gene by PCR reaction
using following cycle conditions (Table 2).
8. Digest the PCR products and pET28a vector with respective
restriction enzymes at 37 C for 3 h and purify the resulting
digested products using agarose gel.
9. Ligate the digested PCR product and plasmid DNA using T4
DNA ligase enzyme at 16 C for 12–16 h.
10. Transform the ligated products into chemically competent
E. coli XL-1 blue cells and plate the transformed cells onto
agar plates containing kanamycin (50 μg/ml) and incubate
the plates for 16 h at 37 C.
11. Next day pick individual colonies and prepare plasmids and
verify the correct insertion of LGALS genes by restriction
digestion, gene sequencing, and bioinformatics analysis.
12. Prepare glycerol stocks of E. coli XL-1 blue cells containing
individual LGALS gene inserts.
3.3 Recombinant Prior to 2015 galectin-14 and galectin-11 were initially produced
Protein Production as recombinant GST-fusion proteins (rGST-galectin-14 and -11).
Using red blood cell agglutination assays, rGST-galectin-14 was
shown to be active but rGST-galectin-11 was not. Cleaving the
GST from rGST-galectin-14 resulted in protein precipitation and
the full fusion protein was therefore used in all functional studies.
However, to produce monoclonal and polyclonal antibodies the
GST tags were removed from both these recombinant galectins.
See previous version for methods involving recombinant
GST-fusion proteins. Nowadays, galectin-11 and -14 are produced
as active soluble recombinant proteins containing an N-terminal
His-tag (rgalectin-11 and rgalectin-14). Activity of recombinant
galectins was confirmed by binding to an immobilized galactose
column and subsequently eluted with galactose [24]. We have used

these recombinant versions from then onwards as they are con-
firmed to be active. The methodology for recombinant proteins
containing an N-terminal His-tag are described below.
For the overexpression of LGALS-11 or LGALS-14 proteins,
transform the recombinant plasmids into E. coli BL21 (DE3) cells.
Screen individual E. coli BL21 (DE3) colonies containing LGALS-
11 and LGALS-14 gene inserts by gene sequencing and inducing
with IPTG.
3.3.1 Protein Expression 1. Inoculate single colony of E. coli BL21(DE3) strain containing
LGALS-11 or LGALS-14 gene construct, in 250 ml of LB
medium containing kanamycin (50 μg/ml) and incubate at
37 C for 16 h.
2. Prepare 1:100 dilution of this initial culture with 800 ml of
fresh LB medium containing kanamycin (50 μg/ml) and grow
at 37 C until the OD600 of 0.6 OD.
3. Reduce the orbital shaking incubator temperature to 16 C and
wait until the bacterial culture to reach the appropriate temper-
ature and induce the cells with 0.5 mM IPTG for another 16 h.
4. Harvest the cells by centrifugation at 4500 g for 10 min and
resuspend the cell pellet in 30 ml/g of lysis buffer.
5. Add DNase (10 μg/ml) and MgCl2 (5 mM) to the resus-
pended bacterial cells and lyse by sonication (Ultrasonics), in
6 30 s bursts in ice.
6. Remove the cellular debris by centrifugation at 25,000 g for
25 min and remove the fine particles by filtering through
0.22 μm filter and store the filtered lysate at 4 C.
7. Recombinant LGALS-11 is prone to aggregation during puri-
fication at high concentration. Hence maintain the purified
recombinant LGALS-11 at less than 5.0 mg/ml concentration
and in reduced state (see Note 13).
3.3.2 Purification of 1. Using a gravity column, add 3 ml of 50% immobilized nickel

Recombinant LGALS-11 or resin (Thermo Fisher Scientific, USA) and wash with 50 ml of
LGALS-14 by Immobilized milli-Q water.
Metal Affinity 2. Equilibrate the resins with 15 ml of buffer A (20 mM Tris–HCl
chromatography (IMAC) pH 8.0, 150 mM NaCl), 0.1% (v/v) Triton X-100, 10 mM
Imidazole, 1 mM DTT, 1 mM PMSF.
3. Apply filtered lysate to the equilibrated IMAC resin at a flow
rate of 1 ml/min. Wash the IMAC resin with 50 ml of buffer B
(50 mM Tris–HCl pH 8.0, 500 mM NaCl, 30 mM Imidazole).
4. Elute the bound recombinant LGALS-11 or LGALS-14 pro-
tein with 20 ml of buffer C (50 mM Tris–HCl pH 8.0, 500 mM
NaCl, 300 mM Imidazole) in 2 ml fractions.
5. Add 5 mM TCEP to the individual protein fractions and con-

centrate to a total volume of 5 mg/ml using Amicon ultracen-
trifugal filters (5000 kDa molecular-weight cutoff; Millipore).
6. Analyze the purified products by SDS-PAGE analysis (see
Subheading 3.4.1).
3.3.3 Removal of 1. Cleave the histidine tag from the purified LGALS-11 or
Hexahistidine Tag from LGALS-14 protein by incubating with human rhinovirus-3C
Recombinant LGALS11 or protease (Thermo Fisher Scientific) at 4 C for 16 h at a
LGALS-14 protease to target protein ratio of 1:100 (w/w).
2. Remove the uncleaved LGALS-11 or LGALS-14 by IMAC as
described in previous section.
3.3.4 Purification of 1. The cleaved LGALS-11 and LGALS-14 was further purified by
Recombinant LGALS-11 or size-exclusion chromatography using an ÄKTA fast protein
LGALS-14 by Size- liquid chromatography system (FPLC).
Exclusion 2. Equilibrate a Superdex S75 16/60 size exclusion column
Chromatography (SEC) (GE healthcare) with 2 column volume of buffer D (20 mM
Tris–HCl pH 8.0, 100 mM NaCl, 10 mM TCEP at 1 ml/min
flow rate.
3. Size exclusion chromatographic trace of recombinant galectin-
11 and galectin-14 corresponds dimer in solution (Fig. 5a(i), b
(i), respectively).
4. Collect the recombinant LGALS-11 or LGALS-14 in 2 ml
fractions and analyze by SDS-PAGE.
5. Pool the eluted fractions containing clear LGALS-11/LGALS-
14 and concentrate using Vivaspin protein concentrator
(MWCO3000, GE healthcare Life Sciences) to a final volume
containing 5 mg/ml.
6. For structural biology studies, further purify the recombinant
LGALS-11 or LGALS-14 by ion-exchange chromatography if
necessary.
3.4 Characterization 1. Prepare 12% polyacrylamide gel and run at 120 V for 90 min.
of Recombinant 2. Following electrophoresis, stain the gel with 50 ml of staining
LGALS-11 and solution (0.22% (w/v) Coomassie Brilliant Blue R, in 40%
LGALS-14 (v/v) ethanol and 7% (v/v) glacial acetic acid) for 2 h in an
3.4.1 SDS-PAGE Analysis
orbital shaker.
3. Destain the polyacrylamide gel for 1 h with destaining solution
(20% (v/v) ethanol and 7% (v/v) glacial acetic acid) and visua-
lize (Fig. 5a(ii), b(ii), respectively).
3.4.2 Western Blot 1. Transfer the protein samples separated by SDS-PAGE gel onto
nitrocellulose membrane (Trans-blot® mini nitrocellulose
membrane, Bio Rad) following the manufacturer’s instruction.
Fig. 5 Size-exclusion chromatography elution profile of recombinant LGALS-11 (a-i) and LGALS-14 (b-i).
Arrows indicate the elution volumes of proteins of known molecular weight (labeled in kDa). (b) SDS-PAGE
analysis of purified human rhinovirus 3c protease treated recombinant LGALS-11 in lane 1 (a-ii) and LGALS-
14 fractions in lane 1 and 2 (b-ii). Molecular weights (lane 1) are indicated on the left in kDa. Western blot
analysis of LGALS-11in lane 1 and LGALS-14 fractions in lane 1 and 2 (a-iii and b-iii, respectively). (Fig. 5 a
(i and ii) originally published in the Acta Crystallographica Section F Journal, volume 71, 2015, page
994, Cloning, expression, purification and crystallographic studies of galectin-11 from domestic sheep (Ovis
aries), (Dhanasekaran Sakthivel, Dene Littler, Adam Shahine, Sally Troy, Matthew Johnson, Jamie Rossjohn,
David Piedrafita and Travis Beddoe, Fig. 1, original copyright notice 2015. Reprinted here). (According to the
Acta F Journal copyright policy, authors are allowed to reuse images in their own publication with original
source acknowledgment)
2. Following LGALS-11 or LGALS-14 transfer onto nitrocellu-

lose membrane, incubate the membrane with 20 ml of 5%
(w/v) Bovine Serum Albumin (BSA) in 1 Phosphate-
buffered Saline (PBS) (137 mM NaCl, 2.7 mM KCL, 10 mM

Na2HPO4, 2 mM KH2PO4 pH 7.4) for 2 h at 4 C.
3. Wash the membrane three times with 1 PBS containing 1%
(v/v) Tween-20 (PBS-T) and incubate with anti-Galectin-11/
14 antibody [25] diluted to 1:1000 in PBS-T for 2 h at room
temperature.
4. Following incubation, wash the membrane three times with
PBS-T and incubate with rabbit anti-mouse IgG horseradish
peroxidase conjugate (Dako, CA) for 1 h at room temperature.
5. Following incubation, wash the membranes three times with
PBS-T and develop the membrane using the PierceTM ECL
plus western blotting substrate (Thermo Fisher) and image
(Fig. 5a(iii), b(iii), respectively).
3.5 Preparation of 1. Prime nematode-free adult merino sheep by oral infection with
Parasite-Infected 5000 H. contortus L3 larvae once a week for approximately
Gastrointestinal 12 weeks (see Notes 14 and 15).
Tissues (H. contortus- 2. Following the manufacturer’s instructions, drench the sheep
Infected Abomasum) with levamisole (10 ml/sheep).
3.5.1 Preparation of 3. Maintain the sheep nematode free for approximately 12 weeks
Parasite-Infected (by keeping in pens and feeding with commercial pellets).
Abomasal Tissue in 4. Challenge by oral administration of 5 104 H. contortus L3
Primed Sheep larvae (see Note 15).
5. At different times after challenge (depending on stage of infec-
tion to be examined) sacrifice the sheep via an overdose of
sodium pentobarbital and collect tissue samples. Open the
abomasum along the greater curvature and gently wash out
the stomach contents with slow-running tap water, before
collecting tissue from the fundic folds (see Notes 16 and 17).
3.5.2 Preparation of 1. Orally infect 3-month-old worm-free sheep once with

Abomasal Tissue from 1–5 104 H. contortus L3 larvae. Using young sheep, particu-
Sheep with a Primary larly those raised in an animal house, makes previous exposure
Infection to the parasite unlikely. Control sheep are uninfected.
2. Sacrifice the sheep at different time post-infection and collect
tissue samples as in Subheading 3.5.1, step 5.
3.6 Detection of 1. Cut tissue into small pieces and embed in OCT. Tissue pieces
Endogenous Galectins can vary in size from approximately 7 mm3 to 1 cm3 as long as
in Tissues they are fully covered by the OCT medium.
(Immunohisto- 2. Freeze embedded tissues by floating samples on the surface of
chemistry in OCT- liquid nitrogen in foil boats.
Embedded Frozen 3. Store frozen tissue blocks at 80 C.
Tissues)
4. Cut 5–10-μm sections of the OCT-embedded frozen tissue

blocks using a cryostat and transfer sections onto microscope
slides.
5. Allow sections to dry at room temperature.
6. Fix sections in 95% ethanol for 10 min at 4 C.
7. Allow fixed sections to dry, wrap slides in tissue and foil, and
store at 80 C.
8. Prior to use, bring slides to room temperature.
9. Quench endogenous peroxidase by submerging slides in per-
oxidise blocking reagent for 10 min.
10. Wash slides by immersing in PBS.
11. Incubate slides with primary anti-galectin antibody in PBS for
1–2 h at room temperature in a humidified chamber. (It is
important from this stage onwards to not allow the slides
to dry).
12. Wash slides in PBS.
13. Incubate slides with horseradish peroxidase-conjugated sec-
ondary antibody in PBS for 1 h at room temperature in a
humidified chamber (see Note 18).
15. Incubate slides in 1.5 mM DAB for 10 min at room
temperature.
16. Counterstain slides with Hematoxylin or Hematoxylin and
Eosin Y (H & E) as directed by the manufacturer and view
under light microscope (see Fig. 6).
3.7 Recombinant 1. Following the manufacturer’s instructions, label the galectin

Galectin Binding to and control probes with the thiol reactive probe
Frozen Tissue Sections 5-iodoacetamidofluorescein (We used rGST-galectin-14, and
rGST alone was the control probe).
3.7.1 Fluorescent
Labeling of Galectin Probes 2. Check that labeling did not interfere with the activity of the
and Assessment of Activity recombinant galectin by performing a hemagglutination assay
by Hemagglutination Assay post-labeling. This assay requires trypsin-treated, glutaralde-
hyde-fixed rabbit erythrocytes that can be prepared as follows
in steps 3–8.
3. Wash erythrocytes four times with 5 vol (see Note 19) of
0.15 M NaCl and resuspend as a 4% suspension (v/v) in PBS
(usually produces 2–2.5 ml of erythrocytes).
4. Add 1 mg/ml trypsin and incubate at 37 C with very mild
shaking for 1 h.
5. Wash four times with 5 vol of 0.15 M NaCl.
6. Fix in 5 vol of 1% glutaraldehyde in PBS for 1 h at room
temperature.
Fig. 6 Immunohistochemistry of OCT-embedded tissue using rabbit polyclonal anti-galectin-11 antibody and
indirect immunoperoxidase staining. H. contortus-infected abomasum collected 5 days post-challenge (a and
c) shows strong galectin-11 staining in the upper epithelial layer. In (a) the thin arrow points to the position of
a L3 larva. In (c) a thick arrow points to nuclear staining. In contrast, parasite-free (primed but not challenged)
abomasum lacks galectin-11 staining (b). “L” indicates the abomasal lumen. Magnification: (a and b) 100;
(c) 200
7. Terminate fixation with 5 vol of 0.1 M glycine in PBS at 4 C.

8. Wash the erythrocytes twice with 0.1 M glycine in PBS and
twice in PBS. The fixed erythrocytes can be stored as a 10%
suspension in PBS for 1 month.
9. For the hemagglutination assay, use V-well microtiter plates
with serial two-fold dilutions of samples in DES. To each well,
in order, add
25 μl of sample diluted in DES.
50 μl of 0.5% BSA in 0.15 M NaCl.
25 μl of 4% erythrocytes in PBS.
Then shake the plate vigorously for 30 s. Read agglutination
after 90 min.
(a) Cells as a dot ¼ no agglutination.
(b) Cells as a mat ¼ agglutination (see Fig. 7).
Fig. 7 Hemagglutination activity using rGST-galectin-14 fusion protein.

Agglutination of rabbit erythrocytes was assayed in a 96-well microtiter plate.
Assays were conducted in the presence (2–5, and) or absence (1 and 6) of
0.28 μM rGST-galectin-14, the minimum concentration to induce agglutination.
Increasing concentrations (0.78–100 mM) of lactose (2), galactose (3), or N-
acetyl-glucosamine (4) were added to inhibit agglutination. In the presence of
rGST-galectin-14 without sugar (5), the rabbit erythrocytes form a “mat” at the
bottom of the wells. The assay buffer DES alone (1) and 3 μM rGST (6) were
included as negative controls
3.7.2 Binding of 1. Prepare OCT-embedded tissue sections as in Subheading 3.6

Fluorescent Galectin to (steps 1–8).
Frozen Tissues Sections 2. Block frozen tissue sections in 3% BSA PBS for 30 min.
3. Incubate sections with fluorescein-labeled recombinant galec-
tin (e.g., rGST-galectin-14) or control (e.g., rGST) diluted to
10 μg/ml in blocking solution for 30 min at room
temperature.
4. Wash in PBS three times for 5 min, and examine under an
epifluorescence microscope (see Fig. 8).
5. To check that the galectin is binding to parasite glycans, include
100 mM lactose or galactose in the blocking solution in step 2;
if glycan-specific this should inhibit galectin binding.
Fig. 8 Binding of galectin-14 on lung and intestinal tissue sections. Frozen tissue sections from healthy sheep
were incubated with recombinant proteins; lung tissue with rGST-galectin-14-fluorescein (a) or rGST-
fluorescein (c) and intestinal tract tissue with rGST-galectin-14 (b) or inactive rGST-galectin-11 (d). Protein
binding was monitored directly by fluorescence microscopy (a and c) or with an anti-GST HRP-conjugated
antibody (b and d). The tissues (b and d) were counter-stained with hematoxylin. Magnification: (a and c),
200; (b and d), 100. (Originally published in the Glycoconjugate Journal, volume 26, 2009, page
430, Functional characterization of an eosinophil-specific galectin, ovine galectin-14, Anna R. Young, Garry
J. Barcham, Joanna M. Kemp, Jillian L. Dunphy, Andrew Nash, Els N. Meeusen, Fig. 5, original copyright
notice 2008. Reprinted here with kind permission from Springer Science and Business Media)
3.7.3 Binding of Non- 1. Prepare tissue section slides as for immunohistochemistry

fluorescent Galectins to (Subheading 3.6, steps 1–10).
Frozen Tissue Sections 2. Incubate slides with recombinant GST-galectin (e.g., rGST-
galectin-14) or control probes (e.g., inactive recombinant
GST-galectin-11 or rGST alone) at 10 μg/ml in a humidity
chamber for 1–2 h at room temperature.
4. Incubate slides with anti-GST HRP-conjugated antibody in
PBS for 1 h at room temperature in a humidified chamber.
6. Incubate slides in 1.5 mM DAB for 10 min at room
temperature.
7. Counterstain slides with hematoxylin or hematoxylin and
Eosin Y (H & E) as directed by the manufacturer and view
under light microscope (see Fig. 8).
3.8 Galectin Binding 1. Aliquot 1 ml of parasites at concentration of 1000/ml in PBS

to Parasites (H. into 1.5 ml eppendorf tubes. For adult H. contortus,
contortus and F. 3–4 worms/ml.
hepatica) 2. Wash parasites three times in PBS by spinning at 500 g for
3.8.1 Fluorescent
30 s. Parasites will pellet at the bottom. Replace supernatant
Staining of H. contortus L3,
with 1 ml PBS.
L4, and Adult Stage 3. To prevent antibody shedding, snap-freeze parasites by placing
Parasites in liquid nitrogen for 1 min.
4. Thaw in water bath at 37 C and wash once in PBS by spinning
at 500 g for 30 s.
5. Incubate parasites with 200 μl of recombinant galectin-11 for
30 min–1 h at room temperature (concentration of recombi-
nant galectin-11 used ranges from 5 μg/ml to 100 μg/ml). For
a negative control, pre-incubate recombinant galectin-11 with
100 mM of D-galactose.
6. Wash parasites to remove any unbound protein as outlined in
step 2.
7. Add 200 μl of secondary antibody, rabbit anti-ovine galectin-
11 at concentration of 1:1000. Incubate for 30 min–1 h at
room temperature.
8. Wash (as outlined in step 2) and incubate with detection
antibody, goat anti-rabbit AlexaFlour647 (1:1000) for
30 min–1 h at room temperature in the dark (wrap in foil).
9. Wash (outlined in step 2) and resuspend in 100 μl of PBS. Place
50 μl onto glass slide and coverslip.
10. Examine fluorescent binding under microscope using a water
lens to enhance focus.
3.8.2 Binding of Non- 1. Embed freshly collected liver flukes in OCT and store frozen
fluorescent Galectin Probes until use (Subheading 3.6, steps 1–6).
to F. hepatica 2. Cut 5–10 μm sections, dry and fix in 95% ethanol for 10 min at
room temperature.
3. Reduce endogenous peroxidase by submerging slides in 1.5%
hydrogen peroxide/PBS for 10 min.
4. Block sections with 3% BSA PBS for 30 min.
5. Incubate for 1 h with 10 μg/ml rGST-Galectin-14, inactive
rGST-Galectin-11, or rGST.
6. Detect recombinant protein binding by incubation with anti-
GST HRP-conjugated antibody and visualize with DAB as
described in Subheading 3.7.3, steps 3–7 (see Fig. 8).
3.9 Galectin Pull- 1. Obtain larval stage of parasite of interest, approximately

Down Assay 0.5–1 g, ensure they have been thoroughly washed in PBS.
3.9.1 Preparation of 2. Remove host-derived lectins by incubating parasites in 50 mM
Parasite Lysate of galactose and 50 mM of lactose for 4 h at 4 C.
3. Remove residual galactose and lactose by rinsing parasites
in PBS.
4. Snap-freeze parasites in liquid nitrogen and grind in a
pre-cooled mortar and pestle.
5. Rinse the ground parasite from the mortar and pestle using
2 ml of RIPA buffer.
6. Ultracentrifuge at 100,000 g for 60 min at 4 C.
7. Filter the supernatant through a 0.22 μm syringe filter.
8. Add supernatant to dialysis tube and dialyze in 10 volumes of
RIPA dialysis buffer for 2 h at room temperature, replace RIPA
dialysis buffer and dialyze overnight at 4 C.
9. Estimate protein concentration using Pierce™ BCA protein
assay kit (Thermo Fisher Scientific) following manufacturer’s
instructions and dilute all replicates to equal concentration,
500 mg/ml with RIPA dialysis buffer.
3.9.2 Conjugation of 1. Recombinantly express galectin of interest (Subheading 3.3)

Galectin to NHS-Activated and buffer exchange into galectin coupling buffer at a concen-
Sepharose and Capture of tration of 1–5 mg/ml (see Note 20).
Galectin-Specific Parasite 2. Wash 1 volume of NHS-activated Sepharose in 10 volumes of
Ligands Milli-Q water, by turning end over end on a rotating wheel for
5 min (see Note 21).
3. Either wait for the resin to settle or centrifuge at a very low
speed ~500 g and remove the supernatant, wash resin again
with another 10 volumes of Milli-Q water and remove
supernatant.
4. Activate the resin by adding 10 volumes of ice cold 1 mM HCl,

mix end over end for 10 min and remove supernatant.
5. Equilibrate the resin in 5 volumes of galectin conjugation
buffer and remove supernatant.
6. Add the recombinant galectin to the resin and add HEPES
buffer to a final concentration of 50 mM, allowing coupling for
16 h at 4 C and an additional 2 h at room temperature and
then remove supernatant (see Notes 22 and 23).
7. Block the remaining uncoupled sites by adding 15 volumes of
blocking buffer and mixing for 2 h at room temperature.
8. Wash the galectin-coupled resin six times, each time alternating
between 10 volumes of wash buffer 1 and wash buffer 2.
9. Kept the galectin coupled resin in storage buffer until pull-
down assay is performed.
10. Add 2 volumes of parasite lysate (350 μg) to the galectin
coupled resin, while mixing, allow binding to be occurred for
3 h at room temperature.
11. Wash resin three times, each time using 2 volumes of RIPA
dialysis buffer and a wash time of 10 min.
12. Elute galectin ligands by adding 2 volumes of galectin ligand
elution buffer.
3.9.3 Mass Spectrometry 1. Dry samples in a SpeedVac Concentrator attached with Savant
Preparation and Analysis Refrigerated Vapor trap.
2. Reconstitute the dried samples in 100 μl of 8 M Urea, 100 mM
of Tris–HCl pH 8.3.
3. Add 1 μl of 200 mM TECP.
4. Incubate at 21 C overnight in a thermomixer.
5. Add 4 μl of 1 M IAA and incubate at 21 C in the dark for
45 min.
6. Add 500 μl of 50 mM Tris–HCl pH 8.3 and 1 μg of
sequencing-grade trypsin.
8. Acidify the samples by adding TFA to a concentration of 1%
(v/v).
9. Desalt the peptides using poly(styrene-divinylbenzene) copol-
ymer (SDB) StageTips following the protocol of [40].
10. Dry samples in a SpeedVac Concentrator attached with Savant
Refrigerated Vapor trap.
11. Reconstitute the trypsin digested samples in 2% (v/v) ACN
and 0.1% (v/v) TFA.
12. Load sample at 5 μl/min and wash the guard column for 6 min
before switching the guard column in line with the analytical
column.
13. Separate the peptides using a flow rate of 300 nl/min using a
non-linear ACN gradient: starting at 5% (v/v) buffer B to 55%
buffer B over 120 min.
14. Collect data on an Orbitrap Elite (Thermo Fisher) in a data-
dependent acquisition mode using m/z 300–1500 as MS scan
range, collision-induced dissociation (CID) MS/MS spectra
and collect for the 20 most intense ions. Set dynamic
exclusions as: count 1, duration 90 s, and set the exclusion
list size at 500 with early expiration disabled. Other instrument
parameters for the Orbitrap include: MS scan at 120,000 reso-
lution, maximum injection time 150 ms, AGC target 1 106,
and CID at 35% energy for a maximum injection time of
150 ms with AGT target of 5000. Operate the Orbitrap Elite
in dual analyzer mode with the Orbitrap analyzer being used
for MS and the linear trap being used for MS/MS.
15. Interrogate the spectral data using MaxQuant, the appropriate
protein database for the parasite being analyzed (see Note 24),
the host species it was derived from, and common known
contaminates. Search parameters include, fixed modification:
carbamidomethylation of cysteines, Variable modifications:
acetylation of protein N-termini and methionine oxidation,
Precursor mass tolerance: 10 ppm. Product ions were searched
at 0.5 Da tolerances, minimum peptide length: 6, maximum
peptide length: 144, and validate peptide spectral matches
(PSM) using Percolator based on q-values at a 1% false discov-
ery rate (FDR).
3.10 Functional 1. Exsheath H. contortus L3 with sodium hypochlorite (Subhead-

Assays Using H. ing 3.1.4, steps 1–5).
contortus 2. Resuspend xL3 at concentration of 1000/ml in tissue culture
3.10.1 Larvae media and transfer 1 ml into a 24-well plate.
Moulting Assay 3. Add 1 ml of rgalectin-11 at concentrations between 5 and
100 μg/ml and 1 ml of appropriate controls.
4. Incubate at 40 C and 10% CO2 for 5 days.
5. Monitor moulting rate to L4 parasites by imaging samples on
day 3 and 5. Remove 500 μl of media containing the parasites
into 1.5 ml eppendorfs and spin down at 1000 g for 1 min to
resuspend parasites into 50 μl of Iodine.
6. Aliquot 10 μl onto microscopic slide and cover with coverslip.
7. Image parasites at 20 magnification, 20–30 parasites per
treatment.
Fig. 9 Galectin binding to the parasite surface. Longitudinal sections of adult F. hepatica were incubated with
rGST-Galectin-14 (a and b), rGST-Galectin-14 + 100 mM lactose (c) or inactive rGST-Galectin-11 (d) and
protein binding detected with an anti-GST HRP-conjugated antibody. The flukes were counter-stained with
hematoxylin. Magnification: (a, c, and d 40; (b) 400. (Reprinted from Veterinary Immunology and
Immunopathology, Vol 145, Anna R. Young, Garry J. Barcham, Hamish E. McWilliam, David M. Piedrafita,
Els N. Meeusen, Galectin secretion and binding to adult Fasciola hepatica during chronic liver fluke infection of
sheep, Pages No. 362–367, Copyright (2011), with permission from Elsevier)
8. Moulting into L4 can be distinctly identified by the morpho-

logical development of mouth parts (see Fig. 9).
3.10.2 Larvae Growth 1. Culture H. contortus L3 to L4 parasites by following culture

Assay Using H. protocol (Subheading 3.1.6). Once L4 is greater than 80%,
contortus L4 begin growth assay.
2. In 24-well plate, add 1 ml of L4 parasites at concentration of
1000/ml in tissue culture medium.
3. Add 1 ml of rgalectin-11 at concentrations between 5 and
100 μg/ml or appropriate controls (e.g., rGST).
4. Incubate parasites at 40 C, 10% CO2 for 5 days.
5. On day 3 and 5, remove 500 μl of media containing the para-
sites into 1.5 ml eppendorfs and spin down at 1000 g for
1 min to resuspend parasites into 50 μl of Iodine.
6. Aliquot 10 μl onto microscopic glass slide and cover with
coverslip.
7. Image parasites at 20 magnification, 20–30 parasites per
treatment.
8. Use software such as Image J to measure the area of the worms
as an indicator of growth.
3.10.3 Larvae Feeding 1. Culture H. contortus L3 to L4 parasites by following culture

Assay Using H. protocol (Subheading 3.1.6). Once L4 is greater than 80%,
contortus L4 begin feeding assay.
2. In 24-well plate, add 1 ml of H. contortus L4 larvae at concen-
tration of 1000/ml in tissue culture medium.
3. Add 1 ml rgalectin-11 at concentrations between 5 and
100 μg/ml or appropriate controls (e.g., rGST).
4. Incubate parasites at 40 C, CO2 10% for 24 h.
5. Add 50 μl of fluorescent protein such as OVA-FITC (2 mg/ml)
for 2 h. (This will be ingested if parasites are feeding.)
6. Transfer parasites into eppendorf tubes and wash three times
with PBS by spinning suspension down at 100 g for 1 min.
Resuspend in 200 μl of PBS.
7. Aliquot 10 μl onto microscopic slide and cover with coverslip.
8. Image on fluorescent microscope and count number of worms
with fluorescent esophagus/intestines (see Fig. 10).
Fig. 10 L4 H. contortus with fluorescently labeled protein located to intestines.

H. contortus L4 was incubated with 50 μl of 2 mg/ml of chicken ovalbumin-
fluorescein isothiocyanate (OVA-FITC) for 2 h and visualized under fluorescent
microscope. White arrow indicates bright fluorescence localized to intestine of a
single L4 indicating ingestion of fluorescent protein. Faint fluorescence of other
L4 represents autofluorescence and in the feeding assay is recorded as L4 which
are not feeding on fluorescently labeled protein. Magnification: 200
4 Notes
1. Ewes can be used but egg hatching is less successful due to

urine contamination. Young (between 3 months and 2 years)
animals are preferred as they are more susceptible to infection
and less likely to be pasture primed.
2. These bags were handmade using shade cloth material and
elastic banding to fit around the sheep.
3. Metacercariae can be purchased or collected from F. hepatica-
infected snails (obtained from laboratory snail cultures). We
used F. hepatica-infected Austropeplea tomentosa snails from
the Elizabeth Macarthur Agricultural Institute (NSW,
Australia). Others have purchased F. hepatica metacercariae
from Compton Paddock Laboratories (UK; VIAS strain) and
Baldwin Aquatics (USA; Oregon strain).
4. L3 are counted by taking 100 μl of solution and pipetting the
volume as microliter droplets onto a clear plastic tray. L3 can be
visualized at 10 magnification and calculated per milliliter.
5. Transfer L3 preparation into gelatin capsules just prior to
administration as capsule will quickly start to dissolve.
6. Dosing gun needs to be thoroughly pushed down the animals’
throat before capsules are released. This is done by allowing the
animal to swallow part of the dosing gun.
7. When many larvae have hatched and developed into L3, a
peanut butter-colored solution will form on the clear plastic
sheets.
8. Exsheathed L3 can be distinguished from unexsheathed L3 by
a rounder tail (see Fig. 3).
9. Adult worms can be incubated to collect eggs but this has not
been done in our laboratory.
10. F. hepatica-infected livers and bile ducts can also be sourced
from local abattoirs for the isolation of juvenile and adult liver
flukes as described in Subheading 3.1.8.
11. In addition to collecting and freezing parasites for immunohis-
tology or exogenous galectin binding, live parasites can be
collected for in vitro culture.
12. We have observed few polymorphisms in the ORF region of
both LGALS-11 and LGALS-14 genes in sheep. Hence for the
structural biology studies, it is advisable to use a synthetic gene
construct for recombinant protein production.
13. The biological activities and glycan binding activity of both
LGALS-11 and LGALS-14 proteins are strictly dependent on
maintaining the recombinant proteins in a reduced state during
recombinant protein production in Subheading 3.3.
14. Primed but not challenged sheep can provide control aboma-
sum samples for comparison.
15. For oral dosing technique, refer to Subheading 3.1.1, steps
1–4.
16. We have also collected gastrointestinal mucus from control and
parasite-infected sheep for analysis as follows. Gently scrape
mucus off the surface of the abomasum, dilute it 1/3 (w/v)
with 10 mM citric acid pH 5, centrifuge ~4000 g for 30 min,
collect the supernatant and store at 20 C.
17. Abomasal tissue samples can be processed for subsequent RNA
preparations, protein extractions, and/or immunohistology to
study galectin expression during parasite infections, or can be
used to test exogenous galectin binding.
18. We have used horseradish peroxidase-conjugated anti-rabbit
IgG to detect primary rabbit polyclonal anti-galectin
antibodies, and horseradish peroxidase-conjugated anti-
mouse secondary antibodies to detect primary murine mono-
clonal anti-galectin antibodies.
19. Collect 5 ml of rabbit blood to 5 ml Alsever’s solution (to pre-
vent coagulation). Then 1 volume is 10 ml and you can use a
50 ml tube for all the 5 vol washes.
20. Concentration-dependant aggregation was evident in LGALS-
11, hence always maintain the protein concentration below
5 mg/ml during protein production.
21. Ensure the buffer does not contain primary amines such as Tris
as this will block the binding sites of the NHS-sepharose resin.
22. The amount of NHS-sepharose resin that will need to be
conjugated will depend on the number of replicates, approxi-
mately 300 mg per replicate.
23. Successful coupling of galectin to NHS-sepharose can be con-
firmed by visualizing the purified galectin, supernatant post
coupling and boiled galectin coupled NHS-sepharose resin,
on an SDS gel, you would expect to see a decrease in concen-
tration of galectin in the supernatant and a reasonable concen-
tration of galectin in boiled resin sample.
24. If analyzing H. contortus proteins, use SwissProt Haemonchus
contortus FASTA database. Alternatively, if analyzing
F. hepatica predicted proteins from the FhD transcriptome
are publicly 1339 available on GenBank (GEVX01000000).
References
1. Young AR, Meeusen EN (2002) Galectins in 2. Kotze AC, Prichard RK (2016) Anthelmintic
parasite infection and allergic inflammation. resistance in Haemonchus contortus: history,
Glycoconj J 19(7–9):601–606. https://doi. mechanisms and diagnosis. Adv Parasitol 93:
org/10.1023/B:GLYC.0000014091. 397–428. https://doi.org/10.1016/bs.apar.
00844.0a 2016.02.012
3. Sangster NC, Cowling A, Woodgate RG galectin, ovine galectin-14. Glycoconj J

(2018) Ten events that defined anthelmintic 26(4):423–432. https://doi.org/10.1007/
resistance research. Trends Parasitol s10719-008-9190-0
34(7):553–563. https://doi.org/10.1016/j. 14. Souza B, Lambert SM, Nishi SM, Saldana GF,
pt.2018.05.001 Oliveira GGS, Vieira LS, Madruga CR,
4. Balic A, Bowles VM, Meeusen EN (2000) The Almeida MAO (2018) Collectins and galectins
immunobiology of gastrointestinal nematode in the abomasum of goats susceptible and resis-
infections in ruminants. Adv Parasitol 45: tant to gastrointestinal nematode infection. Vet
181–241. https://doi.org/10.1016/s0065- Parasitol Reg Stud Reports 12:99–105.
308x(00)45005-0 https://doi.org/10.1016/j.vprsr.2017.
5. Nisbet AJ, Meeusen EN, Gonzalez JF, Piedra- 12.001
fita DM (2016) Immunity to Haemonchus 15. Sakthivel D, Preston S, Gasser RB, Costa T,
contortus and vaccine development. Adv Para- Hernandez JN, Shahine A, Shakif-Azam MD,
sitol 93:353–396. https://doi.org/10.1016/ Lock P, Rossjohn J, Perugini MA, Gonzalez JF,
bs.apar.2016.02.011 Meeusen E, Piedrafita D, Beddoe T (2020)
6. Hein WR, Pernthaner A, Piedrafita D, Meeu- The oligomeric assembly of galectin-11 is criti-
sen EN (2010) Immune mechanisms of resis- cal for anti-parasitic activity in sheep (Ovis
tance to gastrointestinal nematode infections in aries). Commun Biol 3(1):464. https://doi.
sheep. Parasite Immunol 32(8):541–548. org/10.1038/s42003-020-01179-7
https://doi.org/10.1111/j.1365-3024.2010. 16. Preston SJ, Beddoe T, Walkden-Brown S,
01213.x Meeusen E, Piedrafita D (2015) Galectin-11:
7. Robinson N, Pleasance J, Piedrafita D, Meeu- a novel host mediator targeting specific stages
sen EN (2011) The kinetics of local cytokine of the gastrointestinal nematode parasite, Hae-
and galectin expression after challenge infec- monchus contortus. Int J Parasitol
tion with the gastrointestinal nematode, Hae- 45(12):791–796. https://doi.org/10.1016/j.
monchus contortus. Int J Parasitol ijpara.2015.06.003
41(5):487–493. https://doi.org/10.1016/j. 17. Hoorens P, Rinaldi M, Mihi B, Dreesen L,
ijpara.2010.11.006 Grit G, Meeusen E, Li RW, Geldhof P (2011)
8. Andrews SJ (1999) The life cycle of F. hepatica. Galectin-11 induction in the gastrointestinal
CABI Publishing, Wallingford, Oxon, UK, tract of cattle following nematode and proto-
1–30 zoan infections. Parasite Immunol
9. Piedrafita D, Spithill TW, Smith RE, Raadsma 33(12):669–678. https://doi.org/10.1111/j.
HW (2010) Improving animal and human 1365-3024.2011.01336.x
health through understanding liver fluke 18. Guo Z, Gonzalez JF, Hernandez JN, McNeilly
immunology. Parasite Immunol TN, Corripio-Miyar Y, Frew D, Morrison T,
32(8):572–581. https://doi.org/10.1111/j. Yu P, Li RW (2016) Possible mechanisms of
1365-3024.2010.01223.x host resistance to Haemonchus contortus
10. Overend DJ, Bowen FL (1995) Resistance of infection in sheep breeds native to the Canary
Fasciola hepatica to triclabendazole. Aust Vet J Islands. Sci Rep 6:26200. https://doi.org/10.
72(7):275–276. https://doi.org/10.1111/j. 1038/srep26200
1751-0813.1995.tb03546.x 19. Dunphy JL, Balic A, Barcham GJ, Horvath AJ,
11. Kelley JM, Elliott TP, Beddoe T, Anderson G, Nash AD, Meeusen EN (2000) Isolation and
Skuce P, Spithill TW (2016) Current threat of characterization of a novel inducible mamma-
Triclabendazole resistance in Fasciola hepatica. lian galectin. J Biol Chem
Trends Parasitol 32(6):458–469. https://doi. 275(41):32106–32113. https://doi.org/10.
org/10.1016/j.pt.2016.03.002 1074/jbc.M003739200
12. Young AR, Barcham GJ, McWilliam HE, Pie- 20. Leonidas DD, Vatzaki EH, Vorum H, Celis JE,
drafita DM, Meeusen EN (2012) Galectin Madsen P, Acharya KR (1998) Structural basis for
secretion and binding to adult Fasciola hepatica the recognition of carbohydrates by human
during chronic liver fluke infection of sheep. galectin-7. Biochemistry 37(40):13930–13940.
Vet Immunol Immunopathol 145 https://doi.org/10.1021/bi981056x
(1–2):362–367. https://doi.org/10.1016/j. 21. Lewis SK, Farmer JL, Burghardt RC, Newton
vetimm.2011.12.010 GR, Johnson GA, Adelson DL, Bazer FW,
13. Young AR, Barcham GJ, Kemp JM, Dunphy Spencer TE (2007) Galectin 15 (LGALS15):
JL, Nash A, Meeusen EN (2009) Functional a gene uniquely expressed in the uteri of sheep
characterization of an eosinophil-specific and goats that functions in trophoblast
attachment. Biol Reprod 77(6):1027–1036. the excretory/secretory products of sheep

https://doi.org/10.1095/biolreprod.107. abomasal nematode parasites: NCF and ECF
063594 in abomasal nematodes. Parasitol Res
22. Gray CA, Adelson DL, Bazer FW, Burghardt 109(3):627–635. https://doi.org/10.1007/
RC, Meeusen EN, Spencer TE (2004) Discov- s00436-011-2305-8
ery and characterization of an epithelial-specific 30. Wildblood LA, Kerr K, Clark DA, Cameron A,
galectin in the endometrium that forms crystals Turner DG, Jones DG (2005) Production of
in the trophectoderm. Proc Natl Acad Sci U S eosinophil chemoattractant activity by ovine
A 101(21):7982–7987. https://doi.org/10. gastrointestinal nematodes. Vet Immunol
1073/pnas.0402669101 Immunopathol 107(1–2):57–65. https://doi.
23. Sakthivel D, Swan J, Preston S, Shakif-Azam org/10.1016/j.vetimm.2005.03.010
MD, Faou P, Jiao Y, Downs R, Rajapaksha H, 31. Hernandez JN, Meeusen E, Rodriguez F,
Gasser R, Piedrafita D, Beddoe T (2018) Pro- Piedrafita D, Gonzalez JF (2020) Increased
teomic identification of galectin-11 and susceptibility to Haemonchus contortus infec-
14 ligands from Haemonchus contortus. tion by interleukin-5 modulation of eosinophil
PeerJ 6:e4510. https://doi.org/10.7717/ responses in sheep. Parasite Immunol 42(1):
peerj.4510 e12680. https://doi.org/10.1111/pim.
24. Sakthivel D, Littler D, Shahine A, Troy S, 12680
Johnson M, Rossjohn J, Piedrafita D, Beddoe T 32. Stear MJ, Henderson NG, Kerr A, McKellar QA,
(2015) Cloning, expression, purification and Mitchell S, Seeley C, Bishop SC (2002) Eosino-
crystallographic studies of galectin-11 from philia as a marker of resistance to Teladorsagia
domestic sheep (Ovis aries). Acta Crystallogr F circumcincta in Scottish blackface lambs. Parasi-
Struct Biol Commun 71(Pt 8):993–997. https:// tology 124(Pt 5):553–560. https://doi.org/10.
doi.org/10.1107/S2053230X15010195 1017/s0031182002001580
25. Dunphy JL, Barcham GJ, Bischof RJ, Young 33. Alvarez Rojas CA, Scheerlinck JP, Ansell BR,
AR, Nash A, Meeusen EN (2002) Isolation and Hall RS, Gasser RB, Jex AR (2016) Time-
characterization of a novel eosinophil-specific course study of the transcriptome of peripheral
galectin released into the lungs in response to blood mononuclear cells (PBMCs) from sheep
allergen challenge. J Biol Chem infected with Fasciola hepatica. PLoS One
277(17):14916–14924. https://doi.org/10. 11(7):e0159194. https://doi.org/10.1371/
1074/jbc.M200214200 journal.pone.0159194
26. Hirabayashi J, Hashidate T, Arata Y, Nishi N, 34. Alvarez Rojas CA, Ansell BR, Hall RS, Gasser
Nakamura T, Hirashima M, Urashima T, RB, Young ND, Jex AR, Scheerlinck JP (2015)
Oka T, Futai M, Muller WE, Yagi F, Kasai K Transcriptional analysis identifies key genes
(2002) Oligosaccharide specificity of galectins: involved in metabolism, fibrosis/tissue repair
a search by frontal affinity chromatography. and the immune response against Fasciola
Biochim Biophys Acta 1572(2–3):232–254. hepatica in sheep liver. Parasit Vectors 8:124.
https://doi.org/10.1016/s0304-4165(02) https://doi.org/10.1186/s13071-015-
00311-2 0715-7
27. Lopez-Lucendo MF, Solis D, Andre S, 35. Ashman K, Mather J, Wiltshire C, Jacobs HJ,
Hirabayashi J, Kasai K, Kaltner H, Gabius HJ, Meeusen E (1995) Isolation of a larval surface
Romero A (2004) Growth-regulatory human glycoprotein from Haemonchus contortus and
galectin-1: crystallographic characterisation of its possible role in evading host immunity. Mol
the structural changes induced by single-site Biochem Parasitol 70(1–2):175–179. https://
mutations and their impact on the thermody- doi.org/10.1016/0166-6851(94)00210-e
namics of ligand binding. J Mol Biol 36. Hanna RE (1980) Fasciola hepatica: glycocalyx
343(4):957–970. https://doi.org/10.1016/j. replacement in the juvenile as a possible mech-
jmb.2004.08.078 anism for protection against host immunity.
28. Rainbird MA, Macmillan D, Meeusen EN Exp Parasitol 50(1):103–114. https://doi.
(1998) Eosinophil-mediated killing of Hae- org/10.1016/0014-4894(80)90012-0
monchus contortus larvae: effect of eosinophil 37. Smith HV, Quinn R, Kusel JR, Girdwood RW
activation and role of antibody, complement (1981) The effect of temperature and antime-
and interleukin-5. Parasite Immunol tabolites on antibody binding to the outer sur-
20(2):93–103. https://doi.org/10.1046/j. face of second stage Toxocara canis larvae. Mol
1365-3024.1998.00132.x Biochem Parasitol 4(3–4):183–193. https://
29. Reinhardt S, Scott I, Simpson HV (2011) Neu- doi.org/10.1016/0166-6851(81)90017-7
trophil and eosinophil chemotactic factors in
38. Chauvin A, Bouvet G, Boulard C (1995) 11 and -14 ligands from Fasciola hepatica. Int J
Humoral and cellular immune responses to Parasitol 49(12):921–932. https://doi.org/
Fasciola hepatica experimental primary and sec- 10.1016/j.ijpara.2019.06.007
ondary infection in sheep. Int J Parasitol 40. Rappsilber J, Mann M, Ishihama Y (2007,
25(10):1227–1241. https://doi.org/10. 1896) Protocol for micro-purification, enrich-
1016/0020-7519(95)00039-5 ment, pre-fractionation and storage of peptides
39. Swan J, Sakthivel D, Cameron TC, Faou P, for proteomics using StageTips. Nat Protoc
Downs R, Rajapaksha H, Piedrafita D, Beddoe 2(8)
T (2019) Proteomic identification of galectin-
Chapter 27
Evaluation of the Bactericidal Activity of Galectins

Nourine A. Kamili, Anu Paul, Shang-Chuen Wu, Marcelo Dias-Baruffi,
Richard D. Cummings, Connie M. Arthur, and Sean R. Stowell
Abstract
Over a century ago, Karl Landsteiner discovered that blood group antigens could predict the immunologi-
cal outcome of red blood cell transfusion. While the discovery of ABO(H) blood group antigens revolu-
tionized transfusion medicine, many questions remain regarding the development and regulation of
naturally occurring anti-blood group antibody formation. Early studies suggested that blood group anti-
bodies develop following stimulation by bacteria that express blood group antigens. While this may explain
the development of anti-blood group antibodies in blood group–negative individuals, how blood group–-
positive individuals protect themselves against blood group–positive microbes remained unknown. Recent
studies suggest that several members of the galectin family specifically target blood group–positive
microbes, thereby providing innate immune protection against blood group antigen–positive microbes
regardless of the blood group status of an individual. Importantly, subsequent studies suggest that this
unique form of immunity may not be limited to blood group expressing microbes, but may reflect a more
generalized form of innate immunity against molecular mimicry. As this form of antimicrobial activity
represents a unique and unprecedented form of immunity, we will examine important considerations and
methodological approaches that can be used when seeking to ascertain the potential antimicrobial activity
of various members of the galectin family.
Key words Galectin, Blood group–positive microbes, Innate immune lectin, Molecular mimicry,
Antimicrobial
1 Introduction
The initial discovery of ABO(H) blood groups by Karl Landsteiner

over a century ago provided an unprecedented ability to predict the
immunological outcome of red blood cell transfusion [1, 2]. This
was followed by the discovery of additional red blood cell (RBC)
alloantigens that can likewise be a barrier to RBC transfusion [3–
5]. However, in contrast to anti-ABO(H) antibodies, many other
RBC alloantibodies are induced only following RBC alloantigen
exposure following pregnancy or transfusion. Collective, RBC
alloantibodies can make it difficult to find compatible RBCs for
transfusion and can directly increase morbidity and mortality in
517
518 Nourine A. Kamili et al.
transfusion-dependent individuals [4, 6–11]. Such immunological

barriers are not limited to RBC-induced alloantibodies, but similar
alloimmunization events can also make it difficult to adequately
treat patients with hemophilia or other conditions requiring cellular
or protein-based replacement therapy [12, 13]. However, models
available to study the development of RBC-induced alloantibodies
have historically been lacking. Recently established preclinical mod-
els have provided important insight in the development and con-
sequences of RBC alloantibodies [14–23]. Such studies, coupled
with similar studies done clinically, hold promise in the identifica-
tion of new strategies aimed at preventing the development of RBC
alloantibodies and the adverse consequences of hemolytic transfu-
sion reactions [24–28].
In contrast to RBC-induced alloantibody formation, many
questions remain regarding the development and regulation of
naturally occurring anti-blood group antibodies [1, 2]. Several
early studies suggested that naturally occurring antibody formation
results from the colonization of microbes that express blood group
antigens within the first few months of life [29–31]. While the
ability of blood group–positive microbes to stimulate naturally
occurring antibodies provides a mechanism for naturally occurring
antibody formation, the factors that regulate immunity to blood
group–positive microbes in blood group–positive individuals
remains undefined [32]. The ability of tolerance to prevent adaptive
immunity against blood group antigens in blood group–positive
individuals not only reduces the likelihood of autoimmunity but
also leaves an important gap in adaptive immunity. These results
suggest that protection against blood group–positive microbes in
blood group–positive individuals likely resides within the realm of
innate immunity.
As ABO(H) blood group antigens reflect carbohydrate post-
translational modifications [33, 34], factors responsible for
providing immunity toward blood group–positive microbes likely
possess carbohydrate binding activity. As a result, over one hundred
different carbohydrate-binding proteins with putative immunolog-
ical activity were examined on a glycan microarray in an effort to
determine whether innate immune factors exist that possess signifi-
cant blood group binding activity [35–38]. While many carbohy-
drate binding proteins, including galectins, display reactivity
toward non-blood group antigens and can directly regulate host
cell function through the engagement of these carbohydrates [39–
48], several members of the galectin family, in particular galectin-4
(Gal-4) and galectin-8 (Gal-8), exhibited significant preference for
blood group antigens [38, 49–51].
While recent studies suggest that blood group A individuals
may be more likely to not only contract SARS-CoV-2 but likewise
experience a more severe disease course following infection, the
underlying mechanism responsible for this apparent blood group A
Evaluation of the Bactericidal Activity of Galectins 519
preference has remained unknown [52–54]. Interestingly, the

receptor binding domain of SARS-CoV-2 responsible for mediat-
ing viral attachment to host cells possesses galectin-like features,
suggesting the direct engagement of blood group A may influence
this process. Consistent with this, SARS-CoV-2 exhibits a binding
preference for the type I blood group A antigen uniquely expressed
on respiratory epithelial cells [55]. A similar preference for the
receptor binding domain of SARS-CoV, which also appears to
possess a slight blood group A preference, was also observed
[55]. These results suggest that in addition to possible differences
in the development of protective anti-SARS-CoV-2 antibodies
[56–58], direct interaction with host blood group polymorphisms
may influence disease progression.
Although galectins have been implicated in a wide variety of
biological processes [43], including antimicrobial immunity against
a variety of organisms [59–61], whether galectins could directly
impact bacterial viability remained untested. Gal-4 and Gal-8 not
only recognized blood group antigens in a microarray format, but
these proteins specifically engage and killed blood group–positive
microbes [50, 62, 63]. These results suggest that Gal-4 and Gal-8
provide innate immunity against blood group–positive microbes
and thus fill an important gap in adaptive immunity against blood
group–positive microbes in blood group–positive individuals
[64]. In addition to recognizing blood group–positive microbes,
recent studies using microbial glycan microarrays suggest that
Gal-4 and Gal-8 display very specific interactions with a variety of
microbes that share the common feature of expressing mammalian-
like carbohydrate antigens, while failing to recognize microbes with
unrelated antigens [65, 66]. Similar to the impact of Gal-4 and
Gal-8 on blood group–positive microbes, galectin engagement of
additional gram-negative and gram-positive microbes that express
other mammalian-like antigens likewise results in impaired micro-
bial viability [65, 66]. These results strongly suggest that galectins
may provide a unique and broad form of innate immunity against
molecular mimicry. In contrast to their ability to recognize and kill
microbes expressing self-like antigens, Gal-4 and Gal-8 recognize
the same antigenic determinants on mammalian cells, yet fail to
induce detectable changes in membrane integrity or cellular viabil-
ity [38, 50, 66]. While complex pathways, such as those that
regulate complement, typically protect mammalian cells from
injury [67], Gal-4 and Gal-8 appear to intrinsically possess the
unique ability to recognize the exact same antigenic determinant
on a microbe and mammalian cell, while only impacting microbial
viability. As a result, galectin antimicrobial activity represents a
unique form of immunity with significant implications in innate
immune protection against molecular mimicry (Fig. 1).
As Gal-4- and Gal-8-mediated immunity requires recognition
of self-like antigens on microbes, this form of immunity
Fig. 1 Galectins provide innate immunity against blood group–positive microbes. While elimination of self-
reactive immune cells reduces the probability of autoimmunity, a fitness cost would be anticipated due to the
reduced ability of an individual to respond to self-like antigens on pathogens. However, compensatory
mechanisms at the level of innate immunity may exist, whereby individuals protect themselves against
self-like antigens. Gal-4 and Gal-8 uniquely recognize pathogens baring self-antigens, providing a possible
mechanism, whereby individuals protect themselves against self-antigen-positive pathogens despite reduced
adaptive immunity toward these microbes
fundamentally differs from defensins and other cationic antimicro-

bial peptides that engage features unique to an intact microbial
membrane [68, 69]. The requirement of Gal-4 and Gal-8 for
distinct antigen recognition not only suggests their potential role
in immunological protection against molecular mimicry, but also
requires unique considerations when assessing microbial recogni-
tion and killing. In this work, we will describe the methods
employed to examine the antimicrobial activity of Gal-4 and
Gal-8 toward microbes, using blood group–positive microbes as a
model system.
2 Materials
2.1 Bacteria 1. Luria Bertani (LB) media (Difco).

Preparation 2. E. coli O86 (ATCC 12701).
3. E. coli K12 (ATCC 29425).
4. Luria Bertani agar (Difco).
5. 100 15 mm sterile disposable petri dishes (Fisher).
6. 17 100 mm polystyrene tubes, dual position cap, sterile
(USA Scientific).
9. Bacteria cell spreader (sterile).
11. 37 C incubator/shaker.
12. 37 C incubator (stationary).
13. Autoclave.
2.2 Galectin 1. Recombinant galectin.

Preparation 2. PD10 column (GE Healthcare).
3. 0.2-micron filter (Fisher).
4. Phosphate-buffered saline standard pH 7.4 (Hyclone).
2.3 Assessing 1. Luria Bertani media (Difco).

Antimicrobial Activity 2. Luria Bertani agar (Difco).
3. Phosphate-buffered saline standard pH 7.4 (Hyclone).
4. Bacteria cell spreader (sterile).
6. Sterile 96-well, round bottom plate.
7. 12 75 mm polystyrene tubes with dual position polyethylene
caps (USA Scientific).
8. 1.7-mL snap cap microcentrifuge tube (Sigma-Aldrich).
9. 37 C incubator/shaker.
10. 37 C incubator (stationary).
3 Methods
3.1 Bacteria 1. Bacterial glycerol stocks should be stored at 80 C.

Preparation 2. Using a sterilized loop under a sterile hood, transfer a loop of
glycerol stock to 1 mL of sterile LB media in a sterile 15 mL
culture tube (17 100 mm polystyrene tube).
3. Incubate culture at 37 C overnight on an orbital shaker at
250 revolutions per minute (rpm).
4. Following the overnight incubation, dilute culture 1:100 into
LB media by placing 0.05 mL of the overnight culture in 5 mL
of sterile LB media.
5. Allow this culture to incubate at 37 C on an orbital shaker for
2.5–3 h.
6. Remove 0.5 mL of culture from the sample and determine the
OD at 600 nm (OD600). If the OD600 is between 0.3 and
0.5, then proceed to step 7. If the value is below 0.3, continue
to check the culture every 30 min until an OD600 of between
0.3 and 0.5 is achieved. If the culture is over 0.5 OD600, then
repeat step 4 with examination of the OD600 at 30-min inter-
vals until an OD600 of 0.3 is achieved.
7. Dilute the culture 1:100 by placing 0.5 mL of the culture into

50 mL of sterile LB media in an autoclaved 200-mL Erlen-
meyer flask.
8. Every 30 min, remove 0.6 mL of this culture. Place 100 μL of
this sample into a sterile Eppendorf tube and use the rest to
record the OD600 of the culture.
9. To determine the corresponding CFU value of the culture to
the OD600 reading, perform 6 tenfold serial dilutions in sterile
LB media.
10. Plate 50 μL of each dilution onto three separate LB agar plates
and spread the suspension over the entire face of the agar plate
using a sterile bacterial cell spreader.
11. Allow plate to dry for 2–3 min under tissue culture hood.
12. Incubate plates face down at 37 C for 12 h or overnight and
then remove plates from incubator.
13. Repeat steps 7–10 every 30 min to obtain an OD600 value
with a corresponding set of dilutions plated on LB agar for
CFU enumeration.
14. Following step 12, count the number of colony forming units
on the corresponding plate (see Note 1).
15. Use the dilution calculations to extrapolate the number of
CFUs in the starting material of each sample.
16. Once the a relationship between OD600 and CFU has been
established, use the OD600 readings to provide an estimate of
the number of bacteria typically present at a particular OD600
value. Additionally, use this data to determine the normal
growth kinetics of the bacterial culture.
3.1.1 Preparing E. coli 17. Inoculate a fresh overnight culture of E. coli O86 and grow
O86 for Bacteria Killing overnight shaking at 37 C as outlined in steps 1–3 of
Assay Subheading 3.1.
18. Use 100 μL of this overnight culture to re-inoculate 2 mL of
sterile culture media.
19. Grow bacteria to mid-log phase (approximately OD600
0.3–0.5, or as determined empirically in steps 1–16).
20. Re-inoculate fresh sterile culture media for killing assay at a 1:
100 dilution (see Note 2).
21. Grow fresh culture to mid-log phase.
22. Dilute culture in sterile LB mediate to 108 cells/mL (as de-
termined empirically in steps 1–16) (see Note 3).
3.2 Galectin
Preparation
1. Thaw frozen galectin stocks on ice at 4 C.

2. To remove βME and lactose, prepare a PD10 column for gel
filtration by equilibrating the column with 5 column volumes
of cold PBS (pH 7.4) (see Note 4).
3. Add 1 mL of the recombinant galectin solution to the PBS
re-equilibrated PD10 column and collect 0.5 mL fractions.
4. Following complete penetration of the galectin solution into
the PD10 column, add 2 mL of cold PBS.
5. Continue to collect 0.5-mL fractions while adding additional
PBS as needed to prevent the column from drying.
6. To determine which fractions may have galectin protein, exam-
ine protein content by simply measuring the OD at 280 nm
(OD280) of each fraction. This is typically done by diluting
each fraction tenfold (i.e., 10 μL of each fraction in 90 μL PBS)
followed by the examination of the OD280 (see Note 5).
7. Once positive fractions are identified, pool galectin-containing
fractions and re-evaluate the OD280 to determine the final
concentration of pooled galectin.
8. Filter-sterilize galectin using 0.2-μm filter. Maintain galectin
sterility after this point (see Note 6).
9. Once the galectin has been filtered, re-evaluate the OD280 to
determine the final galectin concentration.
10. Dilute galectin to the appropriate concentration with sterile
PBS using aseptic technique under a sterile hood (see Note 7).
11. In addition, hapten inclusion control solutions, which consist
of LB solution containing 100 mM TDG (galectin inhibitor)
or 100 mM sucrose (non-inhibitor control) may be added to
control samples. Control solutions are diluted fivefold to
achieve a final concentration of 20 mM in control samples
(see Note 8).
3.3 Assessing 1. Once galectin is diluted to the appropriate concentration and

Antimicrobial Activity bacterial cultures are grown to mid-log phase, a final bacteria
concentration of 108 bacteria/mL should be obtained.
2. Mix galectin and bacteria and place in optimal growth condi-
tions for the microbe being examined.
3. Following 2 h incubation, bacteria should be immediately
removed (see Note 9).
4. To determine the corresponding CFU value of each condition,
perform 6 tenfold serial dilutions in sterile LB media.
5. Plate 50 μL of each dilution onto three separate LB agar plates
and spread the suspension over the entire face of the agar plate
using a sterile bacteria cell spreader.
6. Allow plate to dry for 2–3 min under tissue culture hood.
7. Incubate plates face down at 37 C for 12 h or overnight and
then remove plates from incubator.
8. Count the number of colony-forming units for each dilution.
9. Use the dilution calculations to extrapolate the number of
CFUs in the starting material of each sample (see Notes 10
and 11).
4 Notes
1. Meticulous care should be employed when enumerating CFU

using these methods to ensure that the potential antimicrobial
activity of galectins is not over- or underestimated. In addition,
the drop plate method can be used in place of the spread
method for CFU enumeration [70]. Each of these methods
possess intrinsic advantages and limitations [70].
2. The number of samples needed to do a particular experiment
will dictate the amount of media used for this step. When
examining alternative microbes, follow recommended guide-
lines for ideal growth conditions and empirically determine the
optimal conditions for examining potential galectin antimicro-
bial activity.
3. Bacteria used in killing assays should be in mid-log phase. This
is the reason for repeat inoculations. Inoculation with over-
night culture will contain mostly dead or stationary post-log
phase bacteria. However, following a second inoculation, bac-
teria that are mostly in the log phase of growth will be present.
This may be important during the killing assay, as many bacteria
are susceptible to particular killing mechanisms during active
growth. In addition, examination of microbes harvested from
post-log growth conditions often results in cultures with sig-
nificant lipopolysaccharide (LPS) contamination. As the blood
group antigen on blood group–positive microbes resides on
LPS, significant increases in LPS concentrations secondary to
overgrowth conditions may artificially inhibit galectin-
mediated antimicrobial activity.
4. βME may interfere with the biological activity of cells in general
and the target microbes in particular [71–74]. In addition,
lactose can readily inhibit galectin–carbohydrate interactions.
Inclusion of lactose can be used as a control but must be
removed to assess the potential biological activity of a particular
galectin in the absence of hapten inhibitors. When assessing the
potential dependency of galectin microbial interactions or anti-
microbial activity, thiodigalactoside (TDG) may be employed
as an inhibitor of galectin–carbohydrate interaction instead of
lactose, as it is relatively inert to metabolism.
5. To extrapolate the protein concentration from the OD280

being examined to calculate actual concentration in
mg/mL. The following websites http://www.basic.northwest
ern.edu/biotools/proteincalc.html and http://web.expasy.
org/protparam/protparam-doc.html offer explanation and
assistance in calculating extinction coefficient and using this
calculation to determine the actual concentration of a given
protein in mg/mL, including caveats as to how these numbers
may differ from the actual values. As methods of calculating the
extinction coefficient only provide estimates, alternative
approaches can be used to empirically determine these values.
These include using a Bradford assay to calculate protein con-
centration or simply re-equilibrating the recombinant protein
directly into water, lyophilizing, weighing, and then resuspend-
ing in the buffer of choice followed by empirically determining
the extinction coefficients for a particular galectin family
member.
6. This should be done before a final concentration is determined
and final dilutions are made since some galectin could be lost
during the sterile filtration step.
7. Typically, a target concentration of five times the desired final
concentration is employed to allow for a 1:5 dilution of galectin
with the target bacteria.
8. Both lactose and TDG are effective inhibitors of galectin car-
bohydrate binding. Lactose is typically easier to obtain, but also
can be metabolized with a variety of potential alterations in the
test system. Incubation of hapten solutions with recombinant
galectins prior to introducing microbes should facilitate hapten
engagement of galectin prior to galectin incubation with
microbes.
9. A 2-h incubation is not necessary to observe the antimicrobial
activity. Shorter intervals can be used and should be empirically
determined. Longer intervals allow easy visual determination of
the potential antimicrobial activity of a particular galectin for a
given microbe before plating.
10. Plating accuracy is the most challenging step of this assay. As a
result, significant care should be taken when plating samples on
LB agar to ensure even distribution of the sample across the
entire agar plate.
11. CFU counts in PBS-treated control samples can be used to
represent 100% of normal growth for a given bacteria. Galectin
antimicrobial activity may be shown as an extrapolation of the
difference in CFU between PBS-treated and galectin-treated
bacteria, typically shown as “CFU remaining (%)” (see Fig. 2).
Fig. 2 Galectin-4 and Galectin-8 killing blood group–positive microbes require carbohydrate recognition.
Quantification of viable bacteria after BGB+ E. coli were mixed with 5 μM Gal-1, Gal-3, Gal-4 or Gal-8 (a), 5 μM
Gal-4 with or without 20 mM lactose (Lac) or 20 mM sucrose (Sucr.) (b), 5 μM Gal-8 with or without 20 mM
lactose (Lac) or 20 mM sucrose (Sucr.) (c) or the indicated concentrations of Gal-1, Gal-3, Gal-4, and Gal-8 (d).
In each experiment, bacteria were quantified by dilution plating. Error bars represent means SD. This
research was originally published in Nature Medicine. Stowell SR, Arthur CM, Dias-Baruffi M, Rodrigues LC,
Gourdine JP, Heimburg-Molinaro J, Ju T, Molinaro RJ, Rivera-Marrero C, Xia B, Smith DF, Cummings
RD. Innate immune lectins kill bacteria expressing blood group antigen. 2010 Mar;16(3):295–301
Acknowledgments

Grants F31 HL 138934, R01 HL154034 to C.M.A. and U01
CA242109 to S.R.S.
References
1. Stowell CP, Stowell SR (2019) Biologic roles 5. Yazdanbakhsh K, Ware RE, Noizat-Pirenne F
of the ABH and Lewis histo-blood group anti- (2012) Red blood cell alloimmunization in
gens part I: infection and immunity. Vox Sang sickle cell disease: pathophysiology, risk factors,
114(5):426–442. https://doi.org/10.1111/ and transfusion management. Blood 120(3):
vox.12787 528–537. https://doi.org/10.1182/blood-
2. Stowell SR, Stowell CP (2019) Biologic roles 2011-11-327361
of the ABH and Lewis histo-blood group anti- 6. Rosse WF, Gallagher D, Kinney TR, Castro O,
gens part II: thrombosis, cardiovascular disease Dosik H, Moohr J, Wang W, Levy PS (1990)
and metabolism. Vox Sang 114(6):535–552. Transfusion and alloimmunization in sickle cell
https://doi.org/10.1111/vox.12786 disease. The cooperative study of sickle cell
3. Ambruso DR, Githens JH, Alcorn R, Dixon disease. Blood 76(7):1431–1437
DJ, Brown LJ, Vaughn WM, Hays T (1987) 7. Vidler JB, Gardner K, Amenyah K, Mijovic A,
Experience with donors matched for minor Thein SL (2015) Delayed haemolytic transfu-
blood group antigens in patients with sickle sion reaction in adults with sickle cell disease: a
cell anemia who are receiving chronic transfu- 5-year experience. Br J Haematol 169(5):
sion therapy. Transfusion 27(1):94–98 746–753. https://doi.org/10.1111/bjh.
4. Vichinsky EP, Earles A, Johnson RA, Hoag 13339
MS, Williams A, Lubin B (1990) Alloimmuni- 8. Habibi A, Mekontso-Dessap A, Guillaud C,
zation in sickle cell anemia and transfusion of Michel M, Razazi K, Khellaf M, Chami B,
racially unmatched blood. N Engl J Med Bachir D, Rieux C, Melica G, Godeau B,
322(23):1617–1621. https://doi.org/10. Galacteros F, Bartolucci P, Pirenne F (2016)
1056/NEJM199006073222301 Delayed hemolytic transfusion reaction in adult
sickle-cell disease: presentations, outcomes, Insight 3(22). https://doi.org/10.1172/jci.

and treatments of 99 referral center episodes. insight.121631
Am J Hematol 91(10):989–994. https://doi. 17. Dean CL, Maier CL, Roback JD, Stowell SR
org/10.1002/ajh.24460 (2018) Multiple hemolytic transfusion reac-
9. Narbey D, Habibi A, Chadebech P, Mekontso- tions misinterpreted as severe vaso-occlusive
Dessap A, Khellaf M, Lelievre JD, Godeau B, crisis in a patient with sickle cell disease. Trans-
Michel M, Galacteros F, Djoudi R, fusion 59(2):448–453. https://doi.org/10.
Bartolucci P, Pirenne F (2017) Incidence and 1111/trf.15010
predictive score for delayed hemolytic transfu- 18. Chonat S, Quarmyne MO, Bennett CM, Dean
sion reaction in adult patients with sickle cell CL, Joiner CH, Fasano RM, Stowell SR (2018)
disease. Am J Hematol 92(12):1340–1348. Contribution of alternative complement path-
https://doi.org/10.1002/ajh.24908 way to delayed hemolytic transfusion reaction
10. Chonat S, Graciaa S, Shin HS, Newton JG, in sickle cell disease. Haematologica 103(10):
Quarmyne MO, Boudreaux J, Tang A, Zerra e483–e485. https://doi.org/10.3324/
PE, Rollins MR, Josephson CD, Brown C, haematol.2018.194670
Joiner CH, Fasano RM, Stowell SR (2020) 19. Arthur CM, Chonat S, Fasano R, Yee MEM,
Eculizumab for complement mediated throm- Josephson CD, Roback JD, Stowell SR (2019)
botic microangiopathy in sickle cell disease. Examining the role of complement in predict-
Haematologica 105(12):2887–2891. https:// ing, preventing, and treating hemolytic trans-
doi.org/10.3324/haematol.2020.262006 fusion reactions. Transfus Med Rev 33(4):
11. Thein SL, Pirenne F, Fasano RM, Habibi A, 217–224. https://doi.org/10.1016/j.tmrv.
Bartolucci P, Chonat S, Hendrickson JE, Stow- 2019.09.006
ell SR (2020) Hemolytic transfusion reactions 20. Dean CL, Maier CL, Chonat S, Chang A, Car-
in sickle cell disease: underappreciated and den MA, El Rassi F, McLemore ML, Stowell
potentially fatal. Haematologica 105(3): SR, Fasano RM (2019) Challenges in the treat-
5 3 9 – 5 4 4 . h t t p s : // d o i . o r g / 1 0 . 3 3 2 4 / ment and prevention of delayed hemolytic
haematol.2019.224709 transfusion reactions with hyperhemolysis in
12. Zerra PE, Cox C, Baldwin WH, Patel SR, sickle cell disease patients. Transfusion 59(5):
Arthur CM, Lollar P, Meeks SL, Stowell SR 1698–1705. https://doi.org/10.1111/trf.
(2017) Marginal zone B cells are critical to 15227
factor VIII inhibitor formation in mice with 21. Mener A, Patel SR, Arthur CM, Stowell SR
hemophilia A. Blood 130(23):2559–2568. (2019) Antibody-mediated immunosuppres-
https://doi.org/10.1182/blood-2017- sion can result from RBC antigen loss indepen-
05-782912 dent of Fcgamma receptors in mice.
13. Arthur CM ZP, Shin S, Wang J, Song X, Doer- Transfusion 59(1):371–384. https://doi.org/
ing CB, Lollar P, Meeks S, Stowell SR (2021) 10.1111/trf.14939
Non-human glycans can regulate anti-FVIII 22. Chonat S, Mener A, Verkerke H, Stowell SR
antibody formation in mice. Blood (in press) (2020) Role of complement in alloimmuniza-
14. Stowell SR, Liepkalns JS, Hendrickson JE, tion and hyperhemolysis. Curr Opin Hematol
Girard-Pierce KR, Smith NH, Arthur CM, 27(6):406–414. https://doi.org/10.1097/
Zimring JC (2013) Antigen modulation con- MOH.0000000000000610
fers protection to red blood cells from antibody 23. Zerra PE, Patel SR, Jajosky RP, Arthur CM,
through Fcgamma receptor ligation. J Immu- McCoy JW, Allen JWL, Chonat S, Fasano RM,
nol 191(10):5013–5025. https://doi.org/10. Roback JD, Josephson CD, Hendrickson JE,
4049/jimmunol.1300885 Stowell SR (2021) Marginal zone B cells medi-
15. Mener A, Arthur CM, Patel SR, Liu J, Hen- ate a CD4 T-cell-dependent extrafollicular
drickson JE, Stowell SR (2018) Complement antibody response following RBC transfusion
component 3 negatively regulates antibody in mice. Blood 138(8):706–721
response by modulation of red blood cell anti- 24. Zimring JC, Spitalnik SL (2015) Pathobiology
gen. Front Immunol 9:676. https://doi.org/ of transfusion reactions. Annu Rev Pathol 10:
10.3389/fimmu.2018.00676 83–110. https://doi.org/10.1146/annurev-
16. Mener A, Patel SR, Arthur CM, Chonat S, pathol-012414-040318
Wieland A, Santhanakrishnan M, Liu J, Maier 25. Sullivan HC, Gerner-Smidt C, Nooka AK,
CL, Jajosky RP, Girard-Pierce K, Bennett A, Arthur CM, Thompson L, Mener A, Patel SR,
Zerra PE, Smith NH, Hendrickson JE, Stowell Yee M, Fasano RM, Josephson CD, Kaufman
SR (2018) Complement serves as a switch RM, Roback JD, Lonial S, Stowell SR (2017)
between CD4+ T cell-independent and Daratumumab (anti-CD38) induces loss of
-dependent RBC antibody responses. JCI. CD38 on red blood cells. Blood 129(22):
3033–3037. https://doi.org/10.1182/ Langerhans cells. Nat Rev Immunol 2(2):

blood-2016-11-749432 77–84. https://doi.org/10.1038/nri723
26. Patel SR, Bennett A, Girard-Pierce K, Maier 36. van Kooyk Y, Rabinovich GA (2008) Protein-
CL, Chonat S, Arthur CM, Zerra PE, glycan interactions in the control of innate and
Mener A, Stowell SR (2018) Recipient priming adaptive immune responses. Nat Immunol
to one RBC alloantigen directly enhances 9(6):593–601. https://doi.org/10.1038/ni.
subsequent alloimmunization in mice. Blood f.203
Adv 2(2):105–115. https://doi.org/10. 37. Stowell SR, Arthur CM, Mehta P, Slanina KA,
1182/bloodadvances.2017010124 Blixt O, Leffler H, Smith DF, Cummings RD
27. Patel SR, Gibb DR, Girard-Pierce K, Zhou X, (2008) Galectin-1, -2, and -3 exhibit differen-
Rodrigues LC, Arthur CM, Bennett AL, tial recognition of sialylated glycans and blood
Jajosky RP, Fuller M, Maier CL, Zerra PE, group antigens. J Biol Chem 283(15):
Chonat S, Smith NH, Tormey CA, Hendrick- 10109–10123. https://doi.org/10.1074/jbc.
son JE, Stowell SR (2018) Marginal zone B M709545200
cells induce alloantibody formation following 38. Stowell SR, Arthur CM, Slanina KA, Horton
RBC transfusion. Front Immunol 9:2516. JR, Smith DF, Cummings RD (2008) Dimeric
https://doi.org/10.3389/fimmu.2018. Galectin-8 induces phosphatidylserine expo-
02516 sure in leukocytes through polylactosamine
28. Sullivan HC, Arthur CM, Thompson L, Patel recognition by the C-terminal domain. J Biol
SR, Stowell SR, Hendrickson JE, Lazarus AH Chem 283(29):20547–20559. https://doi.
(2018) Anti-RhD reduces levels of detectable org/10.1074/jbc.M802495200
RhD antigen following anti-RhD infusion. 39. Karmakar S, Stowell SR, Cummings RD,
Transfusion 58(2):542–544. https://doi.org/ McEver RP (2008) Galectin-1 signaling in leu-
10.1111/trf.14452 kocytes requires expression of complex-type
29. Springer GF, Williamson P, Brandes WC N-glycans. Glycobiology 18(10):770–778
(1961) Blood group activity of gram-negative 40. Stowell SR, Cho M, Feasley CL, Arthur CM,
bacteria. J Exp Med 113(6):1077–1093 Song X, Colucci JK, Karmakar S, Mehta P,
30. Springer GF, Horton RE (1969) Blood group Dias-Baruffi M, McEver RP, Cummings RD
isoantibody stimulation in man by feeding (2009) Ligand reduces galectin-1 sensitivity
blood group-active bacteria. J Clin Invest to oxidative inactivation by enhancing dimer
48(7):1280–1291. https://doi.org/10.1172/ formation. J Biol Chem 284(8):4989–4999.
JCI106094 https://doi.org/10.1074/jbc.M808925200
31. Yi W, Shao J, Zhu L, Li M, Singh M, Lu Y, 41. Liu FT, Rabinovich GA (2010) Galectins: reg-
Lin S, Li H, Ryu K, Shen J, Guo H, Yao Q, ulators of acute and chronic inflammation. Ann
Bush CA, Wang PG (2005) Escherichia coli N Y Acad Sci 1183:158–182. https://doi.org/
O86 O-antigen biosynthetic gene cluster and 10.1111/j.1749-6632.2009.05131.x
stepwise enzymatic synthesis of human blood 42. Poland PA, Rondanino C, Kinlough CL,
group B antigen tetrasaccharide. J Am Chem Heimburg-Molinaro J, Arthur CM, Stowell
Soc 127(7):2040–2041. https://doi.org/10. SR, Smith DF, Hughey RP (2011) Identifica-
1021/ja045021y tion and characterization of endogenous galec-
32. Garratty G (2000) Blood groups and disease: a tins expressed in Madin Darby canine kidney
historical perspective. Transfus Med Rev 14(4): cells. J Biol Chem 286(8):6780–6790. https://
291–301. https://doi.org/10.1053/tmrv. doi.org/10.1074/jbc.M110.179002
2000.16228 43. Cerliani JP, Stowell SR, Mascanfroni ID,
33. Yamamoto F, Clausen H, White T, Marken J, Arthur CM, Cummings RD, Rabinovich GA
Hakomori S (1990) Molecular genetic basis of (2011) Expanding the universe of cytokines
the histo-blood group ABO system. Nature and pattern recognition receptors: galectins
345(6272):229–233. https://doi.org/10. and glycans in innate immunity. J Clin Immu-
1038/345229a0 nol 31(1):10–21. https://doi.org/10.1007/
34. Lee-Sundlov MM, Stowell SR, Hoffmeister s10875-010-9494-2
KM (2020) Multifaceted role of glycosylation 44. Jiang K, Rankin CR, Nava P, Sumagin R,
in transfusion medicine, platelets, and red Kamekura R, Stowell SR, Feng M, Parkos CA,
blood cells. J Thromb Haemost 18(7): Nusrat A (2014) Galectin-3 regulates
1535–1547. https://doi.org/10.1111/jth. desmoglein-2 and intestinal epithelial intercel-
14874 lular adhesion. J Biol Chem 289(15):
35. Figdor CG, van Kooyk Y, Adema GJ (2002) 10510–10517. https://doi.org/10.1074/jbc.
C-type lectin receptors on dendritic cells and M113.538538
45. Robinson BS, Arthur CM, Evavold B, Protti A, Aghemo A, Lleo A, Biondi A,
Roback E, Kamili NA, Stowell CS, Vallecillo- Caballero-Garralda A, Gori A, Tanck A, Car-
Zuniga ML, Van Ry PM, Dias-Baruffi M, reras Nolla A, Latiano A, Fracanzani AL,
Cummings RD, Stowell SR (2019) The Peschuck A, Julia A, Pesenti A, Voza A,
sweet-side of leukocytes: galectins as master Jimenez D, Mateos B, Nafria Jimenez B,
regulators of neutrophil function. Front Quereda C, Paccapelo C, Gassner C,
Immunol 10:1762. https://doi.org/10. Angelini C, Cea C, Solier A, Pestana D,
3389/fimmu.2019.01762 Muniz-Diaz E, Sandoval E, Paraboschi EM,
46. Rodrigues LC, Kabeya LM, Azzolini A, Cerri Navas E, Garcia Sanchez F, Ceriotti F,
DG, Stowell SR, Cummings RD, Lucisano- Martinelli-Boneschi F, Peyvandi F, Blasi F,
Valim YM, Dias-Baruffi M (2019) Galectin-1 Tellez L, Blanco-Grau A, Hemmrich-Stanisak-
modulation of neutrophil reactive oxygen spe- G, Grasselli G, Costantino G, Cardamone G,
cies production depends on the cell activation Foti G, Aneli S, Kurihara H, ElAbd H, My I,
state. Mol Immunol 116:80–89. https://doi. Galvan-Femenia I, Martin J, Erdmann J,
org/10.1016/j.molimm.2019.10.001 Ferrusquia-Acosta J, Garcia-Etxebarria K,
47. Robinson BS, Saeedi B, Arthur CM, Owens J, Izquierdo-Sanchez L, Bettini LR, Sumoy L,
Naudin C, Ahmed N, Luo L, Jones R, Neish A, Terranova L, Moreira L, Santoro L,
Stowell SR (2020) Galectin-9 is a novel regula- Scudeller L, Mesonero F, Roade L, Ruhlemann
tor of epithelial restitution. Am J Pathol MC, Schaefer M, Carrabba M, Riveiro-
190(8):1657–1666. https://doi.org/10. Barciela M, Figuera Basso ME, Valsecchi MG,
1016/j.ajpath.2020.04.010 Hernandez-Tejero M, Acosta-Herrera M,
D’Angio M, Baldini M, Cazzaniga M,
48. Vallecillo-Zuniga ML, Rathgeber MF, Poulson Schulzky M, Cecconi M, Wittig M,
PD, Hayes S, Luddington JS, Gill HN, Ciccarelli M, Rodriguez-Gandia M,
Teynor M, Kartchner BC, Valdoz J, Bocciolone M, Miozzo M, Montano N,
Stowell C, Markham AR, Arthur C, Stowell S, Braun N, Sacchi N, Martinez N, Ozer O,
Van Ry PM (2020) Treatment with galectin-1 Palmieri O, Faverio P, Preatoni P, Bonfanti P,
improves myogenic potential and membrane Omodei P, Tentorio P, Castro P, Rodrigues
repair in dysferlin-deficient models. PLoS One PM, Blandino Ortiz A, de Cid R, Ferrer R,
15(9):e0238441. https://doi.org/10.1371/ Gualtierotti R, Nieto R, Goerg S,
journal.pone.0238441 Badalamenti S, Marsal S, Matullo G, Pelusi S,
49. Carlsson S, Oberg CT, Carlsson MC, Juzenas S, Aliberti S, Monzani V, Moreno V,
Sundin A, Nilsson UJ, Smith D, Cummings Wesse T, Lenz TL, Pumarola T, Rimoldi V,
RD, Almkvist J, Karlsson A, Leffler H (2007) Bosari S, Albrecht W, Peter W, Romero-
Affinity of galectin-8 and its carbohydrate rec- Gomez M, D’Amato M, Duga S, Banales JM,
ognition domains for ligands in solution and at Hov JR, Folseraas T, Valenti L, Franke A, Karl-
the cell surface. Glycobiology 17(6):663–676. sen TH, Severe Covid GG (2020) Genome-
https://doi.org/10.1093/glycob/cwm026 wide association study of severe covid-19 with
50. Stowell SR, Arthur CM, Dias-Baruffi M, respiratory failure. N Engl J Med 383(16):
Rodrigues LC, Gourdine JP, Heimburg- 1522–1534. https://doi.org/10.1056/
Molinaro J, Ju T, Molinaro RJ, Rivera- NEJMoa2020283
Marrero C, Xia B, Smith DF, Cummings RD 53. Zietz M, Tatonetti NP (2020) Testing the
(2010) Innate immune lectins kill bacteria association between blood type and COVID-
expressing blood group antigen. Nat Med 19 infection, intubation, and death. medR-
16(3):295–301. https://doi.org/10.1038/ xiv:2020.2004.2008.20058073. https://doi.
nm.2103 org/10.1101/2020.04.08.20058073
51. Kamili NA, Arthur CM, Gerner-Smidt C, 54. Zhao J, Yang Y, Huang H, Li D, Gu D, Lu X,
Tafesse E, Blenda A, Dias-Baruffi M, Stowell Zhang Z, Liu L, Liu T, Liu Y, He Y, Sun B,
SR (2016) Key regulators of galectin-glycan Wei M, Yang G, Wang X, Zhang L, Zhou X,
interactions. Proteomics 16(24):3111–3125. Xing M, Wang PG (2020) Relationship
https://doi.org/10.1002/pmic.201600116 between the ABO blood group and the
52. Ellinghaus D, Degenhardt F, Bujanda L, COVID-19 susceptibility. medR-
Buti M, Albillos A, Invernizzi P, Fernandez J, xiv:2020.2003.2011.20031096. https://doi.
Prati D, Baselli G, Asselta R, Grimsrud MM, org/10.1101/2020.03.11.20031096
Milani C, Aziz F, Kassens J, May S, 55. Wu SC, Arthur CM, Wang J, Verkerke H,
Wendorff M, Wienbrandt L, Uellendahl- Josephson CD, Kalman D, Roback JD, Cum-
Werth F, Zheng T, Yi X, de Pablo R, Chercoles mings RD, Stowell SR (2021) The SARS-CoV-
AG, Palom A, Garcia-Fernandez AE, 2 receptor-binding domain preferentially
Rodriguez-Frias F, Zanella A, Bandera A, recognizes blood group A. Blood Adv 5(5):
1305–1309. https://doi.org/10.1182/ 63. Robinson BS, Arthur CM, Kamili NA, Stowell
bloodadvances.2020003259 SR (2018) Galectin regulation of host micro-
56. Suthar MS, Zimmerman MG, Kauffman RC, bial interactions. Trends Glycosci Glycotechnol
Mantus G, Linderman SL, Hudson WH, 30(172):SE185–SE198
Vanderheiden A, Nyhoff L, Davis CW, 64. Arthur CM, Baruffi MD, Cummings RD,
Adekunle O, Affer M, Sherman M, Stowell SR (2015) Evolving mechanistic
Reynolds S, Verkerke HP, Alter DN, insights into galectin functions. Methods Mol
Guarner J, Bryksin J, Horwath MC, Arthur Biol 1207:1–35. https://doi.org/10.1007/
CM, Saakadze N, Smith GH, Edupuganti S, 978-1-4939-1396-1_1
Scherer EM, Hellmeister K, Cheng A, Morales 65. Arthur CM, Cummings RD, Stowell SR
JA, Neish AS, Stowell SR, Frank F, Ortlund E, (2014) Using glycan microarrays to under-
Anderson EJ, Menachery VD, Rouphael N, stand immunity. Curr Opin Chem Biol
Mehta AK, Stephens DS, Ahmed R, Roback 18C:55–61. https://doi.org/10.1016/j.cbpa.
JD, Wrammert J (2020) Rapid generation of 2013.12.017
neutralizing antibody responses in COVID-19 66. Stowell SR, Arthur CM, McBride R, Berger O,
patients. Cell Rep Med 1(3):100040. https:// Razi N, Heimburg-Molinaro J, Rodrigues JP,
doi.org/10.1016/j.xcrm.2020.100040 Noll AJ, von Gunten S, Smith DF, Knirel YA,
57. Verkerke H, Horwath M, Saeedi B, Boyer D, Paulson JC, Cummings RD (2014) Microbial
Allen JW, Owens J, Arthur CM, Nakahara H, glycan microarrays define key features of host-
Rha J, Patel K, Wu SC, Paul A, Yasin N, microbial interactions. Nat Chem Biol 10(6):
Wang J, Shin S, Brown D, Normile K, Cole L, 470–476
Meyers M, Lin H, Woods E, Isaac J, Broder K, 67. Stowell SR, Winkler AM, Maier CL, Arthur
Wade J, Kauffman RC, Patel R, Josephson CD, CM, Smith NH, Girard-Pierce KR, Cummings
Reynolds S, Sherman M, Wrammert J, Alter D, RD, Zimring JC, Hendrickson JE (2012) Ini-
Guarner J, Roback JD, Neish A, Stowell SR tiation and regulation of complement during
(2021) Comparison of antibody class specific hemolytic transfusion reactions. Clin Dev
SARS-CoV-2 serology for the diagnosis of Immunol 2012:307093. https://doi.org/10.
acute COVID-19. J Clin Microbiol 59(4): 1155/2012/307093
e02026-20. https://doi.org/10.1128/JCM.
02026-20 68. Zasloff M (2002) Antimicrobial peptides of
multicellular organisms. Nature 415(6870):
58. Verkerke H, Saeedi BJ, Boyer D, Allen JW, 389–395. https://doi.org/10.1038/415389a
Owens J, Shin S, Horwath M, Patel K,
Paul A, Wu S, Wang J, Ho A, Arthur CM, 69. Ganz T (2003) Defensins: antimicrobial pep-
Roback JD, Neish AS, Lough C, Josephson tides of innate immunity. Nat Rev Immunol
CD, Stowell SR (2021) Are we forgetting 3(9):710–720. https://doi.org/10.1038/
about IgA? A re-examination of COVID-19 nri1180
convalescent plasma. Transfusion 61(6): 70. Herigstad B, Hamilton M, Heersink J (2001)
1740–1748 How to optimize the drop plate method for
59. Kohatsu L, Hsu DK, Jegalian AG, Liu FT, enumerating bacteria. J Microbiol Methods
Baum LG (2006) Galectin-3 induces death of 44(2):121–129
Candida species expressing specific beta-1,2- 71. Stowell SR, Karmakar S, Stowell CJ, Dias-
linked mannans. J Immunol 177(7): Baruffi M, McEver RP, Cummings RD
4718–4726 (2007) Human galectin-1, -2, and -4 induce
60. Vasta GR (2009) Roles of galectins in infection. surface exposure of phosphatidylserine in acti-
Nat Rev Microbiol 7(6):424–438. https://doi. vated human neutrophils but not in activated T
org/10.1038/nrmicro2146 cells. Blood 109(1):219–227. https://doi.
org/10.1182/blood-2006-03-007153
61. Vasta GR (2012) Galectins as pattern recogni-
tion receptors: structure, function, and evolu- 72. Stowell SR, Qian Y, Karmakar S, Koyama NS,
tion. Adv Exp Med Biol 946:21–36. https:// Dias-Baruffi M, Leffler H, McEver RP, Cum-
doi.org/10.1007/978-1-4614-0106-3_2 mings RD (2008) Differential roles of galectin-
1 and galectin-3 in regulating leukocyte viabil-
62. Arthur CM, Patel SR, Mener A, Kamili NA, ity and cytokine secretion. J Immunol 180(5):
Fasano RM, Meyer E, Winkler AM, Sola- 3091–3102
Visner M, Josephson CD, Stowell SR (2015)
Innate immunity against molecular mimicry: 73. Stowell SR, Karmakar S, Arthur CM, Ju T,
examining galectin-mediated antimicrobial Rodrigues LC, Riul TB, Dias-Baruffi M,
activity. BioEssays 37(12):1327–1337. Miner J, McEver RP, Cummings RD (2009)
https://doi.org/10.1002/bies.201500055 Galectin-1 induces reversible
phosphatidylserine exposure at the plasma Detection of phosphatidylserine exposure on

membrane. Mol Biol Cell 20(5):1408–1418. leukocytes following treatment with human
https://doi.org/10.1091/mbc.E08-07-0786 galectins. Methods Mol Biol 1207:185–200.
74. Arthur CM, Rodrigues LC, Baruffi MD, Sulli- https://doi.org/10.1007/978-1-4939-1396-
van HC, Cummings RD, Stowell SR (2015) 1_12
Chapter 28
Detection of Phosphatidylserine Exposure on Leukocytes

Following Treatment with Human Galectins
Sean R. Stowell, Marcelo Dias-Baruffi, Richard D. Cummings,
and Connie M. Arthur
Abstract
Cellular turnover represents a fundamental aspect of immunological homeostasis. While many factors
appear to regulate leukocyte removal during inflammatory resolution, recent studies suggest that members
of the galectin family play a unique role in orchestrating this process. Unlike cellular removal through
apoptotic cell death, several members of the galectin family induce surface expression of phosphatidylserine
(PS), a phagocytic marker on cells undergoing apoptosis, in the absence of cell death. However, similar to
PS on cells undergoing apoptosis, galectin-induced PS exposure sensitizes cells to phagocytic removal. As
galectins appear to prepare cells for phagocytic removal without actually inducing apoptotic cell death, this
process has recently been coined preaparesis. Given the unique characteristics of galectin-induced PS
exposure in the context of preaparesis, we will examine unique considerations when evaluating the potential
impact of different galectin family members on PS exposure and cell viability.
Key words Galectin, Phosphatidylserine (PS), Inflammatory resolution, Leukocyte turnover,

Preaparesis
1 Introduction
Appropriate removal of leukocytes during inflammatory resolution

represents a fundamental process in immunological homeostasis
[1, 2]. Although many factors, including members of the TNF
family [3], appear to regulate this process, recent studies suggest
that several members of the galectin family possess the ability to
regulate a wide variety of cellular functions, from directly inducing
microbial death to regulating the turnover of leukocytes [4–
10]. While most immunoregulatory factors induce leukocyte
removal by initiating apoptotic cell death, galectins possess a
unique capacity to trigger phagocytic removal in the absence of
apoptosis [11–13]. Galectin engagement of neutrophil ligands
induces surface expression of phosphatidylserine (PS), similar to
cells undergoing apoptotic cell death [14]. However, unlike
533
apoptosis, galectin-induced PS exposure occurs in the conspicuous

absence of cell death [11–13, 15–18]. While galectins fail to induce
apoptotic cell death in neutrophils, galectin-induced PS exposure
maintains the ability to sensitize cells to phagocytic removal
[13]. Thus, galectins possess a unique ability to prepare cells for
phagocytic removal without inducing cell death; a process recently
termed preaparesis [12].
During cellular turnover, apoptotic cell death often provides a
unique mechanism of enabling cell removal without inducing sig-
nificant inflammation [1, 2]. This largely reflects the ability of cells
undergoing apoptosis to maintain membrane integrity prior to
successful phagocytic removal. However, once an apoptotic pro-
gram is engaged, cells only maintain membrane integrity for a
limited time before membrane integrity becomes compromised
and late apoptosis/necrosis occurs [1, 2]. In the setting of acute
inflammation, not only are inflammatory signals already present,
but the number of neutrophils often far exceed the number of
phagocytic cells needed for rapid removal during inflammatory
resolution [19]. As a result, induction of apoptosis in neutrophils
would be predicted to result in a significant number of late apopto-
tic events prior to efficient phagocytic removal, which would be
predicted to increase inflammatory stimuli [19, 20]. Furthermore,
given the inflammatory environment, the likelihood of maintaining
membrane integrity is likewise reduced [20, 21]. Consistent with
this, previous studies suggested that neutrophils evolved a unique
mechanism of turnover that occurs independent of apoptosis
[19, 22, 23]. The ability of galectins to induce PS exposure in the
absence of cell death, while retaining the ability to sensitize these
cells for phagocytic removal likely enables neutrophils to maintain
membrane integrity until successfully phagocytized [18].
As neutrophils do not typically utilize that same level of
directed antimicrobial activity as T cells and other cells involved in
adaptive immunity, the ability of galectins to induce preaparesis in
neutrophils, while typically inducing apoptosis in T cells [12, 24],
may in part reflect engagement of unique pathways not only impor-
tant in turnover but also important in the spatial and temporal
regulation of neutrophil function. Galectins possess a unique sensi-
tivity to oxidative or proteolytic inactivation once outside the cell
[13, 15, 25–27]. As a result, galectin released from damaged tissue
at the time of an initial inflammatory insult likely undergoes inacti-
vation prior to significant neutrophil accumulation. However, as
neutrophils begin to encroach on viable tissue following the
removal of necrotic debris, release of reduced, and therefore active,
galectin likely results in neutrophil engagement and induction of
preaparesis, thereby facilitating macrophage-mediated removal
(Fig. 1) [15, 28, 29]. In contrast, as T cells often do not accumulate
at the same rate or magnitude as neutrophils and also display a
targeted approach to their activity [30, 31], apoptotic cell death is
Detection of Phosphatidylserine Exposure on Leukocytes Following Treatment. . . 535
Fig. 1 Galectins may contribute to the spatial and temporal regulation of neutrophil turnover. Intracellular
galectin remains reduced and active. However, following cellular injury, intracellular galectin becomes
exposed to the extracellular oxidizing environment. Rapid infiltration of neutrophils following injury allows
for neutralization of potential pathogens and removal of necrotic tissue. During this inflammatory episode,
most of the galectin released during the primary injury likely becomes oxidized, preventing galectins from
inhibiting a productive inflammatory response. Following removal of necrotic tissue and removal of pathogens,
encroachment of neutrophils on surrounding viable tissue results in cellular damage and release of reduced
and therefore active galectin. Galectin engages neutrophil receptors and induces PS exposure without altering
cellular viability, a process termed preaparesis, which allows neutrophils to maintain membrane integrity until
successfully phagocytosed
likely sufficient to induce cellular removal without causing deleteri-

ous consequences in the setting of inflammation.
Given the unique sensitivity of galectins to oxidative inactiva-
tion coupled with their unprecedented ability to induce leukocyte
turnover independent of cell death [13, 15, 25], examination of the
potential impact of galectins on cellular viability requires careful
analysis of galectin activity and cellular function. In this chapter, we
will describe methods utilized to examine the potential impact of
galectins on cellular viability and PS exposure, including potential
pitfalls commonly associated with these approaches.
2 Materials
2.1 Neutrophil 1. Dextran (6% Dextran 70 in 0.9% Sodium Chloride injection

Isolation USP—Braun Medical Inc.).
2. 0.2% NaCl.
3. 1.6% NaCl.
4. Hanks Balanced Salts Solution (HBSS) (standard with Mg and

Ca) (Hyclone).
5. BSA (Fisher Scientific).
6. Ficoll (GE Healthcare).
7. Butterfly needle (BD).
8. 60-mL syringe (Cole-Parmer).
9. 15-mL Falcon tube (BD).
11. Wright-Geimsa stain (Sigma Aldrich).
12. fMLP (Sigma Aldrich).
2.2 Cell Culture 1. HL60 cells (ATCC CCL-240).

2. RPMI (Gibco).
3. FBS (Gibco, Life Technologies).
4. Glutamine (Life Technologies).
5. Penicillin/streptomycin (Sigma Aldrich).
6. Supplemented RPMI media: RPMI with 10% FBS, 1% gluta-
mine, 1% penicillin/streptomycin.
2.3 Galectin 1. Recombinant galectin.

Preparation and 2. PBS standard pH 7.4 (Hyclone).
Incubation with Cells
3. β-Mercaptoethanol (βME) (Fisher).
(GE Healthcare).
5. RPMI media (Gibco).
7. Thiodigalactoside (TDG) (Carbosynth).
8. 0.1 M Lactose in HBSS.
9. Sterile 48-well tissue culture plate.
2.4 Annexin V 1. HEPES/NaOH.

Staining of Galectin- 2. NaCl.
Treated Cells
3. CaCl2.
4. Hanks Balanced Salts Solution (HBSS) (standard with Mg and
Ca) (Hyclone).
5. Annexin V-Alexa Fluor 488 (Life Technologies).
6. Propidium Iodide (PI) (Life Technologies).
Prepared Buffers
7. Annexin V staining buffer: 10 mM HEPES/NaOH, 140 mM

NaCl, 5 mM CaCl2, pH 7.4.
8. Annexin V and propidium iodide (PI) staining solution: Add
2 μL of Annexin V-FITC conjugate and PI (final concentration
of 1 μg/mL) for every 100 μL of Annexin V staining buffer (see
Notes 1 and 2).
3 Methods
3.1 Isolation of 1. Using a 60-mL syringe, draw 300 μL of heparin (1000 U/mL)
Human Neutrophils (see Note 3).
from Whole Blood 2. Draw 30 mL of whole blood into a 60-mL syringe using a
butterfly needle by employing standard phlebotomy proce-
dures (see Note 4).
3. Carefully rotate capped syringe in order to adequately mix
heparin with blood to prevent clotting.
4. Draw 15 mL of dextran into the whole blood heparin mixture
(see Note 5).
5. Carefully mix the dextran solution in with the whole blood
solution.
6. Affix the syringe facing upright against a wall or other support
in a sterile hood. The dextran/whole blood solution will begin
to separate. The top layer, which appears to become devoid of
red blood cells (RBCs), is the layer that contains neutrophils.
7. At 30 min, 45 min, and 1 h after affixing the syringe in the
upright position, remove this top layer (see Note 6).
8. Once the top layer of solution is completely removed, place this
solution in a 50-mL sterile tube and a centrifuge for 7 min at
600 g at RT (see Note 7).
9. After centrifugation, discard the supernatant (see Note 8).
10. Resuspend the pellet in 25 mL of a 0.2% NaCl solution and
incubate at RT for 20 s (see Note 9).
11. Immediately add 25 mL 1.6% NaCl and mix well.
12. Place cells in a centrifuge for 7 min at 600 g at RT.
13. Discard supernatant and resuspend pellet in 50 mL of HBSS
with 0.5% human serum albumin.
14. Place cells in a centrifuge for 7 min at 600 g at RT.
15. Add 6 mL Ficoll to a sterile 15-mL Falcon tube.
16. Resuspend pellet isolated in step 14 in 6 mL of HBSS with
0.5% human serum albumin.
17. Place leukocyte solution on top of the Ficoll by tilting the tube
at 45 and carefully layering the leukocyte solution on top of
the Ficoll (see Note 10).
18. Centrifuge layered leukocyte solution for 30 min at 600 g at
RT (see Note 11).
19. Remove the middle band (this layer primarily consists of
lymphocytes).
20. Remove the pellet (primarily consists of neutrophils) and resus-
pend in 12 mL HBSS with 0.5% human serum albumin.
21. Centrifuge cells for 7 min at 600 g at RT.
22. Repeat steps 20 and 21 two more times to complete the
neutrophil wash.
23. Examine the cells for neutrophil content by staining with
Wright-Geimsa stain per the manufacturer’s protocol, followed
by morphological analysis of neutrophils (see Note 12).
24. Following the final wash, cells can be resuspended in
supplemented RPMI.
25. Count the number of cells per mL using a standard
hemocytometer.
26. For activation, resuspend the cells in RPMI at 1 million/mL
and add 1 μg of fMLP for every mL. Incubate at 37 C for
15 min. Following incubation, wash cells in supplemented
RPMI by centrifuging for 7 min at 600 g at RT twice.
3.2 Maintaining Cell 1. HL60 cells should be purchased from ATCC.

Culture of 2. Frozen aliquots of cells should be prepared and cultured
Nonadherent according the recommended ATCC protocol (see Note 13).
HL60 Cells
3. Culture cells in supplemented RPMI at 37 C in 5% CO2.
4. Typically cells are maintained at a concentration of
2 105 cells/mL (see Note 14).
3.3 Galectin 1. Thaw frozen vial of recombinant galectin by placing on ice

Preparation and (4 C) (see Note 15).
Incubation with Cells 2. Re-equilibrate a PD-10 column with 5 column volumes of
cold PBS.
3. To remove lactose and βME from galectin, add 1 mL of recom-
binant galectin solution to the re-equilibrated PD10 column
(see Note 16).
4. Collect 0.5 mL fractions from the PD10 column following the
addition of recombinant galectin.
5. Once the recombinant galectin solution has completely pene-
trated the column, add 2 mL of cold PBS.
6. Continue collecting 0.5 mL fractions from the PD10 column

following the addition of cold PBS, adding additional PBS as
needed to elute all galectin from the column and to prevent the
column from drying.
7. Examine protein content of each fraction by measuring the OD
280 nm. This is typically done by diluting each fraction 10-fold
(i.e., 10 μL of each fraction in 90 μL PBS) followed by exami-
nation of the OD 280 nm on a standard spectrophotometer (see
Note 17).
8. Place pooled recombinant galectin on ice until ready to use.
9. Prepare galectins in PBS at five times the desired final
concentration.
10. Pipet 200 μL of cells in RPMI at a concentration of
1 106 cells/mL into each well of a 48-well plate (see
Note 18).
11. Pre-warm galectin to 37 C and add 50 μL of recombinant
galectin solution or PBS control to each well (see Note 19).
12. Incubate at 37 C for the indicated time (see Note 20).
3.4 Staining Cells 1. At the end of the incubation time point, add 50 μL of 120 mM
with Annexin V lactose in RPMI to remove bound galectin and disengage cells
(see Note 21).
2. Incubate for 5 min at 37 C.
3. Examine cells for agglutination following incubation (see
Note 22).
4. Once cells appear to be sufficiently disengaged, wash cells three
times in HBSS by centrifuging cells at 600 g for 7 min.
5. Place cells on ice.
6. Resuspend cells in 100 μL Annexin V and PI staining solution
pre-cooled to 4 C.
7. Incubate at 4 C for 15 min in the dark.
8. Add 400 μL of Annexin V staining buffer pre-cooled at 4 C to
each sample.
9. Analyze cells by flow cytometry (see Notes 23 and 24).
4 Notes
1. Following buffer preparation, cool buffers to 4 C prior to

staining the cells.
2. The amount of Annexin V used may vary based on the manu-
facturer of the product and should be determined empirically.
3. Typically 100 μL of heparin stock solution (1000 U/mL) is

used for every 10 mL of whole blood drawn.
4. Institutional review board (IRB) or equivalent approval must
be obtained prior to drawing blood from healthy human
volunteers.
5. The total amount of dextran added typically equals half the
total starting volume of whole blood. This is typically done by
drawing the dextran solution into a separate container followed
by injecting the dextran solution into the whole blood–heparin
mixture to prevent contamination of the Dextran stock.
6. The 30- and 45-min interval removal of the top layer appears to
facilitate additional separation. However, these steps are not
absolutely necessary.
7. Although protocols can differ, we found that placing neutro-
phils on ice, followed by rewarming can result in significant
neutrophil activation. As a result, a concerted effort is made to
avoid cold temperatures during neutrophil isolation. All pro-
cedures should be done under a sterile laminar flow hood.
8. The pellet will look red due to significant red blood cell
contamination.
9. This should lyse the red blood cells. As this is a hypotonic RBC
lysis method, care should be taken to not over incubate the
cells.
10. Care should be taken to not allow the leukocyte HBSS solution
to penetrate the Ficoll solution.
11. Do not use the brakes on the centrifuge as this can disrupt the
desired interfaces generated during centrifugation.
12. Using this protocol typically >90% of the isolated cells are
neutrophils.
13. As HL60 cells will differentiate in the presence of dimethylsulf-
oxide (DMSO) and DMSO is a commonly employed cryopre-
servant, care should be taken to remove DMSO as soon as
possible following initiation of HL60 culture.
14. Care should be taken to not allow cells to exceed 1 106 cells/
mL. Cells can be counted using a standard hemocytometer.
Equally important, the media should be regularly changed to
allow cells to remain in a physiologic pH. Using this approach,
typically >90% of the cells remain viable at baseline. When cells
are allowed to become overcrowded and/or the media is
changed infrequently, cellular viability suffers and the overall
outcome of galectin-HL60 cell interactions can be altered.
15. All galectin preparation should be done under a sterile laminar
flow hood. Potential lipopolysaccharide (LPS) contamination
should be assessed prior to preparing galectin for incubation
with cells. This can be done using the commercially available

limulus amebocyte lysate (LAL) assay (Pierce). If significant
LPS contamination is noted, then LPS removal can be achieved
by passing the endotoxin contaminated recombinant galectin
sample over a polymyxin B column (Sigma Aldrich) according
to the manufacturers protocol. Repeat removal should be
employed until LPS is undetectable.
16. Several galectins display unique sensitivity to oxidative inacti-
vation [11, 15]. This should be considered when assessing the
potential ability of an individual galectin to induce PS exposure
when incubated with a given cell. Methods to alkylate
galectin-1 and protect it from oxidative inactivation have
been developed [15]. However, it should be noted that for
most galectins, oxidative inactivation is a gradual process and
that biological activity of these proteins can be assessed in the
absence of reducing conditions or other chemical modifica-
tions if care is taken to use them immediately following removal
of βME (or other reducing agent) and ligand (see Fig. 2 for
impact of DTT on cellular viability and sensitivity to Gal-1-
induced PS exposure).
values, we use the extinction coefficient for the particular
galectin being examined to calculate actual concentration in
mg/mL. The following websites http://www.basic.northwest
ern.edu/biotools/proteincalc.html and http://web.expasy.
org/protparam/protparam-doc.html offer explanation and
assistance in calculating extinction coefficient and using this
calculation to determine the actual concentration of a given
protein in mg/mL, including caveats as to how these numbers
may differ from the actual number. As methods of calculating
the extinction coefficient only provide estimates, alternative
approaches can be used to empirically determine these values.
These include using a Bradford assay to calculate protein con-
centration or simply re-equilibrating the recombinant protein
directly into water, lyophilizing, weighing, and then resus-
pending in the buffer of choice followed by empirically deter-
mining the extinction coefficients for a particular galectin
family member.
18. When counting cells, use trypan blue (Sigma Aldrich) to deter-
mine the percent of viable and non-viable cells. If the nonviable
cell count is higher than 10%, these cells are not sufficiently
healthy to use in this assay.
19. Control wells containing galectin along with a final concentra-
tion of 20 mM Lactose or 20 mM TDG (both hapten inhibi-
tors of galectin–carbohydrate binding) should be included as
negative controls of galectin signaling. Additionally, it is critical
Fig. 2 DTT sensitizes cells to Gal-1-induced apoptosis. (a) Promyelocytic HL60 cells (cells) were incubated
with PBS, 10 μM Gal-1, or the indicated concentration of DTT for 9 h followed by detection for cellular
fragmentation as indicated by changes in forward (FSC) and side scatter (SSC) profiles of cells. (b) Cells were
incubated with PBS, 10 μM Gal-1, or the indicated concentration of DTT for 9 h followed by detection for cell
death by PS exposure and membrane integrity loss by An-V-FITC and PI staining. (c) Cells were incubated with
PBS, 10 μM iodoacetamide-treated Gal-1 (iGal-1), or 10 μM iGal-1 with 20 mM lactose followed by detection
for PS exposure by An-V-FITC staining and PI exclusion. (d) Cells were incubated with PBS or 10 μM iGal-1 for
the indicated time followed by detection for PS exposure by An-V-FITC staining and PI exclusion. (e) Cells were
incubated with PBS or the indicated concentration of iGal-1 for 8 h followed by detection for PS exposure by
An-V-FITC staining and PI exclusion. (f) Cells were incubated with PBS, 10 μM iGal-1, or 10 μM Camp for 8 h
followed by incubation of peritoneal macrophages for 1 h and microscopic examination of phagocytosis. (This
research was originally published in Molecular Biology of the Cell. Stowell SR, Karmakar S, Arthur CM, Ju T,
Rodrigues LC, Riul TB, Dias-Baruffi M, Miner J, McEver RP, Cummings RD. Galectin-1 induces reversible
phosphatidylserine exposure at the plasma membrane.2009 Mar;20(5):1408–18)
to incorporate an apoptosis positive control. Typically, Fas for

neutrophils and camptothecin for HL60 cells can be used as an
apoptosis positive control, although a variety of pro-apoptotic
agents can be employed.
20. PS exposure typically peaks between 2 and 4 h. A range of
galectin concentrations and times should be employed to
determine the optimal time for PS exposure to be realized.
To determine whether PS exposure occurs in the absence of
cell death, later time points, including 8, 12, and even 24 h are
recommended to determine whether PS exposure continues to
occur in the absence of apoptosis using additional methods in
parallel that are capable of detecting markers of apoptotic cell

death (examination of DNA fragmentation, etc.).
21. Unlike use of antibodies for staining cell surface CD markers,
galectins typically agglutinate cells in suspension. Agglutinated
cells not only possess the ability to occlude the appropriate
fluid flow during flow cytometric analysis, but cellular fragmen-
tation of agglutinated cells during flow cytometric analysis can
result in false-positive results (see Figs. 3, 4, 5). Although most
antibodies used for flow cytometric examination of antigen
reactivity at saturating concentrations do not induce aggluti-
nation [32–36], antibodies capable of inducing agglutination
can induce similar fragmentation and PS positivity [37].
22. Each galectin family member has different affinity for lactose
and leukocyte ligands. As a result, the concentration of lactose
and the duration of incubation. Thiodigalactose (TDG) can be
a more efficient inhibitor of galectins and therefore can occa-
sionally be used when lactose appears to be insufficient to
induce galectin release and disengagement of galectin-induced
agglutinated aggregates.
23. When analyzing cells undergoing apoptosis, it is important to
consider potential changes in the forward and side scatter
profiles that may occur. The changes can be diagnostic of
cells undergoing apoptosis, but may be inadvertently gated
out during data acquisition and analysis. As a result, care
should be taken when analyzing cells by flow cytometry to
insure that apoptosis/late apoptosis positive events are not
gated out during data acquisition or analysis (Fig. 6).
24. Keep cells on ice and analyze as soon as possible (typically <1 h
following completion of staining to avoid potential changes in
viability associated simply with prolonged incubation in
Annexin V staining buffer). Fixation of cells can result in
artificial exposure of PS and should be avoided.
Acknowledgments

Grant R01 HL154034 to CMA.
References
1. Strasser A, O’Connor L, Dixit VM (2000) 3. Iwai K, Miyawaki T, Takizawa T, Konno A,

Apoptosis signaling. Annu Rev Biochem 69: Ohta K, Yachie A, Seki H, Taniguchi N
217–245. https://doi.org/10.1146/annurev. (1994) Differential expression of bcl-2 and sus-
biochem.69.1.217 ceptibility to anti-Fas-mediated cell death in
2. Danial NN, Korsmeyer SJ (2004) Cell death: peripheral blood lymphocytes, monocytes,
critical control points. Cell 116(2):205–219 and neutrophils. Blood 84(4):1201–1208
Fig. 3 Incubation of leukocytes with galectins induces agglutination. Promyelocytic HL60 cells were incubated
with PBS or 10 μM galectin-1 (Gal-1) for 4 h followed by microscopic evaluation for agglutination. Following
incubation with PBS or Gal-1, lactose was added at a final concentration of 20 mM lactose for 15 min at 37 C
followed by microscopic evaluation as indicated
4. Cerliani JP, Stowell SR, Mascanfroni ID, 16(3):295–301. https://doi.org/10.1038/

Arthur CM, Cummings RD, Rabinovich GA nm.2103
(2011) Expanding the universe of cytokines 7. Stowell SR, Arthur CM, McBride R, Berger O,
and pattern recognition receptors: galectins Razi N, Heimburg-Molinaro J, Rodrigues LC,
and glycans in innate immunity. J Clin Immu- Gourdine JP, Noll AJ, von Gunten S, Smith
nol 31(1):10–21. https://doi.org/10.1007/ DF, Knirel YA, Paulson JC, Cummings RD
s10875-010-9494-2 (2014) Microbial glycan microarrays define
5. Robinson BS, Saeedi B, Arthur CM, Owens J, key features of host-microbial interactions.
Naudin C, Ahmed N, Luo L, Jones R, Neish A, Nat Chem Biol 10(6):470–476. https://doi.
Stowell SR (2020) Galectin-9 is a novel regula- org/10.1038/nchembio.1525
tor of epithelial restitution. Am J Pathol 8. Arthur CM, Patel SR, Mener A, Kamili NA,
190(8):1657–1666. https://doi.org/10. Fasano RM, Meyer E, Winkler AM, Sola-
1016/j.ajpath.2020.04.010 Visner M, Josephson CD, Stowell SR (2015)
6. Stowell SR, Arthur CM, Dias-Baruffi M, Innate immunity against molecular mimicry:
Rodrigues LC, Gourdine JP, Heimburg- examining galectin-mediated antimicrobial
Molinaro J, Ju T, Molinaro RJ, Rivera- activity. BioEssays 37(12):1327–1337.
Marrero C, Xia B, Smith DF, Cummings RD https://doi.org/10.1002/bies.201500055
(2010) Innate immune lectins kill bacteria 9. Robinson BS, Arthur CM, Kamili NA, Stowell
expressing blood group antigen. Nat Med SR (2018) Galectin regulation of host
Fig. 4 Failure to disengage cells can result in significant changes in the forward and side scatter profiles
following flow cytometric analysis. Promyelocytic HL60 cells were incubated at the indicated concentrations
for 4 h followed by addition of PBS (no lactose) or lactose at a final concentration of 20 mM (lactose) and
incubated for an additional 15 min. Following incubation, cells were washed and analyzed for potential
alterations in the forward and side scatter profiles as indicated. Gate ¼ % of cells experiencing no
fragmentation The percent of cells in the gate for each condition is shown
microbial interactions. Trends Glycosci Glyco- 12. Stowell SR, Qian Y, Karmakar S, Koyama NS,
technol 30:172 Dias-Baruffi M, Leffler H, McEver RP, Cum-
10. Kohatsu L, Hsu DK, Jegalian AG, Liu FT, mings RD (2008) Differential roles of galectin-
Baum LG (2006) Galectin-3 induces death of 1 and galectin-3 in regulating leukocyte viabil-
Candida species expressing specific beta-1,2- ity and cytokine secretion. J Immunol
linked mannans. J Immunol 180(5):3091–3102
177(7):4718–4726 13. Stowell SR, Karmakar S, Arthur CM, Ju T,
11. Stowell SR, Karmakar S, Stowell CJ, Dias- Rodrigues LC, Riul TB, Dias-Baruffi M,
Baruffi M, McEver RP, Cummings RD Miner J, McEver RP, Cummings RD (2009)
(2007) Human galectin-1, -2, and -4 induce Galectin-1 induces reversible phosphatidylser-
surface exposure of phosphatidylserine in acti- ine exposure at the plasma membrane. Mol
vated human neutrophils but not in activated T Biol Cell 20(5):1408–1418. https://doi.org/
cells. Blood 109(1):219–227. https://doi. 10.1091/mbc.E08-07-0786
org/10.1182/blood-2006-03-007153
Fig. 5 Failure to disengage cells can result in significant Annexin V and Propidium Iodide double-positive
staining. Promyelocytic HL60 cells were incubated at the indicated concentrations for 4 h followed by addition
of PBS (no lactose) or lactose at a final concentration of 20 mM (lactose) and incubated for an additional
15 min. Following incubation, cells were washed and stained with Annexin V (An-V) and propidium iodide
(PI) as outlined. The percent of cells in each quadrant is shown
14. Fadok VA, Bratton DL, Rose DM, Pearson A, 17. Arthur CM, Baruffi MD, Cummings RD,
Ezekewitz RA, Henson PM (2000) A receptor Stowell SR (2015) Evolving mechanistic
for phosphatidylserine-specific clearance of insights into galectin functions. Methods Mol
apoptotic cells. Nature 405(6782):85–90. Biol 1207:1–35. https://doi.org/10.1007/
https://doi.org/10.1038/35011084 978-1-4939-1396-1_1
15. Stowell SR, Cho M, Feasley CL, Arthur CM, 18. Robinson BS, Arthur CM, Evavold B,
Song X, Colucci JK, Karmakar S, Mehta P, Roback E, Kamili NA, Stowell CS, Vallecillo-
Dias-Baruffi M, McEver RP, Cummings RD Zuniga ML, Van Ry PM, Dias-Baruffi M,
(2009) Ligand reduces galectin-1 sensitivity Cummings RD, Stowell SR (2019) The
to oxidative inactivation by enhancing dimer sweet-side of leukocytes: galectins as master
formation. J Biol Chem 284(8):4989–4999. regulators of neutrophil function. Front
https://doi.org/10.1074/jbc.M808925200 Immunol 10:1762. https://doi.org/10.
16. Karmakar S, Stowell SR, Cummings RD, 3389/fimmu.2019.01762
McEver RP (2008) Galectin-1 signaling in leu- 19. Nathan C (2006) Neutrophils and immunity:
kocytes requires expression of complex-type challenges and opportunities. Nat Rev
N-glycans. Glycobiology 18(10):770–778
Fig. 6 Gal-1 induces continuous PS exposure in the absence of cellular fragmentation. (a) Cells were
incubated with PBS, 10 μM iodoacetamide alkylated Gal-1 (iGal-1), or 10 μM Camp for 1 or 2 days as
indicated, followed by detection for PS exposure by An-V-FITC staining and PI exclusion. (b) Cells were
incubated with PBS, 10 μM iGal-1, or 10 μM Camp for 1 or 2 days as indicated, followed by examination for
cellular fragmentation as indicated by changes in forward (FSC) and side scatter (SSC) profiles of cells.
Gate ¼ % of cells experiencing no fragmentation. (c) Quantification of cells treated in A. White bars ¼ % An-V
+, PI-; black bars ¼ % An-V+, PI+. (d) Quantification of cells treated in b. (This research was originally
published in Molecular Biology of the Cell. Stowell SR, Karmakar S, Arthur CM, Ju T, Rodrigues LC, Riul TB,
Dias-Baruffi M, Miner J, McEver RP, Cummings RD. Galectin-1 induces reversible phosphatidylserine
exposure at the plasma membrane. 2009 Mar;20(5):1408–18)
Immunol 6(3):173–182. https://doi.org/10. 20. Mayadas TN, Cullere X, Lowell CA (2013)

1038/nri1785 The multifaceted functions of neutrophils.
Annu Rev Pathol 9:181–218. https://doi.org/ syngeneic or semiallogeneic system. Nature

10.1146/annurev-pathol-020712-164023 248(450):701–702
21. Mittal M, Siddiqui MR, Tran K, Reddy SP, 31. Zinkernagel RM, Doherty PC (1997) The dis-
Malik AB (2013) Reactive oxygen species in covery of MHC restriction. Immunol Today
inflammation and tissue injury. Antioxid 18(1):14–17
Redox Signal 20(7):1126–1167. https://doi. 32. Stowell SR, Liepkalns JS, Hendrickson JE,
org/10.1089/ars.2012.5149 Girard-Pierce KR, Smith NH, Arthur CM,
22. Lagasse E, Weissman IL (1994) Bcl-2 inhibits Zimring JC (2013) Antigen modulation con-
apoptosis of neutrophils but not their engulf- fers protection to red blood cells from antibody
ment by macrophages. J Exp Med through Fcgamma receptor ligation. J Immu-
179(3):1047–1052 nol 191(10):5013–5025. https://doi.org/10.
23. Shi J, Gilbert GE, Kokubo Y, Ohashi T (2001) 4049/jimmunol.1300885
Role of the liver in regulating numbers of cir- 33. Girard-Pierce KR, Stowell SR, Smith NH,
culating neutrophils. Blood 98(4):1226–1230 Arthur CM, Sullivan HC, Hendrickson JE,
24. Perillo NL, Pace KE, Seilhamer JJ, Baum LG Zimring JC (2013) A novel role for C3 in
(1995) Apoptosis of T cells mediated by antibody-induced red blood cell clearance and
galectin-1. Nature 378(6558):736–739. antigen modulation. Blood
https://doi.org/10.1038/378736a0 122(10):1793–1801. https://doi.org/10.
25. Levi G, Teichberg VI (1981) Isolation and 1182/blood-2013-06-508952
physicochemical characterization of electrolec- 34. Stowell SR, Henry KL, Smith NH, Hudson
tin, a beta-D-galactoside binding lectin from KE, Halverson GR, Park JC, Bennett AM,
the electric organ of Electrophorus electricus. Girard-Pierce KR, Arthur CM, Bunting ST,
J Biol Chem 256(11):5735–5740 Zimring JC, Hendrickson JE (2013) Alloanti-
26. Cho M, Cummings RD (1995) Galectin-1, a bodies to a paternally derived RBC KEL anti-
beta-galactoside-binding lectin in Chinese gen lead to hemolytic disease of the fetus/
hamster ovary cells. I. Physical and chemical newborn in a murine model. Blood
characterization. J Biol Chem 122(8):1494–1504. https://doi.org/10.
270(10):5198–5206 1182/blood-2013-03-488874
27. Cho M, Cummings RD (1995) Galectin-1, a 35. Mener A, Arthur CM, Patel SR, Liu J, Hen-
beta-galactoside-binding lectin in Chinese drickson JE, Stowell SR (2018) Complement
hamster ovary cells. II. Localization and bio- component 3 negatively regulates antibody
synthesis. J Biol Chem 270(10):5207–5212 response by modulation of red blood cell anti-
gen. Front Immunol 9:676. https://doi.org/
28. Dias-Baruffi M, Stowell SR, Song SC, Arthur 10.3389/fimmu.2018.00676
CM, Cho M, Rodrigues LC, Montes MA,
Rossi MA, James JA, McEver RP, Cummings 36. Mener A, Patel SR, Arthur CM, Chonat S,
RD (2010) Differential expression of immuno- Wieland A, Santhanakrishnan M, Liu J, Maier
modulatory galectin-1 in peripheral leukocytes CL, Jajosky RP, Girard-Pierce K, Bennett A,
and adult tissues and its cytosolic organization Zerra PE, Smith NH, Hendrickson JE, Stowell
in striated muscle. Glycobiology SR (2018) Complement serves as a switch
20(5):507–520. https://doi.org/10.1093/ between CD4+ T cell-independent and
glycob/cwp203 -dependent RBC antibody responses. JCI
Insight 3(22):e121631. https://doi.org/10.
29. Cerri DG, Rodrigues LC, Stowell SR, Araujo 1172/jci.insight.121631
DD, Coelho MC, Oliveira SR, Bizario JC,
Cummings RD, Dias-Baruffi M, Costa MC 37. Liepkalns JS, Hod EA, Stowell SR, Cadwell
(2008) Degeneration of dystrophic or injured CM, Spitalnik SL, Zimring JC (2012) Biphasic
skeletal muscles induces high expression of clearance of incompatible red blood cells
Galectin-1. Glycobiology 18(11):842–850. through a novel mechanism requiring neither
https://doi.org/10.1093/glycob/cwn079 complement nor Fcgamma receptors in a
murine model. Transfusion
30. Zinkernagel RM, Doherty PC (1974) Restric- 52(12):2631–2645. https://doi.org/10.
tion of in vitro T cell-mediated cytotoxicity in 1111/j.1537-2995.2012.03647.x
lymphocytic choriomeningitis within a
Chapter 29
Detection of Reactive Oxygen Species in Human Neutrophils

Under Various Conditions of Exposure to Galectin
Lilian Cataldi Rodrigues, Daniel Giuliano Cerri,
Cleni M. Marzocchi-Machado, Richard D. Cummings, Sean R. Stowell,
Abstract
Reactive oxygen species (ROS) have been extensively studied in biology in the past years. This class of
molecules can be derived from endogenous sources (e.g., phagocytic cells as neutrophils, eosinophils,
monocytes, macrophages, and organelles as mitochondria and peroxisomes) and participate in physiological
and pathological conditions. The beneficial and harmful effects of ROS depend on redox regulation, which
establishes the balance between their production and the activity of antioxidant systems to prevent oxidative
stress in vivo. Neutrophils are the immune effectors most well depicted with an intense oxidative burst in
response to tissue inflammation. Several proteins and members of the galectin family are involved in this fine
modulation of ROS production by neutrophils. Interestingly, studies have indicated that Galectin-1 (Gal-1)
can up- or downregulate ROS production by neutrophils even when exposed to N-formyl-Met-Leu-Phe
(fMLP) or Phorbol Myristate Acetate (PMA), both of which are potent neutrophil stimulants that trigger
high levels of ROS production. Similarly, Galectin-3 (Gal-3) induces ROS in neutrophils from a sterile or
nonsterile inflammatory environment, possibly creating a negative loop that could control ROS produc-
tion. Besides, superoxide production is also induced by Galectin-8 (Gal-8) and Galectin-9 (Gal-9) in
neutrophils but in a different manner. We describe herein the luminol and lucigenin-dependent chemilu-
minescence technique by using a luminometer as a method of assessment to measure ROS production by
human neutrophils isolated and exposed to purified human recombinant Gal-1. The protocol described
herein could be applied for the investigation of the role of other galectins in the modulation of ROS
production by neutrophils.
Key words Galectin, Neutrophil, ROS, Luminol, Lucigenin
1 Introduction
Most molecules have pairs of electrons, with opposite spins, in their

outer shells that stabilize them. Differently, some chemical species
present one or more unpaired electrons named free radicals. This
feature makes this special class of chemicals highly reactive due to
the tendency to interact with neighboring molecules to pair that
549
550 Lilian Cataldi Rodrigues et al.
single electron, generating more stable species [1]. However, it is

worth noting that the nomenclature “free radicals” is expanded in
biological systems. Other molecules with strong oxidative poten-
tial, but without unpaired electrons, are grouped with free radicals
into two classes of molecules, reactive oxygen species (ROS) and
reactive nitrogen species (RNS), both with radical and non-radical
species [2]. Herein, we will focus our attention on the crucial
biological class of those oxidant molecules: ROS. In phagocytic
cells, molecular oxygen (O2) is the primary source of these oxidant
molecules and their main byproducts are: (1) free radicals: superox-
ide anion (O2); hydroxyl, alkoxyl and peroxyl radicals (OH·, RO·,
and ROO·, respectively); and (2) non-radicals: hydrogen peroxide
(H2O2), ozone (O3), singlet oxygen (1O2), organic peroxide
(ROOH), hypochlorous and hypobromous acids (HOCl and
HOBr, respectively) [3]. Of all the aforementioned molecules, the
most important ones regarding tissue injury are the ROS: O2·,
OH·, and H2O2 [4]. The production of ROS/RNS occurs at low
levels in living systems with a beneficial effect in the immune
function, signaling critical pathways as NF-κB, FOXO3, and
Keap1 [5], in mitogenic events, and in its own redox regulation
[6, 7]. Living organisms present a strict form to neutralize/inhibit
ROS, such as through anti-oxidants of enzymatic (glutathione
peroxidase, catalase, and superoxide dismutase) and nonenzymatic
(vitamins C and E, melatonin, glutathione, flavonoids, carotenoids,
etc.) sources [5].
On the other hand, when ROS accumulates at high levels in any
organism, hazardous reactions will occur due to their wide range of
molecular targets such as DNA [8], protein [9], and lipids
[10]. The reaction of ROS with biological compounds may lead
to a series of pathological conditions such as diabetes mellitus,
rheumatoid arthritis, cataract, cardiovascular and neurodegenera-
tive diseases, and aging [2].
The production of ROS by phagocytes is initiated by the acti-
vation of the enzyme complex nicotinamide adenine dinucleotide
phosphate-reduced form (NADPH) oxidase, which consists of
membrane and cytosolic components. NADPH oxidase donates
an electron to molecular oxygen to form the superoxide anion,
the first ROS that initiates an increase in the oxidative metabolism
of phagocytes, known as oxidative burst [11].
Superoxide anion is produced and rapidly converted to H2O2
by superoxide dismutase, and it will be metabolized by Fenton
reaction to OH.. The oxygen-derived radicals generation by neu-
trophils recruited to an infected site could be essential to kill
bacteria. Although much is known on neutrophils and their ROS
production, for some studies, the very short life of these free
radicals can be a problem [1, 2]. Indirect detection of ROS has
been done by analyzing their scavengers such as superoxide dismu-
tase, catalase, glutathione peroxidase, or vitamin E [12, 13]. Due to
Detection of Reactive Oxygen Species in Human Neutrophils Under Various. . . 551
the crucial role of ROS production by neutrophils for the immune

response against pathogens as well as their implication in the path-
ogenesis of various diseases [14], it is of particular interest to
evaluate the production of ROS by these cells in response to distinct
external agents.
In the literature, it has been reported that members of the
galectin family play an essential role in the regulation of ROS
production in neutrophils at different stages of activation
[15]. Galectins are mammalian β-galactoside-binding proteins
expressed in various tissues and immune cell types such as macro-
phages, neutrophils, and dendritic cells. Most of their functions are
strictly related with their carbohydrate recognition domain (CRD)
composed by a conserved amino acid sequence [16–18]. The
carbohydrate-binding property of Gal-1 significantly contributes
to up- or downregulation of ROS production by neutrophils,
being the prevalence of positive or negative effects depends on
the cell activation stage and the microenvironment (Fig. 1)
[19]. Other galectins, such as Gal-3, present a similar impact of
Gal-1, requiring neutrophil priming to be effective in ROS produc-
tion [20, 21].
In this context, fMLP and PMA are known as neutrophil
priming and/or activators. fMLP is a pathogen-derived formyl
peptide that binds to specific receptors on neutrophil surface,
while PMA is a phorbol ester analog of diacylglycerol. Both are
kinase protein activators that regulate NADPH oxidase complex
resulting in ROS production by neutrophils [22].
The Gal-1 with the ability to induce ROS production in neu-
trophils that have migrated to the inflammation site is associated
with the fMLP mobilization of a Gal-1 ligands pattern from gela-
tinase granules secretory vesicles [23, 24]. In Gal-3-induced con-
text, ROS production from activated neutrophils drives fMLP to
degradation and cell desensitization, enabling further exposure to
increased fMLP concentrations. This effect in the presence of Gal-3
provides a positive loop that contributes to the killing of the bacte-
rial and microbial agents at the site of infection [25].
To balance Gal-3 activation, neutrophils are able to cleave
Gal-3, decreasing its ROS induction capacity. Also, this negative
effect is essential to control high levels of neutrophil activation
which leads to tissue damage [26]. The downregulation of ROS
can also be seen in nonactivated neutrophils exposed to Gal-1 prior
to neutrophil stimulation with fMLP or PMA, in vitro. One of the
possible mechanisms for Gal-1 to down modulate fMLP-
dependent ROS production is that this lectin decreases formyl
peptide receptor 1 (FPR-1) expression in naı̈ve human neutrophils.
Indeed, this effect is lectin-activity dependent [19].
Moreover, authors have described that Gal-8 induces both
intracellular and extracellular superoxide production in primed
human neutrophils due to its CRD-domain, and the levels of
Fig. 1 Galectin-1 plays a role in the dynamics of ROS production in nonactivated and activated neutrophils.
Members of the galectin family are involved in regulating ROS production in neutrophils at different stages of
activation. Gal-1 decreases ROS production in nonactivated (naı̈ve and primed) neutrophils when it is exposed
prior to neutrophil activation. On the other hand, at the inflammation site, Gal-1 induces ROS production in
activated neutrophils. Priming and activating neutrophils were performed using different fMLP concentrations:
109 M and 106 to 108 M, respectively [19]. Therefore, for this galectin, the inflammation and/or infection
microenvironment is determinant for activating ROS production
superoxide are higher than those produced by fMLP in the absence

of this lectin [27, 28]. Additionally, Gal-9 alone is able to induce
superoxide production via NADPH oxidase, and in primed neu-
trophils, its effect is much more evident [29]. Mechanistically,
Gal-8 and Gal-9 stimulate ROS production by different signaling,
through C-terminal domain and a Tim3-dependent pathway,
respectively [28, 29].
Thus, these galectins may be a significant part of the mechan-
isms underlying ROS modulation under pathological processes,
decreasing the potential risk of free radicals causing tissue damage
in the absence of infection, as well as increasing ROS production
that favors the killing of pathogens in a context of infection or
inflammation. Although the same regulatory effect of ROS produc-
tion by neutrophils during exposure to galectin-1 may be present in
other galectins, all these outcomes need to be further investigated.
In this chapter, we will highlight protocols for the measure-
ment of ROS production from human neutrophils under pre- or
post-exposure to recombinant human Gal-1, using the luminol and
lucigenin-dependent chemiluminescence method in a lumin-

ometer. The protocol here presented could be used in studies
involving other galectins.
Lucigenin and luminol are chemical species used as luminescent
probes to amplify the detection of ROS. The ROS produced by
neutrophils induce a state of electronic excitation in these probes,
which then emit light (chemiluminescence) in the form of photons
that are recorded in a luminometer. The cell membrane is imper-
meable to the passage of lucigenin, and therefore, this molecule is
used as a marker of the production of the superoxide anion released
in the extracellular medium. The superoxide anion is the first ROS
produced by the activation of the phagocyte NADPH oxidase
complex, and therefore, lucigenin is used in assays that aim to
evaluate the activity of this enzyme complex. As for luminol, its
chemical structure allows it to cross/permeate the cell membrane,
and thus, this probe is able to detect the total production of ROS,
intracellular and extracellular, including the generation of ROS via
NADPH oxidase as well as that dependent on myeloperoxidase, the
enzyme released from neutrophil granules [30, 31].
2 Materials
2.1 Neutrophil 1. Hanks’ Balanced Salts Solution (HBSS) (standard with Mg2+
Isolation and Ca2+) (HyClonc).
2. 0.15 M NaCl.
3. 0.83% NH4Cl pH 7.4.
4. BD Vacutainer™ EDTA Plasma Tube (ethylenediamine
tetraacetic acid).
5. Histopaque®-1077, Histopaque®-1119.
6. Butterfly needle (BD).
7. 20-mL syringe (BD).
10. 0.4% Trypan Blue Solution (Thermo Fisher Scientific).
2.2 Galectin 1. Recombinant purified human galectin-1 (see Note 10).

Preparation 2. PBS standard pH 7.4 (HyClone).
3. β-Mercaptoethanol (β-ME) (Fisher).
(GE Healthcare).
5. RPMI media (Gibco).
6. 0.1 M lactose in HBSS.
2.3 Additional 1. 102 M Luminol (3-aminophthalhydrazide, 5-amino-2,3-

Reagents dihydro-1,4-phthalazinedione; Sigma-Aldrich A8511).
2. 102 M Lucigenin (N,N0 -dimethyl-9,90 -biacridinium dini-
trate; Sigma-Aldrich M8010).
3. 1 mM DPI (diphenyleneiodonium chloride; Sigma-Aldrich).
4. 106 M to 109 M PMA (phorbol-12-myristate-13-acetate;
Sigma-Aldrich).
5. 106 M to 109 M fMLP (N-formyl-L-methionyl-L-leucyl-L-
phenylalanine; Sigma-Aldrich F3506).
6. 20 mM Lactose (Fisher).
7. 20 mM Sucrose (Fisher).
8. 5 mM Thiodigalactoside (TDG) (Carbosynth).
3 Methods
3.1 Isolation of 1. Collect 6 mL of the whole blood in EDTA (Vacutainer BD®)

Human Neutrophils tube (see Note 1).
from Whole Blood 2. At room temperature, layer 6 mL of whole blood in a 15-mL
conical centrifuge tube containing: 3 mL of Histopaque-1119
plus 3 mL of Histopaque-1077 (see Note 2).
3. Set the brake and acceleration of the centrifuge to the lowest
possible settings (see Note 3).
4. Set the centrifuge temperature to room temperature (see Note
4).
5. Promptly centrifuge the tubes at 700 g for 30 min at room
temperature (see Note 5).
6. After the centrifuge, five layers can be seen: (1) plasma with
platelets on the top; (2) under the plasma one layer with
monocytes; (3) target layer with neutrophils; (4) the gradient
Histopaque-1077/Histopaque-1119; and (5) the bottom
containing red blood cells (RBCs) (see Note 6).
7. Carefully remove the two top layers using a 3-mL pipette
Pasteur.
8. Draw the neutrophil ring and transfer it into a 15-mL
Falcon tube.
9. Fill the rest of the falcon tube with Hanks Solution (HBSS)
containing 0.5% human serum albumin.
10. Centrifuge the cells at 400 g for 10 min at 22 C and discard
the supernatant.
11. Resuspend the pellet in 5 mL of 0.83% NH4Cl pH 7 to lyse the
remaining erythrocytes (see Note 7).
12. Keep the Falcon tube for 5 min at 37 C in a bath water.
13. Fill the 15-mL falcon tube with HBSS containing 0.5% human
serum albumin.
14. Centrifuge the cells at 400 g for 10 min at 22 C and discard
the supernatant.
15. Resuspend the pellet in 12 mL of a HBSS solution with 0.5%
human serum albumin.
16. Wash one more time in 12 mL of a HBSS containing 0.5%
human serum albumin and centrifuge as described on step 14.
17. Resuspend the pellet in 1 mL of HBSS without albumin.
18. Examine the cells for neutrophil content by staining it with
Wright-Giemsa stain according to manufacturer’s protocol,
followed by morphological analysis of neutrophils (see Note 8).
19. Count the number of cells per mL using a standard Newbauer
hemocytometer (see Note 9).
3.2 Galectin 1. Thaw frozen vial of recombinant galectin by placing it on ice

Preparation (4 C) (see Note 10).
2. Re-equilibrate a PD-10 column with 5 column volumes of
cold PBS.
3. To remove lactose and β-ME from galectin, add 1 mL of
recombinant galectin solution to the re-equilibrated PD-10
column (see Note 11).
4. Collect 0.5-mL fractions individually from the PD-10 column
following the addition of recombinant galectin.
5. Once the recombinant galectin solution has completely pene-
trated the column, add 2 mL of cold PBS.
6. Continue collecting 0.5-mL fractions from the PD-10 column
following the addition of cold PBS, adding additional PBS, as
needed to elute all galectin from the column and to prevent the
column from drying out.
7. Examine protein content of each fraction by measuring the OD
at 280 nm. This is typically done by diluting each fraction
tenfold (i.e., 10 μL of each fraction in 90 μL PBS) followed
by the examination of the OD 280 nm on a standard spectro-
photometer (see Note 12).
8. Place pooled recombinant galectin on ice until ready to use.
9. Prepare galectins in PBS at five times the desired final concen-
tration (see Note 13).
3.3 ROS Production 1. Prepare 102 M luminol and 102 M lucigenin stock solutions
by Neutrophils (see Notes 14 and 15).
Followed Galectin
Treatment
Fig. 2 The integrated areas of chemiluminescence (CL) profile represent kinetics

of ROS production by human neutrophils. The amount of ROS is measured during
10 min for the first stimulation and calculated considering 0–3 min (CL area 1).
For the second stimulation, the integrated area corresponds to the following time
interval, 10–13 min (CL area 2). The CL value before cell stimulation at
0 and10 min (indicated as a and b) is considered as the baseline value to
calculate the integrated CL area of the respective step. BCL1: baseline of the
integrated CL Areas 1; BCL2: baseline of the integrated CL Area 2. (Reproduced
from Molecular Immunology: “Galectin-1 modulation of neutrophil reactive
oxygen species production depends on the cell activation state” (2019), with
the permission from Elsevier)
2. Distribute 5 105 neutrophils/200 μL in a 2-mL polystyrene

tubes (diameter of 12 mm and volume from a few microliters
up to 5 mL) (see Note 16 and steps 1–14).
3. Place the tubes containing neutrophils on the equipment (Auto
Lumat LB 953 luminometer; EG&G Berthold, Germany, or
equivalent), add luminol or lucigenin at final concentration
104 M and keep at 37 C with for 2 min. Luminol and
lucigenin should be present in all assay’s tubes (see Note 17).
4. As a control, incubate the cells only with: (1) HBSS medium;
(2) 20 m M lactose or 5 mM TDG; (3) 20 mM sucrose; and
(4) 1 mM DPI (see Note 18).
5. The incubation of the galectin can be performed either before
or after exposure to fMLP or PMA activators. If galectin incu-
bation will be performed before, follow the steps 6–10; if
galectin incubation will be performed after fMLP or PMA
activators, follow the steps 11–13.
6. After 2 min of incubation with luminol or lucigenin, add
galectin at concentration ranging from 1.25 to 20 μM diluted
in PBS or HBSS, in the presence or absence galectin controls
Fig. 3 (a) ROS production in naı̈ve human neutrophils is negatively modulated by Gal-1 and its effect is
reverted partially with TDG. (a) Nonactivated neutrophils (5 105 cells) are treated with Gal-1 (10 μM) for
10 min in the presence of TDG (5 mM) or sucrose (20 mM), and further stimulated with fMLP (106 M) for more
10 min. (b) Nonactivated neutrophils sequentially exposed to Gal-1 (10 μM) followed PMA (107 M) for 10 min
periods each. (a) and (b) Gal-1 was added to the reaction medium at t ¼ 0 min, but the integrated CL area was
calculated only for the second step of reaction (CL area 2), after addition of fMLP or PMA at t ¼ 10 min.
Control: Cells exposed only to HBSS during the whole treatment period (20 min). (c) Schematic representation
of the successive treatment of non-activated neutrophils with Gal-1 plus fMLP activation. Data are expressed
as mean SEM of the integrated CL areas of three independent experiments assayed in duplicate. Values not
sharing the same letter (a–d) are significantly different from each other (unpaired t test). In (a): p ¼ 0.045
(a vs. b), p ¼ 0.0309 (b vs. c). In (b): p ¼ 0.0245 (a vs. b), p < 0.0001 (a vs. c; a vs. d), p ¼ 0.0008 (b vs. c),
p ¼ 0.0001 (b vs. d). (Reproduced from Molecular Immunology: “Galectin-1 modulation of neutrophil reactive
oxygen species production depends on the cell activation state” (2019), with the permission from Elsevier)
(lactose, sucrose, and TDG), into the tube containing 5 105

neutrophils/200 μL (see Note 19).
7. After immediately drawing galectin, controls and galectin con-
trols measure ROS production by neutrophils for 10 min.
8. After 10 min, add fMLP (concentration range: 106 M to
109 M) or PMA (concentration range: 106 M to 109 M)
as neutrophil activators and keep the reaction for more 10 min
while the ROS production is measured.
9. Optional: other steps of activation can be done in sequence
(repeat the following steps 6–8).
10. All the measurements are recorded in photons detected per
unit time, expressed in counts per minute (cpm). The
Fig. 4 Gal-1 induces ROS production in human neutrophils pre-activated with fMLP in a concentration and
lectin activity-dependent manner. (a) Neutrophils sequentially treated with different concentrations of fMLP
(106, 107, and 108 M) and Gal-1 (1.25, 5, 10, and 20 μM). (b) Neutrophils sequentially treated with
107 M fMLP and 10 μM Gal-1, in the presence of lactose (20 mM) or sucrose (20 mM). fMLP was added to
the reaction medium at t ¼ 0 min, but the integrated CL area was calculated only for the second reaction step
(CL area 2), after addition of Gal-1 and/or inhibitors at t ¼ 10 min. Cells exposed only to HBSS during the
whole treatment period (20 min). (c) schematic representation of the activation of non-activated neutrophils
with fMLP followed Gal-1 treatment, in the presence or absence of lactose or sucrose. Data are expressed as
mean SEM of the integrated CL areas of three independent experiments assayed in duplicate. Data are
expressed as mean SEM of the integrated CL areas of three independent experiments assayed in duplicate.
* p < 0.05 vs. HBSS (ANOVA and Tukey post hoc test). (Reproduced from Molecular Immunology: “Galectin-1
modulation of neutrophil reactive oxygen species production depends on the cell activation state” (2019), with
the permission from Elsevier)
integrated areas of chemiluminescence (CL) profiles are calcu-

lated considering intervals: 0–3 min (CL area 1) for this first
treatment (Fig. 2—first peak) and 10–13 min (CL area 2;
Fig. 2—second peak and Fig. 3a, b) for the second stimulation
(see Note 15).
11. If the sequence of the assay is to stimulate the neutrophils
before the treatment with galectins, after 2 min of incubation
with luminol or lucigenin add the activators fMLP (concentra-
tion range: 106 M to 109 M) or PMA (concentration range:
106 M to 109 M) and measure ROS production by neutro-
phils for 10 min.
12. Sequentially, add galectin (range: 1.25–20 μM) and their con-
trols as described above (step 6), and keep the reaction for
more 10 min while the ROS production is measured.
13. All the measurements are recorded in photons detected per
unit time, expressed in counts per minute (cpm). The
integrated areas of chemiluminescence (CL) profiles are calcu-
lated considering an intervals: 0–3 min (CL area 1) for this first
treatment (Fig. 2—first peak) and considering 10–13 min
(CL area 2: Fig. 4a, b) for the second stimulation (see Note
15).
14. Statistical analysis could be performed by analysis of variance
(ANOVA) followed by the parametric Tukey-Kramer test or
unpaired t-test, with the aid of the GraphPad Prism Software
(GraphPad Software Inc., San Diego, CA, USA).
4 Notes
1. Calculate the volume of the whole blood that will be used in

the assay according to the demand for each assay. Remember
that for an amount of 6 mL of whole blood/tube, the yield is
about 103 neutrophils/ mL.
2. Add 3 mL of Histopaque-1119 to a 15-mL conical centrifuge
tube, and then layer 3 mL of Histopaque-1077 onto the
Histopaque-1119. It is important to keep the cells at room
temperature to avoid their activation.
3. It is important to know that abrupt braking or acceleration may
interfere in layer separation.
4. Centrifugation at temperatures lower than 10 C may result in
cell clumping and poor recovery.
5. Taking over 10 min to centrifuge will make the blood (Histo-
paque/whole blood solution) start to improperly separate.
6. Plasma and platelets will be on the top of the separation
medium. It is important to carefully remove the platelets and
plasma from the top and sequentially the monocytes layer to
avoid mixing of all components with neutrophils layer. Failing
to follow procedures may cause absence of layers and gradient
mixture.
7. The pellet will look red due to significant red blood cell
contamination.
8. Using this protocol, typically >90% of the isolated cells are
neutrophils.
9. Use 0.4% trypan blue (Sigma-Aldrich) to determine the per-
cent of viable and nonviable cells. If the nonviable cell count is
higher than 10%, these cells are not sufficiently healthy to use in
this assay.
10. The purification of recombinant human Gal-1 was performed
using affinity chromatography on lactosyl-Sepharose, as previ-
ously described [32]. All galectin preparation ought to be done
in a sterile laminar flow hood to avoid cell contamination.
There may be potential lipopolysaccharide (LPS) contamina-
tion in preparing galectin solutions. Hence, the removal of LPS
can be done using the commercially available limulus
amebocyte lysate (LAL) assay (Pierce). If significant LPS con-
tamination is noted, LPS removal can also be achieved by
passing the endotoxin-contaminated recombinant galectin
sample over a polymyxin B-agarose column (Sigma-Aldrich)
according to the manufacturer’s protocol. Repeat removal
until LPS is undetectable. LPS detection is performed using
Pierce™ LAL Chromogenic Endotoxin Quantitation Kit
(Thermo Scientific™).
11. Several galectins display unique sensitivity to oxidative inacti-
vation [33, 34]. This should be considered when assessing the
potential ability of an individual galectin to induce ROS
production.
being examined to calculate actual concentration in
mg/mL. Alternative approaches can be used to determine
these values, which can include Bradford or BCA assays to
calculate protein concentration.
13. Galectin-1 and -3 can be lyophilized, weighed, and resus-
pended in the choice buffer, followed by the protein concen-
tration determination as indicated in Note 12.
14. (A) Preparation of 102 M luminol stock solution: (1) Add
4990 μL of HBSS to a Falcon tube (tube 1). Keep out of the
light. Reserve. (2) Add 1.77 mg of luminol to an Eppendorf
tube. Add 20 μL of 0.1 M NaOH or dimethyl sulfoxide
(DMSO). Homogenize by vortexing until complete solubili-
zation (tube 2). Keep out of the light. (3) Add 10 μL tube 2 to
tube 1, carefully pouring the luminol solution through the wall
of tube 2. (4) Close tube 2 carefully and homogenize by
vortexing until the mixture becomes transparent. This is the
102 M luminol stock solution. Keep out of the light. (5) The
luminol stock solution will be diluted 1:100 in the reaction
tube (final concentration 104 M in the total reaction volume).
(B) Preparation of 102 M lucigenin stock solution: (1) Add
4500 μL of HBSS in a Falcon tube (tube 1). Keep out of the
light. (2) Add 5.1 mg of lucigenin to an Eppendorf tube. Add
1 mL of HBSS and homogenize by vortexing until complete
solubilization (tube 2). (3) Add 500 μL of the lucigenin solu-

tion from tube 2 to tube 1. Homogenize by vortexing. Keep
out of the light. The lucigenin stock solution will be diluted 1:
100 in the reaction tube (final concentration 104 M in the
total reaction volume).
15. Both lucigenin and luminol solutions are stable at room tem-
perature for 8–12 h if protected from the light. In this proto-
col, the experimental conditions were established in order to
guarantee the sensitivity and stability of the probes during the
assays, which can be observed in the second peaks in Figs. 2, 3a,
b, 4a, b. Thus, concentrations of luminol, lucigenin, fMLP,
PMA, and neutrophils, and the measurement time of ROS
production were based on previously described methods
[35, 36] with modifications for standardization in our labora-
tory. Luminol and lucigenin-dependent chemiluminescence is
a highly sensitive technique used to measure ROS production
by cells [37], but it has limitations like other techniques and
probes described for this purpose. Therefore, the choice of
methodology should consider characteristics such as selectivity,
sensitivity, and biological matrix of the experimental model,
following guidelines and recommendations from the literature.
The limitations of each methodology must be overcome by
conducting controlled experiments and using pharmacological
inhibitors [38].
16. Calculate the number of cells necessary for the total assay and
do the dilution according to the maximal volume that should
be used for the tube (around 200–300 μL). Remember that
the neutrophils need to be resuspended in HBBS medium. The
basic protocol requires one tube with neutrophils in HBBS
medium, four tubes with neutrophils plus galectin-1 (range:
1.25, 5, 10, and 20 μL), three tubes with neutrophils plus
galectin controls (lactose, sucrose, TDG—see Note 18), one
tube with neutrophils plus DPI (see Note 18), three tubes with
neutrophils, galectin, and galectin controls. Apart of the tube
with neutrophils in HBSS medium, all experimental tubes
should be prepared in enough replicates to perform the assay
with or without fMLP or PMA (concentration range: 106 M
to 109 M), before or after galectin-1 exposure. Bear in mind
that each protocol requires the use of either luminol or luci-
genin as luminescent probe.
17. This incubation can be done outside or inside of the lumin-
ometer. Protocols can be applied to luminescent microplate
readers. Consider standardizing cell number adjustments, sti-
muli, and treatments when necessary.
18. Control containing galectin at final concentration in the pres-
ence of 20 mM lactose or 5 mM TDG (both hapten inhibitors
of galectin–carbohydrate binding) are used to verify if galectin

effects are dependent on carbohydrate recognition. 20 mM
sucrose is a hapten sugar that does not inhibit the galectin-
carbohydrate binding. Moreover, it is of utmost importance to
add DPI at 1 mM, as an NADPH oxidase inhibitor. TDG can
be a more efficient inhibitor than lactose for galectins, being
commonly used instead.
19. During galectin preparation, make sure that their concentra-
tions are around 10 times more than the one that will be used.
This method allows avoiding increased dilution in neutrophils
tubes. Also, the dimer configuration of Gal-1 is maintained
when stored in high concentration [19, 33, 34].
Acknowledgments
We are thankful to the Brazilian National Council for Scientific

and Technological Development (CNPq—#312606/2019-2 for
M.D.B.).
References
1. Sies H, Jones DP (2020) Reactive oxygen spe- 7. Nordberg J, Arnér ESJ (2001) Reactive oxygen
cies (ROS) as pleiotropic physiological signal- species, antioxidants, and the mammalian
ling agents. Nat Rev Mol Cell Biol 21(7): thioredoxin system. Free Radic Biol Med
363–383. https://doi.org/10.1038/s41580- 31(11):1287–1312. https://doi.org/10.
020-0230-3 1016/s0891-5849(01)00724-9
2. Phaniendra A, Jestadi DB, Periyasamy L 8. Marnett LJ (2000) Oxyradicals and DNA dam-
(2015) Free radicals: properties, sources, tar- age. Carcinogenesis 21(3):361–370. https://
gets, and their implication in various diseases. doi.org/10.1093/carcin/21.3.361
Indian J Clin Biochem 30(1):11–26. https:// 9. Stadtman ER, Levine RL (2000) Protein oxi-
doi.org/10.1007/s12291-014-0446-0 dation. Ann N Y Acad Sci 899:191–208.
3. Halliwell B (2006) Phagocyte-derived reactive https://doi.org/10.1111/j.1749-6632.2000.
species: salvation or suicide? Trends Biochem tb06187.x
Sci 31:509–515. https://doi.org/10.1016/j. 10. Yl€a-Herttuala S (1999) Oxidized LDL and ath-
tibs.2006.07.005 erogenesis. Ann N Y Acad Sci 874:134–137.
4. Reilly PM, Schiller HJ, Bulkley GB (1991) https://doi.org/10.1111/j.1749-6632.1999.
Pharmacologic approach to tissue injury tb09231.x
mediated by free radicals and other reactive 11. Babior BM, Kipnes RS, Curnutte JT (1973)
oxygen metabolites. Am J Surg 161(4): Biological defense mechanisms. The produc-
488–503. https://doi.org/10.1016/0002- tion by leukocytes of superoxide, a potential
9610(91)91120-8 bactericidal agent. J Clin Invest 52(3):
5. Kurutas EB (2016) The importance of antiox- 7 4 1 – 7 4 4 . h t t p s : // d o i . o r g / 1 0 . 1 1 7 2 /
idants which play the role in cellular response JCI107236
against oxidative/nitrosative stress: current 12. Bravenboer B, Kappelle AC, Hamers FPT et al
state. Nutr J 15(1):71. https://doi.org/10. (1992) Potential use of glutathione for the
1186/s12937-016-0186-5 prevention and treatment of diabetic neuropa-
6. Valko M, Leibfritz D, Moncol J et al (2007) thy in the streptozotocin-induced diabetic rat.
Free radicals and antioxidants in normal physi- Diabetologia 35(9):813–817. https://doi.
ological functions and human disease. Int J org/10.1007/BF00399926
Biochem Cell Biol 39:44–84. https://doi. 13. Cameron NE, Cotter MA, Maxfield EK (1993)
org/10.1016/j.biocel.2006.07.001 Anti-oxidant treatment prevents the
development of peripheral nerve dysfunction in 24. Elola T, Chiesa E, Fink NE (2005) Activation
streptozotocin-diabetic rats. Diabetologia of oxidative burst and degranulation of porcine
36(4):299–304. https://doi.org/10.1007/ neutrophils by a homologous spleen galectin-1
BF00400231 compared to N -formyl- l -methionyl- l -leucyl-
14. Kolaczkowska E, Kubes P (2013) Neutrophil l -phenylalanine and phorbol 12-myristate
recruitment and function in health and inflam- 13-acetate. Comp Biochem Physiol B Biochem
mation. Nat Rev Immunol 13(3):159–175. Mol Biol 141(1):23–31. https://doi.org/10.
https://doi.org/10.1038/nri3399 1016/j.cbpc.2005.01.004
15. Robinson BS, Arthur CM, Evavold B et al 25. Forsman H, Salomonsson E, Önnheim K et al
(2019) The sweet-side of leukocytes: galectins (2008) The β-galactoside binding immuno-
as master regulators of neutrophil function. modulatory lectin galectin-3 reverses the
Front Immunol 10:1762. https://doi.org/ desensitized state induced in neutrophils by
10.3389/fimmu.2019.01762 the chemotactic peptide f-met-Leu-Phe: role
16. Barondes SH, Castronovo V, Cooper DNW of reactive oxygen species generated by the
et al (1994) Galectins: a family of animal NADPH-oxidase and inactivation of the ago-
β-galactoside-binding lectins. Cell 76(4): nist. Glycobiology 18(11):905–912. https://
597–598. https://doi.org/10.1016/0092- doi.org/10.1093/glycob/cwn081
8674(94)90498-7 26. Sato S, Nieminen J (2002) Seeing strangers or
17. Cooper D, Iqbal AJ, Gittens BR et al (2012) announcing “danger”: Galectin-3 in two mod-
The effect of galectins on leukocyte trafficking els of innate immunity. Glycoconj J 19(7–9):
in inflammation: sweet or sour? Ann N Y Acad 583–591. https://doi.org/10.1023/B:
Sci 1253:181–192. https://doi.org/10.1111/ GLYC.0000014089.17121.cc
j.1749-6632.2011.06291.x 27. Carlsson S, Öberg CT, Carlsson MC et al
18. Thiemann S, Baum LG (2016) Galectins and (2007) Affinity of galectin-8 and its carbohy-
immune responses-just how do they do those drate recognition domains for ligands in solu-
things they do? Annu Rev Immunol 34: tion and at the cell surface. Glycobiology
243–264. https://doi.org/10.1146/annurev- 17(6):663–676. https://doi.org/10.1093/
immunol-041015-055402 glycob/cwm026
19. Rodrigues LC, Kabeya LM, Azzolini AECS 28. Nishi N, Shoji H, Seki M et al (2003) Galectin-
et al (2019) Galectin-1 modulation of neutro- 8 modulates neutrophil function via interaction
phil reactive oxygen species production with integrin αM. Glycobiology 13(11):
depends on the cell activation state. Mol 755–763. https://doi.org/10.1093/glycob/
Immunol 116:80–89. https://doi.org/10. cwg102
1016/j.molimm.2019 29. Vega-Carrascal I, Bergin DA, McElvaney OJ
20. Feuk-Lagerstedt E, Jordan ET, Leffler H et al et al (2014) Galectin-9 signaling through
(1999) Identification of CD66a and CD66b as TIM-3 is involved in neutrophil-mediated
the major galectin-3 receptor candidates in gram-negative bacterial killing: an effect abro-
human neutrophils. J Immunol 163(10): gated within the cystic fibrosis lung. J Immunol
5592–5598 192(5):2418–2431. https://doi.org/10.
4049/jimmunol.1300711
21. Alves CMOS, Silva DAO, Azzolini AECS et al
(2013) Galectin-3 is essential for reactive oxy- 30. Dahlgren C, Stendahl O (1983) Role of mye-
gen species production by peritoneal neutro- loperoxidase in luminol-dependent chemilumi-
phils from mice infected with a virulent strain nescence of polymorphonuclear leukocytes.
of toxoplasma gondii. Parasitology 140(2): Infect Immun 39(2):736–741. https://doi.
2 1 0 – 2 1 9 . h t t p s : // d o i . o r g / 1 0 . 1 0 1 7 / org/10.1128/IAI.39.2.736-741
S0031182012001473 31. Pavelkova M, Kubala L (2004) Luminol-, iso-
22. El-Benna J, Dang PM, Gougerot-Pocidalo luminol- and lucigenin-enhanced chemilumi-
MA, Elbim C (2005) Phagocyte NADPH oxi- nescence of rat blood phagocytes stimulated
dase: a multicomponent enzyme essential for with different activators. Luminescence 19(1):
host defenses. Arch Immunol Ther Exp 37–42. https://doi.org/10.1002/bio.754
(Warsz) 53(3):199–206 32. Dias-Baruffi M, Zhu H, Cho M, Karmakar S,
23. Almkvist J, Dahlgren C, Leffler H, Karlsson A McEver RP, Cummings RD (2003) Dimeric
(2002) Activation of the neutrophil nicotin- galectin-1 induces surface exposure of phos-
amide adenine dinucleotide phosphate oxidase phatidylserine and phagocytic recognition of
by galectin-1. J Immunol 168(8):4034–4041. leukocytes without inducing apoptosis. J Biol
https://doi.org/10.4049/jimmunol.168.8. Chem 278(42):41282–41293. https://doi.
4034 org/10.1074/jbc.M306624200
33. Stowell SR, Cho M, Feasley CL et al (2009) lucigenin-dependent chemiluminescence in

Ligand reduces galectin-1 sensitivity to oxida- human neutrophils. Infect Immun 47(1):
tive inactivation by enhancing dimer forma- 326–328. https://doi.org/10.1128/IAI.47.
tion. J Biol Chem 284(8):4989–4999. 1.326-328.1985
https://doi.org/10.1074/jbc.M80892520 37. Dahlgren C, Björnsdottir H, Sundqvist M,
34. Cooper DNW, Barondes SH (2000) Galectins. Christenson K, Bylund J (2020) Measurement
Carbohydr Chem Biol 2000:625–647. of respiratory burst products, released or
https://doi.org/10.1002/9783527618255. retained, during activation of professional pha-
ch77 gocytes. Methods Mol Biol 2087:301–324.
35. Briheim G, Stendahl O, Dahlgren C (1984) https://doi.org/10.1007/978-1-0716-0154-
Intra- and extracellular events in luminol- 9_22
dependent chemiluminescence of polymor- 38. Maghzal GJ, Krause KH, Stocker R, Jaquet V
phonuclear leukocytes. Infect Immun 45(1): (2012) Detection of reactive oxygen species
1–5. https://doi.org/10.1128/IAI.45.1.1-5. derived from the family of NOX NADPH oxi-
1984 dases. Free Radic Biol Med 53(10):
36. Dahlgren C, Aniansson H, Magnusson KE 1903–1918. https://doi.org/10.1016/j.fre
(1985) Pattern of formylmethionyl-leucyl- eradbiomed.2012.09.002
phenylalanine-induced luminol- and
Chapter 30
Analysis of Galectin-Binding Receptors on B Cells

Asmi Chakraborty, Norhan B. B. Mohammed, Angela E. Bernasconi,
and Charles J. Dimitroff
Abstract
The reported roles of the β-galactoside-binding lectin family, known as galectins, in disease development
have been advancing at a remarkable pace. Galectins and their glycan counter-receptor ligands are now
considered key functional determinants in malignant and metastatic progression, tumor immune evasion,
autoimmunity, and immune homeostasis. Their influence in these processes is elicited through coordinated
expression in tumor, immune and stromal cellular compartments. While analysis of galectin levels in related
research efforts is routinely performed through immunoassays and RT-qPCR, detection, and identification
of glycan counter-receptor ligands in their native form on the cell surface has lagged. In this report, we
present methods to detect and identify glycan counter-receptor ligands to galectin (Gal)-3 and Gal-9—two
galectins at the crosshairs of cancer and immunology research. As a model, we will describe (1) isolation of
human B-cell subsets from fresh tonsil tissue, (2) assaying of Gal-3/-9-binding activities on human B cells,
and (3) identifying Gal-3/-9 ligands on human B-cell surfaces. These methods, of course, can be imple-
mented on any cell type to provide a cellular and molecular context capable of transmitting a galectin-
mediated phenotype. Establishing a galectin-binding activity on specific counter-receptor ligand(s) can help
unearth potential critical determinants capable of delivering cellular signals required for disease progression.
These advances open new avenues of research investigation that result in novel therapeutic targets and
approaches.
Key words Galectins, Ligands, Receptors, B cells, T cells, Poly-N-acetyllactosamines
1 Introduction
Immune modulation is controlled by a multitude of microenviron-

mental factors, which potentiate genotypic and phenotypic changes
in immune cells [1–4]. Of note, one of the main leukocyte targets
to microenvironmental cues are cell surface glycoproteins. Immune
cell surfaces present with unique glycoconjugates, wherein a gly-
come signature is the common distinguishing factor on each
immune cell population. Characteristically, there are major glycome
shifts in immune cells, as they mature, differentiate, polarize,
and/or change activation states [5]. One of the major classes of
565
566 Asmi Chakraborty et al.
proteins that interact with the highly dynamic immune glycome is

galectins. There are 15 different galectins that have been identified
in mammals and are subdivided into three types: prototype,
chimera-type, and tandem repeat-type. Galectin (Gal)-1, -2, -5,
-7, -10, -11, -13, -14, and -15 belong to the prototype family,
which harbor one carbohydrate-binding domain (CRD) domain
and form noncovalent homodimers. Gal-3 is the only chimeric-type
galectin with its CRD domain connected to collagen-like oligomer-
ization domain. Due to its structure, Gal-3 can form higher order
pentamers. Gal-4, -8, -9 and -12 belong to the tandem repeat-type
with two CRD domains covalently attached by a linker of variable
length. Although, galectins generally have affinity for β-galactoside-
containing glycans, particularly N-acetyllactosamine (LacNAc) and
derivatives thereof, each galectin exhibits a binding specificity for a
particular glycoconjugate(s). Galectins have well-defined intra- and
extracellular functions; however, even though they have a specific
CRD domain designed to interact with distinct cell surface glyco-
sylated receptors, their mechanism of secretion remains vague
[6]. The intracellular roles of galectins vary from acting as
co-transcriptional regulators, pro/anti-apoptotic factors to gene
splicing [6]. With the regard to extracellular functions, galectins
can oligomerize utilizing hydrogen bonds and van der Waals forces
to form lattice structures, which alter receptor conformation and
downstream cell signaling. Galectin-dependent signaling, thereby,
can help maintain homeostasis or encourage disease formation and
progression [7–10].
Gal-1, -3 and -9 are associated with tumorigenesis as well as
autoimmune disorders, such as rheumatoid arthritis [11, 12], by
forming complex interactions with counter-receptor ligands
between immune-, stromal-, and/or disease-specific cells. The
result of the galectin–receptor ligand interaction is profound as it
conveys important major physiological functions, such as cell acti-
vation, proliferation, death, effector function, differentiation,
and/or quiescence [13]. With regard to immune modulation,
galectins affect multiple immune cell populations, such as dendritic
cells, monocytes, and lymphocytes. In T-cell biology, Gal-1 has
been heavily implicated in shaping T-cell development. Gal-1
engagement and its interactions with CD2, CD7, CD43, and
CD45 have been implicated in T-cell apoptosis. Furthermore,
Gal-1 has been demonstrated to interfere with T cell–endothelium
interactions, hinder cytotoxic T-cell expansion and promote Th2
phenotype in T helper cells [14–17]. The focus of galectins in B-cell
development is overall less studied, though Gal-3 and -9 have been
shown to play key roles in shaping B-cell functions [18–21]. As
with T-cell development, there is progressive remodeling of the
glycome of B-cell surfaces during differentiation, which greatly
affect binding capability to galectins [19, 21]. Human naı̈ve and
memory B cells, for example, have markedly different cell surface
Analysis of Galectin-Binding Receptors on B Cells 567
glycomes highlighted by i-linear poly-LacNAcs that favor Gal-9-

binding specifically to immunoregulator CD45, whereas germinal
center (GC) B cells exhibit weak Gal-9-binding via inhibitory
I-branched poly-LacNAcs [20].
Given the multifaceted role of galectins in pathological pro-
cesses, it is mechanistically, as thus, therapeutically, imperative to
identify the ligands for galectins. In this methods report, we
describe techniques adopted by our laboratory to isolate human B
cells and identify their native Gal-3 and Gal-9 counterreceptor
ligands. We present a co-immunoprecipitation protocol designed
to detect native Gal-3- and Gal-9-binding glycoproteins on human
B-cell subsets.
2 Materials
2.1 Processing of 1. Hanks’ Balanced Salt Solution (HBSS) without Ca2+, Mg2+
Tonsil to Isolate (Thermo Scientific™, #88284), pre-chilled at 4 C.
Tonsillar 2. 1 Phosphate-buffered saline (PBS), pH 7.4 (Thermo Fisher
Mononuclear Cells Scientific, Gibco™, #10010-023).
3. 70 μm nylon mesh cell strainers (Fisher, #22-363-548).
4. Scalpel blades (Exel International, #14-840-00).
5. 5-mL syringe (plungers only) (BD Biosciences, #14-829-45).
6. Ficoll/Histopaque-1077 (Sigma-Aldrich, #10771).
7. Freshly-resected tonsil tissue.
8. Appropriate centrifuge.
2.1.1 Magnetic Cell Sort 1. 1 Phosphate-buffered saline (PBS), pH 7.4 (Thermo Fisher
of Tonsillar GC and Naive Scientific, Gibco™, #10010-023).
B-Cell Subsets from 2. LS Columns (Miltenyi Biotec, # 130-042-401):
Tonsillar Mononuclear
(a) Maximum number of total cells: 2 109.
Cells Using Miltenyi
MACS® Beads (b) Maximum number of labeled cells: 1 108.
3. MACS Buffer with Bovine Serum Albumin (BSA): PBS
pH 7.4 + 0.5% BSA + 2 mM EDTA (Invitrogen, #15-575-
020) (for 500 mL: 498 mL PBS + 2.5 g BSA + 2 mL
0.5 M EDTA).
4. B-cell Isolation Kit II (human) (Miltenyi Biotec, #130-091-
151): Cocktail of biotin-conjugated monoclonal antibodies
against: CD2, CD14, CD16, CD36, CD43, and CD235a
(Glycophorin A) and Anti-Biotin MicroBeads conjugated to
monoclonal anti-biotin antibodies (isotype: mouse IgG1).
5. Anti-FITC Microbeads (Miltenyi Biotec, #130-048-701).
6. Antibodies (Ab):
(a) CD10-Biotin (Miltenyi Biotec, #130-093-451).

(b) CD77-FITC (BioLegend, #357104).
(c) Biotin anti-mouse/rat/human CD27 antibody (BioLe-
gend, #124206).
(d) MACS buffer (mentioned above).
(e) Antibody cocktail provided by B-cell isolation kit II (men-
tioned above).
Surface stain cocktail:
(a) CD77-FITC (BioLegend, #357104).
(b) CD38-PE (BioLegend, #356604).
(c) CD19-PerCP (BioLegend, #302228).
(d) IgD-APC (BioLegend, #348222).
(e) CD3-APC/Cy7 (BioLegend, #300318).
(f) CD14-APC/Cy7 (BioLegend, #325620).
(g) FACS buffer (2% (v/v) FBS in PBS).
2.2 Assaying 1. 1 Phosphate-buffered saline (PBS), pH 7.4 (Thermo Fisher

Galectin-Binding Scientific, Gibco™, #10010-023).
Activities on Tonsillar 2. BSA (Fraction V) (Fisher Bioreagents™, #BP1605100).
B-Cell Subsets Using
3. Beta-D-Lactose (Alfa Aesar, #5965-66-2).
Recombinant Human
Galectin-3 and -9 4. R10 media: RPMI medium 1640 (Gibco™, #11875093), 10%
FBS (R&D Biosystems, #S11150H), 1 Penicillin/Strepto-
(rhGal-3 and rhGal-9)
mycin (Fisher, #MT30002CI), 20 mM HEPES (Gibco, #15-
630-080), 2 mM L-glutamine (Fisher, #MT25005CI).
5. Zombie NIR™ viability dye (BioLegend, #423105).
6. Galectins used for staining:
(a) rhGal-3 (PeproTech, #450-38): 10 μg/mL.
(b) rhGal-9 (R&D, #2045-GA): 1 μg/mL.
7. Antibody staining reagents.
(a) Alexa Fluor® 647 Rat anti-mouse/human Mac-2 (Galec-
tin-3) antibody (BioLegend, #125408): 2.5 μg/mL.
(b) APC mouse anti-human Galectin-9 antibody (BioLegend,
#348908): 1 μg/mL.
2.3 Identifying 1. EZ-Link™ Sulfo-NHS-Biotin, No-Weigh™ Format (Thermo

Galectin-Binding Scientific™, #A39256).
Proteins on B Cells 2. Buffer A: 150 mM NaCl, 0.5 mM Tris–Base (pH 10.4), 1 mM
EDTA, 0.02% NaN3.
3. Rinse Buffer/Lysis Buffer: Buffer A, 2% NP-40 (Sigma,
#18896), 0.1 mM PMSF (ACROS Organics,
#AC215740010), 1 Protease inhibitors (Thermo Scientific,

#PI78440).
4. IP Wash Buffer: Buffer A, 2% NP-40, 0.1 mM PMSF, 1
Protease inhibitors, 1% BSA, 0.1% SDS.
5. 6 Sample Buffer (Laemmli buffer): 375 mM Tris–HCl
pH 6.8, 9% SDS, 50% Glycerol, 0.03% Bromophenol blue.
Add 355 mM BME to the sample buffer before use.
6. Recombinant Protein G (rProtein G) Agarose (Invitrogen™,
#15920-010).
7. Purified mouse anti-human Gal-9 Ab (BioLegend, #348902)
or Gal-3 Ab (BioLegend, #125402).
8. Hamilton Syringe.
9. 4–12% reducing gel for SDS-PAGE (Bio-Rad, #3450124).
10. SDS-PAGE Running and Transfer equipment.
11. PVDF membrane (Sigma, #IPFL00010).
12. Protein Ladder (Bio-Rad, #1610374).
13. XT MOPS running buffer—diluted to 1 (Bio-Rad,
#1610788).
14. Transfer Buffer: 20% methanol + 1 Tris-Glycine (Bio-Rad,
#1610734).
15. TBST: 1 Tris-Buffered Saline (Bio-Rad, ##1706435), 0.1%
Tween 20 (Bio-Rad, #1706531).
16. Intercept® Blocking Buffer (TBS) (LI-COR Biosciences, #
927-60001).
17. IRDye 680RD-Streptavidin (LI-COR, #926-68079).
3 Methods
3.1 Isolation Human 1. Biosafety level 2 practices must be observed when handling
B-Cell Subsets from discarded tonsil tissue. Work should be performed in a tissue
Human Tonsillar culture hood under sterile conditions.
Tissue 2. Tissues can be stored briefly (<1 h) on ice in isotonic saline
3.1.1 Process Tonsil to
solution before being transferred to HBSS for processing.
Isolate Tonsillar 3. Remove tonsil using forceps and place in a petri dish with
Mononuclear Cells 2–3 mL of 1% Penicillin/Streptomycin (Pen/Strep)
(Modified from [19, 22]) in HBSS.
4. Cut a small portion of the tissue, using scalpel blades, and
return remaining tissue in 1% Pen/Strep/HBSS. Mince the
tissue into 1–2 mm pieces. Add cold media as needed to the
other petri dish containing the remaining tonsil tissue to keep
the tissue immersed.
Fig. 1 Isolation of tonsillar mononuclear cell layer from Ficoll density gradient. Minced tonsillar tissues are
processed using the cell strainer. The strained cells are passed through 70-μm nylon mesh to get a single cell
suspension, which layered over Ficoll/Histopaque and then centrifuged. The mononuclear cell layer appears
as a cloudy, thin layer between the Ficoll and wash media. (Cartoon image created using Biorender illustrating
tool (Biorender.com))
5. Place a 70-μm nylon cell strainer in a separate petri dish. Wet

the mesh of the cell strainer with 3 mL of 1% Pen/Strep/
HBSS.
6. Transfer tissue onto the mesh of the cell strainer using a wide
orifice tip, while catching run-off containing mononuclear cells
in a petri dish.
7. Grind tissue to release lymphocytes from stroma using end of a
5-mL syringe plunger.
8. Transfer strained cells into a 50-mL polypropylene conical tube
and rinse the mesh with a 3–5 mL of HBSS, transfer strained
run-off into the conical tube. Repeat rinses until supernatant is
clear.
9. Repeat steps 3–7 until the tonsil is fully disaggregated. Replace
the nylon mesh when necessary.
10. Resuspend and pass the minced tissue through another 70-μm
nylon mesh.
11. Collect and gently layer cells over Ficoll/Histopaque-1077
(35 mL cell suspension over 10 mL Ficoll). Centrifuge at
room temperature—900 g (2300 rpm in a Beckman GS-6
centrifuge) for 20 min with no brake.
12. Using a transfer pipet or P1000, remove the mononuclear cell
layer between the Ficoll and media layers and transfer to a new
50-mL conical tube (Fig. 1).
13. Wash cells twice with 30 mL cold HBSS or PBS, and centrifuge
at 400 g (with brakes), for 10 min at room temperature.
14. After the second wash, resuspend cells in 30 mL HBSS or PBS
(with brakes).
15. Count live cells using trypan blue and hemocytometer. Typical
yields vary based on tonsil size. Range from 1 108 to 2 109
mononuclear cells (see Note 1).
16. Isolated mononuclear cells can be stored in 90% FBS, 10%
DMSO freezing media at 80 C. Alternatively specific B-cell
subsets can be further isolated from these cells using immuno-
magnetic bead separation kits.
3.1.2 Magnetic Cell 1. Thaw vials of mononuclear cells in a 37 C water bath with
Sorting of Tonsillar GC and gentle agitation.
Naive B Cells from Tonsillar 2. In a tissue culture hood, transfer cells to a 15-mL conical,
Mononuclear Cells Using slowly add 5 mL of warm R10 media in a dropwise manner
Miltenyi MACS® Beads and then add remaining media (total media volume ¼ 25 mL
per 100 million cells or max volume of 50 mL).
Cell Preparation (See Note
3. Centrifuge at 1500 rpm (426 g) for 7 min with brakes.
2)
Discard supernatant and resuspend cell pellet in 12.5 mL
MACS® buffer.
4. Pass the cell suspension through a wet 70-μm mesh nylon
strainer into a fresh 50-mL conical tube. Rinse tube and mesh
with 12.5 mL PBS and add to the cell suspension.
5. Count cells and reserve ~500,000 cells for post-sort analysis.
6. Centrifuge at 1500 rpm (426 g) for 7 min at room tempera-
ture, discard supernatant, proceed to isolation.
Isolation of Tonsillar GC B 1. Resuspend cell pellet in 50 μL MACS buffer and FITC anti-
Cells by Positive Selection human CD77 antibody (1:20 dilution to get a final concentra-
tion of 10 μg/mL) per 107 total cells. Mix well and incubate on
ice for 10 min (see Note 2).
2. Wash cells once with 1–2 mL of MACS® buffer per 107 cells to
remove unbound primary antibody. Centrifuge at 1500 rpm
(426 g) for 7 min at 4 C with brake off.
3. Resuspend cell pellet in 50 μL MACS buffer and Anti-FITC
Microbeads (1:10 dilution) per 107 total cells. Mix well and
incubate on ice for 20 min.
4. Wash cells once with 1–2 mL MACS® buffer per 107 cells.
Centrifuge at 1500 rpm (426 g) for 7 min at 4 C.
5. Resuspend up to 108 cells in 3 mL MACS® buffer.
6. Place LS column(s) in the magnetic field of MACS® Separator
and rinse with 3 mL MACS® buffer per column.
7. Add cell suspension onto column. Collect unlabeled mono-
nuclear cells for subsequent naı̈ve B-cell sorting (if needed).
8. Wash three times with 3 mL MACS® buffer. Collect flow-
through from the washing steps for subsequent naı̈ve B-cell
sorting.
9. Remove the column from the magnet and place in 15-mL

conical tube. Add 5 mL MACS® buffer and insert the plunger.
Flow-through media contains the labeled GC B-cell population
(these cells fluoresce green due to FITC labeling).
10. Pool cells and count post-sort (both labeled and unlabeled
cells).
11. Set aside sorted GC B cells and proceed with naı̈ve B cells sort
from unlabeled population.
Isolation of Tonsillar Naive 1. Pellet cells and resuspend in 50 μL MACS® buffer containing
B Cells by Negative naive B-cell enrichment antibody cocktail using the following
Selection (See Note 3) recommended dilutions:
(a) Biotinylated Ab cocktail 1:5.
(b) CD10 biotin (0.5 mg/mL) 1:10.
(c) CD27 biotin (0.5 mg/mL) 1:20.
(d) CD77 FITC (200 μg/mL) 1:20.
2. Gently vortex and incubate on ice for 10 min.
3. Wash cells once with 1–2 mL MACS® buffer per 107 cells.
4. Resuspend cells in 50 μL Microbead cocktail per 107 cells as
follows:
(a) 70% MACS® buffer.
(b) 20% Anti-biotin Microbeads.
(c) 10% Anti-FITC Microbeads.
5. Gently vortex and incubate on ice for 20 min.
6. Wash cells once with 1–2 mL MACS® buffer per 107 cells and
resuspend in 3 mL MACS® buffer.
7. Place LS columns in the magnetic field of MACS® Separator
and rinse with 3 mL MACS® buffer per column.
8. Add cell suspension to column. Collect flow-through contain-
ing unlabeled B cells.
9. Wash each column three times with 3 mL MACS® buffer and
collect flow-through from these washing steps as well.
10. Remove the column from the magnet and place in 15-mL
conical tube. Add 5 mL MACS® buffer and insert the plunger
(only do this step if analyzing labeled cell population).
11. Pool cells and count post-sort.
12. Proceed with the required downstream processing and test for
sort purity (below).
Analysis of Cell Sort Purity 1. Aliquot pre-sort cells into seven compensation tubes, 1 FMO
(fluorescence minus one) tube. Wash cells once with 1%
BSA/PBS (or FACS buffer).
2. Prepare surface stain cocktail as follows: The volumes listed

below are for a single reaction. Prepare a master mix for all
your samples by multiplying the volume of each antibody with
the number of samples. Please add an extra sample number to
account for pipetting errors.
Per 50 μL test:
CD77-FITC (BioLegend 357104) 1.25 μL

CD38-PE (BioLegend 356604) 0.31 μL
CD19-PerCP (BioLegend 302228) 1.25 μL
IgD-APC (BioLegend 348222) 0.50 μL
CD3-APC/Cy7 (BioLegend 300318) 0.50 μL
CD14-APC/Cy7 (Biolegend 325620) 0.31 μL
FACS buffer 45.69 μL
50.00 μL
3. Add surface stain to PE/Cy7 FMO gating control, then add

the following Ab to the remaining cocktail for both pre-sort
and post-sort cells:
Per 50 μL test:
CD27-PE/Cy7 (LG.3A10 clone, Biolegend 124216) 0.31 μL
4. Prepare single color compensation controls as follows:
Single stain controls Per 50 μL test

CD77-FITC 1.25 μL
CD38-PE 0.50 μL
CD19-PerCP 1.25 μL
CD27-PE/Cy7 0.50 μL
IgD-APC 0.50 μL
CD3 + CD14 APC/Cy7 0.5 μL + 0.5 μL
Unstained –
5. Add 50 μL of surface stain to each experimental condition

excluding the compensation controls.
6. Resuspend compensation control and add antibody as indi-
cated above for single stain controls.
7. Incubate on ice for 45 min in the dark.
8. Wash cells twice with 100 μL FACS buffer (1500 rpm

(426 g), 5 min, 4 C, with brakes).
9. Transfer to FACS tubes for acquisition.
3.2 Assaying 1. Thaw one vial of tonsillar mononuclear cells (~1 mL) in a 37 C
Galectin-Binding water bath. Transfer cells to 9 mL of warm R10 media (for
Activities on Tonsillar <25 million). Centrifuge with brakes at 1500 rpm (426 g)
B Cells Using rhGal-3 for 5 min at room temperature and discard supernatant.
or rhGal-9 (See Note 4) 2. Wash cells once with 10 mL PBS and then count. Aliquot
enough cells for each assay (Generally ~1 million cells are
3.2.1 Thawing and
sufficient per each staining condition).
Preparation of Cells
3. Reserve 3–5 million cells for compensation on ice.
3.2.2 Viability Stain 1. Pellet cells and then resuspend in PBS + Zombie viability dye
(BioLegend, #423105) at 1.5 106 cells per 50 μL test
(Recommended dilution 1:500).
2. Incubate for 15 min at room temperature in the dark.
3. Aliquot cells into a 96-well round bottom plate, so that there
are ~1 million cells per each assay group and 200,000 cells
(the reserved unstained cells) for each compensation group.
4. Wash cells with 200 μL 1% BSA/PBS per well. Centrifuge at
1500 rpm (426 g) for 5 min at 4 C. Flick the plate over the
waste container to discard supernatant. Without inverting the
plate, blot the plate onto absorbent paper.
3.2.3 Galectin-Binding 1. Prepare galectin ligand stains in 1% BSA/PBS to final concen-

(Ligand) Staining trations as follows:
(a) rhGal-3 (PeproTech 450-38): 10 μg/mL.
(b) rhGal-9 (R&D 2045-GA): 1 μg/mL.
2. Galectin incubated assay groups should have lactose controls,
with the same concentration of galectins in 50 mM lactose + 1%
BSA/PBS.
3. Add 50 μL galectin stains per well. Mix cells with a multichan-
nel pipette set to 30 μL/well and change tips between wells.
5. Wash cells using 200 μL 1% BSA/PBS or 50 mM lactose + 1%
BSA/PBS per well.
6. Centrifuge at 1500 rpm (426 g) for 5 min at 4 C, with
brakes. Discard supernatant.
3.2.4 Secondary 1. Prepare secondary reagent for galectin stains in 1% BSA/PBS

Reagent Stain to get final concentrations as follows:
(a) Alexa Fluor® 647 Rat anti-mouse/human Mac-2 (Galec-
tin-3) antibody (BioLegend, #125408): 2.5 μg/mL.
(b) APC mouse anti-human Galectin-9 antibody (BioLegend,

#348908): 1 μg/mL.
2. Add 50 μL of the secondary antibody to each corresponding
well. Mix cells with a multichannel pipette set to 30 μL/well
and change tips between wells.
4. Wash cells in plate twice using 200 μL 1% BSA/PBS per well.
Centrifuge at 1500 (426 g) rpm for 7 min at 4 C. Discard
supernatant as mentioned before.
5. Resuspend cells in 200 μL 1% BSA/PBS and transfer to FACS
tubes for acquisition.
3.3 Identifying 1. MACS sort naı̈ve B-cell isolates from tonsil tissue.
Galectin-Binding 2. Count cells in 1:1 dilution of trypan blue using a hemocytom-
Proteins on Naı¨ve B eter and centrifuge at 1500 rpm (426 g) for 5 min at 4 C
Cells (Fig. 2) (See Note with brakes.
5) 3. Wash cells three times with 500 μL ice-cold PBS, pH 7.4, or
3.3.1 B Cell–Galectin MACS® buffer to remove all amine containing proteins in
Incubation and Cell Lysis growth media. Aliquot 1/5 of cells to save for whole-cell
analysis.
4. On the last wash, aspirate all PBS from the cell pellet. Resus-
pend cells in 2 mM sulfo-biotin at 2.5 107 cells/mL (or ac-
cording to product guidelines).
5. Incubate cells on ice for 30 min.
6. Pellet cells and aspirate supernatant. Wash cells twice with
100 mM glycine/PBS to quench excess biotin and byproducts.
When resuspending cells for the final wash, divide cells into
tubes (coated with 1% BSA/PBS) according to the subsequent
desired assay groups.
7. Pellet cells and aspirate supernatant.
8. Resuspend 1 million cells in 50 μL 1%BSA/PBS to block
noncompetitive binding. Add rhGal-3 to a final concentration
10 μg/mL or rhGal-9 to a final concentration of 1 μg/mL.
Incubate cells on ice for at least 30 min.
9. Wash cells twice with 500 μL cold PBS.
10. Lyse cells with NP-40 lysis buffer on ice on a rocker for 1 h,
vortex gently every 5 min (for B cells, add 250 μL lysis buffer
per 108 cells). Spin lysates at top speed in a minicentrifuge to
pellet and remove nuclei/cell debris.
11. Block nonspecific binding sites on agarose beads used for IP.
12. Quantitate lysates using Bradford or BCA according to prod-
uct guidelines.
Fig. 2 Immunoprecipitation of galectin-binding receptors on B cells. Viable B-cell isolates are first biotinylated
and then incubated with rhGalectins to bind specific counter-receptor glycoprotein ligands. Cells are lysed and
incubated with anti-galectin antibodies, forming an antibody–galectin–ligand complex. These complexes are
then pulled down using Protein A or G coupled beads. (Cartoon image created using Biorender illustrating tool
(Biorender.com))
3.3.2 Preparation of 1. Cut the bottom of a P200 tip at an angle with a razor blade.
Recombinant Protein G 2. Vortex beads for 20 s immediately before pipetting. Add 40 μL
(rProtein G) Agarose Beads of rProtein G Agarose beads per reaction into a 1.5-mL
Eppendorf tube.
3. Add 400 μL IP Wash Buffer per tube and vortex (1:10 ratio of
agarose to IP Wash).
4. Incubate for 30 min on a rotator at 4 C. Pellet rProtein G
Agarose beads at 14,000 rpm (14,462 g) for 30 s in a
microcentrifuge.
5. Aspirate with a P1000, then a Hamilton Syringe to remove any
remaining buffer.
6. Wash twice with 400 μL Rinse Buffer (1:10 ratio of agarose to
IP Rinse). Vortex vigorously between washes.
7. Pellet rProtein G Agarose beads and aspirate the supernatant
with Hamilton Syringe.
3.3.3 Incubate Lysate 1. For every 100 μg of protein, add 3–5 μg of anti-human galectin
with Antibody antibody to lysate and bring final volume up to 150 μL with
PBS. Place antibody and lysate mixture to the rProtein G
Agarose beads and incubate at 4 C on rotator overnight.
Cover the lids with parafilm to prevent evaporation and
sample loss.
2. Pellet the rProtein G Agarose beads at 14,000 rpm

(14,462 g) for 30 s and using a Hamilton syringe, collect
supernatant and aliquot into freshly labeled Eppendorf tubes
for immunoblotting.
3. Wash twice using IP wash at 1:10 ratio to the rProtein G
Agarose beads, spin down, and using a freshly rinsed and
wiped off Hamilton syringe, aspirate, and discard supernatant.
4. Wash three times with Rinse Buffer at 1:10 ratio, aspirate, and
discard supernatant with Hamilton syringe at each step.
3.3.4 Elution of Sample 1. To the rProtein G Agarose beads, add 36 μL PBS (without
(Eluate or Ca2+/Mg2+) and 12 μL 6 reducing sample buffer
Immunoprecipitate (IP)) +10% BME.
2. Vortex the agarose solution and boil (95 C) for 10 min. Pellet
rProtein G Agarose beads at 14,000 rpm (14,462 g) for 30 s.
3. Using a Hamilton Syringe, transfer eluate to fresh tubes and
proceed to immunoblotting for desired targets (see Note 6).
3.3.5 Detection of 1. Prepare input samples (eluates (IPs) and lysates) for loading as
Galectin-Binding Receptors follows:
(See Note 7) (a) Whole cell lysate.
(b) Eluate/IP (obtained with control Lactose or without
Lactose).
(c) Supernatant (with and without Lactose).
2. Load equal amounts of protein into the wells along with pro-
tein ladder to reducing 4–20% gel for SDS-PAGE.
3. Run the gel at 150 V for 1–2 h (until the dye front runs off
the gel).
4. Perform wet transfer at 100 V for 1 h using PVDF membrane.
5. Incubate the PVDF membrane with blocking buffer on a
shaker for 1 h at room temperature or overnight at 4 C to
block nonspecific binding sites.
6. 680RD-Streptavidin for 30 min in the dark.
7. Wash three times with 10 mL TBST for 5 min.
8. Wash once with 10 mL deionized water.
9. Acquire 680RD-Streptavidin-reactive (biotinylated) galectin
ligand(s)-stained image on LI-COR Odyssey Scanner (see
Note 7).
4 Notes
1. The average tonsil weighs 1.5 g and yields approximately 1.2

billion cells. More inflamed tonsils can have a yield of up to
4 billion cells.
2. Dilute cells into at least 3 mL of MACs buffer before running
over LS column. Use total cell count when determining
amount of Abs/microbeads rather than live cell count. Keep
all buffers and cells on ice. All reagents must be scaled up
accordingly per 107 cells in both volume and antibody needed.
3. Naive B cells can be isolated by negative selection either from
the unlabeled cells of the positive-selection GC B-cell isolate or
from fresh tonsillar mononuclear cells.
4. It is essential to use 1% BSA/PBS throughout in galectin
experiments for maximum sensitivity due to the presence of
various glycosylated proteins in serum that can act as competi-
tive inhibitors of Galectin binding.
5. Biotinylation should be performed prior to cell lysis and immu-
noprecipitation (IP), allowing for labeling of surface proteins.
Biotin-labeled immunoprecipitated galectin-binding surface
proteins can be separated via reducing SDS-PAGE and visua-
lized by incubation with infrared dye-labeled streptavidin
probe.
6. Galectin counter-receptor ligands can also be eluted from the
beads utilizing a low pH elution buffer; Pierce™ IgG Elution
Buffer, pH 2.0 (Thermo Scientific™, #21028) followed by
neutralization buffer (1 M Tris, pH 8.5). The supernatants
from the immunoprecipitation (IP) step can be used later for
immunoblotting or multiple rounds of exhaustive IP. Further,
the eluate can also be used for mass spectrometry-based pro-
teomic analysis.
7. Detection of galectin counter-receptor ligands can be done in
multipronged manner:
(a) Immunoblotting: After detection of lactose-inhibitable
galectin glycoprotein ligand band(s) by SDS-PAGE/
biotin blotting, the same membrane can be probed for
known galectin ligands observed on other cell types.
l After probing for biotinylated protein with 680RD-
streptavidin, strip the blot using stripping buffer (New-
Blot™IR Stripping Buffer 5; LI-COR; cat#
928-40028) for 5 min at room temperature on a
rocker.
l Wash two times with 10 mL TBST for 5 min.
l Block the membrane for 1 h.
l Incubate in primary antibody at 4 C overnight on a

shaker. Use appropriate primary antibody per manufac-
turer recommendations for immunoblotting. Prepare
the antibody solution in blocking buffer.
l Wash three times in 10 mL TBST for 5 min.
l Incubate the membrane with 1:10,000 dilution of
800CW-secondary Ab (LI-COR) (against animal spe-
cies of anti-protein candidate antibody).
l Wash three time in 10 mL TBST for 5 min.
l Wash once in 10 mL of deionized water.
l Acquire stained image on LI-COR Odyssey Scanner.
(b) Mass spectrometry: To identify unknown galectin ligands,
prepare an identically loaded SDS-PAGE gel and excise
the gel fragments corresponding to the streptavidin-
stained band(s) on the blot and analyze the gel fragment
by tryptic digestion and mass spectrometry. The eluate
may be also be directly analyzed without SDS-PAGE by
tryptic digestion/mass spectrometry to identify all minor
and major galectin glycoprotein ligand candidates.
Acknowledgments
This methods chapter was supported by the National Institutes of

Health (NIH)/National Cancer Institute (NCI) Alliance of Glyco-
biologists for Cancer Research: Biological Tumor Glycomics Labo-
ratory (U01 CA225644 to CJ Dimitroff) and the NIH/National
Institute of Allergy and Infectious Diseases (NIAID) (R21
AI146368 to CJ Dimitroff). The content is solely the responsibility
of the authors and does not necessarily represent the official views
of the NIH.
References
1. Broussard G, Damania B (2020) KSHV: 4. Van De Veen W, Akdis M (2019) The use of
immune modulation and immunotherapy. biologics for immune modulation in allergic
Front Immunol 10:3084 disease. J Clin Invest 129:1452–1462
2. Navegantes KC, De Souza GR, PAT P et al 5. Earl LA, Bi S, Baum LG (2010) N- and
(2017) Immune modulation of some autoim- O-glycans modulate galectin-1 binding,
mune diseases: the critical role of macrophages CD45 signaling, and T cell death. J Biol
and neutrophils in the innate and adaptive Chem 285:2232–2244
immunity. J Transl Med 15:36 6. Cummings RD, Liu FT (2009) Galectins. In:
3. Tan L, Wu H, Liu Y et al (2016) Recent Varki A, Cummings RD, Esko JD, Freeze HH,
advances of exosomes in immune modulation Stanley P, Bertozzi CR, Hart GW, Etzler ME
and autoimmune diseases. Autoimmunity 49: (eds) Essentials of glycobiology. Cold Spring
357–365 Harbor Laboratory Press The Consortium of
Glycobiology Editors, La Jolla, California.,
Cold Spring Harbor (NY)
7. Fouillit M, Joubert-Caron R, Poirier F et al immunoregulatory signature in Th cells func-

(2000) Regulation of CD45-induced signaling tionally defined by IL-10 expression. J Immu-
by galectin-1 in Burkitt lymphoma B cells. Gly- nol 188:3127–3137
cobiology 10:413–419 16. Cedeno-Laurent F, Watanabe R, Teague JE
8. Hoyer KK, Pang M, Gui D et al (2004) An et al (2012) Galectin-1 inhibits the viability,
anti-apoptotic role for galectin-3 in diffuse proliferation, and Th1 cytokine production of
large B-cell lymphomas. Am J Pathol 164: nonmalignant T cells in patients with leukemic
893–902 cutaneous T-cell lymphoma. Blood 119:
9. Perillo NL, Pace KE, Seilhamer JJ et al (1995) 3534–3538
Apoptosis of T cells mediated by galectin-1. 17. Stillman BN, Hsu DK, Pang M et al (2006)
Nature 378:736–739 Galectin-3 and galectin-1 bind distinct cell sur-
10. Rabinovich GA, Riera CM, Landa CA et al face glycoprotein receptors to induce T cell
(1999) Galectins: a key intersection between death. J Immunol 176:778–789
glycobiology and immunology. Braz J Med 18. Clark MC, Pang M, Hsu DK et al (2012)
Biol Res 32:383–393 Galectin-3 binds to CD45 on diffuse large
11. Cedeno-Laurent F, Barthel SR, Opperman MJ B-cell lymphoma cells to regulate susceptibility
et al (2010) Development of a nascent galectin- to cell death. Blood 120:4635–4644
1 chimeric molecule for studying the role of 19. Giovannone N, Antonopoulos A, Liang J et al
leukocyte galectin-1 ligands and immune dis- (2018) Human B cell differentiation is charac-
ease modulation. J Immunol 185:4659–4672 terized by progressive remodeling of O-linked
12. Mendez-Huergo SP, Hockl PF, Stupirski JC Glycans. Front Immunol 9:2857
et al (2019) Clinical relevance of Galectin-1 20. Giovannone N, Liang J, Antonopoulos A et al
and Galectin-3 in rheumatoid arthritis patients: (2018) Galectin-9 suppresses B cell receptor
differential regulation and correlation with dis- signaling and is regulated by I-branching of
ease activity. Front Immunol 9:3057 N-glycans. Nat Commun 9:3287
13. Johannes L, Jacob R, Leffler H (2018) Galec- 21. Giovannone N, Smith LK, Treanor B et al
tins at a glance. J Cell Sci 131:jcs208884 (2018) Galectin-glycan interactions as regula-
14. Cedeno-Laurent F, Dimitroff CJ (2012) tors of B cell immunity. Front Immunol 9:
Galectin-1 research in T cell immunity: past, 2839
present and future. Clin Immunol 142: 22. Johnston A, Sigurdardottir SL, Ryon JJ (2009)
107–116 Isolation of mononuclear cells from tonsillar
15. Cedeno-Laurent F, Opperman M, Barthel SR tissue. Curr Protoc Immunol. Chapter 7:Unit
et al (2012) Galectin-1 triggers an 7.8
Chapter 31
Methods for Assessing the Effects of Galectins on Leukocyte

Trafficking and Clearance
Hannah L. Law and Dianne Cooper
Abstract
Numerous protocols exist for investigating leukocyte recruitment and clearance both in vitro and in vivo.
Here we describe an in vitro flow chamber assay typically used for studying the mechanisms underpinning
leukocyte movement through the endothelium and zymosan-induced peritonitis, an acute in vivo model of
inflammation that enables both leukocyte trafficking and clearance to be monitored. Insight is given as to
how these models can be used to study the actions of galectins on the inflammatory process.
Key words Leukocyte, Neutrophil, Endothelium, Galectin, Transmigration, Efferocytosis
1 Introduction
Inflammation is a complex and coordinated response that has

evolved to maintain tissue homeostasis in response to injury or
infection. The rapid migration of leukocytes from the bloodstream
to the site of inflammation is a critical process in the inflammatory
response, as evidenced in patients with leukocyte adhesion defi-
ciency (LAD), whereby abnormally high susceptibility to infections
results from impaired interactions between leukocytes and the
blood vessel wall [1–4]. Once the inflammatory response has served
its purpose, the ideal outcome is resolution, a process driven by a
range of specialized pro-resolving mediators including lipid media-
tors, such as resolvins and lipoxins [5–7], as well as members of the
galectin family [8–12]. A successful resolution process requires that
leukocyte trafficking is terminated and that migrated cells are
cleared from the tissue, either through migration to the lymphatics
or efferocytosis [13]. Given the pivotal role of leukocyte trafficking
in the inflammatory process and the ever-increasing burden of
inflammation-driven pathologies [14], research into the mechan-
isms driving inflammation and its resolution remains an area of
active investigation.
581
582 Hannah L. Law and Dianne Cooper
Several methods exist for investigating leukocyte recruitment,

and here we describe two assays commonly used in our laboratory:
the in vitro flow chamber assay and zymosan-induced peritonitis. In
vitro flow chamber assays allow analysis of the leukocyte recruit-
ment cascade under flow, and several systems have been designed
for this purpose. The method detailed below describes how to use
Ibidi® chamber slides. This involves culturing endothelial cells in
the chamber slide and flowing blood or freshly isolated peripheral
blood leukocytes over a confluent endothelial monolayer
[15, 16]. Alternatively, a recombinant protein such as galectin-9,
intercellular adhesion molecule-1 (ICAM-1), CD62E or mucosal
addressin cell adhesion molecule-1 (MAdCAM-1) can be used in
place of endothelial cells [15, 17, 18]. Second, we describe the
method of zymosan-induced peritonitis, which allows enumeration
and phenotypical characterization of recruited leukocytes [19]. Fur-
thermore, when performed at multiple timepoints, the self-
resolving nature of this model allows for analysis of both the initial
recruitment and sequential clearance of myeloid cells, thus
providing valuable insight into the resolution of inflammation
[13]. We have utilized both of these methodologies for investigat-
ing the role of galectins on leukocyte recruitment both in vitro and
in vivo [12, 15, 17, 20–22] and have extended the peritonitis assay
to incorporate the subsequent clearance of the trafficked cells. This
chapter provides information of the reagents and technical know-
how required to perform these techniques and will indicate how
they can be adapted to study the effect of proteins such as galectins
on the leukocyte recruitment cascade and the subsequent clearance
of these cells from the inflammatory site.
2 Materials
2.1 In Vitro Flow 1. Human umbilical cord (see Note 1).

Chamber Assay 2. 23-G butterfly needle.
2.1.1 Isolation and 3. 30-ml syringe with Luer lock.
Culture of Human Umbilical 4. Hartman hemostat forceps, straight 10 cm (Fine Science Tools,
Vein Endothelial Cells Inc.).
(HUVEC)
5. Type II collagenase from Clostridium histolyticum (Lorne
laboratories) (see Note 2).
6. Human serum (Sigma Aldrich) (see Note 3).
7. 0.5% bovine gelatin (Sigma Aldrich).
8. T75 tissue culture flasks (see Note 4).
Buffers
9. Medium M199 (Sigma Aldrich) containing penicillin (100 U),
streptomycin (100 mg/ml), and Fungizone (2.5 mg/ml).
Methods for Assessing the Effects of Galectins on Leukocyte Trafficking. . . 583
10. Complete medium: M199 containing 20% human serum,

penicillin (100 U), streptomycin (100 mg/ml), and Fungi-
zone (2.5 mg/ml).
11. Hank’s balanced salt solution (HBSS) (Sigma Aldrich) con-
taining penicillin (100 U), streptomycin (100 mg/ml), and
fungizone (2.5 mg/ml).
12. Dulbecco’s phosphate-buffered saline (DPBS) (without cal-
cium and magnesium) (Life Technologies) containing peni-
cillin (100 U), streptomycin (100 mg/ml), and Fungizone
(2.5 mg/ml).
2.1.2 Subculturing 1. Trypsin (0.05%)/EDTA (0.02%).

of HUVEC
Buffers
2. DPBS (without calcium and magnesium) containing penicillin
(100 U), streptomycin (100 mg/ml), and fungizone (2.5 mg/
ml).
3. Complete medium (see Subheading 2.1.1).
2.1.3 Seeding Cells into 1. Ibidi μ-Slides VI 0.4 (see Note 5).
Ibidi Slides 2. 0.5% bovine gelatin (Sigma Aldrich).
3. Trypsin/EDTA.
4. Subcultured HUVEC (see Subheadings 3.1.1 and 3.1.2).
5. hrTNFα (see Note 6).
Buffers
6. Complete medium (see Subheading 2.1.1).
Special Equipment
7. Humidified incubator with 5% CO2 at 37 C.
2.1.4 Coating Chambers 1. Recombinant protein of choice, e.g., recombinant ICAM-1-Fc

with Recombinant Proteins (R&D) or recombinant galectin (rGal, -1, -3, or -9).
Buffers
1. HBSS containing 1.5% bovine serum albumin (BSA) (Sigma
Aldrich).
2.1.5 Isolation of 1. 15-ml Falcon tubes.

Neutrophils from 2. 50-ml Falcon tubes.
Human Blood
3. Histopaque 1119 (Sigma Aldrich).
4. Histopaque 1077 (Sigma Aldrich).
5. 21-G butterfly needle.
6. Sterile Pasteur pipettes.
Buffers
7. RPMI 1640 medium (Gibco).
8. Double-distilled H2O.
9. DPBS (without calcium and magnesium).
10. Sodium chloride (NaCl) (Sigma Aldrich) (3.6% in dH2O).
11. Sodium citrate (Sigma Aldrich) (3.2% in dH2O).
12. Turk’s solution: 0.01% crystal violet in 3% glacial acetic acid.
Special Equipment
13. Hemocytometer.
14. Inverted microscope.
2.1.6 Preparation of 1. Appropriate syringe/needle for blood withdrawal depending

Whole Blood for Flow whether using murine or human blood.
Assays 2. Eppendorf tubes.
3. 50-ml Falcon tubes.
Buffers
4. Anticoagulant (see Note 7).
5. HBSS.
2.1.7 In Vitro Flow 1. Human PMN (see Subheading 3.1.5).

Chamber Assay 2. 50-ml Falcon tubes.
3. Ibidi μ-Slides VI0.4 (see Note 5).
4. Tygon tubing (Saint Gobain Performance Plastics, T5002-13)
with an internal diameter of 1.6 mm and an external diameter
of 3.2 mm.
5. Luer lock syringes.
Buffers
6. DPBS (with calcium and magnesium) (see Note 8) containing
0.1% BSA.
7. DPBS (without calcium and magnesium) containing
0.1% BSA.
Special Equipment
8. Automated syringe pump (e.g., PHD 2000 programmable
pump, Harvard Apparatus).
9. A heated and humidified chamber with an active gas mixer to
ensure that the temperature and carbon dioxide levels stay
constant at 37 C with 5% CO2 throughout the experiment.
10. Inverted phase contrast microscope fitted with 10 and 20
phase contrast objectives.
11. A video camera (e.g., Q-Imaging Retiga EXi Digital).
12. Software for analysis (see Note 9).
2.2 Zymosan- 1. Zymosan A from Saccharomyces cerevisiae (Sigma Aldrich).

Induced Peritonitis 2. 29-G needles.
2.2.1 Peritoneal 3. 19-G needles.
Inflammation and Lavage 4. Carbon dioxide (CO2).
5. 5-ml syringes.
6. 15-ml Falcon tubes.
Buffers
7. DPBS (without calcium and magnesium).
8. Lavage fluid: DPBS (without calcium and magnesium) con-
taining 2 mM ethylenediaminetetraacetic acid (EDTA) (Sigma
Aldrich).
Special Equipment
9. Surgical tools including Vannas scissors, Iris curved serrated
forceps, dissecting scissors, and Dumont #5 fine forceps
(Mouse dissecting Kit, World Precision Instruments).
2.2.2 Leukocyte 1. 96-well round bottom plate.

Recruitment 2. 4% paraformaldehyde.
Antibodies
3. Anti-mouse CD45 PerCP (clone 30-F11, Biolegend)*.
4. Anti-mouse F4/80 BV650 (clone BM8, Biolegend)*.
5. Anti-mouse Ly6G PE (clone 1A8, BD Pharmingen)*.
6. Anti-mouse 7/4 FITC (clone 7/4, Abcam)*.
7. Anti-mouse CD11b BV785 (clone M1/70, Biolegend)*.
8. Anti-mouse SiglecF (clone ES22-10D8, Mitenyl Biotech)*.
*and relevant isotype control (if applicable).
9. Anti-mouse CD16/CD32 (FcγRIII/II) (clone
93, eBioscience).
Buffers
10. FACS buffer: DPBS (without calcium and magnesium) con-
taining 0.02% BSA.
11. Turk’s solution: 0.01% crystal violet in 3% glacial acetic acid.
Special Equipment
12. Neubauer hemocytometer.

13. Inverted microscope.
2.2.3 Leukocyte 1. 96-well round bottom plate.

Clearance
Antibodies
2. Anti-mouse F4/80 BV650 (clone BM8, Biolegend)*.
3. Anti-mouse Ly6G PE (clone 1A8, BD Pharmingen)*.
*and relevant isotype control (if applicable).
4. Anti-mouse CD16/CD32 (FcγRIII/II) (clone
93, eBioscience).
Buffers
5. FACS buffer: DPBS (without calcium and magnesium) con-
taining 0.02% BSA.
6. Fixation buffer: 4 stock diluted with fix/perm diluent
(eBioscience).
7. Permeabilization buffer: 10 stock diluted with ddH2O
(eBioscience).
3 Methods
3.1 In Vitro Flow 1. Collect cords in HBSS containing antibiotics and store at 4 C
Chamber until processing.
Adhesion Assay 2. Insert a butterfly needle (sheath on) into one end of the umbil-
3.1.1 Isolation of Human ical vein (see Note 10) and clamp around needle to hold in
Umbilical Vein place.
Endothelial Cells 3. Attach a 30-ml syringe containing DPBS to the needle and
flush cord with approximately 30 ml DPBS to wash away
residual blood and identify any perforations in the cord.
4. Clamp the bottom of the cord and infuse vein with approxi-
mately 20–25 ml collagenase (see Fig. 1).
5. Incubate the clamped cord for 15 min in a humidified chamber
in 5% carbon dioxide at 37 C.
6. Transfer the collagenase solution to a 50-ml Falcon tube and
flush the cord with 30 ml DPBS, also adding this to the tube.
7. Push air through the cord to collect any remaining DPBS.
8. Centrifuge the cells at 300 g for 10 min, remove the super-
natant, and resuspend the cell pellet in 15 ml complete
medium.
9. Transfer the cells to a gelatin-coated (see Note 4) T75 flask
(75 cm2) and place in a humidified incubator in 5% carbon
dioxide at 37 C.
Fig. 1 Isolation of human umbilical vein endothelial cells from umbilical cords. (a) Cross section of the
umbilical cord showing the two umbilical arteries and the umbilical vein. (b) The umbilical vein is filled with
collagenase and clamped at either end prior to incubation at 37 ˚C for 15 min
10. Change the culture medium 24 h later and then every other
day until approximately 95% confluent.
11. Once at 95% confluence, the cells can be sub-cultured or used
for experimentation (see Note 11).
3.1.2 Subculturing 1. Remove culture medium and rinse cells in 10 ml sterile DPBS.
of HUVEC 2. Add 2 ml Trypsin/EDTA (warmed to 37 C) to the cells.
3. Monitor cells microscopically for signs of “rounding up.”
4. Remove 1.5 ml Trypsin/EDTA solution from flask and dis-
card. Leave cells a further 2 min at room temperature.
5. Observe cells microscopically to assess detachment. Tap side of
flask gently if required (see Note 12).
6. Add 2 ml of complete culture medium and resuspend cells.
7. Transfer required volume of cells to new culture vessel (~150 μl

cell suspension per chamber for Ibidi μ-Slides VI0.4 flow cham-
ber slides).
3.1.3 Seeding Cells into 1. Coat Ibidi μ-Slides VI0.4 flow chamber slides with 100 μl 0.5%
Ibidi Slides gelatin for at least 30 min at room temperature.
2. Trypsinize one confluent T75 flask of HUVEC and resuspend
in 2 ml complete medium.
3. Add 150 μl cell suspension to each channel, replace the lid and
culture the cells overnight in a humidified incubator with 5%
CO2 at 37 C (see Note 13).
4. Stimulate the HUVEC 4 h prior to carrying out the flow assay
with 10 ng/ml hrTNFα in complete media. Add enough vol-
ume (~100 μl) to fill the chamber and both ports, then incu-
bate in a humidified incubator with 5% CO2 at 37 C for 4 h
(see Note 14).
3.1.4 Coating Chambers 1. Dilute recombinant protein (we routinely use rGal, -1, -3, or
with Recombinant Proteins -9) to desired concentration in HBSS (typically 10 μg/ml).
2. Coat slides with protein solution. Add enough volume
(~100 μl) to fill the chamber and both ports then incubate in
a humidified incubator with 5% CO2 at 37 C for 2 h.
3. Wash out recombinant protein with HBSS, add 1.5% BSA, and
incubate for 1 h at 37 C.
3.1.5 PMN Isolation 1. Prepare 15-ml Falcon tubes containing 3 ml histopaque 1119
overlaid with 3 ml histopaque 1077 (see Note 15).
2. Collect blood from healthy volunteers using a 21-gauge but-
terfly needle and transfer to a 50 ml Falcon tube containing a
1/10 volume of 3.2% sodium citrate to prevent clotting (see
Note 16).
3. Dilute blood 1:2 with warm RPMI.
4. Layer 6 ml blood on top of double histopaque gradient.
5. Centrifuge blood at 400 g for 30 min (see Note 17).
6. Remove plasma and PBMC layer and discard (see Note 18 and
Fig. 2).
7. Remove granulocyte layer using a sterile Pasteur pipette and
transfer to new 15-ml Falcon tubes (maximum volume 6 ml/
tube). Top tubes up to 12 ml with RPMI.
9. During this centrifugation place 50 ml double distilled water at
80 C.
10. Discard supernatant and flick tubes to resuspend neutrophil
pellet.
Fig. 2 Blood separation using a double density gradient of histopaque. Blood is layered on top of the
histopaque and centrifuged at 400 g for 30 min at room temperature (no brake) to separate it into its
constituent parts
11. Add 7.5 ml ice-cold water per tube and ensure cell pellet is fully
resuspended. Invert tube several times for a period of 10 s (see
Note 19).
12. Add 2.5 ml 3.6% NaCl per tube and invert tube twice to mix.
14. Discard supernatant and resuspend cell pellet in residual fluid.
Combine the contents of all tubes into one.
15. Take 10 μl cell suspension and add to 990 μl Turk’s solution.
16. Count number of neutrophils based on nuclear morphology
using a hemocytometer.
17. For flow chamber assays, dilute PMN to a concentration of
1 107/ml (see Note 20) in DPBS without calcium and
magnesium and place cells on ice until use.
3.1.6 Flow Chamber 1. Dilute isolated human PMN in a 50-ml Falcon tube to a
Assay (PMN) concentration of 1 106 cells/ml in DPBS (with calcium
and magnesium) and incubate for 10 min at 37 C (see
Note 21).
2. Place Ibidi slide in heated chamber and attach Tygon tubing to
outlet reservoir.
3. Pre-load 30 ml Luer lock syringes with 10 ml 37 C DPBS
(with calcium and magnesium; see Note 8) and place in
syringe pump.
4. Fill the inlet reservoir of the Ibidi slide with DPBS and then
attach the tubing from the syringes to the Ibidi slide (see Fig. 3)
and pump through DPBS at 1 dyn/cm2, taking extreme care
not to introduce any air bubbles into the system.
Fig. 3 Assembly of Ibidi® μ slides. (a) A Luer lock syringe is connected to a female Luer connector which in
turn is connected to Tygon tubing. (b) Male Luer connectors attach the tubing to the inlet and outlet ports of
the Ibidi slide
5. Once the DPBS has reached the end of the outlet tubing, pause
the syringe pump and transfer the tubing to the 50-ml Falcon
containing the isolated leukocyte suspension.
6. Start the syringe-pump, this time withdrawing the cells over
the HUVEC at 1 dyn/cm2 for 8 min (see Notes 22 and 23).
7. Record random six frames per channel along the centerline,
each lasting 10 s (see Note 24). Frames should be recorded
starting at the end of the slide nearest to the syringe pump to
avoid recording the same leukocyte population twice.
3.1.7 Flow Chamber 1. Dilute freshly drawn blood 1:10 with HBSS and place in water
Assay (Whole Blood) bath at 37 C (see Note 21).
2. Follow steps 2–5 above.
3. Start the syringe-pump, this time withdrawing the blood
through the chamber at 1.010 ml/min for 3 min (see Notes
22 and 23).
4. Flow HBSS for 1 min to enable visualization of recruited
leukocytes and then begin image acquisition (see Note 24).
3.1.8 Data Analysis Most commonly, the analysis parameters include the number of
cells initially captured, which is then divided into two groups: those
cells that are rolling or those that are firmly adherent and remain
stationary for the duration of the recording (10 s) as well as the
number of transmigrated cells (see Note 25).
1. To analyze the frames in Image Pro Plus, use the Measure >
Manual Tag option to count the phase light cells seen in the
Fig. 4 Image from a flow chamber experiment. Phase bright neutrophils can be
observed on the surface of the endothelial monolayer (circled). Neutrophils that
have transmigrated underneath the endothelium appear phase dark (arrow)
Fig. 5 Image from a flow chamber experiment using recombinant proteins as an

adhesion substrate. Individual cell tracks of neutrophils adherent to recombinant
ICAM-1-Fc analyzed as described in Note 9
first still of the frame, which are those that are captured on the
surface of the endothelia.
2. Move to the last frame and count how many of the cells have
moved throughout the recording—these are the rolling cells.
3. The total captured cells minus the rolling cells equal the adher-
ent cells.
4. Finally, use the manual tag to count the transmigrated cells,
which will appear as flattened and phase dark cells underneath
the monolayer (see Fig. 4).
5. To analyze crawling behaviors such as directionality and veloc-
ity, individual cells can be tracked using the manual tracking
plugin in Fiji and then further analyses using the Ibidi chemo-
taxis tool (see Fig. 5).
3.2 Zymosan- To investigate the role of galectins in this model, the experiment is
Induced Peritonitis performed in galectin (-1, -3 or -9) null mice or WT mice treated
with an intra-peritoneal (i.p.) injection of recombinant galectin
(rGal, -1, -3, or -9) typically 10 μg in 0.2–0.5 ml DPBS. To
determine the effects of prophylactic treatment, rGal is adminis-
tered (15 min to 1 h) prior to zymosan administration [23]. Alter-
natively, administration of the protein at the peak of the
inflammatory response (6–8 h post zymosan administration) can
be used to examine if the galectin has therapeutic effects [12].
3.2.1 Peritoneal 1. Inject mice i.p. with 1 mg Zymosan in 0.5 ml sterile DPBS
Inflammation and Lavage using a 29-G needle (see Note 26).
2. At required time-point (see Note 27) sacrifice mice by CO2
inhalation (see Note 28).
3. Perform a midline laparotomy to expose the abdominal mus-
cles, inject 5 ml lavage fluid into the peritoneal cavity using a
23-G needle.
4. Gently massage the abdomen (see Note 29).
5. Collect lavage fluid using a 19-G needle and store in a 15-ml
Falcon tube on ice (see Note 30).
3.2.2 Leukocyte 1. Ensure cells are fully resuspended and then remove 20 μl of
Recruitment fluid and add to 380 μl Turks (1:20 dilution). Count cells using
a light microscope to obtain a total cell count in a Neubauer
chamber.
2. Add 200 μl lavage fluid per well on a 96-well round bottomed
plate for analysis of cell types recruited by flow cytometry (see
Note 31).
3. Wash the cells by centrifuging (30 s at 850 g) and resuspend-
ing the pellet in 200 μl FACS buffer and repeating
centrifugation.
4. Resuspend pellet in 30 μl FACS buffer containing 1:100 anti-
mouse FcgRII/III to block against nonspecific binding and
incubate on ice for 10 min.
ing the pellet in 200 μl FACS buffer and repeating
centrifugation.
6. Resuspend pellet in 50 μl FACS buffer containing relevant
antibodies or buffer only for the wells as outlined below (see
Note 32):
(a) Antibody mix (for n samples).
(b) Unstained (one sample).
(c) Controls (one sample per antibody).
7. Incubate with the FACS buffer ( antibodies) for 30 min on

ice in the dark.
ing the pellet in 200 μl FACS buffer and repeating centrifuga-
tion. Repeat the wash step.
9. Resuspend pellet in 150 μl FACS buffer and transfer to tubes
for analysis by flow cytometry.
10. To fix cells, add 50 μl 4% paraformaldehyde to each tube and
gently resuspend cell solution (see Note 33).
11. Store flow cytometry samples at 4 C and protected from light.
3.2.3 Leukocyte 1. Ensure cells are fully resuspended and follow steps 2–5
Clearance (see Subheading 3.2.2) to block cells.
2. Resuspend pellet in 50 μl FACS buffer containing F4/80 as the
surface antibody or buffer only for the relevant wells as out-
lined below:
(a) F4/80 (2 μg/ml) followed by Ly6G (1 μg/ml) antibody
(for n samples).
(b) Unstained (one sample).
(c) F4/80 (one sample).
(d) Ly6G (one sample).
3. Incubate with the FACS buffer ( F4/80 antibody) for 30 min
on ice in the dark.
ing the pellet in 200 μl FACS buffer and repeating centrifuga-
tion. Repeat the wash step.
5. Resuspend pellet in 200 μl fixation buffer.
6. Incubate with the fixation buffer for 20 min at RT in the dark.
ing the pellet in 200 μl permeabilization buffer and repeating
centrifugation. Repeat the wash step.
8. Resuspend pellet in 50 μl permeabilization buffer containing
anti-Ly6G or buffer only for the relevant wells (as outlined in
step 2).
9. Incubate with the permeabilization buffer (Ly6G (1 μg/ml)
antibody) for 30 min at RT in the dark.
ing the pellet in 200 μl permeabilization buffer and repeating
centrifugation. Repeat the wash step twice.
11. Resuspend pellet in 200 μl FACS buffer and transfer to tubes
for analysis by flow cytometry.
12. Store flow cytometry samples at 4 C and protected from light.
3.2.4 Data Analysis Total cell numbers in the peritoneum can be determined using
Turk’s solution, a Neubauer hemocytometer chamber and inverted
microscope. Total cell counts per mouse can be calculated based on
cells per ml for the 5 ml of lavage fluid within the peritoneal cavity.
Leukocyte infiltrate can be quantified using flow cytometry
from CD45+ cells as the parent population and the subsequent
percentage of positive cells by biomarker by gating on populations
of neutrophils (7/4+ Ly6G+), inflammatory monocytes (7/4+
Ly6G ), eosinophils (SiglecF+), resolving (CD11blow), and
mature (CD11bhigh) macrophages (F4/80+) (see Fig. 6).
Leukocyte clearance by efferocytosis can be determined by flow
cytometry from gating on the cell population expressing the macro-
phage specific surface biomarker (F4/80) followed by assessing this
population for cells positive for the neutrophil specific biomarker
(Ly6G). Here, a quadrant gating strategy can be applied to deter-
mine the percentage of cells double positive (F4/80+ Ly6G+),
indicative of efferocytosis (see Fig. 7). The subsequent departure of
macrophages following their efferocytosis of neutrophils renders
this process particularly challenging to capture in vivo, as such the
percentage of double positive (F4/80+ Ly6G+) cells obtained is
generally low. Analyzing the population for the median fluorescence
intensity (MFI) value for the neutrophil (Ly6G+) staining can be
used to compare the relative number of neutrophils within
macrophages.
3.2.5 Additional Analysis 1. Centrifuge 1 ml remaining lavage fluid (1 min at 850 g) and
transfer to 1.5 ml Eppendorf tubes for additional analysis.
2. Store both the cell pellet and peritoneal lavage fluid at
20 C until subsequent analysis as outlined below
(see Note 34).
(a) Cell pellets can be analyzed by RT-PCR.
(b) Lavage fluid can be analyzed by ELISA.
4 Notes
1. Human umbilical cord cut from placenta of healthy donors

within 24 h post-delivery. Collected into cold HBSS and stored
at 4 C (for a maximum time of 24 h) prior to isolation of
HUVEC.
2. A stock solution of 1 mg/ml collagenase is prepared in medium
M199 and aliquots are stored at 20 C until required. These
aliquots are further diluted to 0.1 mg/ml in M199 for use.
Usually around 30 ml is required per cord.
a b 250K c
250K 250K
200K 200K 200K
SSC-W
150K 150K 150K
FSC-H
SSC-A
Cells
lls
Ce
le
100K 100K ng 100K
Si
50K 50K 50K Single Cells
0 0 0
0 50K 100K 150K 200K 250K 0 50K 100K 150K 200K 250K 0 50K 100K 150K 200K 250K
FSC-A FSC-A FSC-A
d e f
250K Leukocytes Neutrophils
105 105
200K
104
4
150K 10
SSC-A
Ly6G
Ly6G
100K 103 103
50K
0 0
Monocytes
0
3 3 104 105
–10 0 10 0 103 104 0 103 104
CD45 7/4 7/4
g h i
250K Macrophages Mature mac. 250K Eosinophils
105
200K 200K
104 150K
150K
SSC-A
SSC-A
CD11b
3
100K 10 100K
50K 102 50K
1 Resolving mac.
0 10 0
1 2 3 4 5 1 2 3 4 5 3 3 4
10 10 10 10 10 10 10 10 10 10 –10 0 10 10
F4/80 F4/80 SiglecF
Fig. 6 Zymosan-induced peritonitis leukocyte flow cytometry gating strategy. Murine leukocytes collected
from peritoneal lavage are analyzed by flow cytometry to determine leukocyte populations. (a) The sample is
displayed on a forward scatter area (FSC-A) against side scatter area (SSC-A) plot and the leukocytes gated on
to exclude debris. (b) Single-cell moieties are obtained from the population by first plotting FSC-A against
forward scatter height (FSC-H) to exclude doublet cells. (c) This population is then plotted on an SSC-A against
side scatter width (SSC-W) axis to further exclude doublet cells. (d) The population of singlets is plotted on an
axis with the bandpass filter for the CD45 marker against SSC-A to determine the leukocyte (CD45+)
population from other events obtained. The relative cell numbers for specific hematopoietic cell types can
calculated using this population by gating on the subpopulations of cells as follows (e) neutrophils
(7/4+Ly6G+), (f) inflammatory monocytes (7/4+Ly6G ), (g) macrophages (F4/80+), (h) mature
(F4/80+CD11bhigh) and resolving (F4/80+CD11blow) macrophages and (i) eosinophils (Siglec F+). Representa-
tive flow cytometry gating plots are shown. [Created using BioRender (BioRender.com) illustrating tool
(BioRender.com)]
a b c
250K Neutrophils Efferocytosis 5–10% Neutrophils Efferocytosis 0–5%
5
10 105
200K
4 4
10 10
150K
SSC-A
Cells
Ly6G
Ly6G
100K 103 103
50K
0 0
Macrophages Macrophages
0
0 50K 100K 150K 200K 250K
-103
0
103 104 105 0
103 104 105
FSC-A F4/80 F4/80
Fig. 7 Zymosan-induced peritonitis efferocytosis flow cytometry gating strategy. Murine leukocytes collected
from peritoneal lavage are analyzed by flow cytometry to determine levels of efferocytosis by cell surface
staining for macrophage (F4/80+) followed by fixation, permeabilization, and neutrophil (Ly6G+) staining using
fluorescently conjugated antibodies. (a) The sample is displayed on a forward scatter area (FSC-A) against
side scatter area (SSC-A) plot and the leukocytes gated on to exclude debris (note: permeabilization may result
in a reduced SSC and/or FSC, if this occurs appropriately adjust gating to fit leukocyte population). A quadrant
gating strategy is applied to determine the percentage of cells positive for efferocytosis (F4/80+Ly6G+) as
displayed in (b) during the peak of inflammation (~6 h) showing a dense neutrophil (Ly6G+F4/80 ) population
and (c) during resolution of the inflammatory response (~48 h) where there is a notable increase in the
macrophage (Ly6G+F4/80 ) population and corresponding decrease in neutrophils, suggesting they have
been cleared following the arrival of macrophages. Representative flow cytometry gating plots are shown.
[Created using BioRender (BioRender.com) illustrating tool (BioRender.com)]
3. HUVEC can also be cultured using fetal calf serum, though cell
growth and surface protein expression levels should be assessed
in-house.
4. Flasks are coated in a solution of 0.5% bovine gelatin for 30 min
at room temperature prior to use.
5. Comprehensive information about the Ibidi flow chamber sys-
tems including the different chamber slides available can be
found on their website at www.Ibidi.com. Other flow systems
widely used are the Glycotech circular and rectangular gasket
sets, though these require higher numbers of HUVEC per one
chamber (www.glycotech.com/apparatus/parallel.htm).
6. Typically, TNF-α is used to activate endothelial cells; however,
other cytokines such as IL-1β are also commonly used for assays
assessing neutrophil recruitment. Cytokines such as IFN-ɣ may
also be required if studying T-cell recruitment or if endothelial
upregulation of galectin-9 is required. Researchers need to
perform preliminary experiments to optimize conditions.
7. For murine blood, heparinized syringes are commonly used for
blood collection. For human blood, 3.2% sodium citrate is used
at a 1:10 ratio to blood.
8. Use of calcium/magnesium free DPBS will result in detach-
ment of endothelial cells and will impede integrin function.
9. Image Pro Plus (media Cybernetics) enables frame capture and

subsequent analysis using the manual tagging and object track-
ing tools; alternatively, ImageJ/FIJI [24–26] can be used. For
assessing cell crawling dynamics, the Ibidi chemotaxis tool can
be used.
Detailed instructions can be found at www.Ibidi.com.
(https://ibidi.com/img/cms/products/software/chemo
taxis_tool/Manual_ChemotaxisTool_2_0_eng.pdf). Recently,
plugins have been developed to facilitate automation of cell
tracking when using phase contrast microscopy, such as Cell
Adhesion with Supervised Training and Learning Environment
(CASTLE; [17]).
10. The umbilical cord contains three blood vessels, one large vein
and two smaller arteries surrounded by a proteoglycan-rich
matrix known as Wharton’s jelly.
11. Typically, the yield from this procedure is 0.5–1.5 106 cells
per cord. The cells are used in experimental procedures up to
passage 3 when they are discarded due to changes in their cell
surface protein expression profile.
12. If the cells start to detach too quickly, an equal volume of
trypsin deactivating media (complete media) can be added,
the cells collected and centrifuged to remove trypsin-
containing media before resuspension in the required volume
of complete culture medium.
13. The ports of the Ibidi slide should be filled completely with
medium, and the slide must be incubated in a humidified
environment to prevent evaporation.
14. Check confluency of the endothelial monolayer prior to stimu-
lation. Monolayers must be 100% confluent for flow chamber
assays. If not fully confluent, the cells can be incubated in the
Ibidi slides for a further 24 h; however, the culture media
should be renewed every 24 h due to the small volume per
channel. During this incubation stage, endothelial cells can also
be treated with recombinant galectins (rGal, -1, -3, or -9).
15. Histopaque should be allowed to warm to room temperature
prior to use, and tubes should be prepared immediately prior
to blood withdrawal. The second layer of histopaque (1077)
should be layered on top of the bottom layer carefully to avoid
mixing. This is best achieved by using either a sterile Pasteur
pipette or a pipette-boy set to low and gravity.
16. The yield of neutrophils isolated varies between donors but a
typical yield is 1 106 neutrophils/ml of blood taken. Typi-
cally for a flow chamber assay 50 ml of blood is sufficient.
17. The brake on the centrifuge should be switched off for
this spin.
18. At this point, the PBMC layer can be retained if required for
analysis of monocyte and or lymphocyte recruitment.
19. Timing is critical when lysing red cells to avoid activation of
neutrophils.
20. At this point, leukocytes/blood can be incubated with recom-
binant galectins prior to flow. Treatment of leukocytes with
galectins may cause aggregation of cells. Such an effect will
skew data acquired in the flow chamber assay. Lower concen-
trations of the galectin should be used (~30 nM, although this
can vary for each galectin and should be optimized prior to
performing the assay) or cells should be treated with an inhibi-
tor such as lactose (30 mM) to disaggregate leukocytes prior to
flow. When using galectin inhibitors such as lactose in flow
assays, this may be added to the leukocytes/blood prior to
flow or to the endothelial cells (1 h) prior to flow. It may also
be necessary to maintain the lactose concentration in the flow
buffer throughout the experimental period.
21. Leukocyte concentration can be varied depending on the leu-
kocyte subset being studied and ease of isolation.
22. When ready to flow leukocytes, it is important that the syringe-
pump acts to pull as opposed to push the cell suspension
through the flow chamber, as the latter will result in uneven
pulsatile flow.
23. We routinely perform our flow chamber experiments at a flow
rate calculated to give a wall shear stress of 0.1 Pa (1 dyn/cm2).
This is a shear stress widely used in flow assays where the aim is
to mimic conditions in inflamed post-capillary venules
[27]. Shear stress can be varied depending on the scientific
question being asked. The exact dimensions of the chamber
as well as the viscosity of the flow medium are required to
accurately calculate shear stress. If whole blood is being used
rather than a leukocyte suspension, the viscosity of the blood
must be determined in order to accurately determine the flow
rate required for a particular shear stress. Alternatively, the flow
rate can be reported as in Gittens et al. [15]. For Ibidi chamber
slides extensive information is provided in application Note 5
at www.Ibidi.com.
24. Timings for perfusion of leukocytes and data acquisition
should be determined depending on the conditions being
studied and whether the primary aim to study capture, rolling,
crawling, adhesion, and/or transmigration.
25. Ibidi chambers can be fixed after completion of the flow assay
by filling the inlet chamber port (closest to cell solution) with
1% paraformaldehyde while simultaneously removing the
connector attached to the syringe. This avoids air entering
the chamber. Histological analyses can then be performed to

visualize recruited cells.
26. Mix zymosan immediately prior to injection to ensure it is fully
suspended in the DPBS. The dose is injected into the bottom
right-hand side of the abdomen of the mouse to avoid damag-
ing the liver.
27. Typically, peritonitis experiments are performed for up to 96 h
with the leukocyte subtype recruited varying over time. At time
zero F4/80 high resident macrophages are the dominant cell
type. From around 4–12 h, Ly6g-positive neutrophils become
dominant, and the macrophage population diminishes. At later
time-points, the number of neutrophils decline, and mono-
cytes become the dominant cell type. Over time, these mono-
cytes differentiate into macrophages.
28. Do not confirm death with cervical dislocation, this can cause
internal bleeding within the peritoneal cavity, resulting in con-
tamination from peripheral leukocytes. Lavage fluid that is
bloody should be excluded from the analysis as leukocyte
counts may be skewed due to contamination.
29. Can be performed by replacing the sheath over the needle head
and rolling gently across the fluid filled peritoneum to promote
detachment and suspension of cells that are loosely adherent to
the peritoneal wall or other organs.
30. Due to a large dead volume, a portion of the lavage fluid will
remain in the cavity (expect to retrieve ~4 ml of lavage fluid per
mouse).
31. One well of a 96-well plate for each mouse for the antibody
mix, an unstained and control samples. Either isotype controls
or fluorescence minus one (FMO) staining may be used for
flow cytometric analysis.
32. The following panel of fluorescently conjugated antibodies is
recommended for immune cell labeling: CD45 PerCP (clone
30-F11, Biolegend), F4/80 BV650 (clone BM8, Biolegend)
or F4/80 APC (clone BM8, eBioscience), Ly6G PE (clone
1A8, BD Pharmingen), 7/4 FITC (clone 7/4, Abcam),
CD11b BV785 (clone M1/70, Biolegend), and SiglecF
(clone ES22-10D8, Mitenyl Biotech).
33. If washing out fixing reagent prior to analysis, fix cells on ice for
15–30 min and then wash by centrifuging (30 s at 850 g) and
resuspending the pellet in 200 μl FACS buffer and repeating
centrifugation. Repeat the wash step and resuspend cell pellet
in 200 μl FACS buffer.
34. Cell pellets can be analyzed by RT-PCR to determine abun-
dance of RNA molecules as a measure of gene expression (as in
[28]). Lavage fluid can be analyzed by ELISA to determine
levels of inflammatory mediators (such as chemokines CCL-2,

CCL-3, and CXCL1, cytokines TNF-α, IL-1β and IL-6, and
growth factors GM-CSF and VEGF, as in [12]).
References
1. Crowley CA, Curnutte JT, Rosin RE et al 11. Yaseen H, Butenko S, Polishuk-Zotkin I et al

(1980) An inherited abnormality of neutrophil (2020) Galectin-1 facilitates macrophage
adhesion: its genetic transmission and its asso- reprogramming and resolution of inflamma-
ciation with a missing protein. N Engl J Med tion through IFN-β. Front Pharmacol 11:
302:1163–1168. https://doi.org/10.1056/ 901. https://doi.org/10.3389/fphar.2020.
NEJM198005223022102 00901
2. Wagner DD, Frenette PS (2008) The vessel 12. Law HL, Wright RD, Iqbal AJ et al (2020) A
wall and its interactions. Blood 111: pro-resolving role for Galectin-1 in acute
5271–5281. https://doi.org/10.1182/ inflammation. Front Pharmacol 11:274.
blood-2008-01-078204 https://doi.org/10.3389/fphar.2020.00274
3. Badolato R (2013) Defects of leukocyte migra- 13. Norling LV, Perretti M (2013) Control of
tion in primary immunodeficiencies. Eur J myeloid cell trafficking in resolution. J Innate
Immunol 43:1436–1440. https://doi.org/ Immun 5:367–376. https://doi.org/10.
10.1002/eji.201243155 1159/000350612
4. Dana N, Todd RF, Pitt J et al (1984) Defi- 14. Nourshargh S, Alon R (2014) Leukocyte
ciency of a surface membrane glycoprotein migration into inflamed tissues. Immunity 41:
(Mo1) in man. J Clin Invest 73:153–159. 694–707
https://doi.org/10.1172/JCI111186 15. Gittens BR, Bodkin JV, Nourshargh S et al
5. Norling LV, Ly L, Dalli J (2017) Resolving (2017) Galectin-3: a positive regulator of leu-
inflammation by using nutrition therapy: roles kocyte recruitment in the inflamed microcircu-
for specialized proresolving mediators. Curr lation. J Immunol 198:4458–4469. https://
Opin Clin Nutr Metab Care 20:145–152. doi.org/10.4049/jimmunol.1600709
h t t p s : // d o i . o r g / 1 0 . 1 0 9 7 / M C O . 16. Norling LV, Dalli J, Flower RJ et al (2012)
0000000000000353 Resolvin D1 limits polymorphonuclear leuko-
6. Schwab JM, Chiang N, Arita M, Serhan CN cyte recruitment to inflammatory loci:
(2007) Resolvin E1 and protectin D1 activate receptor-dependent actions. Arterioscler
inflammation-resolution programmes. Nature Thromb Vasc Biol 32:1970–1978. https://
447:869–874. https://doi.org/10.1038/ doi.org/10.1161/ATVBAHA.112.249508
nature05877 17. Gilbert SG, Krautter F, Cooper D et al (2020)
7. Serhan CN, Chiang N, Dalli J, Levy BD (2015) CASTLE: cell adhesion with supervised train-
Lipid mediators in the resolution of inflamma- ing and learning environment. J Phys D Appl
tion. Cold Spring Harb Perspect Biol 7: Phys 53:16. https://doi.org/10.1088/1361-
a 0 1 6 3 1 1 . h t t p s : // d o i . o r g / 1 0 . 1 1 0 1 / 6463/ab9e35
cshperspect.a016311 18. Goldberg R, Scotta C, Cooper D et al (2019)
8. Iqbal AJ, Sampaio ALF, Maione F et al (2011) Correction of defective T-regulatory cells from
Endogenous galectin-1 and acute inflamma- patients with Crohn’s disease by ex vivo liga-
tion: emerging notion of a galectin-9 pro-re- tion of retinoic acid receptor-α. Gastroenterol-
solving effect. Am J Pathol 178:1201–1209. ogy 156:1775–1787. https://doi.org/10.
https://doi.org/10.1016/j.ajpath.2010. 1053/j.gastro.2019.01.025
11.073 19. Cash JL, White GE, Greaves DR (2009)
9. Chiang N, Schwab JM, Fredman G et al (2008) Chapter 17 Zymosan-induced peritonitis as a
Anesthetics impact the resolution of inflamma- simple experimental system for the study of
tion. PLoS One 3:e1879. https://doi.org/10. inflammation, 1st edn. Elsevier Inc.,
1371/journal.pone.0001879 Amsterdam
10. Starossom SC, Mascanfroni ID, Imitola J et al 20. Cooper D, Norling LV, Perretti M (2008)
(2012) Galectin-1 deactivates classically acti- Novel insights into the inhibitory effects of
vated microglia and protects from Galectin-1 on neutrophil recruitment under
inflammation-induced neurodegeneration. flow. J Leukoc Biol 83:1459–1466. https://
Immunity 37:249–263. https://doi.org/10. doi.org/10.1189/jlb.1207831
1016/j.immuni.2012.05.023
21. Norling LV, Sampaio ALF, Cooper D, Perretti 25. Schindelin J, Arganda-Carreras I, Frise E et al
M (2008) Inhibitory control of endothelial (2012) Fiji: an open-source platform for
galectin-1 on in vitro and in vivo lymphocyte biological-image analysis. Nat Methods 9:
trafficking. FASEB J 22:682–690. https://doi. 676–682
org/10.1096/fj.07-9268com 26. Cordelières F (2005) Manual tracking—
22. Wright RD, Souza PR, Flak MB et al (2017) ImageJ. In: imagej.nih.gov. https://imagej.
Galectin-3-null mice display defective neutro- nih.gov/ij/plugins/track/track.html.
phil clearance during acute inflammation. J Accessed 1 Oct 2020
Leukoc Biol 101:717–726. https://doi.org/ 27. Lawrence MB, Springer TA (1991) Leukocytes
10.1189/jlb.3A0116-026RR roll on a selectin at physiologic flow rates: dis-
23. Gil CD, Gullo CE, Oliani SM (2011) Effect of tinction from and prerequisite for adhesion
exogenous galectin-1 on leukocyte migration: through integrins. Cell 65:859–873. https://
modulation of cytokine levels and adhesion doi.org/10.1016/0092-8674(91)90393-D
molecules. Int J Clin Exp Pathol 4:74–84 28. Accarias S, Genthon C, Rengel D et al (2016)
24. Rueden CT, Schindelin J, Hiner MC et al Single-cell analysis reveals new subset markers
(2017) ImageJ2: ImageJ for the next genera- of murine peritoneal macrophages and high-
tion of scientific image data. BMC Bioinfor- lights macrophage dynamics upon Staphylo-
matics 18:529. https://doi.org/10.1186/ coccus aureus peritonitis. Innate Immun 22:
s12859-017-1934-z 3 8 2 – 3 9 2 . h t t p s : // d o i . o r g / 1 0 . 1 1 7 7 /
1753425916651330
Chapter 32
Examination of the Contributions of Maternal/Placental-

Derived Galectin-1 to Pregnancy Outcome
Sophia Borowski, Nancy Freitag, Iris Urban, Geert Michel,
Gabriela Barrientos, and Sandra M. Blois
Abstract
Galectin-1 (gal-1), a member of a family of evolutionarily conserved glycan-binding proteins, is differen-
tially expressed at the feto–maternal interface and appears to be functionally polyvalent, with a wide range of
biological activities. However, the contributions of maternal and/or feto-placental gal-1 to the signaling
networks promoting a healthy pregnancy are still being elucidated. This chapter discusses the methods
commonly employed to study the maternal or feto-placental contribution of gal-1 during pregnancy in
mice. The methods described here can be used to decipher the specific role of each source, e.g., maternal
and/or feto-placental derived gal-1 in the orchestration of pregnancy-associated processes.
Key words Galectin-1, Maternal and placental galectin-1 models, Pregnancy outcome, Lectin stain-
ing, In vitro fertilization
1 Introduction
Glycan binding proteins, specifically galectins, regulate a wide vari-

ety of cellular processes that ensure a healthy pregnancy. In partic-
ular, galectin-1 (gal-1) is a chief regulator at the feto–maternal
interface. Tracing back in developmental time, gal-1 expression
occurs as early as the blastocyst stage with the formation of the
trophectoderm [1], which ultimately gives rise to all trophoblast
cell types of the future placenta. After embryo attachment, as the
trophoblast layer differentiates, gal-1 localizes to villous cytotro-
phoblast and promotes syncytium formation; being also highly
expressed in the most invasive trophoblast cells of the placenta
where it regulates trophoblast invasion and the remodeling of
maternal spiral arteries [1–5]. Both the placental structures are
linked to embryo growth, and disruption of trophoblast invasion
and placental blood flow is a major cause of fetal growth restriction
[6]. However, gal-1 participates not only in placentation influen-
cing fetal well-being but also in the maternal adaptation to
603
604 Sophia Borowski et al.
pregnancy. In the maternal compartment, gal-1 is known to dictate

the establishment of immune tolerance to paternal alloantigens
[1, 7–9] and also regulate maternal vascular adaptations early in
gestation [2].
Despite the structural and functional differences between
mouse and human placentas, mouse models have been employed
for gaining insights into the molecular pathways that direct early
maternal/ trophoblast fate decisions as well as for identifying the
individual contributions of either the maternal or placental com-
partments to pregnancy outcome. In mice, gal-1 plays a critical role
driving proper placental development and function as demon-
strated in Lgals1 knock-out pregnancies, which are characterized
by placental insufficiency, fetal growth retardation, and develop-
ment of maternal preeclampsia-like features including de novo
hypertension and proteinuria [2]. However, such gal-1 deficiency
pregnancy model presents the limitation given by the wide range of
biological activities mediated by the lectin in either the maternal or
placental compartments, making it difficult to dissect the specific
contribution of each source of gal-1 expression to the maternal,
placental, and fetal components of the preeclampsia syndrome.
To overcome the inherent limitation imposed by the ubiqui-
tous expression of gal-1 at the murine feto–maternal interface, we
have developed a maternal and a feto-placental gal-1 deficiency
model using conventional in vitro fertilization techniques for
small animals (Fig. 1a). Thus, Lgals1+/+ embryos were transferred
to Lgals1/ dams to create the maternal gal-1 deficiency model
(mKO), whereas a feto-placental deficiency model (fplKO) was
accomplished by transferring Lgals1/ embryos to Lgals1+/+
recipient dams. Studies employing these models have indeed
demonstrated a differential contribution of maternal vs. feto-
placental expression of the lectin driving placental efficiency and
fetal growth, as shown by the significant reduction of fetal weights
in pregnancies carried in the mKO model compared to wild-type
mice (Fig. 1b). Fetal growth retardation in the mKO model was
also evident at embryonic day (E)13 as a differential distribution of
Theiler stages of development, with a greater frequency of mKO
fetuses displaying a TS 21 (with fingers still joined together, pinna
turned backwards covering the auditory meatus, Fig. 1c) as
opposed to wild-type and fplKO fetuses which mostly reached the
TS 22 appropriate for the gestational age. On the other hand, our
results also suggest that the main contribution to gal-1 circulating
levels throughout pregnancy comes from the maternal
compartment, as demonstrated by the different kinetics of gal-1
serum levels observed in each of the pregnancy models (Fig. 1d).
In this chapter, we describe the methods commonly employed
to study the relative contribution of maternal vs. feto-placental
sources of gal-1 during pregnancy in mice. Together with the
assisted reproduction technologies allowing the development of
Examination of the Contributions of Maternal/Placental-Derived Galectin-1. . . 605
Fig. 1 Assessment of the specific contribution of maternal vs. feto-placental sources of gal-1 expression to
pregnancy-associated processes (a) Overview of the in vitro fertilization and embryo transfer procedures for
production of the maternal (mKO) and feto-placental (fplKO) gal-1 deficiency pregnancy models. (b) Summary
of the results for the evaluation of placental (left panel) and fetal weights (right panel) on E13 in the different
gal-1 deficiency pregnancy models. ** denotes p < 0.01 as determined by ANOVA followed by Tukey post hoc
test. (c) Evaluation of fetal development in the different pregnancy models. Upper panel: Representative
the different gal-1 deficiency models, we provide the basis for

histochemical analyses of galectin–glycan circuits, which could
improve our understanding of the galectin-driven signaling net-
works operating in each compartment of the fetal–maternal inter-
face. The methods detailed herein can be applied to decipher the
specific role played by each source of galectin expression (i.e.,
maternal or feto-placental) in the orchestration of pregnancy-
associated processes.
2 Materials
2.1 Superovulation 1. Lgals1/ females (129/P3J background) and Lgals+/+

and In Vitro 129/P3J females.
Fertilization in Mice 2. Pregnant Mare’s Serum Gonadotropin (PMSG; Pregmagon,
Covetrus DE GmbH). Diluted with 1 PBS to 5 I.U. per
100 μl. Stored in Aliquots of 1000 μl at 20 C. Use after
thawing and do not store for more than 4 h.
3. Human Chorionic Gonadotropin (hCG; Ovogest, Intervet
Deutschland GmbH). Diluted with 1 PBS to 2.5 I.U. per
100 μl. Stored in Aliquots of 1000 μl at 20 C. Use after
thawing and do not store for more than 4 h.
4. Petri dishes (35 mm, 60 mm).
5. Micropipettes and tips.
6. Capillary tube (World Precision Instrument No. TW1204).
7. Mineral oil (Mineral Oil Light, Reproline No. 451202).
8. CARD MEDIUM (Cosmo Bio Co. Ltd.).
9. Card Fertiup Preincubation Medium: PM (Cosmo Bio
Co. Ltd.).
10. Incubator 37 C, 5% CO2.
11. Hot Plate 062 (Labotect No. 13854).
12. Stereo microscope.
Fig. 1 (continued) images of E13 fetuses collected from wild-type, mKO and fplKO dams. Arrows indicate the
anatomical features that distinguish Theiler stages 21 and 22: position of the pinna (black) and distal
separation of fingers (red, white; for details, refer to https://www.emouseatlas.org). Bottom panel: Distribution
of Theiler stages observed in the different pregnancy models. There seems to be a developmental delay in the
gal-1 deficiency models since an increased number of fetuses corresponding to lower Theiler stages (i.e.,
TS21, TS20) was observed as compared to wild-type pregnancies. (d) Left panel: Gal-1 levels measured in
maternal circulation by ELISA on E7. Right panel: Schematic representation of the serum gal-1 level kinetics
throughout pregnancy in the different gal-1 deficiency models. (e) Representative microscopy image showing
the immunofluorescence staining of gal-1 at the feto–maternal interface. Arrowheads: Giant cells. Scale bar:
100 μm. (f) Representative microscopy image showing the plant lectin staining for SNA-I and MAA at the
feto–maternal interface. Arrowheads: Giant cells. Scale bar: 100 μm. JZ Junctional zone, Lab Labyrinth
13. Surgical instruments (forceps, dissecting scissors, micro-spring

scissors):
(a) Spring scissors, e.g., Student VannasSpring Scissors (FST
No. 15003-08).
(b) Fine Iris Scissors, e.g., curved (FST No. 14061-09).
(c) Forceps, e.g., Dumont #5—Fine Forceps (FST
No. 11254-20).
(d) Forceps, e.g., Semken, straight serrated (FST No. 11008-
13).
(e) Dissecting scissors, e.g., Student Fine Scissors (FST
No. 91460-11).
(f) Bulldog Type Serrefines e.g. (FST No. 18051-35).
14. 26-gauge needle.
15. 70% Ethanol.
16. Eye and Nose Ointment (Panthenol).
17. Anesthesia working solution: Mix 250 μl Ketamine (100 mg/
ml), 100 μl Xylazine (20 mg/ml), with 1.7 ml isotonic NaCl.
Mix fresh and do not store.
18. Analgesia working solution: Mix 1 ml Rimadyl (50 mg/ml)
with 49 ml isotonic NaCl. Mix fresh and do not store.
2.2 Determination of 1. Surgical instruments (forceps, dissecting scissors).

Placental Weight, 2. Analytical balance.
Pregnancy Outcome,
3. Bouin’s solution (HT10132, Sigma-Aldrich).
Embryo Development
4. Falcon tube (15 ml).
5. Horizontal shaker.
6. 70% EtOH.
7. Stereo microscope.
2.3 Quantitative 1. Microcentrifuge tubes (0.5 ml, 1.5 ml).

Analyses of Circulating 2. Glass Pasteur pipette or capillary tube.
Gal-1 Levels in
3. Microcentrifuge.
Pregnant Mice
4. High binding Costar Assay Plate, 96 well half area (Corning).
5. Mouse Galectin-1 DouSet ELISA (R&D systems DY1245)
containing the reagent diluent, capture antibody, detection
antibody, and Streptavidin-HRP A.
6. PBS pH 7.4.
7. Wash buffer (0.05% Tween 20 in PBS).
8. Eight-channel (or 12-channel) manual pipet (100–300 μl).
9. TMB (3,30 ,5,50 -tetramethylbenzidine) solution. TMB sub-
strate: 240 mg TMB (3,30 ,5,50 -tetramethylbenzidine, Fluka
cat.#87,748) dissolved in 5 ml Absolut Ethanol (Roth

#9056.3) and in 5 ml DMSO 100% (Sigma Aldrich D2650).
Store at 4 C for max. 4 weeks. Gallati Buffer: 42.5 g of citric
acid (PanReac AppliChem #141018.1211) in 800 ml distilled
water adjust pH 3.95 with KOH. Add distilled water to a total
of 1 l and store at 4 C. TMB-Gallati Buffer: 10 ml Gallati
Buffer + 3.4 μl H2O2 30% + 100 μl TMB substrate).
10. Stop solution (4 N H2SO4).
11. ELISA reader (e.g., FLUOStar Optima).
12. Analytical balance.
2.4 Tissue 1. Blister (tablet packaging).

Preparation for Immu- 2. OCT Tissue Freezing Medium (Leica).
nohistochemistry
3. Dry ice.
(Cryo and Paraffin
Sectioning) 4. Ultra-low lab freezer.
5. Cryostat (e.g., CM3050 S from Leica).
6. Microscopic Slides SuperFrost Plus (Assistent).
7. Acetone in coplin jar with glass lid (at 20 C).
8. Fume hood.
9. Freezer.
10. Falcon tube (15 ml).
11. Formalin solution (5% in PBS).
12. Fridge.
13. Horizontal shaker.
14. PBS.
15. Ethanol (50%, 70%, 80%, 96%, 100%).
16. Xylene or substitute (e.g., Neo-Clear).
17. Paraffin.
18. Paraffin embedding station.
19. Paraffin embedding cassettes.
20. Aminosilane.
21. Incubator.
2.5 1. Liquid repellent PAP Pen.

Immunofluorescent 2. Steel staining rack (for 10 slides).
Determination of Gal-1
3. Big glass staining dish.
Expression at the
Feto–Maternal 4. Tris-buffered saline (TBS; 7.4 mM Tris–base, 43.5 mM Tris–
Interface
HCl, 150 mM NaCl, pH 7.5).
5. Orbital shaker.
6. Goat Normal Serum (GNS), 2% solution (Jackson ImmunoR-

esearch). Rehydrate freeze-dried GNS with the indicated vol-
ume of distilled water. Store 100 μl aliquots at 20 C. Freshly
prepare the 2% GNS working solution by dilution with TBS.
Do not store the working solution.
7. Humidity chamber (to prevent drying of tissue sections during
incubation times).
8. Primary anti-galectin-1 antibody, Rabbit anti-galectin-1 IgG
(GeneTex).
9. Alexa488-conjugated secondary antibody, AlexaFluor488
Goat Anti-Rabbit (Jackson ImmunoResearch).
10. 40 ,6-Diamidino-2-phenylindole (DAPI) for staining of nuclei
(Sigma-Aldrich). Storage solution: dilute to 2 mg/ml in water
and store in 100 μl aliquots at 20 C. Working solution:
dilute 100 μl storage solution in 100 ml methanol.
11. Mounting medium, Prolong Gold (Invitrogen).
12. Cover glasses (size depends on the size of the tissue sections
you want to cover).
13. Bright field and fluorescence slide scanner (Pannoramic MIDI
BF/FL, 3DHISTECH Ltd.).
2.6 Lectin Binding to 1. Liquid repellent PAP Pen.

Whole Implantation 2. Steel staining rack (for 10 slides).
Tissue
3. Big coplin jar.
4. Tris-buffered saline (TBS; 7.4 mM Tris–base, 43.5 mM Tris–
HCl, 150 mM NaCl, pH 7.5).
5. Orbital shaker.
6. Biotin Blocking system to suppress endogenous Biotin activity,
Biotin Blocking system (Dako).
7. Humidity chamber (to prevent drying of tissue sections during
incubation times).
8. Protein blocking solution (glycoprotein free) to block nonspe-
cific background staining, Carbo-Free Blocking Solution (Vec-
tor Laboratories).
9. Biotinylated lectins (EY Laboratories): biotin-labeled Sambu-
cus nigra agglutinin (SNA)-I, biotin-labeled Phaseolus vulgaris
lectin (PHA)-L, biotin-labeled Helix pomatia agglutinin
(HPA).
10. Tetramethylrhodamine (TRITC)-conjugated Streptavidin
(Invitrogen).
11. Fluorescein isothiocyanate (FITC)-labeled lectins
(EY Laboratories): FITC-labeled Maackia amurensis lectin
(MAA), FITC-labeled Arachis hypogaea lectin (PNA), Lycoper-

sicon esculentum lectin (LEA).
12. 40 ,6-Diamidino-2-phenylindole (DAPI) for staining of nuclei
(Sigma-Aldrich). Storage solution: dilute to 2 mg/ml in water
and store in 100 μl aliquots at 20 C. Working solution:
dilute 100 μl storage solution in 100 ml methanol.
13. Mounting medium, Prolong Gold (Invitrogen).
14. Cover glasses (size depends on the size of the tissue sections
you want to cover).
15. Bright field and fluorescence slide scanner (Pannoramic MIDI
BF/FL, 3DHISTECH Ltd.).
3 Methods
3.1 Superovulation 1. Superovulation of female mice (8–12 weeks old) with 5 iU

PMSG i.p. 48 h before hCG injection.
2. Inject 2.5 iU hCG i.p. 15 h before oocyte collection.
3.2 In Vitro 1. For the IVF, prepare one petri dish (35 mm) with 200 μl
Fertilization for the CARD MEDIUM and another petri dish (60 mm) with four
Production of Maternal drops of CARD MEDIUM (4 125 μl) covered with mineral
or Fetal-Placental Gal- oil for washing of the oocytes 4–6 h after starting the IVF
1-Deficient Mice (Fig. procedure. Pre-incubate the petri dishes with medium over-
1a) night at 37 C and 5% CO2. Make sure that all drops are
completely covered with mineral oil. Because of the 5% CO2
in the incubator, the oil is used as barrier for gas exchange to
preserve a physiological pH in the medium. Outside of the
incubator, the pH would change very fast without oil.
2. For the collection of spermatozoa, prepare a petri dish
(35 mm) with 90 μl drop of CARD FERTIUP Preincubation
Medium. Cover the drop completely with mineral oil and
preincubate 15–20 min in incubator (37 C, 5% CO2) (see
Note 1).
3. Sacrifice the male mice (3–6 months old), disinfect with 70%
EtOH and remove the cauda epididymidis. Avoid fat, blood,
and tissue fluid.
4. Place the cauda epididymidis on the mineral oil-covered PM
medium that was preincubated overnight at 37 C in the
presence of 5% CO2.
5. Make a cut in the cauda epididymidis with a 26-gauge needle
and introduce the released clots of spermatozoa into the drop
of PM medium.
6. Remove the remaining tissue from the mineral oil and allow the
sperm to capacitate for 40–60 min in incubator (37 C, 5%
CO2) (see Note 2).
7. Sacrifice the superovulated female mice by cervicale dislocation
for oocyte collection, disinfect with 70% EtOH, and dissect the
mouse to open the abdominal cavity.
8. Move the digestive tract to expose the uterine horns, oviducts,
and ovaries.
9. Grasp the uterus with the forceps and cut it close to the
oviduct. Cut above the ovary and transfer the tissue into oil
phase of the 35-mm dish with the oil-covered CARD
MEDIUM. Avoid fat, blood, and tissue fluid.
10. Locate the ampulla which is the swollen part of the oviduct
where the cumulus–oocyte complexes are visible. Tear open
the ampulla and release the cumulus–oocyte complexes and
pull them into the drop of CARD MEDIUM.
11. Remove the remaining tissue from the mineral oil (see Note 3).
12. For IVF, add 2–5 μl of the sperm suspension to the drop of
CARD MEDIUM containing the cumulus–oocyte complexes
(see Note 4).
13. Incubate for 4–6 h in an incubator (37 C, 5% CO2) (see
Note 5).
14. After 1 h, check if the cumulus–oocyte complexes got dis-
solved. If not, add another 10 μl of sperm suspension.
15. After an incubation of 4–6 h wash the cells (zygotes) two times
in separate drops of preincubated CARD MEDIUM. For
washing the embryos, pre-fill the pipette with medium from
the clean drop. The embryos are then drawn into the pipette
with the clean medium and released into the clean drop. This
procedure is repeated twice under the microscope to remove
aberrant embryos and cumulus cells (see Note 6).
16. Count cells. For more than 100 cells, split onto two drops.
17. Incubate overnight in an incubator (37 C, 5% CO2).
18. Count the two-cell stage embryos the next morning.
19. For anesthesia, pseudo-pregnant foster mice (mated with
vasectomized males) at gestation day 0.5 were weighed, and
100 μl of anesthetic is applied intraperitoneally per 10 g of
body weight (Ketamine: 122 mg/kg, Xylazine: 10 mg/kg).
Cover eyes with eye ointment.
20. Analgesia is administered (100 μl of analgesia per 10 g body
weight) under the skin in the neck when the anesthesia is
effective (Rimadyl: 10 mg/kg).
21. Disinfect with 70% EtOH and cut the skin above the spinal
column just below the rib cage. Grasp and lift the skin with
forceps and make a small incision. Insert the scissors and tear
open the skin for 5–8 mm by expanding the scissors.
22. Relocate the opening 5 mm to the left looking for the ovary
(orange) or fat pad (white), grasp and lift the body wall with
forceps and make a small incision with scissors and open further
as described above (5 mm).
23. Attach a serrefine clamp to the fat pad near the ovary and
relocate the distal part of the uterus, oviduct, and ovary onto
the back of the mouse and open the bursa over the infundibu-
lum with micro-spring scissors.
24. Aspirate air and medium in alternate intervals of 2–3 mm into
the glass capillary and later draw six embryos into the same.
25. Insert the tip of the capillary into the infundibulum and gently
push toward the ampulla (see Note 7).
26. Transfer six embryos and 2–3 air bubbles to each oviduct (see
Note 8). Repeat for the other infundibulum (relocate the
opening 5 mm to the right) from step 22.
27. Remove the capillary, release the fat pad, and relocate uterus,
oviduct and ovary into the body cavity and close the muscle
with a single knot suture and the fur with wound clips.
28. Let the mouse recover from the anesthesia on the hot plate (see
Note 9).
3.3 Determination of 1. Take blood under anesthesia before caesarean section by retro-
Placental Weight, bulbar vein plexus puncture with a glass Pasteur pipette or
Pregnancy Outcome, capillary tube.
Embryo Development 2. Transfer blood to 1.5-ml microcentrifuge tube and centrifuge
(Fig. 1b, c) and Tissue for 10 min at 6000 g.
Preparation for Cryo 3. Transfer the serum in aliquots of 0.5-ml microcentrifuge tubes
and Paraffin and store at 80 C for further use (e.g., ELISA to determine
Sectioning circulating gal-1 levels).
4. Euthanize the mouse by cervical dislocation.
5. Dissect out the uterine horns with caesarian section 13 days
after embryo transfer (E13) (see Note 10).
6. Place uterine horns in petri dish (see Note 11).
7. Assess pregnancy outcome by identifying resorbed implanta-
tion sites and calculate fetal loss as follows: fetal loss
(%) ¼ (resorbed implantations 100)/total number of
implantations.
8. Separate the single implantation sites (embryonic-placental
units), which can be used for different experiments (see Note
12):
(a) Embryonic (steps 5–7) and placental (step 8) analysis:

Remove amniotic sac, placenta, and umbilical cord from
the embryo (see Note 13).
(b) Cryo sectioning: Take whole implantation (steps 9–10).
(c) Paraffin sectioning: Take whole implantation (steps 11–
15).
9. Embryo analysis: Put embryo in Bouin’s solution for 24 h with
gentle shaking.
10. Put embryo in and change ethanol every day until it is clear (for
about 7 days).
11. Embryos can be stored in 70% EtOH, and developmental
examination under a stereo microscope can take place at any
time in a Petri dish according to Theiler stages (as described in
https://www.emouseatlas.org).
12. Placenta analysis: Separate the placenta from the decidua with
forceps and weigh the placenta on an analytical balance (see
Note 14).
13. Cryo sectioning: Put the whole implantation in a blister filled
with tissue freezing medium with the embryo on the left side
and the placenta on the right side (top view).
14. Freeze slowly on dry ice, remove the blister and store in a
plastic tube or bag at 80 C.
15. Paraffin sectioning: Put whole implantation in a tube with 5%
formalin solution and incubate for 5–7 days at 4 C.
16. Dehydrate in beakers as follows on horizontal shaker: 1 h in
PBS, 2 h in 50% ethanol, 2 h in 70% ethanol, 2 h in 80%
ethanol, overnight in 96% ethanol, 2 h in 100% ethanol, and
3 h in Neo-Clear.
17. Put implantations in paraffin at embedding station for 3 h at
60 C.
18. Embed in paraffin with the embryo on the left side and the
placenta on the right side (top view) and freeze over night at
20 C.
19. Store paraffin blocks at room temperature.
3.4 Quantitative 1. Pre-coat high binding capacity 96-well micro plates (Corning)
Analysis of Circulating with capture anti-gal-1 antibody at 800 ng/ml diluted in PBS
Gal-1 Levels in overnight at room temperature (50 μl/well).
Pregnant Mice (Fig. 1d) 2. Wash pre-coated microtiter plates two times with 100 μl of
wash buffer (0.05% Tween 20 in PBS) (see Note 15).
3. Block plate by adding 100 μl of reagent diluent and incubate
for 1 h at room temperature.
4. Wash two times with 100 μl of wash buffer (0.05% Tween

20 in PBS).
5. Add 50 μl/well of serum mouse sample (diluted 1/2) or
standards in Reagent diluent and incubate for 2 h at room
temperature. A 7-point standard curve using twofold serial
dilutions in Reagent Diluent is prepared from the Standard
stock solution: 70 ng/500 μl. The standard curve is between
125 and 8000 pg/ml.
6. Wash three times with 100 μl of wash buffer (0.05% Tween
20 in PBS).
7. Add 50 μl/well of the detection antibody (200 ng/ml) diluted
in reagent diluent and incubate for 2 h at room temperature.
20 in PBS).
9. Add 50 μl/well of the Streptavidin-HRP A (1/200) diluted in
reagent diluent and incubate for 20 min at room temperature
(see Note 16).
20 in PBS).
11. Add 50 μl/well of Substrate Solution (TMB) and incubate for
5–7 min at room temperature.
12. Add 25 μl/well of Stop Solution (4 N H2SO4).
13. Determine the optical density immediately using a microplate
reader set to 450 nm (see Note 17).
3.5 Cryo and Paraffin 1. Cryo sectioning: Sections are cut at 8 μm with a cryostat and
Sectioning for Immu- put on SuperFrost Plus glass slides.
nohistochemistry 2. After drying (about 10 min), fixation is accomplished in ace-
tone (at 20 C) in a glass dish for 10 min.
3. Let slides dry for at least 20 min under fume hood.
4. Store slides in freezer until immunohistochemistry.
5. Paraffin sectioning: Before paraffin sectioning, put paraffin
blocks in the freezer for 12–24 h.
6. Coat SuperFrost glass with aminosilane: Put SuperFrost glass
slides with a steel rack in a glass dish with 2% (3-Aminopropyl)
triethoxysilane in distilled H2O for 20 min and let dry over-
night under the fume hood.
7. Sections are cut at 4 μm with a microtome and put on
aminosilane-coated SuperFrost glass slides.
8. Let slides dry for 20 min and incubate over night at 37 C in an
incubator.
9. Store slides at room temperature until immunohistochemistry.
3.6 1. Draw a liquid repellent barrier around the tissue section.

Immunofluorescent 2. For washing, collect slides in a steel staining rack and subse-
Determination of Gal-1 quently put them in a big coplin jar filled with enough TBS to
Expression at the cover all slides (for 10 slides approximately 100 ml TBS). Wash
Feto–Maternal slides three times for 5 min on an orbital shaker and change the
Interface Using Tissue buffer in the staining dish after every 5 min.
Cryosections 3. Block nonspecific background staining by incubating 2% GNS
on the tissue section (use as much solution as needed to fully
cover the whole tissue) for 20 min at room temperature in a
humidity chamber.
4. Incubate slides with anti-galectin-1 antibody (dilution 1:250 in
TBS; prepare as much solution as needed to fully cover the
whole tissue) overnight at 4 C in a humidified chamber.
5. Wash slides in TBS.
6. Incubate slides with Alexa488-conjugated goat anti-rabbit sec-
ondary antibody (dilution 1:400 in TBS; prepare as much
solution as needed to fully cover the whole tissue) for 1 h at
room temperature in a humidity chamber.
8. Incubate slides with DAPI (use as much solution as needed to
fully cover the whole tissue) for 5 min and protected from light.
10. Seal slides with mounting medium. For mounting, put one
drop of mounting solution on the tissue section and subse-
quently spread the mounting solution by slightly pressing the
cover glass on the tissue section (try to avoid air bubbles).
11. Staining of whole implantation sites was digitally scanned by a
high-resolution bright field and fluorescence slide scanner
(Pannoramic MIDI BF/FL, 3DHISTECH Ltd.) using 20
magnification. Filter settings should be adjusted according to
the fluorescent dyes used (e.g., FITC and TRITC filter set-
tings). Excitation time depends on the quality of the staining
and on the type of tissue used for staining. The evaluation of
the staining was done on virtual slides using Pannoramic
Viewer 1.15.4 (3DHISTECH Ltd.).
12. Slides can be stored at 20 C.
3.7 Lectin Binding to 1. Draw a liquid repellent barrier around the tissue section (see
Whole Implantation Notes 18 and 19).
Tissue 2. For washing collect slides in a steel staining rack and subse-
quently put them in a big coplin jar filled with enough TBS to
cover all slides (for up to 10 slides approximately 100 ml TBS).
Wash slides three times for 5 min on an orbital shaker and
change the buffer in the coplin jar after every 5 min.
3. Block endogenous biotin activity by incubating biotin blocking

solution (use as much solution as needed to fully cover the
whole tissue) on the tissue section for 20 min at room temper-
ature in a humidity chamber.
4. Wash slides with TBS.
5. Block nonspecific background staining by incubating protein
blocking solution (use as much solution as needed to fully
cover the whole tissue) on the tissue section for 30 min at
room temperature in a humidity chamber.
7. Incubate slides with the corresponding biotin labeled lectin
(dilution 1:50 in protein blocking solution; prepare as much
solution as needed to fully cover the whole tissue) over night at
4 C in a humidified chamber.
9. Incubate slides with TRITC-conjugated streptavidin (dilution
1:500 in TBS; prepare as much solution as needed to fully cover
the whole tissue) for 1 h at room temperature in a humidity
chamber.
11. Incubate slides with the corresponding FITC-labeled lectin in
protein blocking solution for 2 h at room temperature in a
humidity chamber (lectin combinations (see Note 20): SNA-I
(biotinylated) and MAA (FITC-labeled), PHA-L (biotiny-
lated) and PNA (FITC-labeled), HPA (biotinylated) and LEA
(FITC-labeled)).
13. Incubate slides with DAPI (use as much solution as needed to
fully cover the whole tissue) for 5 min and protected from
light.
15. Seal slides with mounting medium. For mounting, put one
drop of mounting solution on the tissue section and subse-
quently spread the mounting solution by slightly pressing the
cover glass on the tissue section (try to avoid air bubbles).
16. Staining of whole implantation sites was digitally scanned by a

high-resolution bright field and fluorescence slide scanner
(Pannoramic MIDI BF/FL, 3DHISTECH Ltd.) using 20
magnification. Filter settings should be adjusted according to
the fluorescent dyes used (e.g., FITC and TRITC filter set-
tings). Excitation time depends on the quality of the staining
and on the type of tissue used for staining. The evaluation of
the staining was done on virtual slides using Pannoramic
Viewer 1.15.4 (3DHISTECH Ltd.).
17. Slides can be stored at 20 C.
4 Notes
1. We suggest using the sperm of two males at one time, which

will be sufficient for oocytes from about twelve superovulating
females. The developing two-cell stage embryos will be enough
for 24 pseudopregnant foster mice that receive about
14 embryos each. If the mouse strain is rather small, do not
transfer too many embryos. Not all transferred embryos will
develop to viable embryos.
2. The quality of spermatozoa is important. High-fertility sper-
matozoa display a high motility and homogeneity movement in
a vortex at the border of the incubation medium.
3. The collection of oocytes from superovulating mice should not
take longer than 5 min per dish.
4. If frozen sperm is used, add 10–30 μl to the oocytes.
5. After starting the IVF, avoid opening the incubator. Tempera-
ture and pH (CO2) have a strong impact on the
fertilization rate.
6. It is very important to remove the aberrant cells (having only
one or no pronucleus) at this point, because it will not be
possible to distinguish them at the two-cell stage the next day.
7. Adjust the position and direction of the oviduct, so that it is
aligned parallel to the capillary.
8. If the embryos and air bubbles cannot be expelled from the
capillary, push back a little and try again. Air bubbles should be
seen through the wall of the ampulla.
9. Not all foster mice will be pregnant with this technique. We
experienced 50–100% IVF success rates depending mainly on
the experience of the person who is doing the embryo transfer
and the housing conditions of the foster mice. Quiet housing
conditions for the fosters are crucial for the efficiency of
embryo transfers.
10. Cut at the cervix and oviducts with micro scissors. Carefully
remove any fat from the uterine horns.
11. Placing the petri dish on an ice pack might be useful depending
on the planned experiments with the decidual and placental
tissue.
12. Be careful when cutting single implantation sites, since—on
later gestation days—the embryo might pop out from the
uterus.
13. Remove the whole amniotic sac and umbilical cord from the
embryo by gently pushing with forceps or cutting with micro
scissors. After fixation, it is difficult to remove excessive tissue.
14. Placental tissue can be flash-frozen for further experiments.
15. Microtiter plate washing process should be done manually
using eight-channel (or 12-channel) pipet. Fill wells gently
with wash buffer using pipet and empty wells into the waste
by inverting plate. After the last wash, remove buffer carefully
tapping the plate against paper towel.
16. When adding the streptavidin-HRP, avoid placing the plate in
direct light.
17. If wavelength correction is available, set to 540 nm or 570 nm.
If wavelength correction is not available, subtract reading at
540 nm or 570 nm from readings at 450 nm. This subtraction
will correct for optical imperfections in the plate. Readings
made directly at 450 nm without correction may be higher
and less accurate.
18. In our experience, cryosections work better for IF staining
since remaining paraffin is strongly auto-fluorescent.
19. For paraffin-embedded sections, deparaffinization and antigen
retrieval is necessary before staining. Incubate paraffin-
embedded sections overnight at 60 C, and subsequently,
xylene is used for deparaffinization prior to rehydration using
a descending alcohol series. Furthermore, blocking of unspe-
cific background staining is achieved by using a 5% BSA, 1% fish
gelatin in PBS blocking solution.
20. Biotinylated lectins (first lectin incubation) are combined with
FITC-labeled lectins (second lectin incubation) for faster stain-
ing procedures and to save valuable tissue material. When
combining different take into account that the binding epi-
topes of the different lectins should not be too similar and as far
apart on the target glycans as possible. When combining dif-
ferent lectins, do a trial experiment before to check if they are
working in the desired combination.
Acknowledgments
Funding Source: Methods discussed in this chapter were part of the

research supported by the Deutsche Forschungsgemeinschaft
(BL1115/2-1, BL1115/4-1 and Heisenberg program BL1115/
7-1) to S.M.B. and Else Kröner-Fresenius-Stiftung (2017_A123)
to N.F.
References
1. Tirado-Gonzalez I, Freitag N, Barrientos G, 5. You JL, Wang W, Tang MY, Ye YH, Liu AX, Zhu
Shaikly V, Nagaeva O, Strand M, Kjellberg L, YM (2018) A potential role of galectin-1 in
Klapp BF, Mincheva-Nilsson L, Cohen M, Blois promoting mouse trophoblast stem cell differ-
SM (2013) Galectin-1 influences trophoblast entiation. Mol Cell Endocrinol 470:228–239.
immune evasion and emerges as a predictive https://doi.org/10.1016/j.mce.2017.11.003
factor for the outcome of pregnancy. Mol Hum 6. Brosens I, Pijnenborg R, Vercruysse L, Romero
Reprod 19(1):43–53. https://doi.org/10. R (2011) The “great obstetrical syndromes” are
1093/molehr/gas043 associated with disorders of deep placentation.
2. Freitag N, Tirado-Gonzalez I, Barrientos G, Am J Obstet Gynecol 204(3):193–201. https://
Herse F, Thijssen VL, Weedon-Fekjaer SM, doi.org/10.1016/j.ajog.2010.08.009
Schulz H, Wallukat G, Klapp BF, Nevers T, 7. Ramhorst RE, Giribaldi L, Fraccaroli L, Toscano
Sharma S, Staff AC, Dechend R, Blois SM MA, Stupirski JC, Romero MD, Durand ES,
(2013) Interfering with Gal-1-mediated angio- Rubinstein N, Blaschitz A, Sedlmayr P, Genti-
genesis contributes to the pathogenesis of pre- Raimondi S, Fainboim L, Rabinovich GA
eclampsia. Proc Natl Acad Sci U S A (2012) Galectin-1 confers immune privilege to
110(28):11451–11456. https://doi.org/10. human trophoblast: implications in recurrent
1073/pnas.1303707110 fetal loss. Glycobiology 22(10):1374–1386.
3. Kolundzic N, Bojic-Trbojevic Z, Kovacevic T, https://doi.org/10.1093/glycob/cws104
Stefanoska I, Kadoya T, Vicovac L (2011) 8. Blois SM, Ilarregui JM, Tometten M, Garcia M,
Galectin-1 is part of human trophoblast invasion Orsal AS, Cordo-Russo R, Toscano MA, Bianco
machinery--a functional study in vitro. PLoS GA, Kobelt P, Handjiski B, Tirado I, Markert
One 6(12):e28514. https://doi.org/10.1371/ UR, Klapp BF, Poirier F, Szekeres-Bartho J,
journal.pone.0028514 Rabinovich GA, Arck PC (2007) A pivotal role
4. Kolundzic N, Cujic D, Abu Rabi T, Bojic- for galectin-1 in fetomaternal tolerance. Nat
Trbojevic Z, Kadoya T, Vicovac L (2015) Galec- Med 13(12):1450–1457
tin signature of the choriocarcinoma JAr cells: 9. Kopcow HD, Rosetti F, Leung Y, Allan DS,
Galectin-1 as a modulator of invasiveness Kutok JL, Strominger JL (2008) T cell apoptosis
in vitro. Mol Reprod Dev 82(10):765–773. at the maternal-fetal interface in early human
https://doi.org/10.1002/mrd.22515 pregnancy, involvement of galectin-1. Proc
Natl Acad Sci U S A 105(47):18472–18477
Chapter 33
Method to Study the Role of Galectins in Angiogenesis In

Vivo Using the Chick Chorioallantoic Membrane Assay
Abstract
Angiogenesis is a complex multi-step process involving various activities of endothelial cells. These activities
are influenced in vivo by environmental conditions like interactions with other cell types and the microen-
vironment. Galectins play a role in several of these interactions and are therefore required for proper
execution of in vivo angiogenesis. This chapter describes a method to study galectins during physiologic
and pathophysiologic angiogenesis in vivo using the chicken chorioallantoic membrane (CAM) assay.
Key words Chorioallantoic membrane (CAM) assay, Chicken, Angiogenesis, Galectin, Tumor graft,
Blood vessel, Vasculature
1 Introduction
Angiogenesis is a complex multi-step process involving different

activities of endothelial cells. The function of endothelial cells can
be influenced by environmental conditions like changing flow
dynamics, interactions with other cell types, and interactions with
specific extracellular matrix components [1, 2]. Thus, while in vitro
assays can provide insights in the effects of molecules like galectins
and/or galectin inhibitors on endothelial cell function, further
assessment of their role in angiogenesis in vivo is important. A
commonly used assay to study angiogenesis in vivo is the chick
chorioallantoic membrane (CAM) assay. The CAM is a highly
vascularized extra-embryonic membrane that mediates exchange
of gas and nutrients during embryonic chick development. It is
formed between embryonic day of development (EDD) 3–10 of
the 21-day gestation period by fusion of the allantois mesodermal
layer—extending out of the embryo—with the mesodermal layer of
the chorion. Within the resultant double layer, a dense vascular
network develops up to EDD11 after which endothelial cell prolif-
eration drops rapidly allowing further maturation of the vascular
bed [3–5].
621
622 Kitty C. M. Castricum and Victor L. J. L. Thijssen
Apart from the rapid vascular development, there are numerous

advantages that warrant the use of the CAM assay for in vivo
angiogenesis studies. First, the assay is low in cost, reproducible,
reliable, and fairly simple to perform [3]. Furthermore, there are a
variety of methods for the application of testing compounds using
CAM, and several methods are available to monitor the subsequent
response in the vasculature. For example, we have used the CAM
assay to study the effects of galectin-1 and galectin-9 on angiogen-
esis [6–8]. The CAM assay can also be used to test the effects of
other treatment modalities like radiotherapy or photodynamic ther-
apy [9, 10]. Finally, the model does not require a sterile work
environment, and since the immune system of the chicken embryo
is not fully developed until EDD18, the CAM assay can also be
used for grafting xenograft cells and tissues. On the other hand,
nonspecific reactions can occur due to contamination with egg shell
itself or due to the use of reactive carrier vehicles [5]. In addition,
CAM development is sensitive to alterations in environmental con-
ditions like temperature, oxygen tension, and osmolarity. This indi-
cates that experiments using the CAM assay should be carefully
executed. In this chapter, we will describe a method for the topical
application of soluble compounds (galectins and/or galectin inhi-
bitors) on the CAM in order to study their effects on angiogenesis
in vivo. In addition, we describe how galectins, galectin-inhibitors,
or other compounds can be injected intravenously in the CAM
vasculature, and we provide a method to graft tumor cells onto
the CAM which can provide information on the role of galectins in
tumor growth and tumor angiogenesis.
2 Materials
We use fertilized eggs of white leghorn chicken that are purchased

from a local commercial supplier. The eggs can be stored for several
days at 4 C/39.2 F but after more than 1 week, the quality and
viability of the eggs decrease, affecting the quality of the data.
Please be aware that depending on the legislation of your country
regarding animal use in research experiments, a license might be
needed to perform the described experiments.
2.1 Incubation of the 1. Milli-Q water (or equivalent).

Chicken Eggs 2. Sterilized fine point tweezers (e.g., standard 15-cm college
tweezers).
3. Scotch “Magic” adhesive tape (see Note 1).
1. Non-latex elastic rings (see Note 2).

2. Saline (from any commercial vendor).
Method to Study the Role of Galectins in Angiogenesis In Vivo Using the. . . 623
2.2 Application of 3. Galectin of interest and/or galectin inhibitor, either purchased

Galectins/Galectin or home-made.
Inhibitors onto 4. 10–100 μL pipette with sterile filter tips.
the CAM
2.3 Data Acquisition 1. Fridge or cold room.

and Analysis 2. 20-mL syringe.
3. 21 Gauge injection needle.
4. Contrast solution (4 g of zinc oxide (Sigma-Aldrich) in 50 mL
pure vegetable oil (from your local supermarket)).
2.4 Data Acquisition 1. Image analysis software (HetCAM, DCIlabs) or Adobe Photo-
and Analysis shop/MS Office.
2.5 Grafting Tumor 1. Approximately 5 106 cells (see Note 3).

Cells onto the CAM 2. Matrigel (see Note 4).
3. Soft paper tissues.
4. Ice.
5. 100-μL pipette with sterile filter tips.
6. Ruler with mm scaling.
7. Small surgical scissors.
8. Balance.
9. Phosphate-buffered saline (PBS).
10. Fixative, e.g., 4% paraformaldehyde in PBS or zinc fixative.
2.6 Intravenous 1. 33 Gauge point 4 Hamilton injection needle.

Injection of Labeling 2. 100-μL Hamilton syringe.
Agents or Cancer Cells
3. Saline.
2.7 Special 1. Fan-assisted humidified (egg) incubator, 37.8 C/100.04 F

Equipment (see Note 5). We use a FIEM MG140/200 Rural which allows
us to switch between tilting racks and non-tilting racks (see
Note 6). The incubator should be well humidified throughout
the whole experiment to prevent dehydration of the CAM. We
achieve this by putting water basins on the floor of the incuba-
tor (see Note 7).
2. Fiber optic illuminator. We use a Schott KL 1500 Electronic
Light Source (see Note 8).
3. Stereo microscope equipped with a camera (e.g., Leica M125
stereomicroscope with 12.5:1 zoom which is equipped with a
Leica KL1500 LED ring illumination system and a Leica
5 Megapixel DFC425 CCD camera).
3 Methods
A schematic drawing of the chicken gestation period is shown in

Fig. 1a. A typical CAM assay takes approximately 10 days and a
CAM tumor graft experiment takes 17 days. Thus, depending on
the type and frequency of treatment, careful planning of the experi-
ments is important. The time schedules that are used in our lab for a
normal CAM experiment as well as for a tumor graft CAM experi-
ment are shown in Fig. 1b.
3.1 Incubation of the 1. Transfer the eggs from the cold storage to room temperature
Chicken Eggs for at least 12 h prior to the incubation (see Note 9).
2. Clean the shell of each egg with a paper wipe or tissue soaked
with Milli-Q water.
3. Place the eggs horizontally on a 90 tilting rack (provided with
the egg incubator), which rotates minimally six times per 24 h.
Place the rack in a pre-warmed and humidified fan-assisted egg
incubator at 37.8 C/100.04 F (see Notes 5–7). The starting
day of the incubation is regarded as EDD0.
4. On EDD3, put the eggs in an upright position and make a
small hole in the narrow end of the shell with fine tip tweezers.
This will translocate the air compartment in the egg to the top
of the egg. Seal the hole with adhesive tape using as little tape as
possible (see Note 10). Stop the rotation of the racks and place
the eggs back in the incubator, with the sealed hole at the top.
5. On EDD6, check the eggs for fertilization. Point the fiber
optic light source (see Special equipment and Note 8) toward
one side of the egg. Vasculature should become visible at the
opposite side of the egg. If not, the egg is not fertilized and can
be discarded.
6. Create a window of 1 cm3 in the top of the shell with fine tip
tweezers (see Note 11). The CAM vasculature can now be
observed through the window.
7. Proceed with direct application of galectin or galectin of inter-
est onto the CAM (Subheading 3.2) or with grafting of tumor
cells onto the CAM (Subheading 3.5).
8. Intravenous injection into the CAM vessels is possible from
EDD10 (Subheading 3.6).
3.2 Application of 1. On EDD6, carefully place a sterilized non-latex plastic dental

Galectins/Galectin ring through the window on top and in the center of the CAM.
Inhibitors onto Seal the window with adhesive tape (see Note 10) and place the
the CAM egg back in the incubator for at least 2 h. This allows the ring to
settle down on the CAM (see Note 12).
Fig. 1 Time schedule for the CAM assay. (a) Schematic representation of the CAM assay schedule during
embryonic chicken development from embryonic day of development (EDD) 0 until EDD20, i.e., the day before
hatching. The EDD during which extensive angiogenesis takes place in the CAM are shown in bold. The images
show the CAM vasculature at different EDD. The number below the egg indicates the weight of the embryo in
milligrams (mg) or grams (g). (b) Scheme of a standard CAM assay (upper panels) and of the tumor graft assay
on the CAM (lower panel)
2. Prepare the treatment solution by diluting the galectin of

interest with or without the specific inhibitor in saline. Use a
concentration range to identify the optimal concentration. For
most galectins, this will likely be in the mid-micromolar range.
The total amount of solution depends on the number of eggs,
the diameter of the ring, and the duration/frequency of the
treatment (see Note 13).
3. Take an egg out of the incubator. Open the sealed window and
check if the embryo is still alive (you should see the heart
beating). Apply 50–80 μL of galectin/galectin inhibitors at
the desired concentration (usually in the mid-micromolar
range) within the ring without touching the CAM itself. Reseal
the window and place the eggs back in the incubator (see
Note 10).
4. Repeat the addition of galectin/galectin inhibitors depending
on your required treatment schedule. Usually, treatment is
performed on a daily basis until EDD9.
3.3 Data Acquisition 1. On EDD10, place the eggs at 4 C/39.2 F for 30 min to
induce hypothermia (see Note 14).
2. Prepare contrast solution by mixing 4 g of zinc oxide with
50 mL of pure vegetable oil in a 50-mL tube. Shake vigorously
and leave it on a roller platform for 20 min.
3. Fill a 20-mL syringe with the zinc oxide/oil mixture. Make
sure to remove any air bubbles.
4. Open the shell of a hypothermic egg as far as possible without
disrupting the CAM.
5. Carefully inject 1 mL of the contrast solution directly under
the CAM where the ring is located.
6. Use the microscope with camera to acquire images of CAM
vasculature within the ring area (see Note 15).
7. If necessary, the treated CAM area can be collected for further
analysis, e.g., gene expression, immunohistochemistry. Follow-
ing acquisition of images, isolate the CAM area under the ring
using fine tweezers and small surgical scissors. Wash the freshly
isolated CAM in PBS and transfer it to the desired fixation
buffer or liquid nitrogen. Further processing of the tissue is
not described in this chapter.
8. Finally, euthanize the chicken embryo by transferring the egg
to 20 C/ 4 F for 24 h.
3.4 Data Analysis Several methods have been published to analyze the CAM images
[5, 11–13]. Nowadays, software-based image analysis is often used
for rapid, objective, and extensive image analysis. The software uses
specific algorithms to recognize and skeletonize the vascular bed
from which different vascular parameters can be extracted like vessel
length, vessel branchpoints, vessel endpoints, total vessel area, etc.

(Fig. 2a). We use HetCAM software (DCIlabs, Belgium), but other
software packages might be used as well. However, we are aware
that such software is expensive and not always available. Therefore,
we here describe a widely accepted morphometric method to ana-
lyze images of the CAM, using software that is available in most
research labs, e.g., Adobe Photoshop or ImageJ (Fig. 2b) (see
Note 16).
1. Open the desired CAM image in a graphics editing program
like Adobe Photoshop (or any comparable software package).
2. Set the image to grayscale to enhance the contrast between the
vessels and the background. If necessary, use the image
Fig. 2 Analysis of the CAM vasculature. (a) CAM analysis by skeletonization-based method. The HetCAM
software (DCI labs) automatically analyses the skeleton length, the vessel area, the number of endpoints, and
the number of branchpoints in each CAM picture. This method provides a highly objective and precise analysis
of the vascular bed, which is quick and allows for high-throughput analysis. (b) Morphometric CAM analysis. In
this method, five concentric rings are projected over the CAM image and the cross-sections of the vessels with
the rings are counted. This will give insight in the vessel density of the CAM
autocontrast function to improve contrast. Note that this

should be performed for all images within a single experiment
and that this should not be used to obscure or remove any
unwanted data.
3. Place the CAM image in a graphic design program like Adobe
Illustrator or a presentation program like PowerPoint.
4. Project 5 concentric rings over the CAM image and count the
cross-sections between the vessels and the rings. The sum of
these counts is a morphometric measurement for vessels den-
sity in the CAM.
3.5 Grafting Tumor While the method described above provides information on the
Cells onto the CAM direct effects of galectin/galectin inhibitors on angiogenesis, much
galectin research is performed in the context of tumor biology.
Consequently, it is important to determine how galectin expression
in tumor cells or treatment with galectin-targeting compounds
affects tumor growth and tumor angiogenesis. This can be readily
studied using the CAM assay since it is possible to graft (human)
tumor cells onto the CAM, most of which will rapidly grow into
well vascularized tumors.
1. On EDD6, harvest the tumor cells. Count the cells and aliquot
them in separate 2-mL tubes, each tube containing
5 106 cells and spin them down for 5 min at 400 rpm.
Discard the medium.
2. On ice, mix 5 106 cells with 50 μL Matrigel (see Note 3).
3. Carefully “damage” a small area of the CAM by touching it
with a soft tissue causing a small bleeding (see Note 17).
4. Transfer the Matrigel/cell mix onto the damaged area.
5. Close window and place egg back into incubator.
6. Check growth of tumor daily and measure size and width using
a ruler (see Note 18).
7. If necessary, start treatment on EDD10 by applying the galec-
tin/galectin inhibitor of interest topically onto the tumor or by
direct injection into the tumor tissue or injection in the CAM
vasculature (Subheading 3.6). The appropriate concentration
should be determined for each specific galectin or inhibitor by
using a concentration range, e.g., from high nM to high μM
range.
8. Measure size on a daily basis until EDD14 or maximally until
EDD17 (see Note 18).
9. At the end of the experiment, harvest, photograph, and weigh
the tumor and subsequently place it in the appropriate fixative
for further processing (Fig. 3).
10. Discard the eggs as described in Subheading 3.3.
Fig. 3 Tumor Grafts on the CAM. (a) Image of a HT29 tumor on the CAM at EDD14. The tumor cells were
grafted on EDD6. (b) HT29 tumor after resection. The tumor is well vascularized, indicating adequate tumor
angiogenesis. (c) Image of hematoxylin/eosin staining on a standard paraformaldehyde-fixed and paraffin-
embedded HT29 tumor graft. Both the CAM and nest of tumor cells (TC) surrounded by tumor stroma (S) are
clearly visible. (d) Image of vessel staining (CD31, brown) on a paraformaldehyde-fixed and paraffin-
embedded HT29 tumor graft
3.6 Intravenous If necessary, galectins, galectin inhibitors, or other compounds of

Injection of Cancer interest can also be injected intravenously. Please note that this
Cells or Labeling takes quite some practice and include some additional eggs to
Agents compensate for loss of egg due to bleedings. Of note, it is also
possible to inject cells to analyze cell migration and metastatic
growth [3], but this will not be described here. Again, optimal
concentrations should be determined empirically.
1. On EDD6, open the eggs and close window and place eggs
back in the incubator (see Note 19).
2. Injection is possible from EDD10. On earlier time points, the
vessels are too small and bleedings will more frequently occur
reducing the efficacy of the experiment.
Fig. 4 Intravenous injection in the CAM vasculature. The left image shows an overview of the CAM vasculature
to get a sense of the size of vessels that are appropriate to inject. The arrowhead shows the point of entry of
the 33-gauge needle. The middle image shows an enlargement of the needle entry point and the right image
shows a volume of liquid entering the blood vessel (highlighted by dotted line)
3. Under the stereomicroscope, carefully microinject 50 μL intra-

venous using a 33-Gauge Hamilton needle and 100-μL Hamil-
ton syringe (see Note 20 and Fig. 4).
4. Ensure that any air bubbles are removed prior to injection.
5. During injections, it is recommended to gently pulse the
plunger. This will result in more uniform distribution of agents.
Expect an average injection time of 2–5 min per embryo.
6. After injection, slowly pull needle out of the vessel. Clean up
blood by light tapping with a soft tissue.
7. Proceed with the experiment as described in Subheading 3.2 or
Subheading 3.5.
4 Notes
1. Other brands of tape can be used, but we have good experience

with the Scotch Magic tape because it is not too sticky which
makes it easy to repeatedly open and close the window in the
egg shell.
2. Non-latex rings are commercially available, or they can be
custom-made. It is important that the weight of the ring is as
low as possible to avoid nonspecific responses in the vasculature
due to occlusion of vessels by the ring. We have found that
orthodontic dental elastic bands (non-latex, Ø 9.5 mm) are a
good and cheap alternative. Varying the diameter of the rings
allows larger area’s to be treated but also requires more com-
pound. In addition, with increasing diameter the weight of the
ring also increases. A diameter of Ø 9.5 mm typically allows
application of 50–80 μL solution.
3. The number of cells required for adequate grafting depends on

the specific cell line and should be tested. However, we have
found that increasing the cell number increases the success of
grafting and most cell lines tested in our lab (colon, kidney)
show successful grafting when 5 million cells are used. This is
just a matter of trial and error. We usually start by testing
0.5 106, 1 106, 2 106 and 5 106 in five eggs per
group to determine graft rate.
4. We have successfully used both normal and growth
factor-reduced Matrigel. The latter is preferred if the angiosti-
mulatory effect of cells is tested since normal Matrigel already
contains more stimulatory factors by itself.
5. Temperature setting depends on the type of incubator. For a
still-air incubator (no fan): 38.5 C (101.3 F) measured at the
top of the eggs. For a fan assisted incubator: 37.8 C
(100.04 F) measured anywhere in the incubator.
6. If the incubator has no tilting racks, manual rotation of the
eggs is also possible. Turn the eggs through 180 degrees by
hand at least twice a day.
7. Humidity should be maintained at 60% during incubation.
Excessive humidity could result in an increased rate of infec-
tions in the eggs.
8. The fiber optic illuminator is used to check successful fertiliza-
tion on EDD6. Vessel-like tube structures (orange/reddish)
should be visible on the inner side of the egg shell. However, if
no such device is available, successful fertilization can also be
readily checked following opening of the egg shell.
9. We use anywhere between 8 and 10 eggs per treatment condi-
tion. For experienced users, a total number of about 80 eggs
per experiment is manageable which thus allows for 8–10
different groups per experiment.
10. The adhesive tape prevents the CAM from dehydration. How-
ever, the tape also prevents gas exchange through the egg shell
and should therefore be kept to a minimum.
11. The egg shell should be removed carefully since debris of the
shell can induce a response in the CAM.
12. Instead of a ring to define the therapeutic area, some people
prefer to use for example gelatin or methylcellulose discs
[14, 15]. Also, other absorbent materials might be used as
long as these do not induce a response in the CAM.
13. In general, 50–80 μL of compound is applied daily from EDD6
until EDD10. The solution can be prepared fresh daily or
stored at 4 C/39.2 F. Note that long-term storage at
4 C/39.2 F of galectins in solution can affect protein stability
and activity. Furthermore, solutions stored at 4 C/39.2 F
should be allowed to return to room temperature before apply-

ing them to the CAM.
14. Movements of the embryo are very likely to disturb you while
taking pictures of the CAM. The induced hypothermia will
result in less movements, by slowing down the metabolism of
the embryo. However, if the aim is to measure blood flow,
hypothermia should not be applied.
15. The magnification will determine the level of detail that can be
analyzed. At 25-fold magnification (2.5 10) mainly the
larger, mature vessels will be visible while at 100-fold magnifi-
cation (10 10) detailed images of the capillary bed can be
obtained. We usually acquire images at both magnifications in
order to distinguish between both vessel types.
16. This quantification method is laborious, less accurate and more
sensitive to subjective errors. In addition, it will only provide
information about the vascular density. Nevertheless, it is a
cheap method and available to everyone.
17. Preferentially damage the small vessels above the embryo. The
larger vessels less easy to damage and bleedings will more
frequently occur reducing the viability of the embryo.
18. Be aware that not all tumors will grow on top of the CAM. We
have observed that some tumors will grow just underneath
the CAM.
19. We observe that more eggshell debris drops onto the CAM
when embryos are opened at a later time point.
20. We use a 33-gauge Hamilton needle and 100-μL Hamilton
syringe. It is also possible to use a pulled sodium borosilicate
needle and 1-mL disposable syringes. Injection is most easy in
one of the larger vessels or at a bifurcation of a larger vessels.
Also try to inject in the direction of the blood flow.
References
1. Carmeliet P, Jain RK (2011) Molecular 4. Ribatti D (2008) Chick embryo chorioallan-
mechanisms and clinical applications of angio- toic membrane as a useful tool to study angio-
genesis. Nature 473:298–307 genesis. Int Rev Cell Mol Biol 270:181–224
2. Griffioen AW, Molema G (2000) Angiogene- 5. West DC, Thompson WD, Sells PG, Burbridge
sis: potentials for pharmacologic intervention MF (2001) Angiogenesis assays using chick
in the treatment of cancer, cardiovascular dis- chorioallantoic membrane. Methods Mol
eases, and chronic inflammation. Pharmacol Med 46:107–129
Rev 52:237–268 6. Aanhane E, Schulkens IA, Heusschen R et al
3. Ribatti D, Nico B, Vacca A, Roncali L, Burri (2018) Different angioregulatory activity of
PH, Djonov V (2001) Chorioallantoic mem- monovalent galectin-9 isoforms. Angiogenesis
brane capillary bed: a useful target for studying 21:545–555
angiogenesis and anti-angiogenesis in vivo. 7. Thijssen VL, Postel R, Brandwijk RJ et al
Anat Rec 264:317–324 (2006) Galectin-1 is essential in tumor angio-
genesis and is a target for antiangiogenesis
therapy. Proc Natl Acad Sci U S A 103: 12. Nowak-Sliwinska P, Ballini J-P, Wagnières G,
15975–15980 van den Bergh H (2010) Processing of fluores-
8. Thijssen VL, Barkan B, Shoji H et al (2010) cence angiograms for the quantification of vas-
Tumor cells secrete galectin-1 to enhance cular effects induced by anti-angiogenic agents
endothelial cell activity. Cancer Res 70: in the CAM model. Microvasc Res 79:21–28
6216–6224 13. Rizzo V, DeFouw DO (1996) Mast cell activa-
9. Nowak-Sliwinska P, van Beijnum JR, van tion accelerates the normal rate of angiogenesis
Berkel M, van den Bergh H, Griffioen AW in the chick chorioallantoic membrane. Micro-
(2011) Vascular regrowth following photody- vasc Res 52:245–257
namic therapy in the chicken embryo chorioal- 14. Ribatti D, Urbinati C, Nico B, Rusnati M,
lantoic membrane. Angiogenesis 13:281–292 Roncali L, Presta M (1995) Endogenous basic
10. van Beijnum JR, Thijssen VL, L€appchen T et al fibroblast growth factor is implicated in the
(2016) A key role for galectin-1 in sprouting vascularization of the chick embryo chorioal-
angiogenesis revealed by novel rationally lantoic membrane. Dev Biol 170:39–49
designed antibodies. Int J Cancer 15. Ribatti D, Nico B, Vacca A, Presta M (2006)
139(4):824–835 The gelatin sponge-chorioallantoic membrane
11. Irvine SM, Cayzer J, Todd EM et al (2011) assay. Nat Protoc 1:85–91
Quantification of in vitro and in vivo angiogen-
esis stimulated by ovine forestomach matrix
biomaterial. Biomaterials 32:6351–6361
Chapter 34
Untangling Galectin-Mediated Circuits that Control

Hypoxia-Driven Angiogenesis
Nadia Bannoud, P. Alfredo Garcı́a, Julian Gambarte-Tudela,
Victoria Sundblad, Alejandro J. Cagnoni, Camila A. Bach,
Juan M. Pérez Saez, Ada G. Blidner, Sebastián M. Maller,
Karina V. Mariño, Mariana Salatino, Juan P. Cerliani,
Gabriel A. Rabinovich, and Diego O. Croci
Abstract
Development of an aberrant vascular network is a hallmark of the multistep pathological process of tumor
growth and metastasis. In response to hypoxia, several pro-angiogenic factors are synthesized to support
vascularization programs required for cancer progression. Emerging data indicate the involvement of
glycans and glycan-binding proteins as critical regulators of vascular circuits in health and disease. Galectins
may be regulated by hypoxic conditions and control angiogenesis in different physiopathological settings.
These β-galactoside-binding proteins may promote sprouting angiogenesis by interacting with different
glycosylated receptors and triggering distinct signaling pathways. Understanding the role of galectins in
tumor neovascularization will contribute to the design of novel anti-angiogenic therapies aimed at com-
plementing current anti-cancer modalities and overcoming resistance to these treatments. Here we describe
selected strategies and methods used to study the role of hypoxia-regulated galectins in the regulation of
blood vessel formation.
Key words Galectins, Hypoxia, Angiogenesis
1 Introduction
Angiogenesis is a physiologic mechanism that leads to the forma-

tion of new blood vessels from preexisting ones. A complex net-
work of stimulatory and inhibitory soluble factors converges in this
process, together with the coordinated action of several adhesion
molecules and cell surface receptors [1, 2].
The mechanisms involved in both pathological (i.e., tumor-
associated) and physiological (i.e., wound healing) angiogenesis
are similar and entail vessel destabilization, guided endothelial cell
(EC) migration, proliferation, and sprouting of newly formed
635
636 Nadia Bannoud et al.
vessels [3]. However, the stabilization phase, in which EC changes

their metabolism and become quiescent [2], is absent in pathologic
angiogenesis. This dysregulated process results in a highly dense
and aberrant vascular network, which generates a hostile microen-
vironment characterized by low pH, hypoxia, and interstitial
leakage [2].
As a hallmark of cancer and several ischemic diseases [1],
angiogenesis represents an essential target for the development of
novel therapies [4]. With the overarching goal of dissecting the
molecular pathways underlying vascularization, in vitro experi-
ments represent a first approach to identify different extracellular
or intracellular signals governing this process. At least three critical
steps of the angiogenic process should be examined in vitro to
evaluate responses to different extracellular or intracellular stimuli:
endothelial cell proliferation, migration, and vessel sprouting or
tube formation [5].
Galectins (Gals) are highly evolutionarily conserved soluble
lectins that act as decoders of glycan-containing information in a
myriad of glycosylated receptors [6]. Some galectins are expressed
in a wide range of cell types (i.e., Gal1 and Gal3), whereas others,
including Gal7 and Gal12, show a more restricted tissue localiza-
tion [7]. Galectins are defined as a protein family based on con-
served β-galactoside-binding activity located within their
characteristic carbohydrate recognition domains (CRDs). In addi-
tion to their well-defined roles in regulating immune cell homeo-
stasis in several physiological and pathological settings, galectins
have emerged as critical mediators that link tumor vascularization
and immunosuppression [8].
Distinct galectin family members, including Gal-1, -3, -8, -9,
and -12, can influence vascular signaling programs by cross-linking
EC surface glycoproteins and activating different signaling path-
ways [5]. In this regard, Gal1 overexpression in tumor-associated
ECs is associated with the promotion of an angiogenic phenotype
[9]. Besides, ECs can also take up Gal1, which stimulates their
proliferation and migration [10]. Interestingly, hypoxia-driven
Gal1 expression, regulated by either nuclear factor kappa B
(NF-κB)-dependent or hypoxia inducible factor 1 subunit alpha
(HIF-1α)-dependent pathways, promotes angiogenesis in different
tumor models [9, 11–14]. Through interactions with specific gly-
cans on neuropilin-1 (NRP-1) and vascular endothelial growth
factor (VEGF)-receptor-2 (VEGFR2), Gal1 modulates receptor
segregation, internalization, and trafficking, leading to VEGFR2
phosphorylation and signaling [13–15]. Meanwhile, Gal3 binds to
N-glycans on αvβ3 integrin and modulates VEGFR2 retention at
cell surface [16, 17], whereas Gal8 triggers angiogenesis through
binding to the activated leukocyte cell adhesion molecule
(ALCAM; CD166) [18]. Interestingly, Gal8 also contributes to
Untangling Galectin-Mediated Circuits that Control Hypoxia-Driven Angiogenesis 637
pathological lymphatic angiogenesis through binding to VEGF-C,

podoplanin, and integrins α1β1 and α5β1 [19]. Through modula-
tion of EC migration and chemotaxis, Gal9 promotes angiogenic
responses in a murine model of inflammatory arthritis [20], and a
Gal9 splice variant isoform (Gal9Δ5) induces a dose- and context-
dependent effect on EC morphogenesis [21]. Recently, hypoxia-
regulated Gal12 has been proposed to control neovascularization in
adipose tissue through binding to 30 -fucosylated ligands
[22]. Moreover, Gal2, Gal4, and Gal8 showed an indirect role in
modulating angiogenic programs by inducing the secretion of
EC-derived cytokines and chemokines (G-CSF, IL-6, MCP-1,
and GROα) which in turn stimulate EC signaling [23]. This angio-
genic function converges with other biological activities displayed
by these multifunctional endogenous lectins including immuno-
suppression, chemotaxis, and cell adhesion, particularly in tumor
microenvironments [24].
Thus, hypoxia emerges as a significant driving force that influ-
ences galectin expression, leading to the regulation of angiogenesis
in different pathophysiologic settings [25]. Given its central role in
angiogenesis, immunosuppression, and resistance to therapies [26],
hypoxia became a critical target for next-generation cancer
therapies.
In this chapter, we describe a selection of methods commonly
used to study the role of galectins in the modulation of angiogene-
sis and their regulated expression by hypoxic microenvironments.
Understanding the molecular and cellular mechanisms underlying
the pro-angiogenic function of galectins will contribute to delin-
eate novel therapeutic strategies in a wide range of diseases. We
hope that the strategies and methods described here will facilitate
and encourage scientists to further evaluate the role of galectins and
glycans in blood vessel formation in different pathological settings.
2 Materials
2.1 Hypoxia 1. Modular Incubator Chamber (MIC-101; Billups-

Induction Rothenberg).
2. Gas Mixture (1% O2, 5% CO2, 94% N2).
3. P100 Petri Dish (Greiner BioOne).
4. DMEM medium supplemented with 0.1% heat-inactivated
(fetal bovine serum) FBS (Gibco-Thermo Fisher), 100 μg/ml
streptomycin, and 100 U/ml penicillin (Gibco-Thermo
Fisher).
5. Hose closing clamps 1/200 .
6. Cell lines: B16-F0 (ATCC CRL-6322); LLC1 (ATCC
CRL-1642); HCT116 (ATCC CLL247); R1.1 (ATCC
TIB-42); HT-29 (ATCC HTB-38); C3H Clone 8 (ATCC

CCL-226); 3 T3-L1 (ATCC CL-173).
7. Mouse primary cells cultures: Bone marrow-derived dendritic
(BM-DCs) and myeloid suppressor cells (BM-MDSC); Mag-
netically Isolated CD4 and CD8 T cells.
8. Human primary cells cultures: Isolated Human Umbilical Vein
Endothelial Cells (HUVEC); Mesenchymal Stem cells isolated
from umbilical vein (h-MSC).
2.2 HIF-1α Detection 1. Protein extraction buffer (50 mM Tris, pH 7.5; 150 mM NaCl;
10 mM EDTA; 1% v/v NP-40) with protease (1% P8340-
Sigma) and phosphatase (1% P5726; Sigma) inhibitor cocktails.
2. Phosphate-buffered saline (PBS): 137 mM NaCl, 2.7 mM KCl,
10 mM Na2HPO4, 2 mM KH2PO4, pH 7.4.
3. Cell scraper 15 cm (70–1250; Biologix).
4. 96-well F-bottom plate (Greiner Bio-One).
5. Micro BCA Protein Assay Kit (23,235; Thermo Scientific).
6. 2 Laemmli sample buffer (161-0737; BioRad).
7. 4–15% Mini-PROTEAN® TGX™ Precast Protein Gels
(4561083; BioRad).
8. Vertical electrophoresis system cell (Mini-PROTEAN;
BioRad).
9. Universal power supply (PowerPac; BioRad).
10. Amersham Hybond-ECL (GE Healthcare).
11. Tris-buffered saline (TBS): 150 mM NaCl, 50 mM Tris,
pH 7.4.
12. tTBS (TBS with 0.05% Tween 20).
13. Blocking buffer: tTBS with 5% non-fat milk (M7409; Sigma-
Aldrich).
14. Antibody buffer: tTBS with 1% non-fat milk (M7409; Sigma-
Aldrich).
15. HIF-1α monoclonal antibody (MA1-516; Pierce).
16. Horse Anti-mouse IgG antibody (H + L), Peroxidase
(PI-2000; Vector Labs).
17. Immobilon chemiluminescent HRP substrate (WBKLS01-00;
Millipore).
18. 25 mM PBS-EDTA (Sigma).
19. HIF-1α Phycoerythrin (PE)-conjugated antibody (Clone
546-16; BioLegend).
20. Perm/Wash buffer (BD Biosciences).
21. 0.6-ml tubes (Axygen).
2.3 Detection of 1. Hypoxic cells to be evaluated.

Soluble VEGF by ELISA 2. 15-ml conical tubes (Greiner Bio-One).
3. P60 Petri dish (Greiner BioOne).
4. 96-well F-bottom ELISA plate (Greiner Bio-One).
5. Human VEGF DuoSet ELISA Kit (DY293-05; R&D System).
2.4 Isolation and 1. Umbilical cord.

Purification of Human 2. Plastic three-way stopcock (S7521; Sigma-Aldrich).
Umbilical Vein
3. 10-ml syringe (BD).
Endothelial Cells
(HUVECs) 4. 15-ml conical tube (Greiner BioOne).
5. PBS: 137 mM NaCl, 2.7 mM KCl, 10 mM Na2HPO4, 2 mM
KH2PO4, pH 7.4.
6. Cotton yarn (size 10/2).
7. PVDF syringe filter 0.22 μm (Millex GV; Millipore).
8. Physiological saline solution.
9. Culture flasks (Greiner Bio-One).
10. RPMI medium (Gibco-Thermo Fisher).
11. Fetal Bovine Serum (Gibco-Thermo Fisher).
12. Human recombinant VEGF (293-VE; R&D systems).
13. Penicillin/Streptomycin (Gibco-Thermo Fisher).
14. Gelatin (G9391; Sigma-Aldrich).
15. Collagenase type IV (G5138; Sigma-Aldrich).
16. CD31 APC-conjugated antibody (clone WM59; BioLegend).
17. CD144 (VE-Cadherin) FITC-conjugated antibody (clone
55-7H1; BD Biosciences).
18. CD45 APC-conjugated antibody (clone 2D1; BioLegend).
2.5 Assessment of 1. Primary Human Umbilical Vein Endothelial Cells (HUVEC)

Angiogenesis In Vitro passages number 6 or lower are recommended.
2.5.1 Endothelial Cell 2. M199 medium supplemented with 2% heat-inactivated FBS
Migration (Gibco-Thermo Fisher), 100 μg/ml streptomycin, and
100 U/ml penicillin (Gibco-Thermo Fisher).
3. Crystal violet (C0775; Sigma-Aldrich).
4. Tissue culture plate insert 8 μm pore (Greiner Bio-One).
5. 24-well plate (Greiner Bio-One).
6. Distilled water.
7. Cotton swabs 145 mm (421084; Greiner Bio-One).
8. Recombinant Gal1 (rGal1) purified as described [27].

10. Anti-Gal1 blocking monoclonal antibody (described in [27]).
11. β-Lactose (L3750; Sigma-Aldrich).
12. 30 -Fucosyllactose (F9641; Sigma-Aldrich).
13. Human recombinant VEGF (293-VE; R&D).
2.5.2 Endothelial Cell 1. Primary Human Umbilical Vein Endothelial Cells (HUVEC)
Tubulogenesis passages number 6 or lower.
2. Conditioned media (see Notes).
3. M199 medium supplemented with 2% heat-inactivated FBS
(Gibco-Thermo Fisher), 100 μg/ml streptomycin, and
100 U/ml penicillin (Gibco-Thermo Fisher).
4. TripLE Express Enzyme (Gibco-Thermo Fisher).
5. Geltrex™LDEV-free Reduced Growth Factor Basement
Membrane Matrix (A11343-01; Gibco).
6. 96-well F-bottom plate (Greiner Bio-One).
10. β-Lactose (L3750; Sigma-Aldrich).
2.5.3 Endothelial 1. Primary Human Umbilical Vein Endothelial Cells (HUVEC)

Spheroid Sprout Assay passages number 6 or lower.
2. Geltrex™LDEV free Reduced Growth Factor Basement Mem-
brane Matrix (A11343-01; Gibco).
3. M199 supplemented with 2% heat-inactivated FBS (Gibco-
Thermo Fisher), 100 μg/ml streptomycin, and 100 U/ml
penicillin (Gibco-Thermo Fisher).
4. TripLE Express Enzyme (Gibco-Thermo Fisher).
5. Collagen Type I obtained from rat tail (C7661; Sigma-
Aldrich).
6. Methyl cellulose (M0512; Sigma-Aldrich).
7. M199 medium 10 (11825015; Thermo Fisher).
9. Distilled water.
10. Fetal bovine serum (FBS) (Gibco).
11. Sodium hydroxide (NaOH—S7653; Sigma).
12. Ammonium chloride (NH4Cl—A9434; Sigma).
13. 96-well sterile U-bottom suspension culture plate (Greiner

Bio-One).
14. 48-well sterile flat bottom culture plate (Greiner Bio-One).
15. PBS:150 mM NaCl, 10 mM Phosphate, 2.7 mM KCl, pH 7.4.
19. β-Lactose (L3750; Sigma).
22. Paraformaldehyde (4% w/v in PBS; P6148; Sigma-Aldrich).
23. DAPI (40 ,6-diamidino-2-phenylindole dihydrochloride)
(D1306; Thermo Fisher).
24. Phalloidin TRITC-conjugated (P1951; Sigma-Aldrich).
25. Saponin solution (SAE0073; Sigma-Aldrich).
26. AutoFrost Microscope Slides 25 75 mm.
27. Fluorescent mounting media (S3023; Dako).
2.6 Assessment of 1. Matrigel Reduced Growth Factor Basement Membrane Matrix

Angiogenesis In Vivo (354248; BD Biosciences).
2.6.1 Matrigel Plug 2. 1-ml syringe (Neojet).
Assay 3. 25-G needle (BD).
4. Phosphate-buffered saline (PBS): 137 mM NaCl, 2.7 mM KCl,
10 mM Na2HPO4, 2 mM KH2PO4, pH 7.4.
6. Human recombinant FGF-2 (PHG0367; Thermo Fisher).
7. Human recombinant TNF (PHC3016; Thermo Fisher).
8. Heparin (H3393; Sigma-Aldrich).
2.6.2 Inoculation of 1. 25-G needle.

Matrigel Plugs to Evaluate 2. Matrigel Reduced Growth Factor Basement Membrane Matrix
Angiogenesis In Vivo (354,248; BD Biosciences).
3. C57BL/6 Lgals1/ (Gal-1 KO; Jackson Lab).
4. C57BL/6 WT (Jackson Lab).
2.6.3 Determination of 1. P60 Petri dish (GBO).

Angiogenesis In Vivo in 2. Drabkin’s reagent (D5941; Sigma).
Matrigel Plugs
3. Mouse hemoglobin (Sigma).
4. RPMI (Gibco).
5. 50-ml conical tubes (BD/Falcon).
6. Collagenase II solution (0.6% in PBS/Sigma).
7. Anti-Gal1 shRNA expression vector. (pGFP-V-RS; OriGene).
8. 100-μm cell strainer (BD Bioscience).
9. 1% fetal bovine serum (FBS) (Gibco).
10. Paraformaldehyde (4% w/v in PBS; P6148; Sigma-Aldrich).
11. Phosphate-buffered saline with 1% FBS and 0.05% NaN3.
12. Ketamine (Holliday Scott).
13. Xylazine (Richmond).
14. Optimum cutting temperature (OCT) medium (Biopack).
15. PBS with 10% normal rat serum (Sigma).
16. Acetone.
17. Anti-CD31 antibody (clone MEC13.3; BD Biosciences).
18. CD31 APC-conjugated antibody (Clone 390; BioLegend).
19. PBS with 1% BSA.
20. AutoFrost Microscope Slides 25 75 mm.
21. AlexaFluor 596-conjugated goat anti-rat antibody (Cell
Signaling).
22. Fluoromount slide mounting media (Southern Biotech).
23. DAPI (40 ,6-diamidino-2-phenylindole dihydrochloride)
(D1306; Thermo Fisher).
2.7 Special 1. LAS4000 imaging Chemiluminescent capture system (Fuji-

Equipment Film Life Sciences).
2. BD Accury C6-Plus Flow cytometer.
3. Olympus FluoView FV1000—confocal laser scanning
microscope.
4. Cryostat, Leica CM1850.
5. Sorval ST 8 Refrigerated Centrifuge.
3 Methods
1. Grow cell cultures at 37 C at 5% CO2 until it reaches sub

confluency (About 70% approximately).
3.1 Hypoxia 2. Open the chamber incubator, place a petri dish containing
Induction sterile water to provide adequate humidification of the cul-
tures. Place the cells inside the chamber incubator. Close the
3.1.1 Induction of
incubator hermetically.
Hypoxia in Modular
Incubator Chamber 3. An identical cell culture should be grown in normoxic condi-
tions as the control.
4. To generate a hypoxic atmosphere, flush the hypoxic gas mix-
ture at 2 psi during 10 min. Turn off the gas flow and isolate
the chamber by closing clamps (see Notes 1 and 2).
5. Place the chamber at 37 C in a conventional incubator for
18–24 h (see Notes 1 and 2).
3.1.2 Evaluation of 1. Open the chamber and immediately place cell cultures on ice.
Hypoxia: HIF-1α Detection 2. Remove culture medium, and wash with 10 ml PBS twice.
by Western Blot Analysis
3. Add lysis buffer (30 μl for P100) and use a scraper on both
hypoxia-treated and control cells.
4. Collect the total volume and centrifuge at 16,000 g for
20 min in a 4 C pre-cooled centrifuge.
5. Transfer the supernatant to a 0.6-ml fresh tube on ice and
discard the pellet.
6. Set aside a small volume of lysate and dilute it (1:50 to 1:200)
in H2O and calculate protein concentration using BCA Protein
Assay (Thermo-Scientific) according to manufacturer
instruction.
7. Run 20–40 μg of total cell lysate on 8% SDS-PAGE gel and
transfer to a PDVF membrane.
8. Block the membrane with 5 ml TBS 5% non-fat milk blocking
buffer at room temperature for 1 h on constant shaking.
9. Incubate for 18 h with anti-HIF-1α primary antibody diluted
1:500 in 3 ml tTBS (1% non-fat milk) at 4 C on constant
shaking.
10. Wash three times with 10 ml tTBS at room temperature for
10 min.
11. Incubate with HRP-conjugated secondary anti-mouse anti-
body diluted 1:2000 in tTBS for 2 h at room temperature.
12. Repeat step 10 and incubate with Immobilon chemilumines-
cent HRP substrate and capture the luminescent image in a
LAS4000 imaging system (Fuji Film Life Sciences).
3.1.3 Evaluation of 1. Open the chamber and immediately place cell cultures on ice
Hypoxia: HIF-1α Detection (see Note 3).
by Flow Cytometry 2. Remove culture medium, and wash with 1 ml PBS twice.
3. If necessary, detach adherent cells by using cold 25 mM EDTA-

PBS solution.
4. Centrifuge for 5 min at 200 g at room temperature and
5. Stain cells with PE-conjugated anti-HIF-1α antibody diluted 1:
200 in perm/wash buffer for 20 min on ice.
6. Centrifuge for 5 min at 200 g and discard supernatant.
7. Resuspend cells and fix in 1% paraformaldehyde PBS solution
during 10 min at room temperature.
8. Centrifuge as described in step 6 and resuspend cells in 150 μl
of PBS.
9. Analyze samples by flow cytometry.
3.1.4 Evaluation of Hypoxia controls a large number of genes through binding of

Hypoxia: Detection of HIF-1α to Hypoxia Response Element (HRE) sequences. Many
Soluble VEGF by ELISA genes are under the regulation of HRE such as VEGF and erythro-
poietin (EPO) [1, 4]. Therefore, measurement of pro-angiogenic
mediators is an effective and reliable method to evaluate the induc-
tion of HIF-dependent hypoxia.
1. Open the chamber and immediately collect cell supernatant.
Perform a brief and quick centrifugation (spin 40 s, 5000 g)
to eliminate cellular debris. Store supernatants at 80 C
until use.
2. Determine VEGF concentration in the supernatants with a
VEGF ELISA kit following the manufacturer’s instructions
(Human VEGF DuoSet ELISA Kit, DY293B; R&D System).
3.2 Isolation and 1. Prepare solutions.

Purification of Human (a) Dissolve 2% gelatin in physiological salt solution and ster-
Umbilical Vein ilize it immediately by using autoclave. It should be stored
Endothelial Cells at 4 C.
(HUVECs) (b) Prepare a 0.6 mg/ml collagenase solution in RPMI, the
same day the HUVEC purification will be carried out.
(c) Prepare HUVEC culture medium: RPMI supplemented
with 10% FBS and 10 ng/ml of VEGF.
2. Clean the surface of the umbilical cord with 70% ethanol.
3. Hold it by one end and cannulate the vein vessel by a three-way
stopcock. Use the cotton yarn to tie it tightly and wash with
30 ml of sterile prewarmed (37 C) PBS.
4. Afterwards, tie the bottom of the cord with cotton yarn and
perfuse the vein, through the 0.22-μm syringe filter, with 25 ml
of prewarmed 0.6 mg/ml collagenase solution (typically 1 ml/
cm of cord). Close the three-way stopcock and incubate the

umbilical cord for 40 min at 37 C.
5. Collect the released cells in a tube containing 3 ml FBS and
centrifuge for 7 min at 200 g at room temperature.
6. Discard supernatant and resuspend cells in 5 ml of HUVEC
culture medium and seed them in culture flasks precoated with
2% gelatin solution for 15 min at 37 C.
7. HUVECs should be maintained at 37 C in a humidified
incubator with 5% CO2 and observed daily.
8. After reaching confluency, cells should be expanded and sub-
jected to phenotypical characterization by flow cytometry
using anti-CD31, anti-VE-Cadherin (positive markers), and
anti-CD45 (negative marker) antibodies.
3.3 Assessment of 1. Use a 24-well plate and fill each well with 500 μl of M199 2%
Angiogenesis In Vitro FBS containing rGal-1 (1 μM) or rGal12 (3 μM). for other
galectins optimal concentration should be determined.
3.3.1 Endothelial Cell
Migration 2. To evaluate glycan-dependent effects of galectins on endothe-
lial cell migration, preincubate rGal1 with 30 mM β-Lactose or
1.5 μM of anti-Gal1 mAb for 1 h. For Gal12 use 0.4 mM
30 -fucosyllactose.
3. For positive control of EC migration, use M199 10% FBS
supplemented with 20 ng/ml VEGF.
4. Introduce the tissue culture inserts and seed 30,000 HUVECs
in 200 μl M199 2% FBS in the upper chamber of the insert (the
filter pores are small enough ~8 μm to allow passage of actively
migrating cells; otherwise, they rest upon the filter).

5. Incubate for 18–24 h at 37 C, 5% CO2 in humidified
incubator.
6. Stain the inserts with 0.1% crystal violet in 20% methanol
solution for 10 min.
7. Wash the transwells dipping them on distilled water.
8. Carefully, remove non-migrating cells from the inserts by using
cotton swabs.
9. Examine using an inverted microscope and count the number
of cells (Fig. 1a).
Data should be expressed as cells per cm2. A “fold-migra-
tion” value may be calculated as the number of cells migrating
in response to a given chemoattractant relative to the number
of cells in its absence.
3.3.2 Endothelial Cell One of the best-established assays to model the formation of three-
Tubulogenesis dimensional vessels is the tube formation assay. This method is
reliable, technically straightforward, easily quantifiable, and
Fig. 1 Representative images from the methods described. (a) Migration of HUVEC incubated with rGal1
(1 μM). (b) Tube formation of HUVEC cells in Geltrex™-coated plates incubated with rGal1 (1 μM). (c) Spheroid
Sprouting Assay (20) negative control and stimulated with 2 μM of rGal1 during 24 h. DAPI was used for
visualizing nuclei and Phalloidin for actin filaments. (d) Workflow for sprouting quantification from original
image to vectorized model using Image J
physiologically relevant. The following protocol is designed to

evaluate the effects of tumor-conditioned medium (or defined
molecules) in the formation of endothelial cell tubular networks.
1. Seed 50 μl of Geltrex per well in a pre-chilled 96-well plate.
Incubate 30 min at 37 C.
2. Prepare 15,000 HUVECs in 100 μl of D1 and seed in
each well.
3. Add 1 μM rGal1 or rGal12 to evaluate galectin-dependent
tubulogenesis.
4. To selectively block Gal1 or Gal12 functions, preincubate
galectins with 30 mM β-lactose or 0.4 mM 30 -fucosyllactose
respectively for 1 h.
5. For positive controls of tubulogenesis, use M199 medium

supplemented with 10% FBS and 20 ng/ml VEGF-A.
6. Incubate at 37 C and 5% CO2. Observe using phase contrast
microscope every 4 h for 18 h. Tubular structures are apparent
between 10 h and 16 h.
7. When tubular structures are apparent, take photomicrographs
of the whole field (Fig. 1b).
8. Quantification of tubular structures may be performed by dif-
ferent ways. One is based on counting the number of tubes/
cm2 or counting the branching points/cm2. Moreover, mor-
phometric analysis can be performed, and tubes length can be
measured using the Fiji 2.0 software [28].
3.3.3 Spheroid Sprout This three-dimensional in vitro angiogenesis assay based on colla-
Assays gen gel-embedded, size-defined spheroids generated from cultured
HUVECs take advantage of classical 2D techniques because of their
higher sensitivity, cost-efficiency, applicability, and physiologic rele-
vance [29]. The formation of size- and cell number-defined spher-
oids ensure reproducibility, providing a sensitive and versatile tool
to study the impact of galectins on multiple steps of the angiogenic
process (migration, proliferation, and vessel sprout formation)
simultaneously.
Spheroid Formation 1. Prepare a Methocel stock solution by dissolving 1.25 g of

methylcellulose in 100 ml RPMI in sterile conditions.
2. Centrifuge Methocel at 5000 g for 40 min and recover
around 70% from the top.
3. Prepare a spheroid formation solution with Methocel, RPMI,
and FBS in a 20:60:20 ratio and mix carefully by inversion.
4. Mix HUVECs (4000 cell/ml) with spheroid formation solu-
tion supplemented with 15 ng/ml of VEGF. Mix carefully by
inversion.
5. Seed 100 μl of cells on spheroid formation solution per well in a
96-well sterile U-bottom plate and incubate for 48–72 h.
6. Harvest the spheroids using a 1000-μl tip and centrifuge at
500 g for 2 min.
7. Carefully discard the overlaying spheroid formation solution
and wash with PBS. Repeat this process twice.
Sprouting Assay Note that solutions should be handled on ice all the time, and once
you mix both solutions, the next steps should be done rapidly
avoiding bubble formation.
1. Fill with PBS all the wells of a 48-well sterile flat bottom plate
and place it in the incubator at 37 C 5% CO2.
2. Place collagen stock solution (3 mg/ml), Sodium Hydroxide,

and M199 10, Methocel, and FBS on ice.
3. Make a Methocel sprouting solution by mixing Methocel and
FBS in an 80:20 ratio. Keep it on ice.
4. Make a collagen working solution by mixing collagen and
M199 in an 80:10 ratio. The solution will turn yellow since
collagen is usually dissolved in an acidic solution; around
5–10% of sodium hydroxide should be used to recover the
medium normal red color. If less than 10% is used, then com-
plete with sterile distilled water. Keep solution on ice all
the time.
5. Calculate 10 spheroids and 100 μl per well of each solution.
Centrifuge the spheroids and resuspend them in Methocel
sprouting solution, then mix both the solutions in a 1:1 ratio.
6. Discard the PBS in the 48-well plate and add 200 μl of the
mixed solutions per well and incubate for 30 min to let collagen
solidify.
7. Add 200 μl on the gel of: (i) conditioned medium from
selected cell lines (see Note 4), (ii) 3 μM rGal1, (iii) 5 μM
rGal12.
8. To evaluate galectin-dependent sprouting preincubate samples
for 1 h with 30 mM β-lactose or 5 μM blocking anti-Gal1 or
0.4 mM 30 -fucosyllactose (for blocking Gal12).
9. Use 2% FBS enriched medium for a negative control and 15%
FBS with 60 ng/ml VEGF for a positive control. Consider that
all concentrations of the conditioned medium and controls will
be diluted in half due to gel volume. Incubate at 37 C, 5%
CO2.
10. Sprouting should be observed between 12 h and 16 h.
11. Once you observe sprouting, carefully discard conditioned
medium, wash with PBS, and fix with 4% paraformaldehyde
in PBS for 1 h at 4 C.
12. Discard paraformaldehyde and add 250 μl of PBS. The plate
can be stored up to 2–3 days.
Sprouting Staining and 1. Discard PBS and add 300 μl of 25 nM ammonium chloride and
Quantification incubate for 1 h at room temperature.
2. Discard the ammonium chloride and add saponin for 30 min at
room temperature.
3. Discard and add 200 μl of Phalloidin solution 0.2 μM in PBS
and incubate for 1 h at room temperature.
4. Discard and wash with 300 μl of PBS.
5. Add DAPI 1/1000 for 15 min at room temperature.
6. Wash with 300 μl of PBS and mount in 15 μl of antifade

solution.
7. Capture images by confocal microscope and quantified them
using ImageJ (Fig. 1c, d).
8. Quantified Sprout length by using Analyze Skeleton (2D/3D)
function from BoneJ [30] plugin in ImageJ can be used (see
Note 5).
3.4 Methods to Study 1. It is important to keep Matrigel HC as cold as possible (lower

Angiogenesis In Vivo than 10 C). It is recommended to maintain all materials and
reagents (syringes, needles, solutions, pipettes, etc.) on ice
3.4.1 Matrigel Plug
prior to use.
Assay
2. Mix in vortex 300 μl of Matrigel with 200 μl of experimental
Matrigel Preparation solutions:
(a) Positive control: PBS + VEGF (10 ng/ml) + bFGF
(20 ng/ml) + TNF-α (5 ng/ml), + heparin 10,000 IU
(positive control mix).
(b) Serum-Free Conditioned Media (SFCM) from human or
mouse tumor cells exposed or not to a hypoxic atmo-
sphere (1% O2, 5% CO2, and 94% N2 during 18 h) and
transfected or not with shRNA against galectins.
(c) For Gal1 use: PBS + rGal1 (1.5 μM).
(d) PBS + rGal1 (1.5 μM) + anti-Gal1 blocking monoclonal
antibody (1.5 μM).
(e) For Gal12 use: PBS + rGal12 (3 μM).
(f) PBS + rGal12 (3 μM) + 30 -fucosyllactose (400 μM).
Inoculation of Matrigel 1. Using 25-G pre-cooled needle inject 0.5 ml of Matrigel mix
Plugs to Evaluate subcutaneously into anesthetized C57BL/6 WT or C57BL/6
Angiogenesis In Vivo Lgals1/ mice. The injections should be done quickly to
prevent the gel solidifying.
2. After 7 days, euthanize mice and remove the Matrigel plugs.
3. Angiogenesis can be evaluated by studying three independent
parameters:
(a) Hemoglobin content in pellets.
(b) Number of endothelial cells.
(c) Microvascular density.
Determination of 1. Remove Matrigel plugs, weigh, and mechanically disaggregate

Angiogenesis In Vivo in them in 1.5 ml of H2O in a petri dish.
Matrigel Plugs 2. Incubate for 20 min at RT.
3. Centrifuge for 5 min at 10,000 g.
Hemoglobin Content 4. Discard cell pellet.
5. Incubate supernatant with Drabkin’s reagent according to

manufacturer’s instructions.
6. Read absorbance of tubes at 540 nm. A standard curve of
mouse hemoglobin has to be simultaneously determined.
Plot absorbance vs. cyanmethemoglobin concentration
(mg/ml) and interpolate. The final concentration of hemoglo-
bin in Matrigel pellets is calculated as mg/ml per 100 mg of
pellets.
Number of Endothelial Cells 1. Remove Matrigel pellets, and mechanically disaggregate them
by Flow Cytometry in 5 ml of RPMI medium in a petri dish.
2. Transfer to a 50-ml conical tube, add 5 ml of collagenase II
solution, and incubate for 30 min at 37 C in water bath. After
incubation, add 25 ml of PBS and filter the suspension through
a 100-μm strainer.
3. Wash the filtered solution by adding 5 ml PBS and centrifuge
for 5 min at 300 g.
4. Remove the supernatant and wash pellet with PBS containing
1% FBS.
5. Stain cells with 1 μg of anti-CD31 APC-conjugated antibody in
100 μl of PBS containing 1% FBS and 0.05% NaN3 for 30 min
on ice.
6. Wash cells with 1 ml PBS and centrifuge for 5 min at 300 g.
7. Fix cells with 1% paraformaldehyde in PBS.
8. Analyze the number of CD31+ cells by flow cytometry.
Analysis of Microvascular 1. Anesthetize animals (ketamine/xylazine, 140/14 mg/kg) and

Density (MVD) perfuse them with 1 ml PBS and 1 ml 4% paraformaldehyde in
PBS at a rate of 300 μl/min.
2. Remove Matrigel pellets, embed them in 400 μl of frozen
Optimum Cutting Temperature (OCT) medium and freeze
them at 80 C.
3. Cut frozen Matrigel into 40–100 μm sections with a cryostat.
4. Air-dry sections at RT and fix in 200 μl acetone for 10 min at
20 C.
5. Air-dry for 5 min at RT. Wash three times with 200 μl PBS.
6. Block nonspecific binding through incubation for 1 h at RT
with PBS containing 10% normal rat serum.
7. Incubate sections with 1.5 μg of anti-CD31 primary antibody
(clone MEC13.3) in 100 μl of PBS containing 1% BSA 16 h at
4 C.
8. Wash with 200 μl of PBS three times.
9. Incubate for 1 h at RT with 0.2 μg of AlexaFluor

596-conjugated goat anti-rat antibody in 100 μl PBS.
10. Wash slides and mount in 20 μl of Fluoromount slide mount-
ing medium containing DAPI as counterstaining fluorophore.
11. Determine microvessel density (MVD) by counting the num-
ber of microvessels per mm2 in ten randomly selected fields
(200).
4 Notes
1. We recommend the use of oxygen sensors for more precise

measurement of the intra-chamber O2 during the experiment.
2. In order to eliminate the O2 diluted in the medium, it is
recommended to re-gas the chamber once after 2 h or bubbling
the gas into the cell culture medium before starting the
experiment.
3. HIF-1α is a molecule extremely sensitive to oxidation being
degraded as soon as samples are taken out from the hypoxia
chamber. Working fast and keeping samples on ice will contrib-
ute to success of this protocol.
4. To harvest conditioned media from different cell lines or cell
cultures, place cells for 24 h in serum-free conditioned media
(SFCM) or containing as low FBS as possible to ensure cell
viability. Collect media and centrifuge for 5 min at 5000 g to
eliminate cell debris. After that, transfer SFCM to a clean tube
and store a 80 C.
5. There are several ways of quantifying sprouting. Automatic
imaging analysis software, a manual approach will be used by
creating a binarization model. Briefly, draw a circle around the
spheroid perimeter and draw the shortest possible straight lines
from the perimeter to the tip of the sprouts including all
possible branching and use the measure function to get the
lengths of those lines (Fig. 1d). We have found total spheroid
sprout length to be more robust than average sprout measure.
Finally, we used as many total spheroid sprout lengths per
condition to build the statistic comparison (ANOVA or t-test).
Acknowledgments
Work in our laboratory is supported by grants from Argentinean

Agency for Promotion of Science and Technology (D.O.C., G.A.
R., and J.P.C.), National University of Cuyo SIIPJ069 (N.B. and
D.O.C.), Argentinean Council of Scientific and Technical Investi-
gations (M.S. and D.O.C.), National Institutes of Health (NIH)/
National Cancer Institute (NCI) U54 program CA221208 (G.A.

R., D.O.C., and J.P.C.), Sales Foundation (G.A.R.), Bunge & Born
Foundation (G.A.R.), and Richard Lounsbery Foundation
(G.A.R.).
References
1. Carmeliet P, Jain RK (2000) Angiogenesis in 13. Croci DO, Cerliani JP, Dalotto-Moreno T et al
cancer and other diseases. Nature 407: (2014) Glycosylation-dependent lectin-recep-
249–257 tor interactions preserve angiogenesis in anti-
2. Martin JD, Seano G, Jain RK (2019) Normal- VEGF refractory tumors. Cell 156:744–758
izing function of tumor vessels: Progress, 14. Hsieh SH, Ying NW, Wu MH et al (2008)
opportunities, and challenges. Annu Rev Galectin-1, a novel ligand of neuropilin-1, acti-
Physiol 81:505–534 vates VEGFR-2 signaling and modulates the
3. Motz GT, Coukos G (2011) The parallel lives migration of vascular endothelial cells. Onco-
of angiogenesis and immunosuppression: can- gene 27:3746–3753
cer and other tales. Nat Rev Immunol 11: 15. D’Haene N, Sauvage S, Maris C et al (2013)
702–711 VEGFR1 and VEGFR2 involvement in extra-
4. Croci DO, Mendez-Huergo SP, Cerliani JP cellular galectin-1- and galectin-3-induced
et al (2018) Immune-mediated and hypoxia- angiogenesis. PLoS One 8:e67029
regulated programs: accomplices in resistance 16. Markowska AI, Liu FT, Panjwani N (2010)
to anti-angiogenic therapies. Handb Exp Phar- Galectin-3 is an important mediator of
macol 249:31–61 VEGF- and bFGF-mediated angiogenic
5. Ferrara N, Alitalo K (1999) Clinical applica- response. J Exp Med 207:1981–1993
tions of angiogenic growth factors and their 17. Markowska AI, Jefferies KC, Panjwani N
inhibitors. Nat Med 5:1359–1364 (2011) Galectin-3 protein modulates cell sur-
6. Johannes L, Jacob R, Leffler H (2018) Galec- face expression and activation of vascular endo-
tins at a glance. J Cell Sci 131:1–9 thelial growth factor receptor 2 in human
7. Cerliani JP, Blidner AG, Toscano MA et al endothelial cells. J Biol Chem 286:
(2017) Translating the ’Sugar Code’ into 29913–29921
immune and vascular signaling programs. 18. Delgado VM, Nugnes LG, Colombo LL et al
Trends Biochem Sci 42:255–273 (2011) Modulation of endothelial cell migra-
8. Mendez-Huergo SP, Blidner AG, Rabinovich tion and angiogenesis: a novel function for the
GA (2017) Galectins: emerging regulatory "tandem-repeat" lectin galectin-8. FASEB J
checkpoints linking tumor immunity and 25:242–254
angiogenesis. Curr Opin Immunol 45:8–15 19. Chen WS, Cao Z, Sugaya S et al (2016) Patho-
9. Thijssen VL, Barkan B, Shoji H et al (2010) logical lymphangiogenesis is modulated by
Tumor cells secrete galectin-1 to enhance galectin-8-dependent crosstalk between podo-
endothelial cell activity. Cancer Res 70: planin and integrin-associated VEGFR-3. Nat
6216–6224 Commun 7:11302–11312
10. Thijssen VL, Postel R, Brandwijk RJ et al 20. O’Brien MJ, Shu Q, Stinson WA et al (2018) A
(2006) Galectin-1 is essential in tumor angio- unique role for galectin-9 in angiogenesis and
genesis and is a target for antiangiogenesis ther- inflammatory arthritis. Arthritis Res Ther 20:
apy. Proc Natl Acad Sci U S A 103: 31–39
15975–15980 21. Heusschen R, Schulkens IA, van Beijnum J et al
11. Mathieu V, Martin de Lassalle E, Toelen J et al (2014) Endothelial LGALS9 splice variant
(2012) Galectin-1 in melanoma biology and expression in endothelial cell biology and
related neo-angiogenesis processes. J Invest angiogenesis. Biochim Biophys Acta 42:
Dermatol 132:2245–2254 284–292
12. Croci DO, Salatino M, Rubinstein N et al 22. Maller SM, Cagnoni AJ, Bannoud N et al
(2012) Disrupting galectin-1 interactions (2020) An adipose tissue galectin controls
with N-glycans suppresses hypoxia-driven endothelial cell function via preferential recog-
angiogenesis and tumorigenesis in Kaposi’s sar- nition of 3-fucosylated glycans. FASEB J 34:
coma. J Exp Med 209:1985–2000 735–753
23. Chen C, Duckworth CA, Fu B et al (2014) 27. Pérez Sáez JM, Hockl PF, Cagnoni AJ et al
Circulating galectins 2, 4 and 8 in cancer (2021) Characterization of a neutralizing anti-
patients make important contributions to the human galectin-1 monoclonal antibody with
increased circulation of several cytokines and angioregulatory and immunomodulatory
chemokines that promote angiogenesis and activities. Angiogenesis 24(1):1–5. https://
metastasis. Br J Cancer 110:741–752 doi.org/10.1007/s10456-020-09749-3
24. Arthur CM, Baruffi MD, Cummings RD et al 28. Schindelin J, Arganda-Carreras I, Frise E et al
(2015) Evolving mechanistic insights into (2012) Fiji: an open-source platform for
galectin functions. Methods Mol Biol 1207: biological-image analysis. Nat Methods 9:
1–35 676–682
25. Pugh CW, Ratcliffe PJ (2003) Regulation of 29. Heiss M, Hellström M, Kalén M et al (2015)
angiogenesis by hypoxia: role of the HIF sys- Endothelial cell spheroids as a versatile tool to
tem. Nat Med 6:677–684 study angiogenesis in vitro. FASEB J 29:
26. Navarro P, Martinez Bosch N, Blidner AG et al 3076–3084
(2020) Impact of galectins in resistance to anti- 30. Doube M, Kłosowski MM, Arganda-Carreras I
cancer therapies. Clin Cancer Res 26: et al (2010) BoneJ: free and extensible bone
6086–6101 image analysis in ImageJ. Bone 47:1076–1079
Chapter 35
Examination of the Role of Galectins and Galectin Inhibitors

in Endothelial Cell Biology
Abstract
The growth of new blood vessels is a key event in many (patho) physiological processes, including
embryogenesis, wound healing, inflammatory diseases, and cancer. Neovascularization requires different,
well-coordinated actions of endothelial cells, i.e., the cells lining the luminal side of all blood vessels.
Galectins are involved in several of these activities. In this chapter, we describe methods to study galectins in
three key functions of endothelial cells during angiogenesis, i.e., endothelial cell migration, endothelial cell
sprouting, and endothelial cell network formation.
Key words Angiogenesis, Endothelial cell, Migration, Network formation, Sprouting, Galectin
1 Introduction
Blood vessels are part of the infrastructure of an organism, as they

facilitate the transport of molecules and cells throughout the body.
The vascular infrastructure is formed during embryogenesis by two
main processes, vasculogenesis and angiogenesis. Vasculogenesis
refers to the early de novo formation of a primitive vasculature by
angioblasts. During further development, new vessels are continu-
ously sprouting from the existing capillaries in this vasculature by a
process called angiogenesis [1]. Angiogenesis is a complex process
that involves different well-coordinated and stepwise activities of
endothelial cells (EC), i.e., the cells that cover the luminal side of
blood vessels. The angiogenesis cascade is initiated by the activation
of the EC by pro-angiogenic growth factors like vascular endothe-
lial growth factor (VEGF). Once activated, supportive cells (peri-
cytes) detach from the capillary and supportive structures (basal
membrane and the underlying extracellular matrix) are degraded by
proteases. Next, the endothelial cells start to proliferate and migrate
in the direction of the proangiogenic stimulus. The growing EC
sprouts form primitive tubular structures that are eventually stabi-
lized by the deposition of a new basal membrane and the
655
recruitment of pericytes. This allows the EC in these newly formed

capillaries to return to a quiescent state [1, 2].
As evident from their pleiotropic functions in cell biology,
several galectins are involved in different steps of the angiogenesis
cascade. This is not restricted to physiological angiogenesis during,
for example, wound healing and embryogenesis, but extends to
different pathologies that are characterized by dysregulated angio-
genesis, including cancer [3–5]. Galectins have been shown to
activate EC and to promote angiogenic signaling in EC. In addi-
tion, galectins have been associated with EC adhesion, migration,
and proliferation during angiogenesis [6–9]. Thus, galectins are
now recognized as regulators of angiogenesis and even as targets
for angiostatic therapy [3]. In this chapter, we provide methods to
assess the effects of galectins on endothelial cell migration, network
formation and sprouting in vitro. Apart from studying the direct
effects of galectins, these methods can also be used to determine
the effects of galectin-blocking compounds.
2 Materials
2.1 Migration 1. Endothelial cells (see Note 1).

2. Flat bottom 96-well plates (Costar) (see Note 2).
3. Pipetman with 1.0–10 mL sterile pipettes.
4. 10–1000 μL pipettes with sterile tips.
5. Phosphate-buffered saline (PBS).
6. 0.2% gelatin (Merck) in PBS.
7. Recombinant galectins and/or galectin inhibitors.
8. Humidified 5% CO2 incubator at 37 C/99.5 F.
9. A 96-well pin tool scratcher (see Note 3).
10. UniversalGrab 6.2 software (DCIlabs) for image acquirement.
11. Image analysis software, e.g., ScratchAssay 6.2 (DCIlabs), Ima-
geJ or Photoshop.
2.2 Network 1. Flat bottom 96-well plates (Costar).

Formation 2. BD Matrigel basement membrane matrix (see Note 4).
3. Endothelial cells (see Note 1).
4. Recombinant galectins and/or galectin inhibitors.
2.3 Sprouting 1. Endothelial cells (see Note 1).

2. μ-Slide eight-well Ibidi plate.
Examination of the Role of Galectins and Galectin Inhibitors in. . . 657
3. Nonadhesive square petri dishes (10 10 cm).

4. Methocel medium: RPMI with 20% methocel (Sigma-Aldrich)
and 10% heat-inactivated human serum (in-house produced).
5. Sprouting medium: 10% fetal bovine serum (FBS), 0.1% Hep-
arin (Leo Pharma), 14.8% Methocel, 62.3% PureColl (Sigma-
Aldrich), 8.3% 10 199 complete (Gibco/ThermoFisher),
4.5% NaOH.
8. Image analysis software, ImageJ.
2.4 Special 1. Inverted microscope equipped with a camera. We use a Leica

Equipment DMI3000B microscope equipped with an automated xyz-stage
and a Hitachi 1.4 Mb GiGE color camera.
3 Methods
3.1 Migration 1. Coat a flat bottom 96-well plate with 50–80 μL 0.2% gelatin/
PBS for at least 30 min at 37 C/99.5 F or 2 h at room
temperature. Aspirate gelatin before seeding the cells. Alterna-
tively, the wells can be coated with the recombinant galectin of
interest in PBS (optimal concentration should be determined
empirically).
2. Seed cells and grow to confluence in 2–3 days (see Note 5).
3. Scratch the confluent monolayer with a 96-well pin tool (see
Note 3).
4. Aspirate/drain the culture medium.
5. Carefully wash cells one time with 100 μL PBS.
6. Apply the appropriate medium containing recombinant galec-
tin or galectin inhibitors (see Notes 6 and 7).
7. Take images of the scratch at t ¼ 0, t ¼ 2, t ¼ 4, t ¼ 6, and
t ¼ 8 h (see Note 8 and Fig. 1a).
8. Measure wound width or scratch area with automated Scratch-
Analysis software. Alternatively use ImageJ (use straight line
selection tool to measure wound width or free hand selection
tool for measuring scratch area) or Photoshop (use ruler for
wound width measurement or wand tool for scratch area).
Calculate the percentage of wound closure or remaining
wound width/area (% of t ¼ 0).
3.2 Network 1. Coat each well of a flat bottom 96-well plate with 40 μL
Formation Matrigel (see Note 4) and incubate at 37 C/99.5 F for
30 min.
Fig. 1 In vitro angiogenesis assays. Representative pictures of migration (a+d),

network formation (b+e), and sprouting (c+f) of nontreated (control) and galectin
inhibitor-treated (e.g., anginex 5 mg/mL) HUVEC. The first panels represent the
2. Apply the required number of cells in 50 μL medium per well

(see Note 9). Add recombinant galectin of interest or galectin
inhibitors (e.g., lactose or anginex) to the medium.

3. Incubate overnight in a humidified incubator at 37 C/
99.5 F, 5% CO2.
4. Acquire images using a microscope with a camera system.
5. Endothelial cell network formation can be quantified by count-
ing the number of branch points, number of meshes (see
Fig. 1b).
3.3 Sprouting 1. Harvest cells and resuspend to a final concentration of

40,000 cells/mL in methocel medium (see Note 10).
2. Distribute 25 μL drops with a multichannel pipet on the lid of a
nonadhesive 12 12 cm square petri dish (see Note 11).
3. Add 10 mL PBS to the bottom of the petri dish to prevent
evaporation of the drops.
4. Turn the lid with drops carefully upside down and place on the
bottom dish containing PBS.

5. Incubate overnight in a humidified incubator at 37 C/
99.5 F, 5% CO2.
6. Add 100 μL sprouting medium per well to a μ-Slide eight-well
Ibidi plate (see Notes 12 and 13).
7. Incubate the plate for 30 min at 37 C/99.5 F until the
sprouting medium is solidified.
8. Harvest the spheroids using a 1-mL pipet and centrifuge for
1 min at 400 rcf.
9. Remove supernatant and resuspend the spheroids carefully in
sprouting medium at approximately 30 spheroids/100 μL and
transfer 100 μL per well (see Notes 12 and 13).
10. Incubate the plate for 30 min at 37 C/99.5 F until the
sprouting medium is solidified.
Fig. 1 (continued) starting conditions at t ¼ 0. The second and third panels

indicate typical results at the end point of the experiment of nontreated and
galectin inhibitor-treated cells, respectively. The panels e-f illustrate different
ways of analyzing the respective results: (d) For analysis of migration, wound
width (white arrow), or wound area (in black) can be measured at different time
points and compared with t ¼ 0. (e) Network formation can be quantified by
counting the number of meshes (M) and branch points (white dots). For clarity
only part of the image is scored. (f) Sprouting is analyzed by measuring the
sprout length (white lines) and counting the number of sprouts
11. Apply 100 μL medium, containing recombinant galectin pro-

teins and/or galectin inhibitors on top of the sprouting
medium (see Notes 13 and 14).
12. Incubate in a humidified incubator at 37 C/99.5 F, 5% CO2
for 16–24 h and take pictures of the spheroids (see Note 15 and
Fig. 1c).
13. Analyze sprout length and sprout number per spheroid using,
e.g., ImageJ or Adobe Photoshop (see Note 16).
4 Notes
1. Different sources of endothelial cells are available [10].

Human umbilical vein endothelial cells (HUVEC) are fre-
quently used, and these cells can be isolated in house or pur-
chased from a commercial source. Apart from HUVEC, we also
use the EC lines EC-RF24 (HUVEC origin) and HMEC
(human microvascular endothelial cells). For network forma-
tion and sprouting assays, we generally use HUVEC while
migration assays can be performed with any source of EC. If
required, cells can be transfected with galectin-targeting siRNA
or galectin expression constructs according to standard proto-
cols. All EC are cultured in RPMI1640 (Lonza) supplemented
with 10% fetal bovine serum (FBS), 10% human serum, 1% L-
glutamine, and 1% penicillin/streptomycin. At confluency, pas-
sage the cells 1:3.
2. We prefer the use of Costar flat bottom plates since plates from
other suppliers sometimes contain markings on the back of the
plates due to the production process. These markings can
interfere with the scratch analysis.
3. A pin tool scratcher is preferred since this gives low variability
and high reproducibility in wound width. We use the Peira
HTSScratcher. However, it is also possible to apply a scratch
manually using a small pipetting tip, e.g., 10 μL tips. Take a
fresh tip for each well to avoid increasing scratch width due to
cells that stick to the tip.
4. Matrigel is derived from Engelbreth-Holm-Swarm (EHS)
mouse sarcoma and contains several extracellular matrix pro-
teins that facilitate endothelial cell adhesion and migration.
However, standard Matrigel also contains several growth fac-
tors that stimulate the angiogenic activity of endothelial cells.
Therefore, depending on the research question, e.g., test effect
of inhibitors on galectin-induced network formation or sprout-
ing, it is better to use growth factor reduced Matrigel. Note
that Matrigel stock solutions should be kept at temperatures
below 4 C/39.2 F to prevent solidification. Thawing should
therefore be performed on ice in a fridge (this takes approxi-

mately 1 h).
5. It is important to start with a confluent monolayer of cells in
order to induce unidirectional migration. For RF24 cells, seed
15,000 cells/well, and for HUVEC/HMEC, use 5000 cells/
well.
6. Evaluation of inhibitory activity can be performed in normal
EC culture medium. To evaluate stimulatory activity, low
serum (0.5% FBS and/or 0.5% human serum) can be used.
Furthermore, we have observed a biphasic activity of galectin-1
depending on the concentration [8, 11]. Therefore, we suggest
using a broad concentration range to optimize your assay.
7. Always include proper controls, e.g., Sutent (Sunitinib malate)
(Sigma-Aldrich) for inhibition or Fibroblast Growth Factor-
Basic (bFGF) (Sigma-Aldrich) for stimulation of migration.
Use as directed by the manufacturer.
8. For analysis, it is important that the images of the scratch
within the time series are taken at the exact same position. If
no appropriate software and automated xyz table are available
to automate this, make sure to mark the culture plate in such
way that the area in the well where the images are taken can be
easily found. For example, prior to plating the cells, use a
scalpel to make a scratch on the backside of each well in the
plate. Make sure that the scratch is perpendicular to the direc-
tion of the “wound.” The perpendicular scratch can then be
used to identify the region where the wound width is
measured.
9. The optimal number of cells used in this assay should be
individually determined for each cell line. Too high cell num-
bers will result in robust branches or confluent areas at the
center of the well. Too few cells will yield incomplete network
structures. We generally use 20,000 HUVEC per 96-well.
10. Dilute to 40,000 cells/mL to obtain 1000 cells/25 μL drop.
Prepare 60 drops (60,000 cells in 1500 μL methocel medium)
per test condition since not all drops will successfully form a
spheroid.
11. Use reverse pipetting technique to avoid air bubbles disturbing
the spheroid formation.
12. When transferring the spheroid suspension, make sure the
whole well is covered by pipetting the suspension “swirl-
wise,” starting in the middle of the well. Avoid air bubbles.
13. The experiment can also be performed in a 24-well culture
plate. In that case, the volumes in steps 6, 9, and 11 should
be adjusted to 200 μL, 200 μL, and 500 μL, respectively. The
concentration of galectin depends on the activity. For example,

for galectin-9, we used 500 nM [11].
14. As a positive control for the induction of sprouting, add
100 nM of the γ-secretase inhibitor dibenzazepine (DBZ) to
the medium. As a positive control for inhibition of sprouting,
10 μM Sutent can be added to the medium.
15. For reliable analysis of sprouting, at least 10 spheroids per
treatment condition should be photographed and analyzed.
16. Sprout number and/or length can be assessed manually in
Adobe Photoshop by using the “count tool” or “ruler tool”
in Image analysis. ImageJ also provides similar tools or even
plugins to analyze sprouting.
References
1. Carmeliet P, Jain RK (2011) Molecular 7. Croci DO, Cerliani JP, Dalotto-Moreno T et al
mechanisms and clinical applications of angio- (2014) Glycosylation-dependent lectin-recep-
genesis. Nature 473:298–307 tor interactions preserve angiogenesis in anti-
2. De Palma M, Biziato D, Petrova TV (2017) VEGF refractory tumors. Cell 156:744–758
Microenvironmental regulation of tumour 8. Thijssen VL, Barkan B, Shoji H et al (2010)
angiogenesis. Nat Rev Cancer 17:457–474 Tumor cells secrete galectin-1 to enhance
3. Thijssen VL, Rabinovich GA, Griffioen AW endothelial cell activity. Cancer Res 70:
(2013) Vascular galectins: regulators of tumor 6216–6224
progression and targets for cancer therapy. 9. Schulkens IA, Griffioen AW, Thijssen VL
Cytokine Growth Factor Rev 24:547–558 (2012) Angiostatic cancer therapy by targeting
4. Thijssen VL, Poirier F, Baum LG, Griffioen galectins in the tumor vasculature. In: Klyosov
AW (2007) Galectins in the tumor endothe- A (ed) ACS symposium series: galectins and
lium; opportunities for combined cancer ther- disease implications for targeted therapeutics.
apy. Blood 110:2819–2827 ACS Publications, Washington DC, pp
5. Thijssen VL, Heusschen R, Caers J, Griffioen 233–247
AW (2015) Galectin expression in cancer diag- 10. van Beijnum JR, van der Linden E, Griffioen
nosis and prognosis: a systematic review. Bio- AW (2008) Angiogenic profiling and compari-
chim Biophys Acta 1855:235–247 son of immortalized endothelial cells for func-
6. Croci DO, Salatino M, Rubinstein N et al tional genomics. Exp Cell Res 314:264–272
(2012) Disrupting galectin-1 interactions 11. Aanhane E, Schulkens IA, Heusschen R et al
with N-glycans suppresses hypoxia-driven (2018) Different angioregulatory activity of
angiogenesis and tumorigenesis in Kaposi’s sar- monovalent galectin-9 isoforms. Angiogenesis
coma. J Exp Med 209:1985–2000 21:545–555
Chapter 36
Evaluating Therapeutic Activity of Galectin-1

in Sarcolemma Repair of Skeletal Muscle
Mary L. Vallecillo-Zúniga, Matthew Rathgeber, Daniel Poulson,
Braden Kartchner, Jacob Luddington, Hailie Gill, Spencer Hayes,
Matthew Teynor, Caleb S. Stowell, Connie M. Arthur, Sean R. Stowell,
and Pam M. Van Ry
Abstract
Galectin-1 is a small (14.5 kDa) multifunctional protein with cell–cell and cell–ECM adhesion due to
interactions with the carbohydrate recognition domain (CRD). In two types of muscular dystrophies, this
lectin protein has shown therapeutic properties, including positive regulation of skeletal muscle differentia-
tion and regeneration. Both Duchenne and limb-girdle muscular dystrophy 2B (LGMD2B) are subtypes of
muscular dystrophies characterized by deficient membrane repair, muscle weakness, and eventual loss of
ambulation. This chapter explains confocal techniques such as laser injury, calcium imaging, and galectin-1
localization to examine the effects of galectin-1 on membrane repair in injured LGMD2B models.
Key words Galectin-1, Muscular dystrophy, LGMD2B, Sarcolemma, Membrane repair, Localization,
Protein labeling
1 Introduction
Muscular dystrophies comprise a heterogeneous group of genetic

diseases, some of which have compromised sarcolemma stability,
decreased muscle strength, and aberrant membrane repair. Despite
the fact that each type of muscular dystrophy has its own unique
etiology, many forms of muscular dystrophies, such as limb-girdle
muscular dystrophy 2B (LGMD2B) and Duchenne muscular dys-
trophy display compromised sarcolemma repair. The incidence of
LGMD2B disease ranges from 1:1300 to 1:200,000, with certain
geographic locations and ethnic population more heavily impacted
than others. Human patient, mouse, and cell models of LGMD2B
have mutations in the DYSF gene, resulting in non-functional
dysferlin, a key membrane protein. People with LGMD2B
663
664 Mary L. Vallecillo-Zúniga et al.
Fig. 1 Galectin-1 as Potential Therapeutic for LGMD2B. (a) Highlighted regions show the muscles affected by
LGMD2B. (b) Graphical representation of how Gal-1 helps increase membrane repair in LGMD2B
experience decreased mobility, chronic inflammation, and fat infil-

tration into muscle groups, leading to a diminished quality of life.
The protein galectin-1 (Gal-1) is expressed intra- and extra-
cellularly, leading to positive regulation in regeneration, neurode-
generation, and inflammatory diseases, including muscular dystro-
phies [1]. Gal-1 has been shown to increase myogenic potential, the
process of muscle growth, and repair capability in in vitro and
ex vivo models of LGMD2B (Fig. 1) [2]. Additionally, Van Ry
et al. [3] showed that exogenous treatment with Gal-1 improved
sarcolemma stability and disease manifestations in Duchenne mus-
cular dystrophy [4].
Since sarcolemma repair is a vital part of cell homeostasis, our
research uncovers biochemical characteristics of Gal-1 and how it
functions to reduce muscle disease pathologies. Membrane stability
and repair capacity of cells and tissues were evaluated through
modified FRAP assays. This assay allows for precise, reproducible
injuries to membranes through ablation by a UV laser pulse. When
used in skeletal muscle cells, this technique is meant to simulate
micro-injuries that occur during normal muscle use. Furthermore,
this assay allows for the concurrent examination of any fluorescently
labeled protein to examine co-localization and movement during
injury events. We have used these methods to discover the thera-
peutic effect Gal-1 has in membrane repair. The versatility of this
technique allows the combined use of in vitro or ex vivo (explants)
with drugs along with real-time visualization.
The carbohydrate recognition domain (CRD) of Gal-1 is nec-
essary for improved membrane resealing using dysferlin-deficient
Evaluating Therapeutic Activity of Galectinn Sarcolemma Repair. . . 665
myotubes. Functionally, the promiscuous nature of the CRD allows

Gal-1 to have many binding partners and affect several different
cellular processes. In order to better understand the importance of
the CRD during membrane repair, we inhibited activity of the CRD
with lactose, which is a known antagonist of this domain. This
prevented binding of Gal-1 to other molecules via the CRD. We
found that the CRD of Gal-1 is essential for membrane repair, as
there was no improvement in membrane repair after the CRD was
inhibited [2].
Another necessary component of normal muscle repair is cal-
cium regulation [5–7]. To understand the dependence of Gal-1’s
on calcium during membrane repair in LGMD2B models, we
performed laser ablations in the presence and absence of calcium,
calcium chelators, and labeled calcium. Our results showed that
Gal-1 improves membrane repair capability in a calcium-
independent manner [2]. Combined with co-localization imaging,
this suggests that Gal-1 stabilizes the membrane through interac-
tions with the CRD, independent of calcium.
In this chapter, we explain how to use labeled and unlabeled
Gal-1 with varying conditions for live confocal laser injury. These
techniques were employed to evaluate therapeutic benefits of Gal-1
through membrane injury assays. While we explain how to injure
muscle specifically, the assay can be adapted to look at membrane
repair in a variety of cell types with or without the use of pharma-
cological agents to define unknown interactions.
2 Materials
2.1 Production and 1. Human Galectin-1 gblock LGALS1 (Integrated DNA Tech-
Purification of nologies, Coralville, IA).
Recombinant Human 2. pET29b (+) vector (Novagen, Millipore Sigma).
Galectin-1 (Gal-1)
3. NEBuilder® HiFi DNA Assembly Cloning Kit (E552OS New
England Biolabs (NEB), Ipswich, MA).
4. E.Z.N.A.® Plasmid DNA Mini Kit I protocol (Omega Bio-Tek,
Inc. Norcross, GA).
5. BL21 (DE3) competent E. coli cells (High Efficiency, NEB #
C2527H).
6. Phosphate Buffered Saline [PBS] (Gibco).
7. Sepharose 4B (Sigma-Aldrich).
8. Isopropyl β-D-1-thiogalactopyranoside (IPTG) (Promega).
9. Lactose (Alfa Aesar).
10. B-mercaptoethanol (βME).
11. Complete Mini EDTA-free Protease Inhibitor Cocktail tablets
(Thermo Fisher).
12. Lysozyme (Sigma-Aldrich).

13. Universal Nuclease for Cell Lysis, Thermo Scientific, Pierce
Universal Nuclease for Cell Lysis (Thermo Scientific).
14. Sodium bicarbonate, Na2CO3 (Macron Fine Chemicals).
15. Divinylsulfone (Sigma Aldrich).
16. Sodium azide (Thermo Scientific).
17. Tris Base.
18. Glycine.
19. Sodium Dodecyl Sulfate.
20. Loading buffer (Thermo Scientific).
21. 30% Bis-acrylamide (Alfa Aesar).
22. Molecular Weight Marker: BLUEstain™ 2 protein ladder
(Gold Bio).
23. 1.7 mL snap cap microcentrifuge tube.
24. 2 L Erlenmeyer flask.
25. 1 L centrifuge bottles.
26. Pierce™ Coomassie (Bradford) Protein Assay Kit (Thermo
Scientific).
27. Lennox broth (LB) powder.
2.1.1 Special Equipment 1. Centrifuge (Eppendorf).

2. Sonicator (BRANSON Digital Sonifier).
3. Gel apparatus including voltage source (Bio-Rad).
4. Imaging (Licor Odyssey® CLx Imaging System).
2.1.2 Buffers 1. Wash Buffer (1 L) ¼ 1 PBS with 14 mM βME.

2. Lysis Buffer (for 1 L of culture) ¼ Wash Buffer including
1 Complete Mini EDTA-free Protease Inhibitor cocktail tablet,
Lysozyme (1 mg), Pierce Universal Nuclease for Cell Lysis
(Thermo Scientific) (100 μL of nuclease at a concentration of
0.1 μL/mL).
3. Elution Buffer (1 L) ¼ 1 PBS with 14 mM βME and 100 mM
Lactose (36 g).
4. Column Storage Buffer (total volume 500 mL) ¼ 1 PBS
w/0.02% Sodium Azide, 14 mM βME.
5. 1 Electrophoresis Buffer ¼ 3 g Tris Base, 14.4 g Glycine and
1 g Sodium Dodecyl Sulfate in total volume of 1 L dH2O.
6. Destain Solution ¼ 40% Methanol and 10% Glacial Acetic Acid
in dH2O.
2.1.3 Separation 1. PD10 column (GE Healthcare).

2. PBS (Gibco).
3. Econo-Column® Chromatography Columns, 1.5 50 cm
(Bio-Rad).
4. Sand, quartz (Thermo Scientific).
5. Lactose (Alfa Easar).
6. B-mercaptoethanol (Fisher).
7. Sepharose 4B (Sigma-Aldrich).
2.2 H2K A/J/ and 1. Immortalized murine myoblasts H2K A/J/, [A/J/],
H2K A/J+/+ Cell Culture (Clone #13-110/28/09) and H2K A/J WT, [WT], (Clone
#16, 6/9.2010) (Purchased from Center for Genetic Medicine
Research, Washington, DC).
2. Glass-bottomed, collagen coated dishes sterilized with gamma-
irradiation (MatTek, Ashland, MA).
3. T-175 Flasks (Sarsted).
4. Counting slides dual chambers for cell counter (Bio-Rad).
5. Growth Media [DMEM (4500 mg/L glucose, 4 mM L-Gluta-
mine, 110 mg/L Pyruvate) (Gibco), 20% HI-FBS (Invitro-
gen), 2% Chick embryo extract (US Biological Life Sciences),
2% L-glutamine (Sigma), 1% penicillin/streptomycin].
6. BSA 10% Stock Solution (Invitrogen).
7. Filter 0.22 μm (Fisher).
8. Differentiation Media [DMEM (4500 mg/L glucose, 4 mM L-
Glutamine, 110 mg/L Pyruvate) (Gibco), Horse Serum (Invi-
trogen), 1% penicillin/streptomycin].
9. Interferon gamma (Gibco).
10. PBS (Thermo Fisher).
11. Penicillin/Streptomycin (CAISSON).
12. Trypsin EDTA (Gibco).
2.3 Calcium 1. H2K A/J/ and H2K A/J+/+ cells.

Dependency on 2. Gal-1 (0.11 μM, diluted in PBS).
Membrane Repair
3. 35 mm Glass Bottom Microwell Dishes (MatTek, Ashland,
MA).
4. 1 mM Ca2+ (as CaCl2).
5. 1 μM (1,2-Bis(2-aminophenoxy) Ethane-N,N,N0 ,N0 -tetraacetic
acid tetrakis (acetoxymethyl ester); (BAPTA-AM) (EMD
Millipore).
6. DMSO (EMD Millipore).
7. 1 μM ethylene glycol-bis(β-aminoethyl ether)-N,N,N0 ,N0 -tet-

raacetic acid; (EGTA) (Acros Organics).
8. 2.5 μM FM™ 1-43 dye N-(3-Triethylammoniumpropyl)-4-
(4-(Dibutylamino) Styryl) Pyridinium Dibromide)3,5
(Thermo Scientific).
9. Fluo-4AM Cell Permeant (Fisher Scientific).
10. TCS SP2 two-photon confocal scanning microscope (Leica).
11. 405 nm UV laser at 100% power on a HCX PL APO CS
63.0 1.40 oil-objective lens [TCS SP2 two-photon confocal
scanning microscope (Leica)].
2.4 Activity of 1. H2k A/J/ cells.

Carbohydrate 2. Gal-1 (0.11 μM, diluted in PBS).
Recognition Domain in
3. 35 mm Glass Bottom Microwell Dishes (MatTek).
Membrane Repair
4. 2.5 μM FM™ 1-43 dye (N-(3-Triethylammoniumpropyl)-4-
(4-(Dibutylamino) Styryl) Pyridinium Dibromide)3,5
5. TCS SP2 two-photon confocal scanning microscope (Leica).
6. 405 nm UV laser (TCS SP2 two-photon confocal scanning
microscope Leica).
7. HCX PL APO CS 63.0 1.40 oil-objective lens.
8. Lactose (20 mM) (Alfa Aesar) in PBS.
9. Sucrose (20 mM) (Carolina Biological) in PBS.
2.5 Labeling 1. H2k A/J/ cells.

Galectin-1 Through 2. Gal-1 (0.11 μM, diluted in PBS).
Primary Amines
3. Alexa Fluor 647 (Protein labeling kit (Molecular Probes)).
4. Wash buffer, 14 mM β-mercaptoethanol in PBS.
5. PD-10 Desalting Columns (GE Healthcare).
2.6 Confocal 1. CellBrite Green Cytoplasmic Dye (Biotium).

Visualization of 2. Hoechst 33342 (Thermo Scientific).
Labeled Galectin-1
3. EZBlock™ buffer (Biovision).
4. 4% paraformaldehyde (PFA) (Thermo Fisher Scientific).
5. ProLong™ Diamond Antifade Mountant (Invitrogen).
6. 18 mm circle glass coverslip.
7. Clear nail polish (Sinful Colors Professional Polish).
8. Confocal Wash Buffer: 10% Sea Blocking Buffer (Thermo),
90% PBST.
2.7 Muscle Fiber 1. 12-well plate (Greiner Bio-One).

Isolation 2. C57B6 and Dysf/ (B6.129-Dysftm1Kcam/J) mice (Jackson
Laboratories).
3. Mouse Surgical Kit (Kent Scientific Corporation).
4. 35 mm Glass Bottom Microwell Dishes (MatTek).
5. 0.11 μM Gal-1 (diluted in PBS).
6. FM™ 1-43 dye (Invitrogen).
7. Digestion solution: Collagenase Type 2, (40 mg/mL, diluted
in DMEM) (Gibco).
8. DMEM (DMEM (4500 mg/L glucose, 4 mM L-Glutamine,
110 mg/L Pyruvate) (Gibco).
9. PBS (Gibco).
3 Methods
3.1 Production of 1. Linearize the pET-29b(+) vector by mixing the following:

Recombinant Human 20 μL of the vector at 140 ng/μL, 22.5 μL of ultrapure
Galectin-1 water, and 5 μL of 10 CutSmart Buffer.
2. Digest the mixture by adding 1.5 μL of NdeI restriction
enzyme, then incubating for 8 h at 37 C. After 8 h, add to
the mixture 1.0 μL of NdeI restriction enzyme and continue
incubating overnight at 37 C (see Note 1).
3. After 24 h, complete the digestion by heating the mixture at
65 C for 20 min (see Note 2).
4. Resuspend the gBlock gene fragment LGALS1 by centrifuging
the tube for 3–5 s at a minimum of 3000 g. After centrifuga-
tion, add 50 μL of TE buffer to reach a final concentration of
10 ng/mL.
5. Prepare the assembly process. Mix the assembly, restriction
enzyme-digested vector, and the gBlock LGALS1 insert by
setting up the following reaction on ice: 1:2 vector: insert
DNA (molar ratio), 10 μL of NEB Builder HiFi DNA assembly
master mix, and 7 μL of deionized water. Incubate the sample
in the thermocycler at 50 C for 20 min (see Note 3).
6. Confirm the transformation process via agarose gel electropho-
resis. Prepare a 1% agarose gel and run ~5.0 μL of the digested
vector mixed with 1 SDS loading dye and ~5.0 μL of the
linearized vector in separate wells.
7. Transform the cloned vector into BL21(DE3) competent
E. coli cells following the NEB (C2527) heat-shock protocol
(see Note 4).
8. Place and spread ~50–100 μL of transformed bacteria onto an

agar plate previously prepared with kanamycin (50 μg/mL) and
incubate overnight at 37 C.
9. Select a colony from the agar plate and grow overnight in LB
broth. Lyse the bacteria and purify the pET 29b(+) plasmid
DNA by using a Mini Kit Protocol and confirm the sequence
(see Notes 5 and 6).
10. Prepare the starter culture by selecting a colony and placing in
sterile 10 mL LB broth/Kanamycin (50 μg/mL) overnight at
37 C with shaking at 250 rpm (see Note 7).
11. Take 10 mL of the starter culture and pour into 1 L of auto-
claved LB broth/kanamycin (50 μg/mL) and place it in a
shaker incubator set at 37 C and 250 rpm.
12. Check the optical density at 600 nm (OD600) of the culture
every 20–30 min using a spectrophotometer until the cultures
reach an OD of 0.5–0.6 and induce the Gal-1 expression by
using 0.1 mM of IPTG. Continue the IPTG induction over-
night at 30 C and 250 rpm.
13. Next day, collect bacteria in a 1 L plastic centrifuge container
and centrifuge down to a pellet at 14,000 RCF for 30 min at
4 C.
14. Discard the supernatant without disturbing the bacteria pellet
(see Note 8).
15. Add 10 mL of lysis buffer to the bottle and resuspend the pellet
until homogeneous and let the homogeneous mixture sit at
room temperature for 30 min.
16. Pour the resuspended sample in a 150 mL beaker. Place the
beaker on ice and sonicate 5 s on, 5 s off at 80% amplitude for
5 min, or until the sample appears cloudy.
17. After sonification, add Protease Inhibitor (use 100 protease
inhibitor, dilute to 1 in solution).
18. Spin the sonicated lysate at 13,000 RCF for 30–45 min at 4 C.
19. Remove the supernatant without disturbing the pellet. Save the
supernatant on ice or store at 20 C.
3.2 Purification of 1. Prepare a lactosyl Sepharose column by placing 100 mL of

Recombinant Human Sepharose 4B into a Buchner funnel containing filter paper
Galectin-1 under vacuum.
2. Wash the Sepharose with 0.5 M Na2CO3 until pH reaches
11 (see Note 9).
3. Transfer the dried Sepharose into a 250 mL beaker and add
100 mL of 0.5 M Na2CO3 to resuspend.
4. Add 10 mL of divinyl sulfone into the beaker slowly (2 mL

aliquots over 15 min). The reaction mixture should turn
slightly brown. Add a magnetic stir bar and stir lightly for
70 min (see Note 10).
5. Transfer reaction mixture to the Buchner funnel and wash with
100 mL of 0.5 M Na2CO3.
6. Transfer the dried and activated Sepharose into a beaker and
add 100 mL of 0.5 M Na2CO3 containing 10% lactose (w/v).
7. Add the magnetic stir bar and gently stir for 15 h (overnight).
8. After reaction is complete place again on the Buchner funnel
and wash with 100 mL 0.5 M Na2CO3 and 100 mL distilled
water. Wash with 100 mL column storage buffer (1 PBS
containing 0.02% sodium azide and 14 mM βME).
9. Obtain four Econo-Columns, 1.5 50 cm. First add 5–7 cm of
quartz sand and then add 15 cm of lactosyl Sepharose into each
column. Allow the gel to settle while tapping the sides of the
column to remove internal air bubbles from the column.
10. Obtain the bacterial supernatant from the 20 C freezer and
begin to thaw on ice.
11. Wash the lactosyl Sepharose column with 10 column volumes
of wash buffer (1 PBS containing 14 mM βME), approxi-
mately 80 mL (see Note 11).
12. Move the column(s) to a 4 C fridge and gently add the sample
lysate to the column, allowing it to pass through using gravity
(see Note 12).
13. Add 10 column volumes of wash buffer to remove unwanted
protein and cell contaminants from the column.
14. Remove the column from the 4 C fridge and add approxi-
mately 3 column volumes of elution buffer (1 PBS containing
14 mM βME and 100 mM lactose) to the column and begin
collecting 1.0–1.5 mL fractions in 1.5 mL microcentrifuge
tubes.
15. Set the absorbance to 280 nm on the spectrophotometer.
Using 1 cm quartz cuvette, blank the spectrophotometer
using wash buffer. Record each OD at 280 nm for each sample
(see Note 13).
16. Calculate the protein concentration using the formula:
OD280 path length extinction coefficient (see Note 14).
Fractions can be stored at 80 C.
17. To isolate functional Gal-1, prepare a PD-10 column by wash-
ing the column with 10 column volumes of PBS (see Note 15).
18. Carefully place 2 mL of protein containing fractions over the
PD-10. Let the mixture flow onto the gel and then elute
with PBS.
19. Capture 1.0–1.5 mL fractions and measure the OD280. Deter-

mine concentration using methods in step 16.
20. Store PD-10 column fraction in 4 C for no more than six
months.
3.3 H2K A/J/ (A/ H2K A/J cells were cultured as described in Morgan et al. [8] and
J/) and A/J+/+ (WT) Jat et al. [9].
Cell Culture
1. Prepare the growth media: 500 mL of DMEM (High Glucose,
L-Glutamine, Phenol Red, and Sodium Pyruvate), 1% Penicil-
lin/Streptomycin, 20% Hi-FBS, 2% Chick Embryo Extract,
and Interferon gamma (1) (see Note 16).
2. Warm the media at 37 C. Take a frozen cell vial from the liquid
nitrogen storage. Thaw frozen vial in the water bath (37 C)
until just a small ice pellet is visible (~2 min). Under sterile
conditions (hood), pour the cells into a 15 mL conical tube
containing 9 mL of warmed growth media and centrifuge for at
300 g for 5–7 min. Remove the supernatant without dis-
turbing the cell pellet. Resuspend the cell pellet by adding
2–4 mL of growth media and count the cells (see Note 17).
3. Prepare a 175 cm2 flask containing 20–30 mL of the warmed
growth media and 40–60 μL of the interferon gamma stock
(at 10 μg/mL). Seed at a density of 5555 cells/cm2. Place the
flask containing the cells into the incubator set at 33 C and
10% CO2.
4. Change the cell growth media every day.
5. Split the cells at 60–70% confluency. Remove the growth media
and wash the cells twice with 5 mL of warmed PBS. Add 5 mL
of warmed Trypsin/EDTA and incubate for ~5 min in the
incubator at 33 C. Add 3 mL pre-warmed plain DMEM per
1 mL of Trypsin/EDTA to inactivate the Trypsin/EDTA.
6. Place the content in a 50 mL conical tube and centrifuge at
300 g for 5–7 min. Remove the media without disturbing the
cell pellet.
7. Resuspend the cell pellet by adding 2–4 mL of growth media,
count the cells, and plate onto sterile 35 mm glass-bottomed
dishes at a density of 5555 cells/cm2 (total volume 2 mL/
plate). Incubate at 33 C in 10% CO2.
8. Differentiate the confluent myoblasts when they reach ~80%
confluency (see Note 18).
9. To differentiate, wash the confluent myoblasts twice with
warmed PBS.
10. Add differentiation media (2 mL/plate) with the respective
treatment (0.11 μM Gal-1, diluted in PBS) and incubate at
37 C and 5% CO2.
11. Change the differentiation media and treatments every other

day. After 2–4 days in differentiation media and treatment,
myotubes are ready for laser injury.
3.4 Laser Injury 1. Culture A/J WT and A/J/ cells as described in the cell
Assay and culture section and prepare for laser injury.
Quantification 2. Obtain fully differentiated myotubes in 35 mm glass bottom
dishes.
3. Prepare the FM1-43 or FM4-64 dye stock solution (200 μg/μ
L): Add 500 μL of cold ultrapure water to one vial (100 μg) of
the dye (see Note 19).
4. Prepare the TCS SP2 two-photon confocal scanning micro-
scope for the live cell imaging experiment: The live cell cham-
ber should have a temperature of 37 C and 5% ambient CO2.
5. Carefully wash one myotube dish with 1 mL of warm 1 PBS.
Add 25 μL of FM1-43 or FM4-64 dye to dish (for 2.5 μM final
concentration of dye) and incubate at room temperature for
5 min. Place 35 mm dish into confocal incubation chamber.
Open the Leica LASX Imaging Software and select the FRAP
application.
6. Select the following settings modified from Carmeille et al.
[10]: Dimensions: 512 512; pinhole: <0.8; zoom: 1; lens
selection: 63/1.40 oil; argon laser: 25%; 488 nm (FM1-43)
or 647 nm (FM4-64) laser power: on; intensity: 1.0; PMT: on;
Live image: On and focus on myotube or myofiber and
stop; Gain: 400–800. Bleach Settings: Visible light: off;
405 nm UV laser:100%; Select ROI:0.01 by placing cursor on
the membrane; Bleach Duration: myotubes (15 s), myofibers
(3 s). Time course settings: Pre-bleach image:1; Bleach:15 s or
3 s; Post-bleach image: 30; Image intervals: 3–5 s; Select Run
Experiment.
7. Select up to five different myotubes to be injured in each dish.
8. Export images as .TIFF. Open ImageJ and create the following
macro:
(a) Line 1: run(“Set Scale...,” “distance¼137 known¼50
pixel¼1 unit¼um global”);
(b) Line 3: n ¼ 31.
(c) Line 5: for(i ¼ 1; i < ¼ n; i++) {.
(d) Line 6: run(“Measure”).
(e) Line 7: if (i < n) {.
(f) Line 8: run(“Open Next”).
(g) Line 9: }.
(h) Line 10: }.
Fig. 2 Laser injury workflow. (a) A/J/ myoblasts are differentiated to form (b) myotubes. (c) BLA/J mice are
processed to yield (d) myofibers. (e) The myotubes or myofibers are injured via laser on a confocal
microscope, causing internal fluorescence change as dye enters into the cell. (f) The fluorescence change
is quantified using ImageJ
9. Open pre-image and under “analyze” tab select set scale. . .;

Distance in pixels: 137; known distance: 50 μm; check global.
Select plugin open Laser injury macro and select “run” (see
Note 20). Use free hand tool to outline region of interest (ROI
abbreviation used in ImageJ software) around total injured area
and select “run.” Copy “label,” “area,” and “mean gray value”
from output table and calculate the fluorescence change
(Δmean gray value area) in Microsoft Excel.
10. Analyze fluorescence change differences between treatment
groups using a multiple comparisons test in preferred statistical
program (Fig. 2).
3.4.1 Activity of 1. Follow the method for laser injury and quantification but
Carbohydrate Recognition replace wash in step 5 with PBS enriched with 20 mM lactose
Domain in Membrane (to bind CRD), or 20 mM sucrose (as a control) and incubate
Repair the myotubes for 10 min before injuring (Fig. 3).
3.4.2 Examination of 1. Refer to the laser injury section and replace step 3 with the
Galetin-1 Calcium following: Wash the myotubes with PBS and incubate the
Dependency in Membrane treatment groups in PBS enriched with: 1 mM Ca2+ (as CaCl2),
Repair 1 μM intracellular BAPTA-AM in DMSO as a vehicle, or 1 μM
EGTA, for 10 min.
2. Resuspend Fluo-4 AM by adding DMSO to a final working
concentration of 1–5 μM and incubate with myotubes for
15–60 min at 37 C before laser injury. Incubate with
FM4-64 for 5 min prior to injury.
3. Quantify Fluo-4 fluorescence change using the ImageJ meth-
ods mentioned above (Figs. 4 and 5).
Fig. 3 Laser injury assay in myotubes and myofibers under varying conditions. (a) Representative images of
A/J/ myotubes or (b) myofibers during laser injury assay at specific time points (0 s represents end of injury,
25 s represents 25 s before the end of injury). Quantification of change in fluorescence inside A/J myotubes
(c), (e) and myofibers (d), (f) after injury. Notice that these specific assays measure the change in internal
fluorescence among varying treatment groups. This research was originally published in PLoS ONE (Reference
[2])
Fig. 4 Galectin-1 influences membrane repair independent of calcium. (a) Laser injury quantification of
myotubes supplemented with or without Ca2+. (b) Laser injury experiment using EGTA as an extracellular Ca2+
chelator. (c) Laser injury experiment using BAPTA-AM as a cytosolic Ca2+ chelator. (d) Representative images
showing the Ca2+ independence of Galectin-1. Notice the lack of Ca2+ recruitment at the injury site (indicated
by arrow) in A/J/ myotubes. This research was originally published in PLoS ONE (Reference [2])
3.5 Labeling 1. Conjugate purified Gal-1 with Alexa Fluor 647 or fluorophore
Galectin-1 Amines and of choice using the protocol provided with the protein labeling
Confocal Visualization kit (Molecular Probes) with a few alterations as described in
Stowell et al. [11].
3.5.1 Labeling Galectin-1
Amines 2. Alterations in labeling Alexa Fluor 647 Gal-1:
(a) Desalt Gal-1 (3–5 mg/mL) through the PD-10 column
that has been equilibrated with PBS.
(b) Take a 400 μL aliquot of Gal-1 and mix it with 50 μL of
1 M lactose solution and 50 μL of 1 M NaHCO3 solution
(final concentrations 100 mM lactose and 100 mM
NaHCO3) (see Note 21).
Fig. 5 Calcium-dependent repair in A/J WT myotubes. (a) Representative images

of A/J WT myotubes. Fluor-4AM fluorescence at site of injury (shown by arrow)
indicates Ca2+ accumulation. (b) Quantification of the change in Fluor-4 AM
fluorescence over time at the site of injury. Error bars indicate SEM. This
research was originally published in PLoS ONE (Reference [2])
(c) Mix this solution with Alexa Fluor 647 at pH 7.4 and
incubate on a stir plate for 90 min.
(d) After the 90 min, pass through a Sephadex G25 column
that was previously equilibrated with 0.2 mM NaN3
in PBS.
(e) Test activity of the carbohydrate recognition domain by
running the labeled Gal-1 on a lactose-Sepharose column.
Any Gal-1 that does not adhere to the column did not
retain its carbohydrate recognition domain activity. Deter-
mine the concentration of Gal-1 and Alexa Fluor
647 labeled Gal-1 with absorbance at 280 nm or a
Pierce™ Coomassie (Bradford) Protein Assay Kit

3. Treat confluent myoblasts or differentiated myotubes with
Alexa Fluor 647 labeled Gal-1 Use for laser injury assay with
the following settings: apply FRAP settings as described above
and turn 633 nm laser on power 1 (Fig. 6).
3.5.2 In Vitro Confocal 1. Seed 5555 A/J/ cells/cm2 on a 35 mm glass culture dish.
Immunofluorescence Using Grow myoblasts to 80% confluent and differentiate and treat
Alexa Fluor 647 Labeled with 0.11 μM Alexa Fluor 647 labeled Gal-1. All steps includ-
Gal-1 ing Alexa Fluor 647 labeled Gal-1 treatment should be pro-
tected from light.
2. In a separate vial add 5 μL CellBrite™ Green Cytoplasmic
Membrane Dye per 1 mL of differentiation media. Prior to
fixation, remove the differentiation media and add 300 μL of
CellBrite™ solution to each plate and incubate for 20 min at
37 C.
3. Remove CellBrite™ staining media and wash two times with
500 μL of warm PBS.
4. Fix cells by adding just enough warm 4% PFA to cover the cells
(300–500 μL) for 10 min at 37 C at various time points during
differentiation (10 min, 4 h, 8 h, 12 h, 24 h, and 48 h).
5. Aspirate the 4% PFA and wash by adding 500 μL of PBS and
rock at room temperature for 5 min. Repeat wash two more
times.
6. Add 300 μL of Hoecht staining solution (0.1 μg/mL) and
incubate 10 min at room temperature. Wash with
300–500 μL of confocal wash buffer for 5 min at room tem-
perature, racking. Repeat wash two more times.
7. Mount cells by adding a small drop (~10 μL) of ProLong™
Diamond Antifade mountant and carefully cover using an
18 mm circle glass coverslip. Let the mountant dry overnight
and seal the coverslip sides with clear nail polish to secure the
coverslip in place.
8. Image on the confocal microscope using the appropriate set-
tings: Open Leica Imaging software and select TCS SP8 appli-
cation; Speed: >600; Frame Average: 3.00; Pinhole: 0.3; Z
Stack: 0.1–0.5 μm each plate; Turn laser on; Argon: 25%, [Set
Sequence 1: 405 nm UV laser: on; Power: 1.0%; Visible Light:
on, 633 nm laser power: 1.0; HyD 1: Hoechst, HyD 3: Alexa
Fluor 647 labeled Gal-1]. [Set sequence 2: Visible Light: on,
488 laser power: 1.0; HyD 2: CellBrite™ Green]; Zoom: 1.0;
Search Dimension: 512 512; Zoom: 4.0; Image Dimension:
2048 2048 (Fig. 6).
Fig. 6 In vitro galectin-1 localization using confocal microscopy. Representative images visualizing Galectin-1
after 10 min (a) and 48 h (b) myotube treatment and subsequent laser injury. Notice accumulation of Galectin-
1 fluorescence at site of injury. The white box indicates the area that is zoomed in on the images, and the
white arrow indicates the injury site. Zooms are approximately 2.75 larger than the original. (c) Immunoflu-
orescence experiment showing Galectin-1 lattice formation at 8 h post-treatment in A/J/ myotubes. The
gal-1 is shown in green, forming a lattice between fusing myotubes. Blue indicates the nuclei and red
indicates the cellular membrane. The x-z and y-z slices are also shown, indicating that the Gal-1 is uptaken by
the myotubes at some point between treatment and the time point indicated. This research was originally
published in PLoS ONE (Reference [2])
3.6 Muscle Fiber 1. Prepare a 12-well digestion plate in the following manner: For
Isolation each extracted muscle, fill 2 wells with 1 mL of plain DMEM
and 1 well with 1 mL of PBS. (This leads to a total of three wells
per digested muscle [12])
2. Prepare 100 μL of digestion solution for each muscle.
3. After preparation of digestion plate, euthanize mice in accor-
dance with approved protocols.
4. After euthanization, extract desired muscle groups (see Notes
22 and 23).
(a) Flexor Digitorum Brevis: Use pins to place hindfoot in
complete dorsiflexion. Make small incision at heel and
remove skin from bottom of footpad. Cut the proximal
flexor digitorum brevis (FDB) tendon as close to the heel
as possible and lift FDB body up and away from the foot.
Cut all tendon from digits.
(b) Tibialis Anterior: Place mouse supine and remove skin
from lower leg. Slide forceps under tibialis and separate
tibial attachments. Cut at proximal and distal tendons.
5. After extraction, place the muscle into a well with DMEM.
6. After all muscle groups are extracted, add an additional 100 μL
of digestion solution to each well containing a muscle.
7. Digest muscles for 60–120 min at 37 C (see Note 24).
8. Transfer muscles to a well with plain DMEM and allow to sit at
for 10–15 min at 37 C.
9. Transfer muscles to a well with PBS to allow for better visuali-
zation of the individual fibers. Under a dissection microscope,
locate properly digested myofibers and use a small-bore pipette
to transfer to a 35 mm Glass Bottom Microwell Dish. After
transferring the desired number of myofibers, bring the final
volume of the dish to 300 μL with warm PBS.
10. Allow the myofibers to attach to 35 mm dish for 10–15 min.
11. Add 15 μL of FM1-43 and allow to incubate for 5–7 min
before beginning injury.
12. Select at least three different myofibers in each dish to be
injured as described in Subheading 3.4. Viable muscle fibers
will have intact membranes and visible striations (see Note 25).
13. Measure the total change in fluorescence intensity of FM™
1-43 dye at the site of the wound for each time point relative to
the pre-injury fluorescent intensity using ImageJ (Fig. 3) [13].
4 Notes
1. Avoid thermocycling the restriction digest enzyme to keep at

optimum level of activity. Keep on ice or at 20 C, per
manufacturer’s instructions.
2. Make sure to set the right temperature and time during the
entire digestion process to maximize the effectiveness of
expression of the gene of interest.
3. The total volume of the assembly reaction must be 20 μL.
4. See the full transformation protocol for BL21(DE3) at https://
www.neb.com/protocols/0001/01/01/transformation-pro
tocol-for-bl21-de3-competent-cells-c2527.
5. DNA purification was performed with E.Z.N.A Plasmid DNA
Mini Kit I Protocol.
6. Sequence was confirmed by ETON Bioscience (San Diego,
CA).
7. Preparation of LB broth: For each liter of media, use 20 g of LB
broth powder. To sterilize the mixture, autoclave on a
20 min cycle and then allow it to cool to room temperature.
8. You can stop here. If so, keep the bacterial pellet at 80 C
(up to 2 weeks).
9. Measure the pH with a pH indicator strip. The mixture is solid
and can damage a pH meter probe.
10. Divinylsulfone is very toxic and all steps including it should be
conducted in a fume hood. Avoid skin contact. Sepharose
mixture will turn slightly pink.
11. For best results use gravity washing. To reduce washing time,
you can pressurize the column with compressed air.
12. To increase protein yield, pass the first flow through the col-
umn again after the elution step. The sepharose can become
saturated during the first flow through.
13. Be sure to use a quartz cuvette, plastic absorbs at 280 nm
which will cause false positive OD readings.
14. Protein concentrations can be determined using Bradford assay
(detection sensitivity limit around 2 mg/mL) or Beer’s Law.
Do not conduct a BCA assay to determine protein concentra-
tion. BCA assay is a very common protein assay which reduces
Cu2+ to Cu1+, resulting in color change. The elution buffer
contains two reducing reagents, 0.1% β-mercaptoethanol
(βME) and 100 mM lactose, which will create an artificial
signal and falsely increase the protein concentration. Know
the specific concentration at which your galectin will precipi-
tate out of solution and dilute in PBS as necessary.
15. PD-10 is a desalting column which will remove lactose and

βME.
16. Important: Interferon gamma (500) needs to be added at the
moment of plating. Add 2 μL of Interferon gamma/mL of
growth media.
17. Resuspended cells were counted by using the TC20 Auto-
mated Cell Counter (Bio-Rad) and Courting Slides, Dual
Chambers for Cell Counter (Bio-Rad).
18. We have observed myoblasts overcrowding when differentiat-
ing >90% in 35 mm glass-bottomed plates. This inhibited the
formation of mature myotubes.
19. The FM1-43 dye is light sensitive. Prepare and store the dye on
ice and under dark conditions.
20. This is to locate the injury site to accurately draw ROI around
total fluorescence change.
21. Lactose was used to prevent binding of Alexa Fluor 647 to the
carbohydrate recognition domain in order to maintain its func-
tional activity.
22. It is possible to digest larger muscle groups, such as the gas-
trocnemius or quadriceps, but the user must be careful not to
over digest. This can be mitigated by lower concentrations of
collagenase 2 in the digestion solution. We have found more
even digestion using smaller, thinner muscle groups.
23. When removing muscle groups, take care not to stretch the
muscle. This can result in broken myofibers.
24. When properly digested, the muscle bundles will have a frayed
or mop-like appearance.
25. Avoid injuring fibers with bends and kinks. Do not injure near
nuclei. FM1-43 will stay viable for up to 20 min.
Acknowledgments
This work was supported by the following Brigham Young Univer-

sity awards; Roland K. Robbins Research Fellowship to
M.L.V.Z. and Earl M. Wooley Research Innovation Award to
P.M.V.R. The authors also thank Dr. Sean Stowell and the Jain
Foundation for their support to P.M.V.R. Figures 1 and 2 images
were provided by BioRender.com.
References
1. Camby I, Le Mercier M, Lefranc F, Kiss R 7. Demonbreun AR, Quattrocelli M, Barefield
(2006) Galectin-1: a small protein with major DY, Allen MV, Swanson KE, McNally EM
functions. Glycobiology 16(11):137R–157R. (2016) An actin-dependent annexin complex
https://doi.org/10.1093/glycob/cwl025 mediates plasma membrane repair in muscle. J
2. Vallecillo-Zúniga ML, Rathgeber MF, Poulson Cell Biol 213(6):705–718
PD, Hayes S, Luddington JS, Gill HN, 8. Morgan J, Beauchamp J, Pagel C, Peckham M,
Teynor M, Kartchner BC, Valdoz J, Stowell C Ataliotis P, Jat P, Noble M, Farmer K, Partridge
(2020) Treatment with galectin-1 improves T (1994) Myogenic cell lines derived from
myogenic potential and membrane repair in transgenic mice carrying a thermolabile T anti-
dysferlin-deficient models. PLoS One 15(9): gen: a model system for the derivation of
e0238441 tissue-specific and mutation-specific cell lines.
3. Van Ry PM, Wuebbles RD, Key M, Burkin Dev Biol 162(2):486–498
DJJMT (2015) Galectin-1 protein therapy pre- 9. Jat PS, Noble MD, Ataliotis P, Tanaka Y,
vents pathology and improves muscle function Yannoutsos N, Larsen L, Kioussis D (1991)
in the mdx mouse model of Duchenne muscu- Direct derivation of conditionally immortal
lar dystrophy. Mol Ther 23(8):1285–1297 cell lines from an H-2Kb-tsA58 transgenic
4. Wuebbles RD, Cruz V, Van Ry P, Barraza- mouse. Proc Natl Acad Sci 88(12):5096–5100
Flores P, Brewer PD, Jones P, Burkin DJ 10. Carmeille R, Croissant C, Bouvet F, Bouter A
(2019) Human Galectin-1 Improves Sarco- (2017) Membrane repair assay for human skel-
lemma Stability and Muscle Vascularization in etal muscle cells. In: Skeletal muscle develop-
the mdx Mouse Model of Duchenne Muscular ment. Springer, pp 195–207
Dystrophy. Mol Ther Methods Clin Dev 13: 11. Stowell SR, Dias-Baruffi M, Penttil€a L,
145–153. https://doi.org/10.1016/j.omtm. Renkonen O, Nyame AK, Cummings RD
2019.01.004 (2004) Human galectin-1 recognition of
5. Chandra G, Defour A, Mamchoui K, poly-N-acetyllactosamine and chimeric poly-
Pandey K, Mishra S, Mouly V, Sreetama S, saccharides. Glycobiology 14(2):157–167
Ahmad MM, Mahjneh I, Morizono H (2019) 12. Demonbreun AR, McNally EM (2015) DNA
Dysregulated calcium homeostasis prevents electroporation, isolation and imaging of myo-
plasma membrane repair in Anoctamin 5/ fibers. J Vis Exp (106):e53551
TMEM16E-deficient patient muscle cells. Cell 13. Schindelin J, Arganda-Carreras I, Frise E,
Death Dis 5(1):1–15 Kaynig V, Longair M, Pietzsch T, Preibisch S,
6. Cheng X, Zhang X, Yu L, Xu H (2015) Cal- Rueden C, Saalfeld S, Schmid B (2012) Fiji: an
cium signaling in membrane repair. In: Semi- open-source platform for biological-image
nars in cell and developmental biology. analysis. Nat Methods 9(7):676
Elsevier, pp 24–31
Chapter 37
Exploring the Role of Galectins in Cancer: In Vitro

and In Vivo Approaches
Neus Martı́nez-Bosch, Noemı́ Manero-Rupérez, Mireia Moreno,
and Pilar Navarro
Abstract
Galectins have been linked to tumorigenesis since 1975, even before this family of proteins was given its
name. Since then, hundreds of papers have analyzed the role of different galectins in cancer development
and progression, deciphering their involvement in many different pathological events, from the regulation
of cell cycle, to angiogenesis, metastasis, and immune attack evasion. Importantly, the tumor galectin profile
is often altered in many cancers and aberrant levels of some of the members of this family have been
considered in diagnosis and frequently correlated with patient prognosis and clinicopathological character-
istics. In this chapter, we summarize most frequent techniques employed in cancer research to interrogate
the role of galectins, using Gal-1 to illustrate one member of the family and pancreatic cancer as an
experimental model. We will cover from techniques employed to detect their expression (tissue and
blood samples) to the most frequent tools used to change expression levels and the cell line-based
in vitro studies and murine preclinical models used to explore their role in tumor progression and/or
clinical translation.
Key words Galectin-1, Cancer, Knockdown, Downregulation, Overexpression, Immunohistochem-

istry, ELISA, Biomarker, Migration, Invasion, Viability, Proliferation, Anchorage independent growth,
PCR, Genetically engineered mouse
1 Introduction
Altered patterns of glycosylation in cancer were recognized as early

as 1949 and have gained researchers’ attention since then
[1]. Changes in O- and N-glycan core structures as well as terminal
glycan motifs occur upon neoplastic transformation [2], coating
the cell with a unique fingerprint that will modify physiological
ligands and will recruit novel partners, triggering cancer develop-
ment and progression [3]. Galectins are a family of proteins that
recognize N-acetyllactosamine residues and are thus key in the
recognition of altered cancer-associated glycoproteins [4]. They
are structurally organized in three different subgroups including
685
686 Neus Martı́nez-Bosch et al.
prototype galectins (Gal-1, Gal-2, Gal-5, Gal-7, Gal-10, Gal-11,

Gal-13, Gal-14, and Gal-15), chimeric (Gal-3) and tandem-repeat
galectins (Gal-4, Gal-6, Gal-8, Gal-9, Gal-12) [4]. Abnormal galec-
tin expression has been reported in most tumor types [5] and
thousands of papers have linked galectins to cancer biology, with
special emphasis on particular members of the family, including
Gal-1 and Gal-3, and more recently Gal-9. Importantly, galectins
have been related to the major hallmarks of cancer [6], playing
critical roles in key events such as proliferation, angiogenesis, and
immune evasion mechanisms [7]. Gal-1 was the first galectin mem-
ber identified and since then several groups have proven its impor-
tance in cancer, highlighting its potential in cancer diagnosis,
prognosis, and therapy [8, 9]. Tissue and blood enhanced levels
of Gal-1 have been reported in many tumors and frequently related
to poor prognosis, including pancreatic, glioma, lung, lymphoid,
colon, and thyroid cancer [5, 10]. Importantly Gal-1 is highly
expressed by activated fibroblasts in the tumor microenvironment
and can be secreted by a non-conventional mechanism which
results in its accumulation in the stroma, where it orchestrates the
crosstalk between fibroblasts, tumor, endothelial, and immune
cells, favoring tumor development and progression [7, 11–
15]. Remarkably, Gal-1 has been shown to promote tumor cell
transformation, proliferation, migration, fibroblast activation,
metastasis, angiogenesis, and immune evasion during cancer pro-
gression [7, 16]. Basic cellular biology and preclinical studies using
cancer cell lines and mouse models have revealed some of the
molecular mechanisms responsible for these pathological effects
triggered by Gal-1, including Ras interaction and downstream
signaling pathway activation triggering cell transformation
[17, 18], binding to integrins, laminin, and fibronectin modulating
cell/cell adhesion, migration, and invasion [19], interaction with
CD45, CD43, and CD7 in the membrane of T cells to induce
specific apoptosis of Th1 and Th17 CD4+ as well as CD8+ T
cells, driving tumor immune evasion [20, 21], and activation of
VEGFR2 to induce angiogenesis [22], among many others. Many
of these reports have used downregulation or upregulation strate-
gies as tools to analyze the role of Gal-1 in cancer progression. This
chapter will focus on the methods necessary to study the role of
galectins in cancer, illustrated using Gal-1 as a reference member of
the family and pancreatic cancer as tumor type. We will review the
techniques to analyze galectin expression in human tissue and
blood samples, the tools to change expression levels of galectins
in vitro, and the functional assays that can shed light on how they
can regulate several key events for tumor progression. We will also
address in vivo experiments to elucidate the importance of galectins
in mouse models of tumorigenesis to offer a whole picture on how a
complete study of galectins in cancer can be approached.
Exploring the Role of Galectins in Cancer: In Vitro and In Vivo Approaches 687
2 Materials
2.1 Galectin 1. Microtome (Leica).

Immuno- 2. Superfrost adhesion glass slides (Thermo Scientific).
histochemistry in
3. Xylene (Merck KGaA), Ethanol (Merck KGaA), Distilled
Paraffin-Embedded
water.
Tissue Sections
4. Sodium Citrate Buffer 10 mM pH 6.0 (Merck KGaA).
5. PBS and PBS with 1% BSA (Merck KGaA).
6. 3% Hydrogen peroxide (Merck KGaA).
7. Rabbit monoclonal primary antibody against Gal-1 (Abcam,
0.3 μg/mL) (Table 1).
8. Envision System anti-rabbit-HRP (Dako).
9. DAB+ chromogenic substrate (Dako).
10. 20% Hematoxylin (VWR).
11. DPX Mounting Media (Merck KGaA).
12. Glass coverslides (Menzel-Gl€aser).
2.2 Gal-1 1. Sterile PBS.

Downregulation and 2. DMEM with 10% FBS, L-Glutamine 2 mM, Sodium pyruvate
Overexpression 1 mM, Penicillin 100 U/mL (Gibco).
Strategies
3. 0.25% Trypsin EDTA solution for cell culture (Gibco).
2.2.1 Gal-1 Transient 4. 0.4% Trypan Blue solution (Merck KGaA).
Downregulation in
5. Lipofectamine 3000 reagent (Invitrogen).
Adherent Human Cancer
Cell Lines 6. Opti-MEM, serum-free media (Gibco).
7. Gal-1 specific human ON-TARGETplus SMARTpool siRNA,
ON-TARGETplus Non-targeting Control Pool (Dharmacon).
2.2.2 Gal-1 Stable 1. DMEM with 10% FBS, L-Glutamine 2 mM, Sodium pyruvate
Downregulation in 1 mM, Penicillin 100 U/mL (Gibco).
Adherent Human Cancer 2. Puromycin, at 1 mg/mL (Merck KGaA).
Cell Lines
3. pLKO.1-puro vector, MissionRNAi, TRCN0000057423-427,
non-targeting shRNA control (SHC002) (Merck KGaA).
4. Polyethylenimine (PEI), 1 mg/ mL (Merck KGaA).
5. Sterile filtered NaCl 150 mM (Merck KGaA).
6. Lentiviral packaging vectors: pRSV-Rev, pCMV-VSV-G,
pMDLg/pRRE.
7. Polybrene 0.8 mg/mL (Merck KGaA).
2.2.3 Stable 1. DMEM with 10% FBS, L-Glutamine 2 mM, Sodium pyruvate
Overexpression of Gal-1 in 1 mM, Penicillin 100 U/mL (Gibco).
Adherent Human Cell Lines 2. Polyethyleneimine (PEI), 1 mg/mL (Merck KGaA).
Table 1
Antibody details
Application (antigen Secondary

Antibody Commercial Reference Concentration retrieval) antibody
Galectin-1 Abcam 138513 1:1000 IHC Anti-rabbit
(citrate pH 6.0) envision
1:10,000 Western blot Anti-rabbit-
HRP
Ki67 BD 550609 1:200 IHC Anti-mouse
Biosciences (citrate pH 6.0) envision
P-Histone Millipore 04-817 1:300 IHC Anti-rabbit
H3 (citrate pH 6.0) envision
(Ser10)
α-SMA Sigma A2547 1:400 IHC Anti-mouse
CD31 Abcam ab231436 1:100 IHC Anti-rabbit
vWF Neomarkers 1:200 IHC (pepsin 0.1% in HCl Anti-rabbit
(Factor VIII 0.1 M, 20 min 37 C) envision
Ag Ab-1)
CD3-e Santa Cruz SC-1127 1:50 IHC Im-press
(citrate pH 6.0) anti-goat
CD8 Abcam ab203035 1:250 IHC Anti-rabbit
MPO Dako A-0398 IHC Im-press
(citrate pH 6.0) anti-goat
F4/80 Bio-rad MCA497 1:35 IHC Anti-rat
(citrate pH 6.0) HRP (1:
50)
Cleaved R&D AF-835 1:150 IHC Anti-rabbit
Caspase 3 (citrate pH 6.0) envision
BrdU Santa Cruz SC- 1:400 ICF Alexa-488
32323
IHC immunohistochemistry
3. OptiPro Serum-Free Media (Gibco).

4. pBabe plasmid.
5. Retroviral packaging vector pCMV-VSV-G.
6. Puromycin 1 mg/mL (Merck KGaA).
7. Polybrene 0.8 mg/mL (Merck KGaA).
2.3 In Vitro 1. PBS with 0.25% Trypsin EDTA solution for cell culture
Functional Analysis (Gibco).
2.3.1 Cancer Cell 2. DMEM with 10% FBS, L-Glutamine 2 mM, Sodium pyruvate
Proliferation 1 mM, Penicillin 100 U/mL (Gibco).
3. 0.4% Trypan Blue solution (Merck KGaA).
Crystal Violet Staining 4. 4% Paraformaldehyde solution (in PBS) (Merck KGaA).
5. 0.05% Crystal violet solution (Merck KGaA).
6. 10% Acetic acid (Merck KGaA).
BrdU Incorporation Assay 1. Sterile coverslips (Paul Marienfeld GmbH & Co.).
2. PBS with 0.25% Trypsin EDTA solution for cell culture
(Gibco).
3. DMEM with 10% FBS, L-Glutamine 2 mM, Sodium pyruvate
1 mM, Penicillin 100 U/mL (Gibco).
5. 4% Paraformaldehyde solution (in PBS) (Merck KGaA).
6. BrdU at 40 μM (Merck KGaA).
8. HCl 4 M (Merck KGaA).
9. 0.3% Triton X-100 prepared in PBS (Merck KGaA).
10. 5% BSA and 0.1% Tween 20 prepared in PBS (Merck KGaA).
11. Anti-BrdU (0.5 μg/mL) (Santa Cruz Biotechnology)
(Table 1).
12. Anti-Mouse Alexa Fluor 488 (Invitrogen): 10 μg/mL in PBS
(protect from light).
13. DAPI (0.25 μg/mL) in PBS (Merck KGaA).
14. Fluoromount-G (SouthernBiotech).
15. Microscope slides (Thermo Scientific).
2.3.2 Cancer Cell 1. PBS with 0.25% Trypsin EDTA solution for cell culture
Migration and Invasion (Gibco).
Cell Migration Using 1 mM, Penicillin 100 U/mL (Gibco).
Scratch Assay
4. 200 μL pipette tips.
5. DMEM with 1% BSA (Merck KGaA).
6. Image J software analysis.
Cell Migration Assay in 1. PBS with 0.25% Trypsin EDTA solution for cell culture
Transwells (Gibco).
4. DMEM with 1% BSA (Gibco).
5. Transwell plates with 8.0 μm membrane pore (12/plate:
6.5 mm diameter) (Cultek).
6. 1% Glutaraldehyde in PBS (Merck KGaA).
7. 0.2% Crystal violet (Merck KGaA).
8. Cotton ear sticks.
Cell Invasion Assay in 1. PBS with 0.25% Trypsin EDTA solution for cell culture
Transwells (Gibco).
4. DMEM with 1% BSA (Gibco).
5. Transwell plates with 8.0 μm membrane pore (12/plate:
6.5 mm diameter) (Cultek).
6. 1% Glutaraldehyde in PBS (Merck KGaA).
7. 0.2% Crystal violet (Merck KGaA).
8. Cotton ear sticks.
10. Matrigel diluted in PBS (1:20) (Corning).
2.3.3 Soft Agar Colony 1. PBS with 0.25% Trypsin EDTA solution for cell culture
Formation Assay (Gibco).
3. 2% Sterile noble agar solution (Merck KGaA).
5. 0.005% Crystal violet or MTT 1 mg/mL (Merck KGaA).
2.4 Preclinical 1. PBS with 0.25% Trypsin EDTA solution for cell culture
Strategies (Gibco).
2.4.1 Tumor Generation 2. DMEM with 10% FBS, L-Glutamine 2 mM, Sodium pyruvate
by Cell Line Injection in 1 mM, Penicillin 100 U/mL (Gibco).
Mice 3. 0.4% Trypan Blue solution (Merck KGaA).
4. Matrigel (Corning).
5. pLHC-luciferase plasmid.
6. Xenolight D-Luciferin potassium salt: 100 μL of Luciferin
(16 mg/kg).
7. Balb/c Nude Mouse (JAX Mice Strain).
2.4.2 Development of 1. Proteinase K Extraction buffer: Tris HCl 20 mM, EDTA

Tissue-Specific Oncogene- 5 mM, pH 8.0, 0.5% SDS, NaCl 200 mM (Merck KGaA),
Driven Tumors in Galectin proteinase K (500 μg/mL) (Roche).
KO Background 2. Cold isopropanol (Merck KGaA).
3. 70% Ethanol (Merck KGaA).
4. TE Buffer: Tris HCl 10 mM, EDTA 1 mM pH 8.0
(Merck KGaA).
5. PCR: dNTPs 200 μM Invitrogen, EcoTaq polymerase (1.5 U),
Transcription buffer 1, MgCl2 2.5 mM buffer (Biotools),
galectin specific primers (0.8 μM 50 Gal-1, 0.4 μM for 30
Gal-1 and 30 Neo) (Merck KGaA).
6. Ela-myc transgenic mice [23], Ela-KRas transgenic mice [24],
Gal-1 knockout mice [25].
2.4.3 Data Collection and Tumor processing

Analysis for Preclinical
l Formalin (Leica).
Studies
l Xylene (Merck KGaA), Ethanol (Merck KGaA), Distilled water.
l Paraffin (VWR).
l Optimal cutting temperature compound (OCT compound)
(Tissue-Tek).
l 30% Hematoxylin (VWR), Acid alcohol solution: 80% ethanol,
0.15% HCl, 0.2% ammonia water solution (Merck KGaA).
l Antibody information is detailed in Table 1.
l RNAlater (Merck KGaA).
l PBS EDTA 1 μM (Merck KGaA), DNase type I 2 U/mL (Merck
KGaA), collagenase IV 1 mg/mL (Life Technologies).
l 70 μm cell strainer (Falcon).
l FBS with 10% DMSO (Merck KGaA).
l Antibodies for FACS analysis.
2.5 Galectin-1 1. EDTA-treated tubes (Sarstedt).

Determination in Blood 2. ELISA Human/Mouse Galectin-1 Immunoassay (R&D
Samples Systems).
3 Methods
3.1 Galectin 1. Section paraffin-embedded tissue blocks to 3 μm thickness and

Immuno- place them over superfrost adhesion slides.
histochemistry in 2. Place slides in a rack and into an oven at 60 C for 30 min.
Paraffin-Embedded
3. Perform deparaffination of the tissue by incubating slide racks
Tissue Sections in solvents of increasing polarity, starting with xylene (twice,
15 min each), and following with a sequence of ethanol/water
mixture (from 100% ethanol to 95%, 70%, 50%), 5 min each.
End by placing slides in distilled water to completely rinse off
alcohol.
4. Perform antigen retrieval by preheating 2.5 L of citrate buffer
pH 6.0 in a pressure cooker and incubating the slides at 120 C
for 15 min.
5. Temper slides to room temperature.
6. Wash three times in water and mount slides in cover plates,
placing them in immunostaining racks.
7. Check cover plates are well assembled by adding 2.5 mL of
water and ensuring water falls gradually and slowly through the
slide.
8. Quench endogenous peroxidase by incubating slides with
2 mL of 3% hydrogen peroxide (dissolved in water), for 10 min.
9. Wash twice with water, once with PBS.
10. Block slides with PBS containing 1% BSA.
11. Incubation with rabbit monoclonal antibody (mAb) anti-Gal1
(0.3 μg/mL), 4 C, overnight. Ensure adding negative con-
trols with irrelevant anti-rabbit antibody at the same dilution.
12. Wash three times with PBS.
13. Add 150 μL of secondary antibody (Envision anti-rabbit, 1 h).
15. Develop with DAB+ substrate chromogen system (see Note 1).
16. Counterstain tissue with hematoxylin.
17. Dehydrate slides in an inverse sequence of ethanol solutions
[50%, 70%, 95%, 100% (twice)], 5 min each, followed by two
incubations with xylene (15 min each).
18. Mount slides with a drop of DPX Mounting Media.
19. Let slides dry in the laboratory hood for 48 h before observing
them under a light microscope (Fig. 1).
3.2 Gal-1 Gal-1 overexpression and downregulation in cells in culture have

Downregulation and been very useful in order to interrogate the role of the lectin in cell
Overexpression physiology in vitro and in vivo. Different strategies are possible for
Strategies decreasing the levels of Gal-1 in cells including transient or stable
Fig. 1 Galectin immunohistochemistry in tumors. Gal-1 expression in different

tumors: pancreatic ductal adenocarcinoma (PDA), glioblastoma, prostate cancer,
and bladder cancer. Scale bars, 200 μm
downregulations using siRNA and shRNA through Dicer and

RNA-induced silencing complex (RISC) as well as CRISPR-Cas9
knockouts. Gal-1 overexpression has been achieved by means of
transfection or infection of eukaryotic expression vectors encoding
the galectin.
3.2.1 Gal-1 Transient 1. Human cancer cell lines are cultured using DMEM with 10%
Downregulation in FBS 2 mM L-Glutamine, 1 mM sodium pyruvate, 100 U/mL
Adherent Human Cancer Penicillin/streptomycin (complete DMEM), 5% CO2, 37 C
Cell Lines (see Note 2).
2. Wash adherent cells with PBS and detach them using trypsin
0.25% trypsin EDTA solution.
3. Neutralize trypsin by resuspension with complete DMEM.
4. Centrifuge cells at 120 g for 5 min.
5. Resuspend cells with reduced volume of complete DMEM and
count cells using a Neubauer chamber after Trypan blue
staining.
6. Seed cells around 50–70% confluence (per well with
40,000–150,000 cells in 24 well plates) (see Note 3).
7. After 24 h, if cells are well spread and confluent as expected,
perform transfection with SMARTpool siRNA (Dharmacon)
against Gal-1 (or irrelevant siRNA as a control) (see Note 4).
8. Dilute Lipofectamine™ (1 μL) in 25 μL of serum-free medium

(Opti-MEM) and mix vigorously (see Note 5).
9. Dilute siRNA (50 nM final concentration in the well) in 25 μL
of serum-free medium and mix.
10. Combine lipofectamine and siRNA dilutions and mix gently.
Incubate for 15 min at room temperature.
11. Dilute the mixture with 250 μL of serum-free medium.
12. Wash cells twice with serum-free media and add the mixture
containing lipofectamine–siRNA complexes.
13. Incubate cells at 37 C in a 5% CO2 incubator for 3 h and
replace transfecting medium with complete DMEM.
14. Knockdown starts to be observed after 24 h and may last up to
5 days depending on cell division rate and transfection con-
fluency. Efficiency can be assessed by collecting protein extracts
in Laemmli Buffer and performing electrophoresis in 15%
acrylamide gels and Western blot analysis (WB) incubating
overnight with anti-Gal1 antibody (Table 1) (see Note 6)
(Fig. 2).
Fig. 2 Modulation of Gal-1 levels in pancreatic cancer cell lines. (a) Gal-1 stable downregulation. PANC-1 and
SK-PC-1 control cells (Ctl), cells infected with lentiviral particles delivering a scramble shRNA (shSC) or
5 different shRNAs sequences targeting Gal-1 were analyzed by electrophoresis and Western blot. (b) Gal-1
stable overexpression: RWP-1 pancreatic cancer cells were transfected with Galectin-1 (Gal-1, 2 clones) or
with the empty vector (pcDNA3) (Ctl, 2 clones) and the expression levels were detected by Western blot.
Tubulin was used for normalization. Band densitometries, quantified by Image J analysis, are shown in the bar
charts below
3.2.2 Gal-1 Stable 1. Cancer cell lines ready for Gal-1 downregulation and HEK
Downregulation in 293 T cells used for lentiviral delivery are cultured upon regular
Adherent Human Cancer cell culture conditions, using complete DMEM (see Note 8).
Cell Lines (See Note 7) 2. Before transfection, a puromycin titration must be performed
for each cell line to be infected, to ensure proper selection after
shRNA insertion. For this, seed cells at different confluences in
24 well plates and treat them with increasing amounts of the
antibiotic (can start from 0.5 to 5 μg/mL). Selection should be
achieved after treating cells with antibiotic for 3–5 days.
3. Seed 293 T cells to be around 50% confluency in 6 well plates
(one well for each shRNA to be delivered (see Note 9) includ-
ing an shRNA scramble as a control) (see Notes 10 and 11).
4. Transfect 293 T by means of polyethylenimine (PEI): Dilute
lentiviral packaging vectors (0.2 μg pRSV-Rev, 0.2 μg pCMV-
VSV-G, 0.6 μg pMDLg/pRRE, and 1 μg MISSION®
pLKO.1-pura shRNA pLKO.1) in 190 μL NaCl 150 mM.
Mix gently.
5. Add 78 μg/mL of PEI, vortex, and incubate at room tempera-
ture for 30 min.
6. Add the mixture into the wells.
7. After 24 h, add fresh complete DMEM.
8. Seed the human cell line to be infected in 6 well plates (one for
each shRNA, one for the shRNA scramble control, one to be
left as non-infected control, and an additional one as a control
for effective puromycin selection).
9. 48 h after transfection, collect 293 T supernatants and replace
with fresh medium. Filter virus suspensions with 0.45 μm filters
and add polybrene (8 μg/mL).
10. Add filtered supernatants over human cancer cell line to be
infected.
11. After 8 h, discard virus supernatants and replace it by fresh
medium.
12. Lentivirus collection and supernatant processing can be
repeated 72 h after transfection to reinfect cells and improve
infection efficiency. Alternatively, virus can be centrifuged and
kept frozen for future experiments.
13. Add puromycin at the optimized concentration. Ensure a con-
trol well is left without antibiotics (as a non-infected control).
14. Replace puromycin every 48 h until complete selection is
achieved (5 days). Split cells if necessary, seed a 24 well plate
for protein extraction and quantification of Gal-1 levels by
western blot (see Note 12). Freeze low passage cells to avoid
compensatory mechanisms in future functional assays.
For CRISPR/Cas9 information, see Note 13.
3.2.3 Stable 1. Cancer cell lines ready for Gal-1 overexpression are cultured
Overexpression of Gal-1 in upon regular cell culture conditions, using complete DMEM.
Adherent Human Cell Lines 2. If cells are easily transfected, an expression vector like
pcDNA3.1 encoding Gal-1 sequence can be inserted by simple
PEI mediated transfection into cells similar to what has been
detailed in Note 5 (Fig. 2).
3. If infection is necessary, culture the second-generation retrovi-
rus producer Phoenix-AMPHO cell line and grow them in
complete DMEM.
4. Seed Phoenix-AMPHO cell line to achieve 90% confluence in
6 well plates.
5. Transfect them using PEI by preparing:
(a) PEI mix in OptiPro Serum-Free Medium (300 μL).
(b) Plasmid mix (C824 (1 μg) and empty pBABE-puro (con-
trol) or pBabe-Gal-1 (10 μg)) in OptiPro SFM Media.
6. Mix PEI and plasmid mix and incubate 30 min at RT.
7. Change Phoenix-AMPHO medium to OptiPro Serum-Free
Medium.
8. After 30 min, add the PEI/plasmid mixture into cells (see Note
11).
9. Replace medium after 5 h of infection to complete DMEM.
10. Change cell culture medium after 24 h and 48 h.
11. Seed cells to be infected at 50% confluency.
12. Collect Phoenix-AMPHO medium, centrifuge 5 min 120 g,
and retain the supernatant.
13. Filter with 0.45 μm filter and add polybrene to a final concen-
tration of 8 μg/mL.
14. Add supernatants to cells to be infected.
15. Infection can be repeated 24 h afterwards.
16. Add puromycin at the concentration optimized after titration.
17. Split, seed for protein extraction and freeze low passage cells.
18. Clones should be isolated for proper functional performance
(Fig. 2). Cell cloning can be performed following different
strategies depending on the cell line. If cells tolerate low cell
number seeding, limited dilution is the preferable option. In
this case, one cell per well is placed in 96 well plates and wells
are screened and labeled as positives if the following day they
just contain 1 single cell attached. Discard empty wells or wells
containing more than one cell. Cells are left to divide and grow
to full in the well and then cells are split into two wells (one for
protein extraction and Gal-1 detection by WB, and a second
one for maintenance). If single cells are unable to divide and
grow, then 300 cells can be seeded in a P100 plate, but ensure
that cells are well distributed and separated. Isolated macro-
scopic clones will be formed after 2–4 weeks. Under the micro-
scope (inside the hood), aspirate clones with a pipette and place
them in separate wells in 24-well plates. Once grown, split and
seed two wells, as explained above.
19. Select clones according to Gal-1 expression levels, assessed by
15% acrylamide gel electrophoresis and WB analysis. Band
densitometry can be quantified using ImageJ analysis (Fig. 2).
3.3 In Vitro Both downregulation and upregulation strategies are key to analyz-
Functional Assays ing the role of Gal-1 in several functional assays that set the basis to
measure the tumorigenic properties of cancer cells. Typical func-
tional assays in this context include proliferation, migration, inva-
sion, and anchorage independent growth.
3.3.1 Cancer Cell Several reports have described how Gal-1 regulates cell growth [26]
Proliferation and indeed Gal-1, Gal-3, and Gal-7 have been reported to regulate
sustained proliferation while evading growth suppressors
[7]. There are different analyses that allow monitoring tumor cell
growth in vitro. Thousands of papers have relied on measuring cell
viability through MTT assays which is an absorbance-based assay
that measures the mitochondrial function of live cells. Though, as
metabolism alterations have been reported in tumor cells, results
could lead to misinterpretations. In this chapter, we will address
crystal violet staining and BrdU incorporation.
Crystal Violet Staining The best and easiest way to monitor cell proliferation in adherent
cells is by counting cells seeded on a plate during a determined time
lapse. This can be technically simplified by staining cells attached to
the well with crystal violet. This dye stains proteins and DNA and
measurement can be easily done by absorbance at 590 nm. In most
tumor cells in culture, which rarely undergo apoptosis, differences
in cell viability can be attributed to changes in cell proliferation.
1. Trypsinize control cells and cells with Gal-1 levels up/downre-
gulated, count cells using a Neubauer chamber, and discard
dead cells stained with trypan blue. Seed 10,000 cells in 12-well
plates in complete DMEM, in triplicates (one plate for each day
to be measured, for instance, 4 plates) (see Notes 14–16).
2. After attachment and before cell division has had time to occur
(depending on the cell type but usually between 30 min and
6 h), check cells under the microscope to ensure cell number
was accurate between different cell types (see Note 17).
3. After 24 h, process the first plate. Start by checking cells under
the microscope and aspirating the supernatant, taking a record
with pictures is highly recommended.
4. Wash cells with PBS (1 mL).

5. Fix cells in wells with 4% paraformaldehyde for 10 min
(400 μL/well).
6. Aspirate fixation solution and wash cells twice with PBS.
7. Add crystal violet staining solution (0.05%, in water), 400 μL/
well, for 30 min.
8. Rinse wells with water until clear.
9. Extract crystal violet with acetic acid (10% in water), 400 μL/
well, 10 min in an orbital shaker.
10. Transfer 50 μL of the well content to a 96-well plate and read
absorbance at 590 nm in a multiplate reader. Perform known
dilutions if saturated signals are detected.
BrdU Incorporation In most tumor cells in culture, which rarely undergo apoptosis,
differences in cell viability can be attributed to changes in cell
proliferation. Otherwise, a method directly interrogating tumor
cell cycle must be performed such as BrdU staining by immuno-
cytofluorescence (Fig. 3).
1. Trypsinize control cells and cells with Gal1 levels up/downre-
dead cells stained with trypan blue. Seed 30,000 cells in 24-well
plates containing glass cover slides in complete DMEM, in
triplicates.
2. After attachment and before cell division has had time to occur
(depending on the cell type but usually between 30 min and
6 h), check cells under the microscope to ensure cell number
was accurate between different cell types (see Note 17).
3. Replace medium by fresh complete DMEM containing 40 μM
BrdU and incubate for 10 min at 37 C in the incubator (see
Note 18).
4. Wash cells with PBS (0.5 mL).
5. Fix cells in wells with 4% paraformaldehyde for 10 min
(300 μL/well).
6. Aspirate fixation solution and wash cells twice with PBS.
7. Add HCl 4 M to hydrolyze DNA for 10 min.
8. Wash cells three times with PBS (0.5 mL).
9. Permeabilize cells with Triton X-100 0.3% in PBS, for 15 min.
10. Block cells with PBS with 5% BSA, 0.1% Tween20, for 1 h.
11. Incubate 1 h with primary antibody with mouse mAb anti-
BrdU [0.5 μg/mL final dilution, Santa Cruz Biotechnology
(Table 1)]. To optimize antibody use, a single drop of 30 μL of
the antibody dilution can be placed in an empty 24 well plate
and then place the coverslip upside down.

13. Incubate for 1 h with secondary antibody using anti-mouse
Alexa Fluor 488 (Invitrogen) at a concentration of 10 μg/mL.
Place a 30 μL drop in a 24 well plate and place the coverslip
upside down, protected from light.
14. Stain cell nuclei with DAPI (0.25 μg/mL) in PBS, for 3 min
(300 μL/slide).
15. Wash coverslips twice with PBS and once with water. Mount
them with 30 μL of Fluoromount-G over glass slides, pro-
tected from light.
16. Observe cells under the microscope with the appropriate
filters.
17. Manually or automatically quantify BrdU positive cells versus
total number of cells (DAPI positive) to obtain the fraction of
cells that are in S-phase in each condition.
These experiments can be directly performed with stably trans-
fected cells expressing Gal-1 or Gal-1 shRNA (as explained above)
or could be also performed by first performing siRNA transfection
previously and then following BrdU incorporation or crystal violet
staining.
3.3.2 Cancer Cell Galectin binding to glycan conjugates found in cell membrane
Migration and Invasion receptors and proteins from the extracellular matrix has been exten-
sively linked to the regulation of migration, invasion, and metastasis
[7]. There are different assays that allow monitoring cell migration.
Cell Migration Using This assay allows assessing migration in a very simple and inexpen-
Scratch Assay sive way by wounding the surface of a confluent monolayer and
analyzing how cells cover the gap over the time (Fig. 3).
1. Trypsinize control cells and cells with Gal1 levels up/downre-
dead cells stained with trypan blue. Seed 80,000 cells (see Note
19) in 24-well plates in complete DMEM, in triplicates.
2. When cells reach confluency, scratch the monolayer with a
200 μL micropipette tip (defining a straight line).
3. Perform two additional perpendicular straight lines to form
two X shapes per well. This will label two exact points (both
crossings) that will be easily found under the microscope dur-
ing cell migration.
4. Wash cell debris with DMEM without serum, twice (500 μL).
5. Add DMEM with 1% BSA to the wells and take pictures of the
X zone (assigned to time zero) under the microscope. Two
pictures per well will be taken. FBS is removed from the migra-
tion medium to avoid differences in proliferation that interferes
with the assay.
Fig. 3 In vitro functional assays to analyze cell proliferation, migration, and anchorage independent growth in
cell lines with altered levels of Gal-1. (a) PANC-1 control cells (Ctl) or cells downregulated for Gal1 by infection
with lentivirus encoding Gal-1 shRNA (shGal-1) were studied for proliferation, using BrdU immunofluorescence
(a), for migration, using wound healing experiments after 24 h (b), and for anchorage independent growth
using soft agar colony formation experiments processed by MTT staining after 30 days (c)
6. Leave cells in the incubator and take pictures of the crossing

areas every 24 h until one condition covers the entire gap.
7. Determine migration rate by quantifying the area free of cells of
each picture and relating them to the time zero area. ImageJ
free software analysis offers an easy tool to quantify areas.
This experiment can be directly performed with stably trans-
fected cells expressing Gal-1 or Gal-1 shRNA (as explained above)
or could be also performed by first performing siRNA transfection
in 24 well plates, allowing confluency and following with the
migration experiment.
Some cells establish strong cell–cell adhesion and a clear scratch
cannot be made as the wound quickly reseals with transiently
detached cells. In this case, transwell migration is a good option.
Cell Migration Assay in 1. Trypsinize control cells and cells with Gal1 levels up- or down-
Transwells regulated, count cells using a Neubauer chamber, and discard
dead cells stained with trypan blue.
2. Wash cells twice with DMEM with 1% BSA.
3. Seed 40,000 cells (see Note 19) in 150 μL DMEM with 1%
BSA over the transwell insert of 24-well plates, in triplicates.
4. Add 500 μL of DMEM with 10% FBS in the lower

compartment.
5. After 48 h, wash transwell insert and plate wells twice with PBS.
6. Fixate cells over the transwell insert (150 μL) and in the plate
well (500 μL) with 1% Glutaraldehyde (in PBS) 10 min (see
Note 20).
7. Wash with PBS both transwell insert and plate well. If neces-
sary, fixed cells can be kept at 4 C with PBS with 0.02% sodium
azide.
8. Stain the plate well and lower side of the transwell for 5 min
with crystal violet (0.2% in H2O). Use a volume of 150 μL for
transwell and 300 μL for plate well.
9. Wash with dH2O both transwell insert and plate well, three
times.
10. Dry the transwell insert carefully from the inside part with an
ear stick.
11. Scan if desired.
12. Cut the transwell membrane and place it in the Eppendorf
containing 150 μL of acetic acid 10%.
13. Allow proper extraction for 15 min in an orbital shaker.
14. Transfer 50 μL of extraction solution, place it in a multi-well
plate, and read absorbance at 590 nm.
Cell Invasion Assay in For cell invasion assays, migration protocol (see Subheading “Cell
Transwells Migration Assay in Transwells”) is adapted to matrigel coated
transwells. For this, the day before seeding cells, an aliquot of
matrigel is carefully defrozen on ice and using pre-cold pipette
tips, matrigel is diluted in ice-cold PBS (1:20). 75 μL of this diluted
matrigel is added over the transwells and the plate is left under UV
light overnight. Usually, double number of cells are seeded com-
pared to transwell migration assays and incubation of 72 h is
required, but these data must be optimized for each cell line
under evaluation.
All these experiments can be directly performed with stably
transfected cells expressing Gal-1 or Gal-1 shRNA (as explained
above) or could be also performed by first performing siRNA
transfection in 24 well plates, allowing cells to attain confluency
followed with the migration experiment.
3.3.3 Soft Agar Colony Several galectins, including Gal-1, Gal-3, and Gal-7, regulate
Formation Assay in vitro anchorage independent growth, a feature of tumorigenesis
[14, 27, 28]. This assay is recognized as one of the best ways to
measure transformation ability in cancer cells in vitro as a way to
predict tumor formation capacity later in murine models. As the
assay lasts around a month, stable up- or downregulation of the

galectin is essential in order to maintain the same levels of the
galectin throughout all the experiment.
1. The day before starting the experiment, a solution of 2% Noble
Agar in water must be prepared and autoclaved. The solution
can be kept in a 50 C oven to prevent solidification.
2. Prepare the bottom layer by cooling the 2% agar to 40 C and
diluting 1:1 with complete DMEM. Pour 2 mL to each well in
6-well plate, in triplicates for each condition. Allow solidifica-
tion by setting aside the plates in the hood for 10 min. Leave
plates in the incubator while preparing cells.
3. Trypsinize control cells and cells with Gal1 levels up- or down-
regulated, count cells using a Neubauer chamber, and discard
dead cells stained with trypan blue. 25,000 cells (see Note 19)
will be seeded per 6-well plate embedded in a low dilution
of agar.
4. For the top layer: Ensure 2% agar solution is at 40 C, resus-
pend cells in complete DMEM, and add agar solution to get
0.3% of agar. Quickly pour 1.5 mL over the bottom layer in the
6-well plates. Allow jellification in the hood for 10 min.
5. Incubate for 30 min at 37 C.
6. Add 2 mL of complete DMEM over the top layer.
7. Incubate plates for 30 days in the incubator (depending on the
cell line), feeding them as required (usually twice in a week)
with 0.5 mL of fresh complete DMEM. Follow anchorage
independent growth regularly under the microscope.
8. To help visualize and quantify colonies, cells can be stained
with 1 mL of 0.005% crystal violet or MTT 1 mg/mL, for 1 h
in the incubator (Fig. 3).
3.4 Preclinical One of the features of galectins is that they have the ability not only
Strategies to regulate functions in tumor cells but also they can be secreted
and modulate the stromal compartment including fibroblasts,
endothelial cells, and immune cells. Therefore, in vivo studies are
required in order to see the whole picture of galectin functions in
tumor development and progression.
3.4.1 Tumor Generation Injection of cancer cell lines in mice is a simple and well-recognized
by Cell Line Injection in methodology to study tumor development and progression
Mice [29]. To study the role played by a particular galectin in these
events, tumor cells with up- or downregulated levels of the galectin
under study are injected in the animal and compared to injection of
control cells (the same cell line untransfected, or transfected with an
empty vector, or with a scramble sequence). When injected cells are
of human origin they should be injected in immunodeficient mice
(xenograft experiments), while if they are mouse tumor cell lines

they can be injected in mice with the same genetic background
(syngeneic mice), which allows working with an immunocompe-
tent system. All animal procedures must be carefully designed
following the principle of the 3 Rs (replacement, refinement, reduc-
tion) and previously approved by an ethical committee of animal
experimentation.
1. Trypsinize control cells and cells with Gal1 levels up- or down-
regulated, count cells using a Neubauer chamber, and discard
dead cells stained with trypan blue. one million cells (see Note
19) can be injected per subcutaneous flank, intraperitoneally or
preferably (if mastering surgery) orthotopically in the tissue of
cancer cell origin depending on the tumor to be studied.
2. Wash cells with DMEM without serum and resuspend the
pellet in appropriate volume of DMEM (100 μL per injection).
Performing injections with Matrigel at 1:1 might be necessary
for cell lines with low implantation capabilities.
3. Inject cells in 6-week-old BALB/c nude mice (other strains can
be used depending on the immune background desired, which
will determine the results of the experiment). If possible, lucif-
erase expressing cells can be used for tumor development
in vivo, especially by intraperitoneal and orthotopic injections.
Luciferase can be inserted into any cell line following the
protocol for retroviral injections (see Subheading 3.2.3) with a
plasmid encoding luciferase gene (pLHCX vector backbone).
4. Follow tumor progression weekly or every 2 days as tumor
develops. If working with a luciferase expressing cell line,
tumor monitoring can be performed by injecting luciferin
intraperitoneally (16 mg luciferin/kg) and measuring biolumi-
nescence after 12 min in an IVIS Imaging System (Xenogen).
3.4.2 Development of Xenograft experiments require immunodeficient mice to avoid

Tissue-Specific Oncogene- human cell rejection, which is a major drawback for galectin studies,
Driven Tumors in Galectin due to their key role in tumor immune regulation [30]. Therefore,
KO Background immune proficient models are a better option for galectin-related
preclinical trials. Still, although syngeneic mice may offer an alter-
native, tumor cells implanted in mice usually fail to recapitulate the
complex stromal reaction found in spontaneous tumors. Among
the different possibilities for targeting Gal-1 in vivo in immune
competent mice, our group has used crossings of transgenic mice
developing tumors by tissue-specific oncogene expression with
Gal-1 KO mice, in order to explore how Gal1 deficiency influences
tumor development and progression. For this, male transgenic
mice carrying oncogenic stimuli (in our case c-myc overexpression
[14], or a mutation in KRas (KRasG12V) under Elastase promoter,
driving constitute activation of the oncogene in pancreatic tissue
Fig. 4 Animal breeding to achieve the required genotypes for studying Gal-1 role in tumor development and
progression (Myc+/TGal1+/+ and Myc+/TGal1/). In F0 generation, transgenic male animals overexpressing
c-myc oncogene under elastase promoter to generate pancreatic tumors (Myc+/T) are bred to galectin
knockout females (not developing tumors as they do not have the c-myc transgene) (Myc+/+ Gal1/). Half
of the progeny will harbor c-Myc overexpression (Myc+/T) in a galectin heterozygote background, whereas the
other half will still be galectin heterozygotes but will not develop pancreatic tumors (Myc+/+). By breeding
these heterozygote animals (male always harboring c-Myc overexpression, female wild type), F2 generation
will already provide the genotypes necessary for the analysis. Half of the progeny will again be Myc+/+ and not
develop tumors, whereas the other half will. Nevertheless, this time 1/8 animals will develop tumors in a
galectin wild type background, 1/4 will do so in a heterozygote background, whereas 1/8 will form galectin
knockout tumors
[15]) are crossed with female Gal-1 KO [25], obtaining hetero-

zygotes in the first generation and knockouts in the second one
(Fig. 4).
For mice genotyping:
1. A tail fragment of approximately 5 mm is cut at 5 weeks age.
2. Samples are digested for 12 h at 55 C in 0.5 mL of proteinase
K extraction buffer (500 μg/mL), with gentle agitation.
3. Samples are centrifuged 10 min at 16,000 g.
4. The supernatant is collected, and 0.5 mL of cold isopropanol is
then added to precipitate DNA.
5. Centrifuge 10 min 16,000 g.
Fig. 5 Animal genotyping. PCR analysis for mice genotyping after DNA extraction
from a tail fragment of Gal-1 control animals (Gal-1+/+), heterozygotes (Gal-1+/

), and knockout animals (Gal-1/). Gal-1+/+ animals show a band at 478 bp,
whereas knockout animals amplify the Neo cassette generating a band at
694 bp. Accordingly, heterozygotes mice show a doublet of 694/478 bp bands
6. Wash the pellet with ethanol 70%.

7. Resuspend the pellet with 200 μL of TE buffer.
8. Perform PCR analysis with 0.5 U EcoTaq Polymerase and
specific primers designed to detect Gal-1 or the Neo cassette
in the KOs: Fw (CTC AGT GGC TAC ATC TGT AAA ATGG)
and Rv [TTC TTT GAC ATT TGA ACC CTA TACC (30 neo)
or TTC TTT GAC ATT TGA ACC CTA TACC (30 Gal)],
expecting a 478 bp band for the wt allele or a 694 bp band
for the targeted one.
9. PCR conditions were as follows: after 2 min at 94 C, 35 cycles
of denaturation at 94 C for 45 s, annealing at 60 C for 1 min,
extension at 72 C for 1 min, followed by a final extension at
72 C for 10 min. DNA bands were observed in a 2% agarose
gel (Fig. 5).
These animals are genotyped and carefully monitored for
tumor progression, as will be explained in the following section.
3.4.3 Data Collection and Two types of experiments can be performed both when using
Analysis for Preclinical xenografts or genetically engineered mice: (a) endpoint analysis:
Studies Animals are sacrificed gradually depending on their well-being,
with an endpoint established criteria following ethical guidelines
including tumor size (1 cm3) or compromised animal general state,
considering significant weight loss, antalgic positions, weakness,
reduced activity, ascites, and jaundice. (b) Selected time-point anal-
ysis: All animals are sacrificed at the same time point, which is
defined following researcher criteria depending on tumor develop-
ment and the goal to be studied (tumor initiation, progression,
etc.). First strategy will allow building Kaplan–Meier survival curves

and analyze survival rates, whereas the second strategy will allow
researchers compare tumors in different conditions at the same
time point (for example, wild type versus galectin knockouts, or
tumors from cell lines with different galectin levels).
1. Perform animal necropsy: Sacrifice animals by CO2 asphyxia-
tion. Weight and measure tumors through the largest 3 perpen-
dicular diameters to calculate tumor volume. Annotate major
macroscopic features of animals and tumors such as spleno-
megaly, hemorrhages, tumor localization, shape, vasculariza-
tion, consistence, and metastatic niches.
2. Fragment the whole tumor with a scalpel in representative parts
to maximize tissue preservation for different later applications:
(a) For histology, a part of the tumor is fixed in buffered
formalin for 24 h, dehydrated and embedded in paraffin.
A second piece is embedded in OCT in plastic molds and
directly frozen (see Note 21). This tissue will offer sections
that can be:
l Stained with hematoxylin and eosin and analyzed by an
expert pathologist to quantify necrosis, differentiation,
and other important features.
l Stained with specific antibodies to make a profound
histo-pathological analysis following adapted immuno-
histochemistry protocol described in Subheading 3.1
(Fig. 6), according to the following markers to measure
different hallmarks (Table 1):
– Tumor proliferation: Ki67, P-Histone H3 (Ser10).
– Angiogenesis: CD31, von Willebrand Factor.
– Stroma abundance and activation: a-SMA,
vimentin.
– Immune infiltrates: CD3, CD4, CD8, MPO,
F4/80.
– Apoptosis: Cleaved caspase 3.
(b) For RNA extraction: Another tumor fragment is placed in
RNAlater to ensure RNases inactivation.
(c) For protein extraction: A tumor fragment is placed in an
Eppendorf and directly frozen in liquid nitrogen.
(d) For analysis of immune cell infiltrates by flow cytometry.
In this case:
l Place the piece of tumor in cold PBS.
l Digest it by incubating for 30 min at 37 C with 5 mL
of collagenase IV (1 mg/mL in PBS EDTA 1 μM,
2 U/mL DNase type I), shaking vigorously every
5 min.
Fig. 6 Immunohistochemistry characterization of tumors developed by control cells (using an irrelevant shRNA,
shCtl) or cells knocked down for Gal-1 (shGal-1). (a, e) Gal-1 immunohistochemistry to determine the
downregulation of the protein in tumor cells (positive staining in e, correspond to host fibroblasts, which
express high levels of this lectin); (b, f), staining with Ki67 for detection of cell proliferation; (c, g), analysis of
stroma composition by α-SMA staining to label activated fibroblasts; (d, h), study of tumor angiogenesis by
staining of endothelial cells using von Willebrand Factor (vWF). Scale bars, 100 μm
l Pass through a cell strainer filter.

l Centrifuge (120 g, 3 min).
l Immune cells can in this step be frozen in FBS with
10% DMSO or follow antibody incubation according
to the panel of antibodies designed for determining
different cell populations [15] and analyzed by flow
cytometry (BD Biosciences).
3.5 Galectin-1 As galectins can be secreted from cells they can be found in blood
Determination in Blood and several reports have shown their potential use as diagnostic
Samples and/or prognostic biomarkers [5] by measuring their levels in
biological fluids.
For human samples the following protocol can be used:
1. Collect blood using EDTA-treated tubes.
2. Centrifuge for 10 min 480 g.
3. Aliquot the supernatant and keep at 80 C.
4. Take 25 μL of plasma for duplicates.
5. Dilute plasma 1:10 using the calibrator diluent of the Quanti-
kine ELISA Human Galectin-1 Immunoassay.
6. Follow Quantikine ELISA Human Galectin-1 Immunoassay
protocol and interpolate measures into its standard curve to
obtain the concentration of Gal-1 (ng/mL) in the blood.
Similarly, in mice, submandibular blood can be collected in

EDTA-treated tubes at different steps during tumor progression
to evaluate the potential of galectin as a prognostic marker by
ELISA using Quantikine ELISA kit specific for murine samples, as
explained above.
4 Notes
1. Development time can be set up by adding DAB and following

staining under the microscope. It usually takes a few seconds.
Ensure reaction is stopped by immersing in PBS before the
whole slide background gets stained.
2. These maintenance conditions work for most of cancer cell
lines, but this must be optimized and adapted to particular
cell line incubation conditions.
3. A setup will be necessary to check number of cells and conflu-
ence achieved for each particular cell line. siRNA is diluted
upon cell division, but transfection efficiency is higher when
cells are growing exponentially, so a particular cell number for
each cell line must be found. Transfections can be performed in
24 well plates if later protein extraction is needed but will be
recommended in 6 well plates if RNA is required. Please scale
up and adjust amount if necessary according to differences in
plate surface.
4. Cells treated with transfection reagent but without siRNA are
also included in the experiment as controls.
5. Alternatively, cells can be easily transfected with PEI (polyethy-
lenimine) (78 μg/mL) by simply mixing siRNA and PEI, vor-
texing and allowing 30 min incubation, RT. Mixture is then
added directly into the wells (in complete medium) and left
until the following day, when fresh medium is added.
6. 40 ng of human recombinant Gal-1 can be added as positive
control.
7. Many cancer cell lines can uptake siRNA by simple transfection
as detailed in Subheading 3.3 but infection is required when
bigger DNA fragments have to be uptaken. For some cells,
though, as in pancreatic cancer RWP-1, simple transfection
using PEI can be used for shRNA delivery. As this is the
minority of cases, we will describe how to infect human cancer
cell lines for stable downregulation.
8. 293 T cells have low attachment over cell culture dishes and
must be washed with PBS very carefully. Optionally, they can be
seeded over poly-lysine coated wells to improve adherence.
9. We recommend checking several shRNA sequences to compare

knockdown efficiency and select the couple of shRNAs with the
best downregulation for performing functional assays in order
to ensure knockdown specificity and avoid off target effects.
10. Transfection and infections can be scaled up or down according
to the number of cells required.
11. Lentiviral generation must be developed in a BSL-2 room
provided with special allowance to work with genetically mod-
ified organisms level 2.
12. In the case of Gal-1, stable downregulation has proven to be
very robust, accepting cell freezing/thawing cycles and passag-
ing. This is not the case of other proteins and it is something
that must be well taken into account. Still, in the case of Gal-1
we strongly recommend checking knockdown efficiency by
analyzing Gal-1 levels by Western Blot in every functional
experiment designed and tracking carefully cell passaging and
freezing/thawing cycles. Stable downregulation and maintain-
ing reduced levels of Gal-1 do not imply that compensatory
mechanisms will not happen over time, reducing observed
phenotypes and interfering with experiments.
13. Alternatively, since the discovery of genome editing using
CRISPR/Cas9 technology in 2013, total knockouts in eukary-
otic cells have also become a possibility [31, 32]. Several
reports have described CRISPR/Cas9 methodology to be
used in cellular biology [33, 34]. LGALS1 (Galectin-1) knock-
out has been efficiently performed in Tu167 head and neck
cancer cell line [35]. Galectin CRISPR/Cas9 methods as well
as gene knockdowns with morpholinos have also been
described to be used in the zebrafish model [36].
14. Cell number must be optimized depending on the cell growth
rate. A setup seeding different amount of cells (5000/10,000/
20,000 cells/well) and monitoring cell proliferation by simply
observing cells daily under a microscope during a week is
highly recommended. These data will help establishing the
correct amount of cells and time points to process.
15. Accurate cell counting is key for proper measurement and
reproducibility of the experiment. Ensure all cells are seeded
in the same number, and homogeneously distributed into the
well (ensure they do not concentrate in the well center as this
would affect cell division). Proper resuspension while cell
counting and while seeding will help.
16. Leave a control well without cells. The background reading
may vary between days and must be removed from the
measurements.
17. Depending on the cell line, if mild differences occur upon
changing Gal-1 levels, performing the experiment in complete
DMEM (10% FBS) may not enable significant differences to be

detected. Otherwise, cells can be left in DMEM with low
serum conditions (a setup can be done with 0.2/0.5/1%
FBS) 24 h after cell seeding and during cell growth.
18. BrdU incubation time depends a lot on the cell line and its
division rate ranging from a few minutes to some hours. A
setup is highly recommended in order to find out the frame
of time in which a partial percentage of cells is
incorporating BrdU.
19. Optimize depending on the cell line.
20. Please use glutaraldehyde under the chemical hood with special
precaution.
21. Some antigens are not well preserved after formalin fixation
and cannot be recognized with some antibodies, for which
OCT embedded tissue may offer an alternative.
References
1. Stowell SR, Ju T, Cummings RD (2015) Pro- et al (2014) Unraveling galectin-1 as a novel

tein glycosylation in cancer. Annu Rev Pathol therapeutic target for cancer. Cancer Treat Rev
10:473–510. https://doi.org/10.1146/ 40:307–319. https://doi.org/10.1016/j.ctrv.
annurev-pathol-012414-040438 2013.07.007
2. Pinho SS, Reis CA (2015) Glycosylation in 9. Rabinovich GA (2005) Galectin-1 as a poten-
cancer: mechanisms and clinical implications. tial cancer target. Br J Cancer 92:1188–1192
Nat Rev Cancer 15:540–555. https://doi. 10. Martinez-Bosch N, Barranco LE, Orozco CA,
org/10.1038/nrc3982 Moreno M, Visa L, Iglesias M et al (2018)
3. Rabinovich GA, Croci DO (2012) Regulatory Increased plasma levels of galectin-1 in pancre-
circuits mediated by lectin-glycan interactions atic cancer: potential use as biomarker. Onco-
in autoimmunity and cancer. Immunity 36: target 9:32984–32996. https://doi.org/10.
322–335 18632/oncotarget.26034
4. Yang RY, Rabinovich GA, Liu FT (2008) 11. Méndez-Huergo SP, Blidner AG, Rabinovich
Galectins: structure, function and therapeutic GA (2017) Galectins: emerging regulatory
potential. Expert Rev Mol Med 10:e17. checkpoints linking tumor immunity and
h t t p s : // d o i . o r g / 1 0 . 1 0 1 7 / angiogenesis. Curr Opin Immunol 45:8–15.
S1462399408000719 https://doi.org/10.1016/j.coi.2016.12.003
5. Thijssen VL, Heusschen R, Caers J, Griffioen 12. Cerliani JP, Blidner AG, Toscano MA, Croci
AW (2015) Galectin expression in cancer diag- DO, Rabinovich GA (2017) Translating the
nosis and prognosis: a systematic review. Bio- “Sugar Code” into immune and vascular sig-
chim Biophys Acta 1855:235–247. https:// naling programs. Trends Biochem Sci 42:
doi.org/10.1016/j.bbcan.2015.03.003 255–273. https://doi.org/10.1016/j.tibs.
6. Hanahan D, Weinberg RAA (2011) Hallmarks 2016.11.003
of cancer: the next generation. Cell 144: 13. Martı́nez-Bosch N, Navarro P (2020) Galec-
646–674. https://doi.org/10.1016/j.cell. tins in the tumor microenvironment: focus on
2011.02.013 galectin-1. In: Birbrair A (ed) Tumor microen-
7. Girotti MR, Salatino M, Dalotto-Moreno T, vironment, Advances in experimental medicine
Rabinovich GA (2020) Sweetening the hall- and biology, vol 1259. Springer, Cham
marks of cancer: galectins as multifunctional 14. Martı́nez-Bosch N, Fern̊andez-Barrena MG,
mediators of tumor progression. J Exp Med Moreno M, Ortiz-Zapater E, Munn̊e-
217:e20182041. https://doi.org/10.1084/ Collado J, Iglesias M et al (2014) Galectin-1
jem.20182041 drives pancreatic carcinogenesis through
8. Astorgues-Xerri L, Riveiro ME, Tijeras- stroma remodeling and hedgehog signaling
Raballand A, Serova M, Neuzillet C, Albert S activation. Cancer Res 74:3512–3524.
https://doi.org/10.1158/0008-5472.CAN- 11:291–302. https://doi.org/10.1016/j.ccr.

13-3013 2007.01.012
15. Orozco CACA, Martinez-Bosch N, Guerrero 25. Poirier F, Robertson EJ (1993) Normal devel-
PEPE, Vinaixa J, Dalotto-Moreno T, Iglesias opment of mice carrying a null mutation in the
M et al (2018) Targeting galectin-1 inhibits gene encoding the L14 S-type lectin. Develop-
pancreatic cancer progression by modulating ment 119:1229–1236
tumor–stroma crosstalk. Proc Natl Acad Sci U 26. Camby I, Le Mercier M, Lefranc F, Kiss R
S A 115:E3769–E3778. https://doi.org/10. (2006) Galectin-1: a small protein with major
1073/pnas.1722434115 functions. Glycobiology 16:137R–157R
16. Liu FT, Rabinovich GA (2005) Galectins as 27. Nangia-Makker P, Balan V, Raz A (2012)
modulators of tumour progression. Nat Rev Galectin-3 binding and metastasis. Methods
Cancer 5:29–41 Mol Biol 878:251–266. https://doi.org/10.
17. Paz A, Haklai R, Elad-Sfadia G, Ballan E, 1007/978-1-61779-854-2_17
Kloog Y (2001) Galectin-1 binds oncogenic 28. Ueda S, Kuwabara I, Liu F-T (2004) Suppres-
H-Ras to mediate Ras membrane anchorage sion of tumor growth by galectin-7 gene trans-
and cell transformation. Oncogene 20: fer. Cancer Res 64:5672–5676. https://doi.
7486–7493 org/10.1158/0008-5472.CAN-04-0985
18. Chung L-Y, Tang S-J, Sun G-H, Chou T-Y, 29. Ireson CR, Alavijeh MS, Palmer AM, Fowler
Yeh T-S, Yu S-L et al (2012) Galectin-1 pro- ER, Jones HJ (2019) The role of mouse
motes lung cancer progression and chemore- tumour models in the discovery and develop-
sistance by upregulating p38 MAPK, ERK, and ment of anticancer drugs. Br J Cancer 121:
cyclooxygenase-2. Clin Cancer Res 18: 101–108. https://doi.org/10.1038/s41416-
4037–4047. https://doi.org/10.1158/ 019-0495-5
1078-0432.CCR-11-3348 30. Rabinovich GA, Conejo-Garcı́a JR (2016)
19. van den Brûle F, Califice S, Garnier F, Fernan- Shaping the immune landscape in cancer by
dez PL, Berchuck A, Castronovo V (2003) galectin-driven regulatory pathways. J Mol
Galectin-1 accumulation in the ovary carci- Biol 428:3266–3281. https://doi.org/10.
noma peritumoral stroma is induced by ovary 1016/j.jmb.2016.03.021
carcinoma cells and affects both cancer cell 31. Cong L, Ran FA, Cox D, Lin S, Barretto R,
proliferation and adhesion to laminin-1 and Habib N et al (2013) Multiplex genome engi-
fibronectin. Lab Investig 83:377–386 neering using CRISPR/Cas systems. Science
20. Toscano MA, Bianco GA, Ilarregui JM, Croci 339:819–823. https://doi.org/10.1126/sci
DO, Correale J, Hernandez JD et al (2007) ence.1231143
Differential glycosylation of TH1, TH2 and 32. Mali P, Yang L, Esvelt KM, Aach J, Guell M,
TH-17 effector cells selectively regulates sus- DiCarlo JE et al (2013) RNA-guided human
ceptibility to cell death. Nat Immunol 8: genome engineering via Cas9. Science 339:
825–834 823–826. https://doi.org/10.1126/science.
21. Liu SD, Tomassian T, Bruhn KW, Miller JF, 1232033
Poirier F, Miceli MC (2009) Galectin-1 tunes 33. Ran FA, Hsu PD, Wright J, Agarwala V, Scott
TCR binding and signal transduction to regu- DA, Zhang F (2013) Genome engineering
late CD8 burst size. J Immunol 182: using the CRISPR-Cas9 system. Nat Protoc
5283–5295. https://doi.org/10.4049/ 8:2281–2308. https://doi.org/10.1038/
jimmunol.0803811 nprot.2013.143
22. Croci DO, Cerliani JP, Dalotto-Moreno T, 34. Sanjana NE, Shalem O, Zhang F (2014)
Méndez-Huergo SP, Mascanfroni ID, Improved vectors and genome-wide libraries
Dergan-Dylon S et al (2014) Glycosylation- for CRISPR screening. Nat Methods 11:
dependent lectin-receptor interactions preserve 783–784. https://doi.org/10.1038/nmeth.
angiogenesis in anti-VEGF refractory tumors. 3047
Cell 156:744–758. https://doi.org/10.1016/
j.cell.2014.01.043 35. Maybruck BT, Pfannenstiel LW, Diaz-
Montero M, Gastman BR (2017) Tumor-
23. Sandgren EP, Quaife CJ, Paulovich AG, Palmi- derived exosomes induce CD8+ T cell suppres-
ter RD, Brinster RL (1991) Pancreatic tumor sors. J Immunother Cancer 5(1):65. https://
pathogenesis reflects the causative genetic doi.org/10.1186/s40425-017-0269-7
lesion. Proc Natl Acad Sci U S A 88:93–97
36. Feng C, Nita-Lazar M, González-
24. Guerra C, Schuhmacher AJ, Cañamero M, Montalbán N, Wang J, Mancini J, Ravindran
Grippo PJ, Verdaguer L, Pérez-Gallego L et al C et al (2015) Manipulating galectin expres-
(2007) Chronic pancreatitis is essential for sion in zebrafish (Danio rerio). Methods Mol
induction of pancreatic ductal adenocarcinoma Biol 1207:327–341. https://doi.org/10.
by K-Ras oncogenes in adult mice. Cancer Cell 1007/978-1-4939-1396-1_22
Chapter 38
Galectin-3–U1 snRNP Complexes Initiate Splicing Activity

in U1-Depleted Nuclear Extracts
Patricia G. Voss, Kevin C. Haudek, Ronald J. Patterson, and John L. Wang
Abstract
Fractionation of HeLa cell nuclear extracts by glycerol gradient centrifugation separates endogenous uracil-
rich small nuclear ribonucleoprotein complexes (U snRNP) into numerous particles sedimenting from 7S
to greater than 60S. Complexes sedimenting at 10S contain a single U snRNP (U1 snRNP) and galectin-3.
Addition of antibodies specific for galectin-3 to fractions containing these 10S complexes coprecipitates U1
snRNP, indicating that a fraction of the U1 snRNP is associated with this galectin. Galectin-3 has been
shown by depletion-reconstitution studies to be an integral splicing component involved both in spliceo-
some assembly and splicing activity. The first step in initiation of spliceosome assembly is binding of U1
snRNP to the 50 splice site of the premessenger RNA substrate. The finding that U1 snRNP and galectin-3
are associated in splicing extracts hints that this complex affords a potential entry point for galectin-3 into
the splicing pathway. Addition of U1 snRNP-galectin-3 complexes immunoselected from the 10S region of
glycerol gradients to a U1-depleted nuclear extract initiates splicing activity with the formation of splicing
intermediates and mature mRNA. This chapter describes the materials and methods for these experiments
that document galectin-3–U1 snRNP complexes initiate the splicing reaction in a U1-depleted nuclear
extract.
Key words Pre-mRNA splicing, Spliceosome, Cell-free splicing assay, Glycerol gradient fractionation,
U1 snRNP complex
1 Introduction
Nuclear extracts (NE) [1] have been used to characterize both the
components and steps involved in premessenger RNA (pre-mRNA)
splicing. Five spliceosomal ribonucleoproteins (snRNPs) contain-
ing the U1, U2, U4, U5, and U6 snRNAs and U-specific and core
Sm proteins with additional splicing factors are assembled in an
ordered fashion to precisely excise introns and join exons to pro-
duce mature mRNA [2, 3]. The initial step in assembling an active
spliceosome is the binding of U1 snRNP to the pre-mRNA 50 splice
site. Subsequently U2 snRNP binds near the 30 splice site followed
by incorporation of the U4, U5, U6 tri-snRNP complex [4].
713
714 Patricia G. Voss et al.
Several lines of evidence indicate that galectin-3 (Gal3) and

galectin-1 (Gal1) are functionally redundant components of the
splicing pathway: (a) depletion of both Gal3 and Gal1 from NE
abrogates spliceosome assembly and splicing activity at an early step
[5, 6]; (b) addition of either galectin restores splicing to the
depleted NE [5, 6], and (c) immunoprecipitation of active splicing
extracts with Gal1- or Gal3-specific antisera results in the copreci-
pitation of pre-mRNA substrate, splicing intermediates, and the
mature mRNA product [7]. In addition, Gal3 associates with
snRNPs outside the splicing pathway [8].
In NE preincubated at splicing permissive temperatures to
dissociate any preformed spliceosomes [9], U1 snRNPs are found
in several particles that can be separated by glycerol gradient centri-
fugation. A monoparticle sedimenting at 10S contains U1 snRNA,
its core and U1-specific proteins and Gal3 [8]. Immunoprecipita-
tion of the fractionated 10S complex with antibodies specific for
Gal3 coprecipitates U1 snRNP and Gal3, indicating some U1
snRNP monoparticles are complexed with Gal3.
These findings indicate Gal3 and Gal1 are integral splicing
components and suggest an entry point for galectin incorporation
into the spliceosomal assembly pathway at the earliest step—bind-
ing of the Gal3–U1 snRNP monoparticle to the pre-mRNA sub-
strate. Our ability to isolate the Gal3–U1 snRNP complex provided
the opportunity to test this hypothesis. We documented that Gal3–
U1 snRNP complexes bound to anti-Gal3 beads are competent to
restore splicing activity to a U1-depleted NE (U1ΔNE) [10].
In Chapter 28 of the volume of methods and protocols dedi-
cated to galectins, we had described the preparation of NE, the
splicing substrate, the assembly of the splicing reaction mixture,
and the analysis of products in our documentation of the role of
galectins in pre-mRNA splicing [11]. We now describe, as an
update to that chapter, the experimental procedures for fraction-
ation of nuclear extracts to obtain a fraction enriched in galectin-3–
U1 snRNP complexes and for immunoselection of these complexes
to reconstitute splicing activity in a U1-depleted nuclear extract
(Fig. 1).
2 Materials
2.1 Preparation of The materials and methods for the preparation of NE and for the
Nuclear Extract (NE) preparation of the MINX splicing substrate have been described in
and the Splicing Chapter 28 of the prior volume dedicated to galectins; see Sub-
Substrate headings 2.1, 2.2, 3.1, and 3.2, as well as the associated Notes 1–3
of that chapter [11]. NE, as initially prepared, is in Buffer C and will
be designated hereafter as NE(C). NE(C) dialyzed against and
equilibrated with Buffer D will be designated as NE(D).
Galectin-3–U1 snRNP Complexes Initiate Splicing Activity in U1-Depleted. . . 715
Fig. 1 Schematic diagram illustrating the key steps in restoring splicing activity in nuclear extract depleted of
U1 snRNP by a Gal3-U1 snRNP complex trapped on beads. (a) NE in Buffer C (NE(C)) is incubated with Protein
A-Sepharose beads derivatized with antibodies specifically against U1 snRNP (αU1 beads). The unbound
fraction is depleted of U1 snRNP, concomitant with loss of splicing activity (U1ΔNE). (b) NE in Buffer D (NE(D))
is fractionated by ultracentrifugation over a 12–32% glycerol gradient. Fractions corresponding to the 10S
region (fractions 3–5) are combined and immunoprecipitated with beads derivatized with anti-Gal3 antibodies
(αGal3 beads). The material bound to the beads contains a Gal3-U1 snRNP complex. (c) U1ΔNE from Part
(A) is mixed with the Gal3-U1 snRNP complex from Part (B) and 32P-labeled MINX pre-mRNA substrate under
splicing assay conditions and RNA corresponding to the intermediates and products of the splicing reaction are
analyzed by gel electrophoresis and autoradiography
Buffers (See Notes 1–3)

1. Buffer C: 20 mM HEPES, pH 7.9, 25% (vol/vol) glycerol,
0.42 M NaCl, 1.5 mM MgCl2, 0.2 mM EDTA, 0.5 mM
phenylmethyl-sulfonyl fluoride (PMSF), 0.5 mM DTT.
2. Buffer D: 10 mM HEPES, pH 7.9, 20% glycerol, 0.1 M KCl,
0.2 mM EDTA, 0.5 mM PMSF, 0.5 mM DTT.
2.2 Preparation of 1. DEPC-treated water (see Notes 1–3).

Nuclear Extract 2. Protein A-Sepharose CL-4B (Amersham Biosciences).
Depleted of U1 snRNP
3. IEC Model PR-6 centrifuge (International Equipment Co.).
(U1ΔNE)
4. Human autoimmune serum specific for U1 snRNP
(Hu ENA-RNP, #33471) (The Binding Site).
5. Dimethylpimelimidate (Sigma-Aldrich).
6. Labquake Shaker (rotator/rocker) 400-110 (Lab Industries,

Inc.).
7. Nuclear extract NE(C) (Subheading 2.1).
8. RNasin (40 U/μl) (Promega).
9. Hamilton syringe.
10. Ethanol.
11. Dry ice.
12. HEPES wash buffer: 20 mM HEPES, pH 7.9, 0.5 M NaCl.
13. HEPES coupling buffer: 20 mM HEPES, pH 7.9.
14. 0.2 M sodium borate (pH 9).
15. 0.2 M ethanolamine (pH 8).
16. Buffer D.
17. 60%D: 60% Buffer D and 40% H2O.
18. TX wash buffer: 60%D + 0.05% Triton X-100 (TX).
19. SDS Sample buffer: 0.0625 M Tris, pH 6.8, 2% SDS, 10%
glycerol, 5% 2-mercaptoethanol, bromophenol blue (0.5 mg/
50 ml).
Special equipment
20. Microdialyzer (12-well) with 6–8 K molecular weight cutoff
membrane (Pierce).
2.3 Glycerol Gradient 1. 20-gauge needle.

Fractionation of NE 2. 5-ml nitrocellulose tubes.
2.3.1 Preparation of Buffers (See Notes 1–3)
Gradients
3. Buffer D.
4. 60%D; 60% Buffer D, 40% H2O.
5. 60%D–12% glycerol: 60% Buffer D, 40% H2O.
6. 60%D–32% glycerol: 60% Buffer D, 20% glycerol, 20% H2O.
Special equipment
7. Dual chamber gradient maker (Bio-Rad).
8. Overhead stirrer (Buchler Instruments).
2.3.2 Fractionation of NE 1. Nuclear extract NE(D) (Subheading 2.1).

on a 12–32% Glycerol 2. MgCl2.
Gradient
3. ATP.
4. Creatine phosphate.
5. Six 5-ml nitrocellulose tubes containing preprepared 12–32%

glycerol gradient in 60%D (Subheading 2.3.1).
6. 1-ml pipetteman.
Special equipment
7. Beckman SW50.1 rotor and 6 buckets.
8. Beckman Optima LE-80K ultracentrifuge.
2.4 1. DEPC-treated water (see Note 3).

Immunoprecipitation 2. Protein A-Sepharose CL-4B (Amersham Biosciences).
of 10S Fractions of
3. IEC clinical centrifuge (International Equipment Co.).
Glycerol Gradient by
Anti-Gal3 4. Polyclonal rabbit antisera against Gal3: #49 (see Note 4).
5. Dimethylpimelimidate (Sigma-Aldrich).
6. Labquake rotator/rocker.
7. 10S fractions (Fractions 3–5) from glycerol gradient fraction-
ation of NE(D) (Subheading 2.3).
8. Hamilton syringe.
9. HEPES wash buffer.
10. HEPES coupling buffer.
11. 0.2 M sodium borate (pH 9).
12. 0.2 M ethanolamine (pH 8).
13. Buffer D.
14. 60%D.
15. TX wash buffer.
16. SDS Sample buffer.
32
2.5 Reconstitution of 1. P-labeled MINX (Subheading 2.1).
Splicing Activity in 2. Nuclear extract NE(D) (Subheading 2.1).
U1ΔNE by Anti-Gal3
3. Nuclear extract depleted of U1 snRNP (U1ΔNE)
Precipitate of 10S
(Subheading 3.1.2).
Fractions
4. Anti-Gal3 precipitate of the 10S fractions of glycerol gradient
(Subheading 3.3.2).
5. MgCl2.
6. ATP.
7. Creatine phosphate.
8. Dithiothreitol (DTT).
9. RNasin (40 U/μl) (Promega).
10. Proteinase K (20 mg/ml).
11. Sodium acetate.

12. Buffer D.
13. 60%D.
14. SDS Sample buffer.
15. TE: 1.25 mM Tris, pH 8, 1 mM EDTA.
3 Methods
3.1 Preparation of 1. Preswell Protein A-Sepharose CL-4B beads in DEPC-treated

Nuclear Extract H2O and then wash in HEPES wash buffer.
Depleted of U1 snRNP 2. For this wash and all subsequent washes, the beads are pelleted
(U1ΔNE) by gentle centrifugation (1000 g in a swinging bucket rotor
3.1.1 Preparation of Anti-
at 4 C for 10–15 s) and the unbound wash removed using a
U1 Beads for
micropipettor and discarded.
Immunoadsorption 3. Mix 150 μl of washed beads with 150 μl of human autoimmune
serum specific for U1 snRNP.
4. Based on the total volume, adjust the mixture to 20 mM
HEPES, pH 7.9 (HEPES coupling buffer); incubate for
60 min with continuous rocking at room temperature.
5. Wash the antibody-bound beads with 1 ml of 0.2 M sodium
borate (pH 9) and resuspend in the same buffer.
6. Add dimethylpimelimidate to a final concentration of 20 mM
to covalently cross-link the antibody to the Protein-G
Sepharose.
7. After 1-h incubation at room temperature with rocking, wash
the beads with 0.2 M sodium borate (pH 9).
8. To block the unreacted cross-linker, add 1 ml of 0.2 M etha-
nolamine (pH 8) and incubate for 1 h at room temperature
with rocking.
9. Wash antibody-coupled beads, hereafter designated as anti-U1
beads, twice with 0.5 ml of TX wash buffer.
3.1.2 Depletion of U1- 1. Incubate 200 μl of NE(C) with 100 μl of anti-U1 beads. Add
snRNP from NE (Fig. 1A) 5 μl of RNasin (Promega) to the mixture.
2. Rotate microtube head-over-tail at 4 C for 1 h.
3. Pellet the mixture by centrifugation (1000 g in a swinging
bucket rotor at 4 C for 10–15 s) and collect the unbound
material (U1ΔNE) using a Hamilton syringe.
4. Dialyze the entire volume of U1ΔNE, along with a separate
50 μl aliquot of the original, nondepleted NE(C) in separate
compartments of a microdialyzer, with stirring at 4 C, for
75 min against 60% D using a dialysis membrane with 6–8 K
molecular weight cutoff.
5. Immediately after dialysis, these preparations (U1ΔNE and NE

in 60%D) are divided into 20 μl aliquots, snap frozen in a dry
ice/ethanol bath, then stored at 80 C.
3.1.3 Analysis of the RNA 1. After removal of the unbound material (U1ΔNE), wash the
and Protein Content of material bound to the anti-U1 beads by adding 0.5 ml of TX
U1ΔNE and Material Bound wash buffer.
on Anti-U1 Beads 2. Pellet the mixture by centrifugation (1000 g in a swinging
bucket rotor at 4 C for 10–15 s) and remove the supernatant
using a micropipettor and discard the wash. Repeat the wash
steps 1 and 2 twice.
3. Remove the material bound to the anti-U1 beads by adding
100 μl 2 SDS Sample Buffer to 100 μl of the beads and
incubate for 10 min at room temperature.
4. Pellet the mixture by centrifugation (1000 g in a swinging
bucket rotor at room temperature for 10–15 s), remove the
supernatant by Hamilton syringe, and snap-freeze in a dry ice–
ethanol bath. Store at 80 C.
5. The nondepleted NE, the depleted NE (U1ΔNE), and the
material bound to the beads (removed from the beads by
SDS-PAGE sample buffer) can be compared in terms of RNA
and protein components (see Note 5).
3.2 Glycerol Gradient Routinely, six 5-ml gradients (12–32% glycerol) are prepared
Fractionation of NE sequentially, using a dual chamber gradient maker.
3.2.1 Preparation of 1. Prepare 50 ml of 60%D-12% glycerol and 50 ml of 60%D
Gradients containing 32% glycerol.
2. Set up gradient maker with an overhead stirrer lowered close to
the bottom of the mixing chamber. A 20-gauge needle is
attached below the outlet valve of the mixing chamber with
the beveled edge resting on the side and near the top of the
collecting 5-ml nitrocellulose tube.
3. With all valves closed, pipette 2.6 ml of 60%D-32% glycerol
into the mixing chamber. Pipette 2.6 ml of 60%D containing
12% glycerol into the adjacent chamber.
4. Turn on stirrer and open valve between the two chambers.
Using a thumb, apply pressure at the top of the chamber
containing 60%D - 12% glycerol to push air through the open
valve to start the flow for mixing. Open the outlet valve below
the mixing chamber to collect the gradient. The levels in both
chambers should decrease equally as the gradient mixes and
drips down the side of the nitrocellulose tube.
5. When the dripping stops, close both valves. Replace gradient

tube with a new nitrocellulose tube and repeat steps 3–5 for
the remaining five gradients.
6. Store gradients at 4 C overnight.
3.2.2 Fractionation of NE 1. Chill a SW50.1 rotor and 6 buckets to 4 C.

on 12–32% Glycerol 2. Prepare a 2.5 ml reaction mix of 1.5 ml NE(D), 3 mM MgCl2,
Gradient (Fig. 1B) 0.5 mM ATP, and 20 mM creatine phosphate. Incubate at
30 C for 30 min.
3. Slowly layer 415 μl of the reaction mix on top of each of six
glycerol gradients.
4. The gradients are carefully loaded into the SW50.1 rotor buck-
ets and centrifuged at 44,000 rpm at 4 C for 4 h.
5. Starting from the top of the gradient, fractions of 250 μl are
collected manually using a 1 ml pipetteman and dispensed into
labeled tubes.
6. Approximately 22 fractions are collected from each gradient;
15 μl samples from each fraction are routinely collected for
analysis of protein components by Western blotting.
7. Gradient fractions are immediately frozen and stored at
80 C.
3.3 1. The procedure for the preparation of anti-Gal3 beads is identi-

Immunoprecipitation cal to that described for the preparation of anti-U1 beads
of 10S Fractions of (Subheading 3.1.1), with the exception that in step 3, the
Glycerol Gradient by ratio of antiserum (anti-Gal3, #49) to beads is 3:1.
Anti-Gal3 2. Just before use, wash the antibody-coupled beads, hereafter
designated as anti-Gal3 beads twice with 0.5 ml of TX wash
3.3.1 Preparation of Anti-
buffer.
Gal3 Beads for
Immunoadsorption
3.3.2 1. Combine and mix glycerol gradient fractions 3, 4, and 5 (num-

Immunoprecipitation of bered from the top of the gradient) which include the 10S
Glycerol Gradient Fractions region of the gradient.
by Anti-Gal3 (Fig. 1B) 2. From combined gradient fractions, remove 150 μl aliquot and
place in 50 μl of anti-Gal3 beads. In parallel, aliquot 150 μl of
fraction 1 (containing Gal3 not in complex with U1 snRNP)
and place in 50 μl of anti-Gal3 beads. As a control, place 150 μl
of 60%D in another microtube of 50 μl anti-Gal3 beads.
3. Mix gently, then rotate microtube head-over-tail at 4 C for
1 h.
4. Pellet the mixture by gentle centrifugation (1000 g in a
swinging bucket rotor at 4 C for 10–15 s).
5. The unbound material is completely removed using a Hamil-

ton syringe. The beads are not washed and are immediately
used for the addition of the splicing reactions
(Subheading 3.4.1).
3.3.3 Analysis of the RNA 1. For analysis of the components of the bound and unbound
and Protein Content in the material from anti-Gal3 precipitation of the 10S gradient frac-
Unbound and Bound tions, collect the unbound material (supernatant after step 4 of
Material from the Anti-Gal3 Subheading 3.3.2), transfer into a fresh microtube, and freeze
Precipitation of 10S at 20 C.
Gradient Fractions 2. Wash the precipitated beads from step 4 of Subheading 3.3.2
(containing material bound to anti-Gal3) by adding 0.5 ml of
TX wash buffer.
3. Pellet the mixture by gentle centrifugation (1000 g in a
swinging bucket rotor at 4 C for 10–15 s); remove the super-
natant using a micropipettor and discard. Repeat the wash
steps 2 and 3 twice more.
4. Add 50 μl 2 SDS Sample Buffer to the washed and pelleted
anti-Gal3 beads.
5. Mix the beads gently and incubate for 10 min at room
temperature.
6. Pellet the mixture by gentle centrifugation (1000 g in a swing-
ing bucket rotor at room temperature for 10–15 s), collect the
supernatant by Hamilton syringe and store in a fresh microtube at
20 C.
7. The unbound material (step 1 of Subheading 3.3.3) and the
bound material (step 6 of Subheading 3.3.3) of the anti-Gal3
precipitation can be compared in terms of RNA and protein
components (see Note 5).
3.4 Reconstitution of 1. The preparation of the splicing substrate, 32P-labeled MINX,

Splicing Activity in has been described [11] (Subheading 2.1).
U1ΔNE by Anti-Gal3 2. The preparation of U1ΔNE is described in Subheading 3.1.2
Precipitate of 10S of this article.
Fractions 3. Splicing reactions are assembled in a total volume of 24 μl
3.4.1 Reconstitution of (6 μl U1ΔNE, 3.5 mM MgCl2, 1.5 mM ATP, 20 mM creatine
Splicing Activity in U1ΔNE phosphate, 0.5 mM DTT, 20 units RNAsin, 4 μl 32P-labeled
(Fig. 1C) MINX splicing substrate (104 cpm/μl), 60%D) and added to
each tube of beads from step 5 of Subheading 3.3.2. A control
splicing reaction is also prepared in a 12 μl total volume (3 μl
NE(D), 3.5 mM MgCl2, 1.5 mM ATP, 20 mM creatine phos-
phate, 0.5 mM DTT, 20 units RNAsin, 2 μl 32P-labeled MINX
splicing substrate (104 cpm/μl), 60%D).
4. The tubes are mixed gently and rotated end-over-tail at 30 C
for 90 min.
Fig. 2 Representative results from analysis of the anti-Gal3 precipitate of the

10S region of the glycerol gradient: RNA, protein components and restoration of
splicing activity to a nuclear extract depleted of U1 snRNP (U1ΔNE). (a) Northern
blotting analysis of U1 snRNA bound to beads coupled with anti-Gal3 (αG3; lane
2) or to beads coupled with preimmune serum (PI; lane 3) after
5. The mixtures are pelleted by gentle centrifugation (1000 g in

a swinging bucket rotor at room temperature for 10–15 s).
6. The reactions are stopped and proteins eluted off the beads by
adding 24 μl of 2 SDS sample buffer to the tubes containing
beads, and 12 μl of 2 SDS sample buffer to the control tube
containing NE but no beads. The tubes are heated 100 C for
7 min.
7. The tubes are centrifuged gently (1000 g in a swinging
bucket rotor at room temperature for 10–15 s).
8. The supernatants (elutions) are transferred to fresh microtubes:
approximately 48 μl from the bead tubes and 24 μl from the NE
control tube.
9. Proteinase K (20 mg/ml) is added to digest and solubilize the
proteins: 5 μl is added to the 48 μl elution from beads and
2.5 μl is added to the 24 μl NE control.
10. The tubes are incubated at 37 C for 40 min.
11. The tubes are gently centrifuged (1000 g in a swinging
bucket rotor at room temperature for 10–15 s).
12. The bead elutions are diluted with 39.5 μl TE and 10 μl 3 M
sodium acetate. The NE control is diluted with 63.5 μl TE and
10 μl 3 M sodium acetate.
Fig. 2 (continued) immunoprecipitation of fractions 3–5. Bound material is 25%

of the total amount eluted from the beads. Lane 1 represents 10% of the amount
of combined glycerol gradient fractions 3–5 subjected to immunoprecipitation.
(b) Western blotting analysis of polypeptides bound to beads coupled with anti-
Gal3 (lane 2) or to beads coupled with preimmune serum (lane 3). Protein
identities are indicated at right; arrows at left highlight the specific protein
band. Bound material is 75% of the total amount eluted from the beads. Lane
1 represents 10% of the amount of combined glycerol gradient fractions 3–5
subjected to immunoprecipitation. (c) Analysis of splicing activity in U1ΔNE
when mixed with anti-Gal3 (αGal3) beads after precipitation of glycerol gradient
fraction 1 (FR 1), fractions 3–5 (FR 3–5), or 60% Buffer D (60%D). The presence
or absence of each of these precipitates in the splicing reaction mixture is
indicated by the + or – sign above the bolded solid line. The presence or absence
of NE in Buffer D or U1ΔNE is indicated by the + or – sign below the bolded solid
line. Lane 1: 3 μl NE in a splicing assay volume of 12 μl, all of which were
subjected to gel analysis. For lanes 2–4, the total splicing assay volume was
24 μl, half of which were subjected to gel analysis. Lane 2: 6 μl U1ΔNE and
beads from precipitation of 60%D. Lane 3: 6 μl U1ΔNE and beads from
precipitation of Fr 3–5. Lane 4: 6 μl U1ΔNE and beads from precipitation of Fr
1. The positions of migration of the pre-mRNA substrate, the intermediates, and
the products are indicated on the left and the products, intron lariat and ligated
exons, are highlighted by the arrows on the right
13. The RNA is extracted and analyzed; see Subheading 3.4.2 (see
Note 6 and Fig. 2).
3.4.2 RNA Extraction, The procedures for carrying out RNA extraction and denaturing
and Analysis of Product gel electrophoresis to analyze the RNA intermediates and products
of the splicing have been described in Chapter 28 (Subheadings 2.3
and 3.3) of the prior volume dedicated to galectins [11].
4 Notes
1. All chemicals (buffer components, enzymes, etc.) are to be kept

free of ribonuclease (RNase). The commercially purchased
reagent bottles are sequestered from general lab use. Use
RNase-free spatulas (see Note 2) to aliquot chemicals from
these reagent bottles. Wear gloves for all procedures.
2. All glassware (beakers, flasks, bottles, pipettes, etc.) are baked
for a minimum of 4 h at 350 C. Other utensils (spatulas, stir-
bars, etc.) are wrapped in aluminum foil before baking under
the same conditions.
3. DEPC (diethylpyrocarbonate) was added to double-distilled
water (ddH2O) to final concentration of 0.1% vol/vol. The
solution was stirred overnight using a magnetic stir bar at room
temperature and then autoclaved. All solutions containing Tris
are prepared using DEPC treated H2O and then filter steri-
lized. Other solutions (without Tris) can be made with regular
ddH2O and then subsequently treated with DEPC and then
autoclaved.
4. The derivation and characterization of rabbit polyclonal anti-
sera directed against Gal3 have been reported previously: rabbit
#24 [12] rabbit #49 [8]. Preimmune serum from rabbit #49
was used as control.
5. We used general laboratory procedures for analysis of RNA and
protein components. The snRNAs were separated by gel elec-
trophoresis through 13% polyacrylamide–8.3 M urea gels and
then were either stained with ethidium bromide or subjected to
northern blotting [8, 10]. To analyze proteins, immunoblot-
ting was carried out after SDS polyacrylamide gel electropho-
resis [6, 12].
6. A typical result documenting that the anti-Gal3 (#49) immu-
noprecipitate of fractions 3–5 of the glycerol gradient can
reconstitute splicing of U1ΔNE is illustrated in Fig. 2. In
addition to the cognate antigen, Gal3 (Fig. 2B, lane 2), beads
bearing this anti-Gal3 immunoprecipitated fraction contain
components of the U1 snRNP particle: U1 snRNA (Fig. 2A,
lane 2) as well as the U1-specific protein U1-70K and the Sm
B/B0 core polypeptides (Fig. 2B, lane 2). In contrast, beads

bearing the preimmune control did not yield any of these
components (Fig. 2A, B, lane 3).
NE depleted of U1 snRNP (U1ΔNE from Subheading
3.1.2 step 5) was mixed with beads containing the anti-Gal3
precipitate of fractions 3–5 (Subheading 3.3.2, step 5) in a
splicing reaction. Splicing activity in U1ΔNE could be recon-
stituted by the bead-bound Gal3–U1 snRNP complex
(Fig. 2C, lane 3); both products, ligated exons and excised
intron lariat (highlighted by the arrows on the right), as well
as intermediates of the splicing reaction (free exon 1 and lariat-
exon 2) were observed. If fractions 3–5 were replaced by either
60% buffer D or fraction 1 in the immunoprecipitation, we did
not observe any splicing activity (Fig. 2C, lane 2 and lane 4).
Inasmuch as previous studies had shown that the Gal3 in
fraction 1 represented free Gal3 protein not in complex with
any RNP [8], these results indicate that the 10S Gal3–U1
snRNP particle in fractions 3–5 were critical in restoring splic-
ing to U1ΔNE.
Additional control experiments were carried out to support
this conclusion (data not shown here but documented in refer-
ence [10]). First, beads from anti-Gal3 precipitation of frac-
tions 3–5 (or fraction 1) failed to exhibit splicing activity when
assayed in the absence of U1ΔNE. Second, immunoprecipita-
tion of fractions 3–5 using beads coupled with preimmune
serum from rabbit #49 was unable to reconstitute splicing in
U1ΔNE. Finally, beads coupled with anti-Gal3 from another
rabbit (#24) yielded the same results as rabbit anti-Gal3
(#49) [10].
Acknowledgments
The work has been supported by National Science Foundation

Grant MCB-0092919 and Michigan State University Intramural
Research Grant 09-CDFP-2001 (to R.J.P.) and by National Insti-
tutes of Health Grant GM-38740 and Michigan AgBioResearch
Project MICL02455 (to J.L.W.).
References
1. Dignam JD, Lebovitz RM, Roeder RG (1983) 3. Maniatis T, Reed R (1987) The role of small
Accurate transcription initiation by RNA poly- nuclear ribonucleoprotein particles in
merase II in a soluble extract from isolated pre-mRNA splicing. Nature 325:673–678
mammalian nuclei. Nucleic Acids Res 11: 4. Hoskins AA, Friedman LJ, Gallagher SS, Craw-
1475–1489 ford DJ, Anderson EG, Wombacher R,
2. Lerner M, Steitz JA (1981) Snurps and scyrps. Ramirez N, Cornish VW, Gelles J, Moore MJ
Cell 25:298–300 (2011) Ordered and dynamic assembly of sin-
gle spliceosomes. Science 331:1289–1295
5. Dagher SF, Wang JL, Patterson RJ (1995) 9. Conway GC, Krainer AR, Spector DL, Roberts
Identification of galectin-3 as a factor in RJ (1989) Multiple splicing factors are released
pre-mRNA splicing. Proc Natl Acad Sci U S A from endogenous complexes during in vitro
92:1213–1217 pre-mRNA splicing. Mol Cell Biol 9:
6. Vyakarnam A, Dagher SF, Wang JL, Patterson 5273–5280
RJ (1997) Evidence for a role for galectin-1 in 10. Haudek KC, Voss PG, Wang JL, Patterson RJ
pre-mRNA splicing. Mol Cell Biol 17: (2016) A 10S galectin-3-U1 snRNP complex
4730–4737 assembles into active spliceosomes. Nucleic
7. Wang W, Park JW, Wang JL, Patterson RJ Acids Res 44:6391–6397
(2006) Immunoprecipitation of spliceosomal 11. Patterson RJ, Haudek KC, Voss PG, Wang JL
RNAs by antisera to galectin-1 and galectin-3. (2015) Examination of the role of galectins in
Nucleic Acids Res 34:5166–5174 pre-mRNA splicing. Methods Mol Biol 1207:
8. Haudek KC, Voss PG, Locascio LE, Wang JL, 431–449
Patterson RJ (2009) A mechanism for incor- 12. Agrwal N, Sun Q, Wang SY, Wang JL (1993)
poration of galectin-3 into the spliceosome Carbohydrate-binding protein 35.
through its association with U1 snRNP. Bio- I. Properties of the recombinant polypeptide
chemistry 48:7705–7712 and the individuality of the domains. J Biol
Chem 268:14932–14939
INDEX
A Cancers ..................................................... 2, 9, 13, 15–17,

19–30, 171, 172, 205, 285, 291, 308, 368, 391,
Affinities................................ 8, 9, 17, 43, 47, 52, 56, 79, 439, 579, 623, 629–630, 636, 637, 656, 685–710
84, 90, 92, 115, 117, 125, 126, 129, 132, 138, Carbohydrate binding proteins
139, 145, 151, 165, 170, 172–174, 176–180,
(CBPs).................................................... 1–30, 151,
183, 190, 198, 199, 206, 211, 215, 227, 233, 205, 218, 236, 518
311, 317, 320, 419, 455, 480, 485, 543, 566 CD16a ........................................................ 218–220, 222,
Affinity chromatography................................... 14, 55–72,
223, 225, 227, 228
80, 106, 126, 128, 140, 143, 152, 172, 173, 182, Chicken....................................................... 61, 91, 92, 95,
206, 239, 241, 242, 255, 256, 269, 283, 316, 97, 308, 310, 311, 445, 446, 448–455, 459, 491,
317, 344, 485, 497, 560
510, 622, 624–626
Agrocybe cylindracea............................................. 233–244 Chicken angiogenesis...................................622, 624–626
Alkylation.................................................... 76–78, 80, 81, Chorioallantoic membrane (CAM)
85, 273, 332
assay.................................................................... 621
Analytes....................................... 125, 126, 129, 132–134 Cis-trans tautomerism .................................................. 236
Anchorage independent growth .................................697, Clathrin Independent Endocytosis ..................... 391–409
700–702 Coprinopsis cinerea galectin-2 (CGL2) ......................... 56,
Angiogenesis.................................. 19, 20, 368, 621–632,
57, 59, 60, 71, 235
635–651, 655, 656, 658, 686, 706, 707 CRISPR-Cas.................................................427–429, 438
Angiogenesis endothelial cell ......................................621, Cross-linking .............................................. 139, 140, 142,
635, 636, 638–640, 644–647, 649, 650, 655,
149, 171, 172, 180, 303, 414, 636
656, 658, 707 Crystallin........................................................................ 446
Antibody Internalization Assay ........................... 394, 401
Anti-microbial ............................................................... 524 D
APEX2-enhanced electron
microscopy ...................... 354–356, 359, 360, 363 Divinyl sulphone-coupled
Autophagy ..................................2, 27, 30, 248, 354, 363 lactosyl-sepharose .............................61–62, 66–67
Downregulations........................................ 291, 460, 551,
B 686–688, 692–697, 702, 707–709
B cells ................................. 4, 18, 25, 216, 311, 565–579 E

β-galactoside ........................................2, 9, 234–236, 464
Bind and jump model ................................................... 178 Efferocytosis .................................................................. 581
Binding affinity.......................................9, 126, 138, 148, ELISA .................................................................. 316, 317,
181, 188, 190, 193, 196, 198, 199, 414, 477 395, 396, 404, 594, 599, 606–608, 611, 639,
Biomarker .................................9, 15, 291, 339, 594, 706 644, 691, 706, 708
Biotinylation ........................................................ 157, 158, Endosomal sorting complex required for transport
161, 191, 193, 315, 578 (ESCRT) ...............................................16, 29, 416
Blood group positive microbes .......................... 518–520, Endosomes ............................................................ 16, 253,
524, 526 353–356, 358–360, 363, 414
Blood vessels...........................................19, 25, 581, 597, Endothelial cells .................................19, 21, 23–25, 290,
630, 635, 637, 655 582–583, 586–587, 596–598, 621, 635, 636,
638–640, 644–646, 649, 650, 655–662, 702, 707
C Endothelium..................................... 17, 19, 24, 566, 591
Enterocytes ..........................................369–371, 374, 382
Calnexin................................................... 4, 253, 279, 286 Entropy ................................................................. 183, 320
https://doi.org/10.1007/978-1-0716-2055-7, © Springer Science+Business Media, LLC, part of Springer Nature 2022
727
GALECTINS: METHODS AND PROTOCOLS
728 Index
Evolution ................................3–11, 13, 21, 30, 109, 113 464, 478–481, 502, 511, 566, 603, 606, 618,
Exosomes..................... 16, 414–416, 418, 419, 421–423 636, 637, 685, 699
Glycoclusters .......................................308, 316, 327, 328
F Glycoengineering ................................................. 205–212
Glycogenes ..........................................207, 208, 210, 211
Flow cell (fc)......................................................... 125, 133
Fluorescence labeling .................191, 192, 194, 195, 313 Glycoproteins (GPs) ............................................. 2, 9, 10,
Fluorescence microscopy .................................... 307–335, 16, 17, 23, 29, 42, 58, 89–91, 93, 97, 99–100,
354, 355, 358, 363, 415, 418, 504 102, 138, 139, 163, 169, 170, 172–174, 179,
206, 208, 209, 212, 215–229, 238, 353, 367,
G 368, 481, 565, 567, 576, 578, 579, 609, 636, 685
Glycoproteomic............................................................. 215
Galectin............................................. 2–23, 25–30, 41–53, Glycosaminoglycans (GAG) .......................................... 89,
55–68, 71, 72, 75, 76, 79, 81–86, 105–107, 109, 137, 138, 142
115, 125–134, 151–165, 169, 170, 173, 179, Glycosides ............................................128–131, 133, 174
187–200, 205–212, 215–229, 233–244, Glycosylation ..........................................4, 9, 17, 29, 106,
247–286, 289–304, 307–335, 339–351, 188, 190, 200, 206, 208, 216, 218, 685
353–364, 367–370, 391–409, 414, 422, Glycosyltransferases..........................................4, 208, 216
425–439, 445–460, 464, 465, 475–512, GST-galectin................................. 45, 47, 49, 51–53, 505
517–526, 533–544, 549–562, 566–568, 574,
576–579, 581–600, 603, 606, 621–632, H
636–651, 655–662, 681, 685–710
Helicobacter pylori.........................................29, 354, 357
Galectin-1 (Gal-1)................................................... 2, 3, 8,
9, 16–18, 21, 23, 25, 29, 75–86, 126, 128–133, Heteronuclear single quantum multiple-bond correlation
165, 172, 190–192, 194, 198, 199, 206, (HSQMBC) ....................................................... 107
Histochemistry .............................................................316,
234–236, 251, 269, 279, 281, 290–292, 301,
308, 445, 446, 475, 541, 542, 544, 547, 552, 323–325, 328, 331
553, 556–558, 560, 561, 603–618, 622, 661, HIV .............................. 29, 227, 463–465, 467, 469–472
HIV persistence............................................................. 463
663–682, 688, 691, 694, 706, 708, 709
Galectin-3 (Gal-3)......................................................9, 14, Hypoxia ...................................... 636, 637, 643–644, 651
17, 25–29, 89–102, 128, 129, 137–149, 172,
I
206, 291, 293, 301, 341, 342, 347, 350, 354,
356, 357, 360, 361, 363, 367–387, 391–400, Imide bond........................................................... 235, 236
404, 413–423, 445, 446, 568, 574 Immobilization.............................................................. 190
Galectin-3 (Gal-3) secretion Immunofluorescence microscopy ...............................354,
assay.......................................................... 395–396, 356–358, 360–363
398–399, 404, 407–408 Immunohistochemistry ........................................ 11, 308,
Galectin-9 (Gal-9).............................................17, 27, 57, 322, 324, 502, 505, 608, 614, 626, 687, 688,
463–472, 479, 480, 568, 575, 582, 596, 622, 662 692, 693, 706, 707
αGalNAc ............................................................... 235, 236 Inflammatory resolution ...................................... 533, 534
GBP-glycan interactions ............................................... 162 Innate immune lectins ........................................... 29, 526
Gene editing .................................................................. 206 Internal diffusion .......................................................... 179
Gene expression ............................. 4, 426, 427, 599, 626 Intestine ...................................... 290, 367–387, 477, 510
Genetically engineered mice......................................... 704 Invasion ..................................26–28, 371, 603, 686–701
Glutathione-sepharose affinity purification of GST-tagged In vitro fertilization............................. 604–607, 610–612
human galectin-7...........................................56, 59 Isothermal titration calorimetry
Glycan binding protein (GBP) ................... 162, 163, 190 (ITC)...................... 139, 308, 318, 319, 322, 323
Glycan microarrays ..................................... 8, 29, 58, 126,
151–165, 187, 518, 519 J
Glycans..................................... 2–4, 8–10, 14, 17, 23, 24, J-Lat cell line ................................................................. 467
26–29, 43, 52, 53, 89, 105–107, 126, 138, 139,
152, 154–155, 159, 162–165, 170, 173, K
187–191, 198, 199, 205–209, 212, 215, 216,
218, 224–226, 228, 229, 242, 248, 289, 307, Knockdowns ............................................. 19, 20, 42, 392,
308, 339, 353, 354, 363, 367, 392, 445, 463, 401, 429, 694, 709
Index 729
L Morpholino (MO) .......................................................430,
432–435, 438, 439
Lactosyl-sepharose affinity Mucins ....................................15, 24, 169–183, 371, 380
chromatography ............................................56, 60 Mucus ......................................................... 369–371, 375,
Lactotransferrin .................................................... 371, 374 378–380, 386, 426, 478, 479, 481, 512
Latent HIV transcription..................................... 463–472 Multivalent carbohydrates .......................... 174, 176, 179
Lectin and glycoprotein purification.............................. 90 Muscular dystrophies ........................................... 663, 664
Lectin engineering ........................................................ 234
Lectins............................................ 2–4, 9, 23, 26, 27, 55, N
56, 58, 89–91, 93–102, 106, 107, 115, 126,
137–140, 144, 146, 147, 169–183, 190, 192, N-acetyllactosamine (LacNAc)....................................8, 9,
195, 233–244, 248, 257, 258, 269–271, 285, 14, 126, 131, 132, 139, 172–174, 198, 199,
290, 307–309, 311, 320, 331, 367, 368, 393, 215–217, 225, 226, 229, 236, 313, 318, 367,
401, 414, 415, 426, 445, 463, 481, 506, 551, 392, 566, 685
552, 558, 604, 609–610, 615–618, 636, 637, Natural killer cells................................................... 13, 225
692, 707 Network formation .............................................. 656–660
Lectin staining ............................................................... 606 Neutrophils................................................... 8, 13, 24, 25,
Leukocytes .................................................... 4, 8, 23, 163, 56, 289, 479, 533–538, 540, 542, 549–562,
164, 189, 218, 292, 479, 533–544, 565, 583–584, 588, 589, 591, 594–599
581–600, 636 N-glycosylation ...............................................9, 216, 218,
Leukocyte turnover....................................................... 535 228, 229, 249, 265
Ligands ............................................ 2, 3, 8, 9, 14, 16, 17, Nickel NTA-affinity purification of histidine-tagged
21, 28, 56, 75–77, 79, 80, 86, 90–92, 95–97, 102, human galectin-7...........................................56, 58
105–120, 125–127, 129–131, 133, 138, 142, Non-covalent cross-linking.................................. 139, 142
146, 147, 152, 158, 162, 165, 173, 174, 176,
O
179, 187, 188, 190, 191, 215–229, 313,
318–320, 331, 334, 339, 340, 353, 370, 371, Overexpression ................................................15, 16, 392,
382, 414, 445, 478, 489–490, 506, 507, 533, 401, 497, 636, 687–688, 692–697, 703, 704
541, 543, 551, 566, 567, 574, 576–579, 637, 685 Oxidation............................................8, 9, 76, 77, 84–86,
Limb-girdle muscular dystrophy 2B 269, 284, 301, 508, 651
(LGMD2B)............................................... 663–665
Listeria monocytogenes ................................. 354, 356, 371 P
L-Maf ............................................................................. 445
Pax6 ............................................................................... 445
Localization ............................................ 2, 10–14, 16, 19,
Phagosomes......................................................... 353, 354,
21, 27, 30, 216, 253, 301, 310, 311, 367, 374,
356–357, 360–362
416, 636, 679, 706
Phosphatidylserine (PS) ............................... 25, 163, 164,
Luciferase........................................... 340, 341, 344, 347,
189, 533–543, 547
350, 451, 452, 457–460, 703
Photosensitizers................................................... 354–356,
Lysosomes .................................................... 27, 248, 353,
358–360, 362, 363
354, 357–358, 363
Polymerase chain reaction
(PCR) .............................................. 42, 43, 45, 52,
M
236–238, 240, 243, 244, 264, 265, 268, 276,
Mass Spectrometry................................. 79–81, 102, 218, 277, 282, 313, 318, 319, 341, 343, 350, 427,
220, 223, 225, 251, 258, 489, 507, 579 429, 431, 433, 437, 439, 451, 452, 456, 458,
Maternal and placental galectin-1 459, 469–471, 484, 495, 496, 691, 705
models ....................................................... 603–618 Poly-N-acetyllactosamines ............................................ 565
Membrane repair................................................. 663–665, Preaparesis ............................................................ 534, 535
667, 668, 674, 676 Pregnancy outcome ............................................. 603–618
Microinjections.................................................... 426, 428, Proliferation.......................................................15, 19, 20,
432–435, 437–439 260, 275, 276, 427, 566, 621, 635, 636, 646,
Migration.................................................. 19, 20, 25, 367, 656, 686, 689, 697–700, 706, 707, 709
368, 370, 478, 581, 629, 635–637, 639, 645, Proline................................................................................ 9
646, 656–661, 685–690, 697, 699–701 Promoters ........................................................9, 427, 439,
Molecular mimicry .................................... 2, 29, 519, 520 446, 448, 451, 452, 454, 455, 457, 460, 703, 704
730 Index
Protein labeling .................................................... 668, 676 Streptavidin (SA) sensor
Proteoglycans (PGs) .........................................9, 89, 138, chip................................................... 126, 128, 130
139, 142, 147 Substrate specificity ....................................................... 205
P(S/T)AP motif .............................................................. 16 Surface Plasmon Resonance
Pull-down assay ................................................... 478, 480, (SPR)........................................125–134, 152, 173
489–490, 506–508
T
R
T cells ........................................... 13, 15, 17, 21–24, 171,
Receptors .................................................. 2, 9, 10, 16–18, 172, 190, 216, 446, 463, 464, 466–472, 534,
22–26, 29, 106, 169, 172, 173, 188, 215–218, 566, 596, 638, 686
295, 296, 308, 331, 340, 370, 413, 414, 445, Thermodynamics................................................. 139, 152,
464, 519, 535, 551, 565–579, 635, 636, 699 169–183, 320
Recombinant galectin Thiodigalactoside (TDG) ....................................... 14, 59,
purification..................................... 55–72, 84, 498 107, 524, 536, 554
Recombinant galectins...................................... 13, 55–72, Transcription .....................................................9, 11, 293,
82, 84, 155, 157–159, 161, 191, 207, 209, 251, 341, 427, 433, 437, 446–449, 451, 455–457,
311, 320, 350, 481, 487–488, 496, 501–505, 459, 464, 469, 471, 691
507, 521, 523, 525, 536, 538, 539, 541, 555, Transcytosis .......................................................... 367–387
560, 583, 592, 597, 598, 656, 657, 659, 660 Transmigration .............................................................. 598
Reducing agents ...................................23, 58, 76, 77, 81, Tumor grafts................................................ 624, 625, 629
84, 86, 161, 332, 541
U
S
Unconventional secretion........................... 367, 414, 415
Sarcolemma .......................................................... 663–682
Secretion ................................................... 2, 9, 13, 16, 22, V
75, 77, 395, 398, 404, 407, 413–423, 464, 478,
Vasculature........................... 25, 622, 624–627, 630, 655
480, 481, 509, 566, 637 Viability......................................................... 22, 210, 212,
Selenoglycosides ................................................... 107, 115 219, 221, 481, 519, 535, 540, 541, 543, 568,
77Se NMR............................................................ 105–121 574, 622, 632, 651, 697, 698
Sialylation ...................................................................... 206
Solid phase assay................................................... 190, 198 Z
Sprouting .................................................... 635, 636, 646,
648, 651, 655–660, 662 Zebrafish .......................................................425–440, 709

Galectins: Methods and Protocols

Uploaded by

Document Information

Original Description:

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Galectins: Methods and Protocols

Uploaded by

Copyright:

Available Formats

Methods in

Molecular Biology 2442

For further volumes:

Methods and Protocols

Sean R. Stowell and Connie M. Arthur

ISSN 1064-3745 ISSN 1940-6029 (electronic)

© Springer Science+Business Media, LLC, part of Springer Nature 2015, 2022

Boston, MA, USA Sean R. Stowell

Sean R. Stowell, Connie M. Arthur, Richard D. Cummings

11 Examination of Whole-Cell Galectin Binding by Solid Phase

23 Manipulating Galectin Expression in Zebrafish (Danio rerio) . . . . . . . . . . . . . . . . . 425

35 Examination of the Role of Galectins and Galectin Inhibitors

JOSÉ ABAD-RODRÍGUEZ • Membrane Biology and Axonal Repair Laboratory, Hospital

SOPHIA BOROWSKI • Department of Obstetrics and Fetal Medicine, University Medical

SOFIA NASCIMENTO DOS SANTOS • Departamento de Radiofarmácia, Instituto de Pesquisas

MICHAEL HINGAR • Pharma Biotech Development Penzberg, Roche Diagnostics GmbH,

FU-TONG LIU • Institute of Biomedical Sciences, Academia Sinica, Taipei, Taiwan

RONALD J. PATTERSON • Department of Microbiology and Molecular Genetics, Michigan State

VICTORIA SUNDBLAD • Laboratorio de Inmunopatologı́a, Instituto de Biologı́a y Medicina

Galectins: An Ancient Family of Carbohydrate Binding

Key words Galectin, Immunology, Intracellular, Extracellular, Glycobiology, Evolution, Microbes,

CBP Carbohydrate binding protein

GALNts GalNAc transferases

Galectins are an ancient family of soluble β-galactoside carbohy-

(GA) of eukaryotic cells. Terminal and internal glycan and

development since Gregor Mendel elaborated his theory of inheri-

galectins have evolved specialized and accessory functions within

Supporting the latter timeline is the unique sequence and dimeric

recently evolved mono-CRD mammalian galectins (5, 6, 7, 10,

functional groups of galactose and the 3 hydroxyl functionality of

to determine when Gal-1 is released in vivo and whether its activity

1.2.4 Synthesis of Analogous to promoters, transcription factor binding sites, or con-

modifications at PNG sequons is both encoded and conditional—

[30]. Expression and secretion of proteins baring the Tn antigen is

2 Mammalian Galectin Expression and Localization

Gal-1, Gal-3, and Gal-9 are first produced during mammalian

expression can also be hormonally regulated during placental devel-

experimental system and in which compartments they are localized

3 Modern Galectin Functions

3.1 Nuclear and While posttranslational glycan modifications are predominantly

4 (Gemin4) and additional snRNP components as Gal-1 and Gal-3

cases correlating with disease progression and resistance to therapy

3.4 Galectin Genes encoding conventional secretory proteins bare a leader or

at the cell surface to form a functional lattice. In vitro binding

stabilization may also be at play on the surface of B cells during

3.6 Homeostasis Because many studies of cytoplasmic galectins in intracellular sig-

counter-receptors by which Gal-1 is secreted under homeostatic

4.2 Tumor Tumor infiltration by lymphocytes has been linked to a survival

Gal-1 and Gal-3 are often overexpressed in the tumor microenvi-

types. In Fig. 6, we summarize major roles for galectins in the

directly to the T-cell receptor (TCR) to decrease antigen sensitivity

4.3.2 Regulation of Neutrophils and neutrophil-like phagocytic immune cells execute

to sites of inflammation through blood vessels. Egress of neutro-

4.3.3 Modulation of As in cancer signaling networks, inflammatory signaling pathways

4.3.4 Intracellular In addition to interacting with signaling complexes, cytoplasmic

formation of bactericidal MBL attack complexes or recruitment of

immune response attributable to the missing galectin [145]. In

4.3.6 Bacteria The role of galectins in direct immunity against microorganisms

structures on LPS. Use of glycan microarrays and genetic variants of

[159]. While the aforementioned role of Gal-8 in intracellular

5 Summary and Conclusions

Many of the evolving mechanisms of galectin function point to a

13. Sodium azide, ReagentPlus®, 99.5%, CAS 26628-22-