Microbial Sulfur Metabolism

Christiane Dahl Cornelius G.
Friedrich (Editors)
Microbial Sulfur Metabolism

Christiane Dahl
Cornelius G. Friedrich (Editors)
Microbial Sulfur
Metabolism
With 65 Figures, 11 in Color and 27 Tables
Dr. Christiane Dahl Professor Dr. Cornelius G. Friedrich
Institute for Microbiology Chair Technical Microbiology
& Biotechnology Department of Biochemical and Chemical
Rheinische Friedrich-Wilhelms- Engineering
Universitt Bonn University of Dortmund
Meckenheimer Allee 168 D-44221 Dortmund
D-53115 Bonn Germany
Germany
Library of Congress Control Number: 2007929727
ISBN-13 978-3-540-72679-1 Springer-Verlag Berlin Heidelberg New York
This work is subject to copyright. All rights are reserved, whether the whole or part of the material is
concerned, specifically the rights of translation, reprinting, reuse of illustrations, recitation, broadcasting,
reproduction on microfilm or in any other way, and storage in data banks. Duplication of this publication
or parts thereof is permitted only under the provisions of the German Copyright Law of September 9,
1965, in its current version, and permission for use must always be obtained from Springer. Violations
are liable to prosecution under the German Copyright Law.
Springer-Verlag is a part of Springer Science+Business Media
springer.com
Springer-Verlag Berlin Heidelberg 2008
The use of general descriptive names, registered names, trademarks, 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.
Editor: Dr. Christina Eckey, Heidelberg

Desk Editor: Anette Lindqvist, Heidelberg
Production: SPi
Typesetting: SPi
Cover Design: Design & Production, Heidelberg
Printed on acid-free paper 149/3152-HM 5 4 3 2 1 0

Preface
Sulfur is an essential element for the living cell. Sulfur occurs in oxidation states of
+2 to 6, is highly reactive and is used by prokaryotes not only to build up cell
constituents but also for energy transformation. Basic research has revealed in
recent years an increasing rate of information on prokaryotic reactions, proteins and
genes involved in sulfur transformations. New insights are emerging concerning
enzyme systems involved in sulfur metabolism, genomics and proteomics of sulfur-
metabolizing bacteria and archaea, and the ecology of prokaryotes oxidizing and
reducing sulfur compounds. Furthermore, new methods have been developed and
are being applied to study microbial sulfur metabolism.
To summarize the fast-moving developments of recent years, to exchange
knowledge and to discuss future developments and research needs, the International
Symposium on Microbial Sulfur Metabolism was held in Mnster, Germany, from
29 June to 2 July 2006. This symposium brought together 85 scientists from 16
countries and was felt to be timely after a previous meeting on bacterial sulfur
metabolism which took place in London in 1982.
The symposium in Mnster focused on prokaryotic sulfur energy metabolism,
the biochemistry of the enzymes involved, the molecular genetics of such enzyme
systems as well as on the ecosystems harboring sulfur-metabolizing prokaryotes.
Pathways of sulfur metabolism present in different physiological groups were
compiled. A collection of invited lectures presented the state of the art regarding
biochemistry, ecology, proteomics, genomics and evolution of chemotrophic and
phototrophic sulfur-oxidizing bacteria, anaerobic sulfate-reducing bacteria and
hyperthermophilic sulfur-metabolizing archaea. The symposium setting and time
schedule encouraged informal discussion and exchange between young and estab-
lished scientists.
This proceedings volume presents the essence of the symposium represented by
23 invited lectures which introduce and report cutting-edge research in the various
fields. The efforts of the authors have created a book which compiles the state of
the art on sulfur metabolism in phototrophic and chemotrophic bacteria and
archaea. The book is organized according to the seven major topics of the
symposium: (1) sulfate-reducing bacteria, (2) genomics/proteomics of sulfur-
metabolizing prokaryotes, (3) biochemistry of sulfur-compound oxidation, (4)
metabolism of organosulfur compounds, (5) dissimilatory sulfur metabolism in
archaea, (6) ecology of sulfur bacteria and (7) specific methods and applied aspects.
v
vi Preface
To each topic leading experts in the field contribute several chapters to yield a
detailed picture of the current state of the art.
Preparing the symposium, we were not aware of all the fascinating research in
these fields. Also, we had limited space and could not include the contents of fasci-
nating lectures which were selected from submitted abstracts. Also, some topics are
not covered by this volume, like sulfur activation and assimilation pathways in
prokaryotes and higher eukaryotes, transport of sulfur compounds, and biosynthe-
sis of sulfur-containing cell constituents. The reader is, however, referred to the
volume Sulfur Metabolism in Phototrophic Organisms (R. Hell, C. Dahl, D. Knaff
and T. Leustek, 2008, eds, Springer, New York, in press).
The organizers of this symposium thankfully acknowledge the substantial finan-
cial support from the Federation of the European Microbiological Societies
(Brussels), the Deutsche Forschungsgemeinschaft (Bonn), the Fonds der Chemischen
Industrie (Frankfurt), the Vereinigung fr allgemeine und angewandte Mikrobiologie
(Frankfurt) and the Gesellschaft fr Biochemie und Molekularbiologie (Frankfurt).
Their support enabled a scientifically lively meeting, and the attending community
decided on a follow-up meeting which will be organized by and Ins Pereira and
Christiane Dahl will take place in Portugal in 2009.
Bonn, Dortmund, March 2007 Christiane Dahl and Cornelius Friedrich
Contributors
Adam P. Arkin, Howard Hughes Medical Institute, Department of Bioengineering,

University of California, Berkeley, CA 94720, USA, Physical Biosciences
Division, E.O. Lawrence Berkeley National Laboratory, Berkeley, CA 94720,
USA, and Virtual Institute of Microbial Stress and Survival, Berkeley, CA 94710,
USA
Nurgul C. Balci, Biochemistry Department, University of Missouri, Columbia,
MO 65211, USA
Frank Bardischewsky, Lehrstuhl fr Technische Mikrobiologie, Fachbereich Bio-
und Chemieingenieurwesen, Universitt Dortmund, 44221 Dortmund, Germany
Simn Beard, Laboratory of Molecular Microbiology and Biotechnology,
Department of Biology, Faculty of Sciences, University of Chile, Santiago, Chile
Anke Behrens, Fachbereich Biologie, Universitt Konstanz, 78457 Konstanz,
Germany
Donald A. Bryant, Department of Biochemistry and Molecular Biology, The
Pennsylvania State University, University Park, PA 16802, USA
Thomas Bchert, Fachbereich Biologie, Universitt Konstanz, 78457 Konstanz,
Germany
Donald E. Canfield, Nordic Center for Earth Evolution and Institute of Biology,
University of Southern Denmark, 5230 Campusvej, 5000 Odense M, Denmark
Leong-Keat Chan, College of Marine and Earth Studies and Delaware
Biotechnology Institute, University of Delaware, Newark, DE 19711, USA
An Chi, Laboratory of Molecular Microbiology and Biotechnology, Department of
Biology, Faculty of Sciences, University of Chile, Santiago, Chile
Alasdair M. Cook, Department of Biology, Universitt Konstanz, 78457 Konstanz,
Germany
Christiane Dahl, Institut fr Mikrobiologie & Biotechnologie, Rheinische
Friedrich-Wilhelms-Universitt Bonn, Meckenheimer Allee 168, 53115 Bonn,
Germany
vii
viii Contributors
Karin Denger, Department of Biology, The University, 78457 Constance, Germany

Lina De Smet, Laboratory of Protein Biochemistry and Protein Engineering,
Department of Biochemistry, Microbiology and Zoophysiology, Ghent University,
K.L. Ledeganckstraat 35, 9000 Ghent, Belgium
Stephan Duller, Department of Microbial Ecology, University of Vienna,
Althanstrasse 14, 1090 Vienna, Austria
Ulrich Ermler, Max-Planck-Institut fr Biophysik, 60438 Frankfurt, Germany
Giovanni Esposito, Subdepartment of Environmental Technology, Agricultural
University of Wageningen, Biotechnion, Bomenweg 2, P.O. Box 8129, 6700 EV
Wageningen, The Netherlands
Jrg Fischer, Lehrstuhl fr Technische Mikrobiologie, Fachbereich Bio- und
Chemieingenieurwesen, Universitt Dortmund, 44221 Dortmund, Germany
Bettina Franz, Institut fr Mikrobiologie und Biotechnologie, Rheinische Friedrich-
Wilhelms-Universitt Bonn, Meckenheimer Allee 168, 53115 Bonn, Germany, and
Fachbereich Oecotrophologie, Mikrobiologie und Lebensmittelhygiene, Hochschule
Niederrhein, Rheydter Strae 277, 41065 Mnchengladbach, Germany
Cornelius G. Friedrich, Lehrstuhl fr Technische Mikrobiologie, Fachbereich
Bio- und Chemieingenieurwesen, Universitt Dortmund, 44221 Dortmund,
Germany
Niels-Ulrik Frigaard, Copenhagen Biocenter, Department of Molecular Biology,
University of Copenhagen, Ole Maales Vej 5, 2200 Copenhagen N, Denmark
Gnter Fritz, Fachbereich Biologie, Universitt Konstanz, 78457 Constance,
Germany
Frauke Grimm, Institut fr Mikrobiologie und Biotechnologie, Rheinische
Germany
Thomas E. Hanson, College of Marine and Earth Studies and Delaware
Petra Hellwig, Laboratoire dlectrochemie, Institut Chimie, Universit Lous
Pasteur, 4 Rue Blaise Pascal, 67000 Strasbourg, France
Joost Hoek, Nordic Center for Earth Evolution and Institute of Biology, University
of Southern Denmark,5230 Campusvej, 5000 Odense M, Denmark
Michael Hgler, Biology Department, Woods Hole Oceanographic Institution,
Woods Hole, MA 02543, USA, and Leibniz Institute of Marine Sciences, 24105
Kiel, Germany
Donald F. Hunt, Department of Chemistry, University of Virginia, Charlottesville,
VA 22904-4319, USA
Contributors ix
Carlos A. Jerez, Laboratory of Molecular Microbiology and Biotechnology,

Eric F. Johnson, Virginia Bioinformatics Institute , Virginia Polytechnic Institute
and State University, Washington Street, MC 0477, Blacksburg, VA 24061, USA
Ulrike Kappler, School of Molecular & Microbial Sciences, Centre for Metals in
Biology, The University of Queensland, Brisbane, QLD 4072, Australia
Arnulf Kletzin, Institute of Microbiology and Genetics, Darmstadt University of
Technology, Schnittspahnstrasse 10, 64287 Darmstadt, Germany
Peter M.H. Kroneck, Fachbereich Biologie, Universitt Konstanz, 78457
Constance, Germany
Piet N.L. Lens, Subdepartment of Environmental Technology, Agricultural
Shuang-Jiang Liu, State Key Laboratory of Microbial Resources, Institute of
Microbiology, Chinese Academy of Sciences, Beijing 100080, Peoples Republic
of China
Alexander Loy, Department of Microbial Ecology, University of Vienna,
William Martin, Institut fr Botanik III, Heinrich Heine Universitt Dsseldorf,
Universittsstr. 1, 40225 Dsseldorf, Germany
Roel J.W. Meulepas, Subdepartment of Environmental Technology, Agricultural
Rachael Morgan-Kiss, College of Marine and Earth Studies and Delaware
Biswarup Mukhopadhyay, Virginia Bioinformatics Institute, Department of
Biochemistry and Department of Biology, Virginia Polytechnic Institute and State
University, Washington Street, MC 0477, Blacksburg, VA 24061, USA
Grazyna Orawski, Lehrstuhl fr Technische Mikrobiologie, Fachbereich Bio- und
Ins A. Cardoso Pereira, Instituto de Tecnologia Qumica e Biolgica, Universidade
Nova de Lisboa, Av. da Repblica EAN, 2784505 Oeiras, Portugal
Alexander Prange, Fachbereich Oecotrophologie, Mikrobiologie und
Lebensmittelhygiene, Hochschule Niederrhein Niederrhein University of
Applied Sciences, Rheydter Strae 277, 41065 Mnchengladbach, Germany, and
Center for Advanced Microstructures and Devices, Louisiana State University,
6980 Jefferson Highway, Baton Rouge, LA 70806, USA
x Contributors
Armin Quentmeier, Lehrstuhl fr Technische Mikrobiologie, Fachbereich Bio-

und Chemieingenieurwesen, Universitt Dortmund, 44221 Dortmund, Germany
Barbara Rapp-Giles, Biochemistry Department, University of Missouri, Columbia,
MO 65211, USA
Dagmar Rother, Lehrstuhl fr Technische Mikrobiologie, Fachbereich Bio- und
Ricardo Sampaio, Subdepartment of Environmental Technology, Agricultural
University of Wageningen, Biotechnion, Bomenweg, 2, P.O. Box 8129, 6700 EV
Savvas Savvides, Laboratory of Protein Biochemistry and Protein Engineering,
Alexander Schiffer, Fachbereich Biologie, Universitt Konstanz, 78457 Konstanz,
Germany
Jeffrey Shabanowitz, Department of Chemistry, University of Virginia, Charlottesville,
VA 22904-4319, USA
Stefan M. Sievert, Biology Department, Woods Hole Oceanographic Institution,
Woods Hole, MA 02543, USA
Theo H.M. Smits, Department of Biology, Universitt Konstanz, 78457 Konstanz,
Germany
Dimitry Y. Sorokin, Winogradsky Institute of Microbiology, Russian Academy
of Sciences, Prospect 60-let Octyabrya 7/2, 117811 Moscow, Russia, and
Department of Biotechnology, Delft University of Technology, Julianalaan 67,
2628 BC Delft, The Netherlands
Jan Stout, Laboratory of Protein Biochemistry and Protein Engineering, Department
of Biochemistry, Microbiology and Zoophysiology, Ghent University, K.L.
Ledeganckstraat 35, 9000 Ghent, Belgium
Craig D. Taylor, Biology Department, Woods Hole Oceanographic Institution,
Ursula Theissen, Institut fr Botanik III, Heinrich Heine Universitt Dsseldorf,
Universittsstr. 1, 40225 Dsseldorf, Germany
Hans G. Trper, Institut fr Mikrobiologie & Biotechnologie, Rheinische
Germany
Lissette Valenzuela, Laboratory of Molecular Microbiology and Biotechnology,
Contributors xi
Marcus Vallero, Subdepartment of Environmental Technology, Agricultural

Jozef Van Beeumen, Laboratory of Protein Biochemistry and Protein Engineering,
Bjorn Vergauwen, Laboratory of Protein Biochemistry and Protein Engineering,
Gerrit Voordouw, Department of Biological Sciences, University of Calgary,
Calgary, AB T2N 1N4, Canada
Michael Wagner, Department of Microbial Ecology, University of Vienna,
Judy D. Wall, Biochemistry Department, University of Missouri, Columbia, MO
65211, USA, and Virtual Institute of Microbial Stress and Survival, Berkeley, CA
94710, USA
Carl O. Wirsen, Biology Department, Woods Hole Oceanographic Institution,
Contents
Preface. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . v
Contributors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . vii
1 Genetics and Genomics of Sulfate Respiration in Desulfovibrio. . . . 1

Judy D. Wall, Adam P. Arkin, Nurgul C. Balci, Barbara
Rapp-Giles
1.1 Introduction. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1
1.2 Approach. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2
1.3 Sulfate Metabolism . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5
1.4 Lactate Oxidation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6
1.5 Hydrogenases . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7
1.6 Transmembrane Electron-Conducting Complexes . . . . . . . . . . . . . 8
1.7 Conclusions. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11
2 Living on Sulfate: Three-Dimensional Structure

and Spectroscopy of Adenosine 5-Phosphosulfate Reductase
and Dissimilatory Sulfite Reductase . . . . . . . . . . . . . . . . . . . . . . . . . . 13
Gnter Fritz, Alexander Schiffer, Anke Behrens, Thomas Bchert,
Ulrich Ermler, Peter M.H. Kroneck
2.1 Introduction. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13
2.2 Adenosine 5-Phosphosulfate Reductase . . . . . . . . . . . . . . . . . . . . 14
2.2.1 Molecular Properties of APSR . . . . . . . . . . . . . . . . . . . . . . 15
2.2.2 Three-Dimensional Structure of APSR . . . . . . . . . . . . . . . 17
2.2.3 Reaction Mechanism of APSR. . . . . . . . . . . . . . . . . . . . . . 17
2.3 Dissimilatory SIR . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 20
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 21
3 Respiratory Membrane Complexes of Desulfovibrio . . . . . . . . . . . . . 24

Ins A. Cardoso Pereira
3.1 Introduction. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 24
3.2 Membrane Complexes Conserved in Sulfate Reducers . . . . . . . . . 25
xiii
xiv Contents
3.2.1 The Qmo Complex . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 26

3.2.2 The Dsr Complex . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 29
3.3 Membrane Complexes Found Only in Desulfovibrio spp. . . . . . . . 30
3.3.1 The Hmc and 9Hc Complexes . . . . . . . . . . . . . . . . . . . . . . 31
3.3.2 The Tmc Complex . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 31
3.4 Conclusions. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 32
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 33
4 Biochemical and Evolutionary Aspects of Eukaryotes

That Inhabit Sulfidic Environments . . . . . . . . . . . . . . . . . . . . . . . . . . 36
Ursula Theissen, William Martin
4.1 Introduction. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 36
4.2 Animals in Sulfidic Environments . . . . . . . . . . . . . . . . . . . . . . . . . 37
4.3 Sulfide-Oxidizing Enzymes in Eukaryotes. . . . . . . . . . . . . . . . . . . 38
4.4 The Possible Functions of SQR-Related
Genes in Eukaryotes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 39
4.5 Sulfide and Eukaryotic Evolution. . . . . . . . . . . . . . . . . . . . . . . . . . 40
4.6 Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 42
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 43
5 Evolution and Ecology of Microbes Dissimilating

Sulfur Compounds: Insights from Siroheme
Sulfite Reductases . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 46
Alexander Loy, Stephan Duller, Michael Wagner
5.1 Introduction. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 46
5.2 Evolution of Dissimilatory Sulfite Reductases. . . . . . . . . . . . . . . . 47
5.2.1 Sulfate/Sulfite-Reducing Microorganisms . . . . . . . . . . . . . 47
5.2.2 DsrAB-Containing Syntrophs:
Former Sulfate/Sulfite-Reducing Microorganisms? . . . . . . 50
5.2.3 Sulfur-Oxidizing Bacteria. . . . . . . . . . . . . . . . . . . . . . . . . . 51
5.2.4 The Root and Major Branches of the DsrAB Tree . . . . . . 52
5.2.5 Other Non-DsrAB Dissimilatory Sulfite Reductases . . . . . 52
5.3 Molecular Insights into the Ecology of DsrAB-Employing
Microorganisms . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 53
5.3.1 PCR-Based Surveys . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 53
5.3.2 Metagenomics. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 54
5.4 Conclusions. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 55
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 56
6 Genomic and Evolutionary Perspectives on Sulfur

Metabolism in Green Sulfur Bacteria . . . . . . . . . . . . . . . . . . . . . . . . . 60
Niels-Ulrik Frigaard, Donald A. Bryant
6.1 Introduction. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 60
Contents xv
6.1.1Green Sulfur Bacteria . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 61

6.1.2Genome Sequencing Projects of Green
Sulfur Bacteria . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 62
6.2 Compounds Oxidized by Green Sulfur Bacteria . . . . . . . . . . . . . . 63
6.3 Enzymes Involved in Sulfur-Compound Oxidation . . . . . . . . . . . . 65
6.3.1 Overview of the Putative Sulfur Compound
Oxidation Enzymes. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 66
6.3.2 Dissimilatory Sulfite Reductase . . . . . . . . . . . . . . . . . . . . . 66
6.3.3 Sulfide:Quinone Reductase. . . . . . . . . . . . . . . . . . . . . . . . . 67
6.3.4 Flavocytochrome c . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 69
6.3.5 Sulfite Oxidation. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 69
6.3.6 Thiosulfate Oxidation by the Sox System . . . . . . . . . . . . . 70
6.3.7 A Novel Complex: SoyYZ . . . . . . . . . . . . . . . . . . . . . . . . . 71
6.4 Assimilatory Sulfur Metabolism. . . . . . . . . . . . . . . . . . . . . . . . . . . 71
6.5 Possible Phage-Mediated Lateral Gene Transfer . . . . . . . . . . . . . . 72
6.6 Conclusions. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 72
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 73
7 Differential-Expression Proteomics for the Study

of Sulfur Metabolism in the Chemolithoautotrophic
Acidithiobacillus ferrooxidans . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 77
Lissette Valenzuela, An Chi, Simn Beard, Jeffrey Shabanowitz,
Donald F. Hunt, Carlos A. Jerez
7.1 Introduction. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 77
7.2 Sulfur Metabolism in A. ferrooxidans . . . . . . . . . . . . . . . . . . . . . . 78
7.3 Proteomics of A. ferrooxidans Grown
in Sulfur Compounds . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 78
7.4 Thiosulfate Sulfur Transferases from A. ferrooxidans . . . . . . . . . . 79
7.5 Other Proteins Involved in Sulfur Metabolism. . . . . . . . . . . . . . . . 82
7.6 High-Throughput Proteomics of Periplasmic Proteins
Induced by Growth of A. ferrooxidans
on Sulfur Compounds . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 83
7.7 Conclusions. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 85
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 85
8 Sulfur and Light? History and Thiology

of the Phototrophic Sulfur Bacteria. . . . . . . . . . . . . . . . . . . . . . . . . . . 87
Hans G. Trper
8.1 Introduction. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 88
8.2 Discovery of Sulfur-Oxidizing Microorganisms . . . . . . . . . . . . . . 88
8.3 Identification of Conspicuous Inclusions as Sulfur . . . . . . . . . . . . 89
8.4 Enrichment Cultures First Taxonomy
and the Question of Photosynthesis . . . . . . . . . . . . . . . . . . . . . . . . 91
8.5 Pure Cultures of Phototrophic Sulfur Bacteria at Last! . . . . . . . . . 92
xvi Contents
8.6 The Age of Enzymology and Isotope Labeling . . . . . . . . . . . . . . . 93

8.7 Advent of Molecular Genetics . . . . . . . . . . . . . . . . . . . . . . . . . . . . 95
8.8 Further Reading . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 96
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 97
9 Thiosulfate and Sulfur Oxidation in Purple Sulfur Bacteria . . . . . . 101

Frauke Grimm, Bettina Franz, Christiane Dahl
9.1 Introduction. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 101
9.2 Oxidation of Thiosulfate in A. vinosum . . . . . . . . . . . . . . . . . . . . . 103
9.2.1 sox Genes in A. vinosum. . . . . . . . . . . . . . . . . . . . . . . . . . . 104
9.2.2 Sox Proteins in A. vinosum. . . . . . . . . . . . . . . . . . . . . . . . . 105
9.2.3 Inactivation and Complementation of sox Genes
in A. vinosum . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 106
9.3 Oxidation of Stored Sulfur in A. vinosum . . . . . . . . . . . . . . . . . . . 107
9.3.1 The dsr Operon and Proteins Encoded Therein . . . . . . . . . 107
9.3.2 Distribution of dsr Genes in Organisms
with Dissimilatory Sulfur Metabolism
and Phylogenetic Analysis . . . . . . . . . . . . . . . . . . . . . . . . . 108
9.3.3 Model of the Sulfur Oxidation Pathway
in A. vinosum . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 111
9.4 Conclusions. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 113
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 113
10 Sulfur Oxidation in Chlorobium tepidum

(syn. Chlorobaculum tepidum): Genetic
and Proteomic Analyses . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 117
Leong-Keat Chan, Rachael Morgan-Kiss, Thomas E. Hanson
10.1 Introduction. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 117
10.1.1 Background . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 117
10.1.2 Sulfur-Compound Dynamics in C. tepidum
Batch Cultures. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 118
10.2 Genetic Analyses . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 120
10.2.1 Organization of Genes Encoding Putative
Sulfur Oxidation Functions. . . . . . . . . . . . . . . . . . . . . . . 120
10.2.2 Mutations Affecting Sulfur Oxidation
Have Secondary Effects on Light Harvesting . . . . . . . . 120
10.2.3 Additional Genetic Techniques Are Needed . . . . . . . . . 121
10.3 Proteomic Analysis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 122
10.3.1 Why Proteomics? . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 122
10.3.2 Proteomic Analysis of Subcellular Fractions . . . . . . . . . 122
10.4 Conclusions. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 124
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 125
Contents xvii
11 Structural Insights into Component SoxY

of the Thiosulfate-Oxidizing Multienzyme System
of Chlorobaculum thiosulfatiphilum . . . . . . . . . . . . . . . . . . . . . . . . . . . 127
Jan Stout, Lina De Smet, Bjorn Vergauwen, Savvas Savvides,
Jozef Van Beeumen
11.1 Introduction. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 127
11.2 SoxY Structure . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 128
11.2.1 Overall Structure. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 128
11.2.2 SoxY Monomer. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 130
11.2.3 SoxY Dimer . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 130
11.2.4 SoxY Tetramer . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 131
11.2.5 Location of the Disulfide Bridges
and the Potential Sulfur Binding Site . . . . . . . . . . . . . . . 132
11.3 Discussion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 133
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 136
12 Redox Control of Chemotrophic Sulfur

Oxidation of Paracoccus pantotrophus. . . . . . . . . . . . . . . . . . . . . . . . . 139
Cornelius G. Friedrich, Armin Quentmeier,
Frank Bardischewsky, Dagmar Rother, Grazyna Orawski,
Petra Hellwig, Jrg Fischer
12.1 The Sulfur-Oxidizing Enzyme System
of Paracoccus pantotrophus . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 139
12.2 Abundance of the sox Genes in Bacteria . . . . . . . . . . . . . . . . . . . 142
12.3 The Physiological Function of the Flavoprotein SoxF. . . . . . . . . 144
12.4 The Periplasmic Partners of SoxV for Transfer of Electrons. . . . 146
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 148
13 Bacterial Sulfite-Oxidizing Enzymes Enzymes

for Chemolithotrophs Only?. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 151
Ulrike Kappler
13.1 Introduction Sulfite in the Environment
and in Cell Metabolism. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 151
13.2 Sulfite-Oxidizing Enzymes. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 152
13.3 Structure and Function of Sulfite-Oxidizing Enzymes . . . . . . . . 153
13.4 Phylogeny of Sulfite-Oxidizing Enzymes . . . . . . . . . . . . . . . . . . 154
13.5 Diversity of Enzymes Within the Sulfite Oxidase Family . . . . . . 157
13.5.1 Group 1 SOE Like Enzymes Originating from
Pathogenic Microorganisms . . . . . . . . . . . . . . . . . . . . . . 157
13.5.1.1 Group 1A Enzymes: YedY
and Related Proteins . . . . . . . . . . . . . . . . . . . 158
13.5.1.2 Group 1B 30-kDa Mo-Domain Proteins . . 159
xviii Contents
13.5.2 Group 2: Classic Sulfite-Oxidizing Enzymes

and Nitrate Reductases . . . . . . . . . . . . . . . . . . . . . . . . . . 161
13.5.2.1 Group 2A: Sulfite Oxidases
and Plant Nitrate Reductases . . . . . . . . . . . . . 161
13.5.2.2 Group 2B: SoxCD-Like Enzymes
Sulfur Dehydrogenases . . . . . . . . . . . . . . . 162
13.5.2.3 Group 2C: SorAB-Like
Sulfite Dehydrogenases . . . . . . . . . . . . . . . . . 163
13.5.2.4 Other Sulfite-Oxidizing
Enzymes in Group 2 . . . . . . . . . . . . . . . . . . . 164
13.5.3 Group 3: Sulfite-Oxidizing Enzymes
Enzymes from Archaea, Phototrophic
and Soil Bacteria. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 164
13.6 Conclusions. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 165
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 166
14 Sulfonates and Organotrophic Sulfite Metabolism . . . . . . . . . . . . . . 170

Alasdair M. Cook, Theo H.M. Smits, Karin Denger
14.1 Introduction. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 170
14.2 Biosynthesis of Organosulfonates . . . . . . . . . . . . . . . . . . . . . . . . 173
14.3 Dissimilation of Organosulfonates . . . . . . . . . . . . . . . . . . . . . . . . 175
14.4 The Detoxification or Fate of Sulfite . . . . . . . . . . . . . . . . . . . . . . 176
14.5 Sulfite Dehydrogenases in Sulfonate Metabolism . . . . . . . . . . . . 178
14.6 Conclusions. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 180
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 181
15 Oxidation of Sulfur and Inorganic Sulfur Compounds

in Acidianus ambivalens. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 184
Arnulf Kletzin
15.1 Introduction. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 184
15.2 Sulfur and Sulfur Oxidation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 186
15.3 A. ambivalens and A. tengchongensis SORs . . . . . . . . . . . . . . . . 188
15.3.1 SOR 3D Structure . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 190
15.3.2 SOR Subunit and Active-Site Structure . . . . . . . . . . . . . 194
15.3.3 SOR Reaction Mechanism . . . . . . . . . . . . . . . . . . . . . . . 194
15.4 Oxidation of Soluble Sulfur Compounds in Acidianus . . . . . . . . 196
15.4.1 Sulfite:Acceptor Oxidoreductase . . . . . . . . . . . . . . . . . . 196
15.4.2 Thiosulfate:Quinone Oxidoreductase . . . . . . . . . . . . . . . 197
15.4.3 Tetrathionate Hydrolase . . . . . . . . . . . . . . . . . . . . . . . . . 197
15.4.4 Sulfide:Quinone Oxidoreductase . . . . . . . . . . . . . . . . . . 198
15.5 Conclusions. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 198
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 199
Contents xix
16 A Novel Coenzyme F420 Dependent Sulfite Reductase

and a Small Sulfite Reductase in Methanogenic Archaea . . . . . . . . . 202
Eric F. Johnson, Biswarup Mukhopadhyay
16.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 202
16.2 Incompatibility of Methanogenesis and Sulfate Reduction,
Sulfite As the Key Determinant . . . . . . . . . . . . . . . . . . . . . . . . 206
16.3 Inevitable Exposure of a Methanogen to Sulfite
in Hydrothermal Vents and on Early Earth. . . . . . . . . . . . . . . . 206
16.4 Use of Sulfite As a Sulfur Source by
Methanocaldococcus jannaschii and Other Methanogens . . . . 207
16.5 Expression of a Novel Coenzyme F420
Dependent Sulfite Reductase in Methanocaldococcus
jannaschii During Growth on Sulfite . . . . . . . . . . . . . . . . . . . . 207
16.6 Fsr, Combining Structural Components of Two Different
Dissimilatory Metabolic Machineries to Bring
About a Sulfite Reduction Function . . . . . . . . . . . . . . . . . . . . . 209
16.7 Purified Fsr Exhibits Properties Predicted from
the Primary Structure . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 210
16.8 Fsr, a Sulfite Detoxification Tool and
an Assimilatory Enzyme . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 211
16.9 Homologs of Fsr in Other Organisms. . . . . . . . . . . . . . . . . . . . 212
16.10 Small Sulfite Reductases in Methanogens . . . . . . . . . . . . . . . . 212
16.11 Conclusion and Hypotheses . . . . . . . . . . . . . . . . . . . . . . . . . . . 213
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 214
17 Archaeal and Bacterial Sulfur Oxygenase-Reductases:

Genetic Diversity and Physiological Function . . . . . . . . . . . . . . . . . . 217
Shuang-Jiang Liu
17.1 Introduction. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 217
17.2 Diversity of Archaeal SORs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 219
17.2.1 SORAb from A. brierleyi . . . . . . . . . . . . . . . . . . . . . . . . . 219
17.2.2 SORAa from A. ambivalens . . . . . . . . . . . . . . . . . . . . . . . 220
17.2.3 SORAt from A. tengchongensis . . . . . . . . . . . . . . . . . . . . 220
17.2.4 SORSt from S. tokodaii . . . . . . . . . . . . . . . . . . . . . . . . . . 220
17.2.5 SORSm from S. metallicus . . . . . . . . . . . . . . . . . . . . . . . . 221
17.3 Efforts To Identify Bacterial SORs. . . . . . . . . . . . . . . . . . . . . . . . 221
17.3.1 SORAqa from A. aeolicus . . . . . . . . . . . . . . . . . . . . . . . . . 221
17.3.2 SORAct from Acidithiobacillus sp. strain SM-1 . . . . . . . 221
17.4 SOR Links Elemental Sulfur Oxidation to ATP Synthesis
via Sulfite:Acceptor Oxidoreductase and Thiosulfate:Acceptor
Oxidoreductase . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 222
17.5 Physiological Regulation of SOR Activity in Archaea . . . . . . . . 223
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 223
xx Contents
18 Diversity of Halophilic Sulfur-Oxidizing Bacteria

in Hypersaline Habitats. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 225
Dimitry Y. Sorokin
18.1 Introduction. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 225
18.2 Description of Habitats Investigated. . . . . . . . . . . . . . . . . . . . . . 226
18.3 Enrichment Strategy . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 227
18.4 Moderately Halophilic Aerobic SOB . . . . . . . . . . . . . . . . . . . . . 228
18.5 Extremely Halophilic Aerobic SOB . . . . . . . . . . . . . . . . . . . . . . 232
18.6 Moderately Halophilic Thiodenitrifyers . . . . . . . . . . . . . . . . . . . 232
18.7 Extremely Halophilic Denitrifying SOB . . . . . . . . . . . . . . . . . . 233
18.8 Oxidation of Thiocyanate at High Salt. . . . . . . . . . . . . . . . . . . . 234
18.9 Fatty Acids in the Membrane Lipids . . . . . . . . . . . . . . . . . . . . . 234
18.10 Conclusions and Future Perspectives . . . . . . . . . . . . . . . . . . . . . 235
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 236
19 Sulfur Oxidation at Deep-Sea Hydrothermal Vents. . . . . . . . . . . . . . 238

Stefan M. Sievert, Michael Hgler, Craig D. Taylor, Carl O. Wirsen
19.1 Introduction. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 238
19.2 Types of Sulfur-Oxidizing Bacteria . . . . . . . . . . . . . . . . . . . . . . . 239
19.2.1 Symbiotic Sulfur-Oxidizing Bacteria . . . . . . . . . . . . . . . 239
19.2.2 Free-Living Sulfur-Oxidizing Bacteria. . . . . . . . . . . . . . 241
19.2.2.1 Gammaproteobacteria . . . . . . . . . . . . . . . . . . 241
19.2.2.2 Epsilonproteobacteria . . . . . . . . . . . . . . . . . . 241
19.2.2.3 Aquificaceae. . . . . . . . . . . . . . . . . . . . . . . . . . 246
19.2.2.4 Carbon Metabolism in Sulfur-Oxidizing
Bacteria . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 246
19.3 Sulfur Oxidation Pathways . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 247
19.3.1 Types of Pathways . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 247
19.3.2 Endosymbionts . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 248
19.3.3 Free-Living Sulfur-Oxidizing Bacteria. . . . . . . . . . . . . . 248
19.3.3.1 Beggiatoa . . . . . . . . . . . . . . . . . . . . . . . . . . . . 248
19.3.3.2 Thiomicrospira crunogena
and Epsilonproteobacteria. . . . . . . . . . . . . . . 249
19.3.3.3 Oxidation of H2 by Sulfur-Oxidizing
Bacteria . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 250
19.4 Snowblower Vents As Signs for Sulfide
Oxidation in the Subseafloor . . . . . . . . . . . . . . . . . . . . . . . . . . . . 251
19.4.1 The Subseafloor Biosphere. . . . . . . . . . . . . . . . . . . . . . . 251
19.4.2 Filamentous-Sulfur Formation in the Laboratory. . . . . . 251
19.4.3 Snowblowers . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 252
19.4.4 Diversity of Filamentous-Sulfur-Forming Bacteria . . . . 252
19.5 Conclusions and Outlook . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 253
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 254
Contents xxi
20 Speciation Analysis of Microbiologically Produced

Sulfur by X-ray Absorption Near Edge Structure Spectroscopy . . . 259
Alexander Prange
20.1 Introduction. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 259
20.2 XAS: X-ray Absorption Near-Edge Structure
and Extended X-ray Absorption Fine Structure . . . . . . . . . . . . . . 260
20.2.1 Experimental . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 260
20.2.2 Advantages of XANES Spectroscopy . . . . . . . . . . . . . . 264
20.2.3 Sample Preparation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 264
20.2.4 Quantitative Analysis of XANES Spectra . . . . . . . . . . . 265
20.3 Sulfur K-Edge XANES Spectroscopy and Speciation
of Microbiologically Produced Sulfur . . . . . . . . . . . . . . . . . . . . . 266
20.3.1 Speciation of Sulfur in Sulfur Globules
of Phototrophic and Chemotrophic Sulfur Bacteria . . . . 267
20.3.2 Speciation of Elemental Sulfur
Taken Up by A. vinosum. . . . . . . . . . . . . . . . . . . . . . . . . 269
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 270
21 Controls on Isotope Fractionation During Dissimilatory Sulfate

Reduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 273
Joost Hoek, Donald E. Canfield
21.1 Introduction. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 273
21.2 Sulfur Isotope Fractionation During Dissimilatory
Sulfate Reduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 275
21.2.1 Pure Cultures. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 275
21.2.2 Natural Populations. . . . . . . . . . . . . . . . . . . . . . . . . . . . . 276
21.3 Stepwise Reduction of Sulfate and Sulfur Isotope
Fractionation Models . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 277
21.4 Multiple Sulfur Isotopes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 279
21.5 Conclusions and Future Research. . . . . . . . . . . . . . . . . . . . . . . . . 282
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 283
22 Bioprocess Engineering of Sulfate Reduction

for Environmental Technology . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 285
Piet N.L. Lens, Roel J.W. Meulepas, Ricardo Sampaio,
Marcus Vallero, Giovanni Esposito
22.1 Introduction. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 285
22.2 Sulfate Reduction in Methanogenic Wastewater Treatment. . . . . 286
22.3 Sulfate-Reducing Bioreactors . . . . . . . . . . . . . . . . . . . . . . . . . . . . 288
22.3.1 High-Rate Sulfate-Reducing Bioreactors . . . . . . . . . . . . 288
22.3.1.1 Inocula . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 288
22.3.1.2 Electron Donor. . . . . . . . . . . . . . . . . . . . . . . . 289
22.3.2 Passive Sulfate-Reducing Systems . . . . . . . . . . . . . . . . . 291
xxii Contents
22.4 Sulfate Reduction for Metal Recovery/Reuse . . . . . . . . . . . . . . . 292

22.4.1 Metal Sulfide Precipitation . . . . . . . . . . . . . . . . . . . . . . . 292
22.4.2 Biogenic Sulfide for Metal Sulfide Precipitation . . . . . . 292
22.4.3 Selective Metal Precipitation . . . . . . . . . . . . . . . . . . . . . 293
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 293
23 Impact of Nitrate on the Sulfur Cycle in Oil Fields . . . . . . . . . . . . . . 296

Gerrit Voordouw
23.1 Introduction. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 296
23.2 The Oil Field Sulfur Cycle . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 297
23.3 Effect of Nitrate Injection on SRB Physiology . . . . . . . . . . . . . . 298
23.4 Mechanism of Souring Control . . . . . . . . . . . . . . . . . . . . . . . . . . 300
23.5 Prospects for Nitrate Injection . . . . . . . . . . . . . . . . . . . . . . . . . . . 301
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 301
Index . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 303
Chapter 1
Genetics and Genomics of Sulfate Respiration
in Desulfovibrio
Judy D. Wall, Adam P. Arkin, Nurgul C. Balci, Barbara Rapp-Giles
Abstract Bacteria that have evolved to use sulfate as a terminal electron acceptor
must commit to spending energy for sulfate activation before there is a return on
the investment allowing net energy gain. How sulfate is used and how electron flow
is controlled have provided challenging topics for research for many years. Having
the complete genome sequences of several of these bacteria is a monumental step
in the elucidation of these questions. This information has provided the tools for
determining the quantity of transcripts for genes under defined growth conditions,
not just the relative changes in transcripts in two growth conditions. A comparison
of the hybridization signal of messenger RNA with that of genomic DNA with oli-
gonucleotide microarrays of all open reading frames reveals the differences in
steady-state levels of transcripts for each gene. Growth of Desulfovibrio vulgaris
Hildenborough on defined medium with lactate as a carbon and reductant source and
with sulfate as the electron acceptor has been examined by this procedure for levels
of gene expression. Relative functional importance was inferred from the levels of
gene transcription, in spite of the recognized limitations of this interpretation. Not
surprisingly, genes encoding established functions for sulfate reduction were highly
expressed. However, the high molecular mass c-type cytochrome genes thought to
encode a most important transmembrane electron conduit for sulfate reduction were
expressed at quite low levels.
1.1 Introduction
Sulfate-reducing bacteria (SRB) are Gram-negative deltaproteobacteria, ubiqui-

tously present in soils, that are able to obtain energy by the dissimilatory reduction
of sulfate. These bacteria are considered anaerobes, although their genome
sequences have revealed multiple genes putatively encoding enzymes that reduce
oxygen or detoxify its products (Klenk et al. 1997; Heidelberg et al. 2004; Rabus
et al. 2004). The ever-evident and offensive end product of sulfate reduction,
sulfide, has brought much attention to the activities of the SRB in the environment.
The black precipitates in sediments and shorelines, discoloration of paper during
the milling processes (Postgate 1984) and the corrosion of ferrous metals (Hamilton
1
C. Dahl and C.G. Friedrich (eds.), Microbial Sulfur Metabolism.
Springer 2008
2 J.D. Wall et al.
2003) are a few of the less desirable effects. On the other hand, the low redox
potentials achieved by these bacteria also provide them with the capacity of reducing
a number of toxic metals, thereby changing their solubilities, and offering a potential
method for remediation of metal-contaminated environments (Lovley et al. 1991;
Gorby and Lovley 1992).
Members of the genus Desulfovibrio are perhaps the most easily and rapidly
cultured of the SRB and, therefore, have been the subject of the most intensive bio-
chemical and molecular research (Postgate 1984; Peck 1993; Voordouw 1993).
Still, there are gaps in understanding energy generation by these anaerobes. For
example, what is the role of hydrogen, formate, carbon monoxide, or ethanol during
the respiration of sulfate with lactate or pyruvate as an electron donor? The genome
sequences available for a few of the SRB are offering us the boundaries, the parts
list, for our inquiries into these questions. Of course, it does not help that all the
parts are not definitively labeled.
Hydrogen metabolism has played a prominent role in the metabolism of many
anaerobes and the SRB are no exception. Hydrogen can support sulfate respiration,
is produced during fermentative growth, and is apparently also involved in the
metabolism of a number of organic acids. A hydrogen transient is observed upon
inoculation of Desulfovibrio strains into medium containing organic acids and sul-
fate (Hatchikian et al. 1976; Tsuji and Yagi 1980). A controversial role for this
production and consumption of hydrogen was proposed by Odom and Peck (1981a)
to be an obligate chemiosmotic vectorial electron transfer for energy supplementa-
tion, called hydrogen cycling. In this model, the oxidation of organic substrates
generates protons and electrons that are substrates for cytoplasmically located
hydrogenase(s). Hydrogen produced in the cytoplasm then diffuses across the cyto-
plasmic membrane, where the periplasmic hydrogenases oxidize the hydrogen,
recapturing the electrons for transfer back to the cytoplasm for sulfate reduction
and liberating the protons to contribute to the proton motive force. Alternative
explanations for this burst of hydrogen have been offered, such as a necessary redox
adjustment of electron transport components (Tsuji and Yagi 1980) or the need for
fermentative ATP production to initiate sulfate activation by ATP sulfurylase
(Lupton et al. 1984). The complete genome sequence of several SRB allows a
closer examination of this model.
1.2 Approach
As a part of a collaborative effort to understand how the SRB reduce toxic metals
and how environmental stresses impact this ability, a number of experiments have
been undertaken to examine the changes in transcription and protein expression
during stress in Desulfovibrio vulgaris Hildenborough (Virtual Institute for
Microbial Stress and Survival 2002). To examine the differentially expressed
genes, microarray analysis of transcripts of putative open reading frames (ORFs)
have been used (Li et al. 2005). To normalize the data for comparison of expression
1 Genetics and Genomics of Sulfate Respiration in Desulfovibrio 3
levels across stresses, the transcript hybridization to microarrays for all experi-
ments has been compared with genomic DNA hybridization. As a result, a sizeable
data set has been obtained that provides the expression level of all genes in the
microarray.
Experiments for stress analyses were performed with defined medium contain-
ing sodium lactate (60 mM) as an electron donor and sodium sulfate (50 mM) as
an electron acceptor (LS4D medium; Mukhopadhyay et al. 2006). D. vulgaris
was grown from freezer stocks to an optical density at 600 nm of 0.3 (approxi-
mately 1 108 cells per milliliter) and the stress was imposed. Table 1.1 lists the
various treatments for which data were collected. Triplicates for the control and the
treated cultures were sampled at the initiation of the treatment and at specified
intervals, usually not exceeding 4 h, following the treatment. The average of the
triplicates was a single data point.
In all, 173 data points comparing transcript and genomic DNA hybridization to
ORF probes have been analyzed. From these data, the relative abundance of tran-
scripts present for a given gene in an exponentially growing culture of D. vulgaris
respiring sulfate with lactate at 30C was determined. Figure 1.1 illustrates the
distribution of log2 of the hydridization signal for transcripts divided by that of
genomic DNA for two different genes. Data points that were not significantly
above the experimental noise were not included in the average calculation. It should
be pointed out that at least 50% of the data points were from untreated control
cultures. In addition, any given treatment or stress resulted in the differential
expression of only a few hundred genes out of about 3,600 ORFs in the genome.
Thus, for regulated genes, the average expression would not expect to be biased
by the various stresses, but regulation would be evident in an increased standard
deviation of the average.
In the following discussion, expression levels of genes involved in various aspects
of metabolism are presented. However, to obtain a reference for the meaning of the
data, Table 1.2 provides expression levels for comparison genes (operon predictions
A B
Number of Observations
Number of Observations
Log2 RNA/ Genomic DNA

Log2 RNA/ Genomic DNA
Fig. 1.1 Histogram of transcripts of a hisD, DVU0796, average log2(RNA/genomic DNA) =

13.6 0.7 (173 total observations) and b sat, DVU1295, average log2(RNA/genomic
DNA) =9.2 0.8 (171 total observations). Cultures were grown as described (Table 1.1) for
RNA preparation and hybridization to microarrays (Mukhopadhyay et al. 2006)
Table 1.1 Treatments examined for transcriptional responses in Desulfovibrio vulgaris
Hildenborough
Treatment Concentration or condition Comparison culture
Cold 8C 30C
Heat 50C 37C
Oxygen 0.1% No O2
Alkaline pHa pH 10 pH 7
Acid pHb pH 5.5 pH 7
Nitritec 2.5 mM No NO2
Nitrate 105 mM No NO3
Sodiumd 250 mM No added Na+
Potassium 250 mM No added K+
Chromate 0.45 M No CrO42
Stationary phase 0.8 OD600 Mid exponential phase 0.3 OD600
a
pH was adjusted by addition of KOH.
b
pH was adjusted by addition of H2SO4.
c
Growth occurred after cells reduced the nitrite concentration below 0.5 mM.
d
Total concentration of sodium in the treatment was about 462 mM; other components
were present as sodium salts.
Table 1.2 Expression levels of D. vulgaris Hildenborough reference genes during exponential
growth phase of cells respiring sulfate with lactate as an electron donor
Operon DVU numbersa Putative gene name Average log expb SD
Ribosomal proteins DVU1302 rpsJ 10.1 0.8
DVU1303 rplC 10.8 0.8
DVU1304 rplD 10.4 0.8
DVU1305 rplW 11.5 0.8
DVU1306 rplB 11.1 0.8
DVU1307 rpsS 11.0 0.8
DVU1308 rplV 11.2 0.7
Tryptophan biosynthesis DVU0465 trpE 14.6 0.9
DVU0466 trpG 15.0 0.9
DVU0467 trpD 14.2 0.9
DVU0468 trpC 14.4 1.0
DVU0469 trpF-1 14.3 1.1
DVU0470 trpB-2 13.7 1.1
DVU0471 trpA 13.7 0.9
High molecular mass DVU0529 rrf2 15.1 1.4
cytochrome c DVU0530 rrf1 15.0 1.4
DVU0531 hmcF 15.6 1.5
DVU0532 hmcE 15.2 1.4
DVU0533 hmcD 15.6 1.4
DVU0534 hmcC 14.7 1.2
DVU0535 hmcB 14.9 1.5c
DVU0536 hmcA 15.1 1.8
SD standard deviation.
a
DVU numbers from TIGR annotation (Heidelberg et al. 2004).
b
Average log exp is the average log2 of the RNA to genomic DNA signal from whole genome
transcript microarrays from cultures treated as in Table 1.1. In calculating average expressions,
fewer than 10% of the 173 data points available for each gene were eliminated because of poor
signal-to-noise ratio unless otherwise indicated.
c
For DVU0535, 27 data points were below the cutoff criterion.
were as described by Price et al. 2005). Ribosomal protein genes are expected to be
rather highly expressed during exponential growth and the log2 of the ratio of the
messenger RNA to DNA (a large negative number because of the greater quantity
of DNA used for hybridization) was in the range of 10.9. That for the tryptophan
operon was 14.3, an operon that must function in medium lacking tryptophan, yet,
because large quantities of this amino acid are not needed, would not be expected
to be highly expressed. Finally the operon for the high molecular mass cytochrome
c was expressed at a still lower level, 15.2. The latter was an unexpectedly low
value for transcription of this operon thought to encode an important conduit for
electrons for sulfate reduction.
1.3 Sulfate Metabolism
The enzymology of sulfate reduction by Desulfovibrio strains is rather mature (Peck

and LeGall 1982; Peck 1993) and enzymes involved were reported to be constitutively
present in sulfate-respiring cells (Odom and Peck 1981b). Four cytoplasmic enzymes
are sufficient for conversion of sulfate to sulfide in an eight electron reduction pathway.
Annotated in the D. vulgaris genome, ATP sulfurylase (DVU1295, sat) activates the
sulfate-generating adenosine 5-phosphosulfate (APS) in preparation for the first two-
electron reduction. Inorganic pyrophosphate is released which is cleaved by an inor-
ganic pyrophosphatase (DVU1636, ppaC) to pull the reaction. APS is then reduced
by APS reductase, a two-subunit enzyme (DVU0846/0847, apsBA). Sulfite is the
reduced product that becomes the substrate for the six-electron reduction by sulfite
reductase, also known as desulfoviridin in the Desulfovibrio strains (DVU04020404,
dsrABD; DVU2776, dsrC). In vitro, this enzyme is capable of producing sulfide as
the final end product of sulfate reduction.
Table 1.3 shows the remarkably high level of expression of the genes for sulfate
respiration that exceeds that of ribosomal protein genes. Of the candidate ORFs for
Table 1.3 Expression levels of putative genes coding for enzymes of sulfate reduction in
D. vulgaris Hildenborough
Putative gene Average log exp
Protein function DVU number name SD
Sulfate adenylyltransferase DVU1295 sat 9.2 0.8
Adenosine 5-phosphosulfate DVU0846 apsB 8.8 0.7
reductase DVU0847 apsA 8.9 0.8
Sulfite reductase DVU0402 dsrA 9.3 0.9
DVU0403 dvsB 9.5 0.7
DVU0404 dsrD 8.8 1.0
DVU2776 dsrC 10.3 1.1a
Inorganic pyrophosphatase DVU1636 ppaC 11.1 1.0
a
Of the 173 data points available, 50 were below the cutoff for the signal-to-noise ratio.
6 J.D. Wall et al.
putative sulfate permeases (DVU0053/0279/0746/0747/1999) none was found to be

expressed at these high levels nor did one have higher expression than others. Thus,
the data did not point to a gene responsible for transport of this substrate, a function
that remains to be specifically identified.
The role of other possible intermediates during sulfite reduction such as thiosul-
fate or trithionate has received significant attention through the years (Drake and
Akagi 1978; Peck and LeGall 1982; Postgate 1984). Recent deletion analysis in D.
gigas proposes these polysulfides as intermediates in sulfite reduction (Broco et al.
2005). Although genes predicted to encode thiosulfate reductases have been anno-
tated in the genomes available, enzymes designed to handle other polysulfides have
not become evident.
1.4 Lactate Oxidation
Electron donors preferred by Desulfovibrio tend to be strain-specific, although

most incomplete oxidizers grow readily with organic acids such as lactate and
pyruvate and most grow with hydrogen or formate. The enzymology of the oxida-
tion of these substrates is much less well deciphered than that for sulfate reduction.
Multiple genes have been annotated as lactate permeases and dehydrogenases in the
Desulfovibrio genomes. Genes annotated as glycolate dehydrogenase may actually
function as lactate dehydrogenases since the structures of substrates and products
are quite similar (see EC 1.99.14 in BRENDA 1987). However, it is clear from
biochemical experiments that the primary lactate dehydrogenase is a membrane-
bound flavoprotein that is extremely unstable in air (Hansen 1994).
Multiple annotations for genes encoding pyruvate ferredoxin oxidoreductase
and for formate acetyltransferase also occur in the genome databases. However,
conserved genes have been annotated for phosphate acetyltransferase and acetyl
kinase. Those two genes appear in a region of the genome encoding two conserved
operons that together possibly contain information for complete metabolism of lac-
tate, i.e., in D. vulgaris a putative lactate permease (DVU3026), two genes anno-
tated as glycolate oxidase and an FeS-cluster-binding protein for electron handling
(DVU3027/3028) that could be a lactate dehydrogenase, pyruvate ferredoxin oxi-
doreductase (DVU3025), the phosphate acetyltransferase (DVU3029) and the
acetyl kinase (DVU3030). The proximity of genes for the potentially complete
metabolism of lactate suggests common regulation.
Table 1.4 shows the expression levels for the genes in these two operons of
D. vulgaris growing with lactate as electron and carbon sources. The transcripts for
these genes are almost as abundant as those for ribosomal proteins. Transcripts for
other annotated lactate dehydrogenases or pyruvate ferredoxin oxidoreductases are
detected at lower levels (data not shown). The interpretation of this differential
expression awaits mutagenesis studies, although it is tempting to speculate that the
region encoding an apparently full complement of enzymes for lactate oxidation to
acetate might be essential for lactate growth of the cells.
Table 1.4 Expression levels of the gene region for lactate oxidation to acetate in
DVU number Putative gene name Average log exp SD
DVU3025 por 10.8 0.9
DVU3026 lldP 12.9 1.0
DVU3027 glcD 11.7 1.0
DVU3028 glpC 12.2 1.0
DVU3029 pta 11.8 1.0
DVU3030 ackA 11.9 1.0
DVU3031 COG-Pta 11.8 0.9
DVU3032 NA 12.1 1.0
DVU3033 NA 12.2 1.0
Arrows indicate operon arrangement and transcription direction of genes.
NA not annotated with gene name.
1.5 Hydrogenases
For the hydrogen cycle to function, it is necessary to have hydrogenases located

on either side of the cytoplasmic membrane. Early biochemical analyses were
ambiguous in establishing the exact position of the known hydrogenases of various
Desulfovibrio strains because of the difficulty in achieving clean cell fractionations
(Odom and Peck 1981b). However, with the sequence of the encoding genes
available, it became clear that the biochemically identified Fe-only, NiFe and
NiFeSe hydrogenases of D. vulgaris were periplasmically located (Peck 1993;
Voordouw 2000). Only when the genome sequence of D. vulgaris was completed
were candidates for cytoplasmically located hydrogenases convincingly revealed.
The operon for carbon monoxide dehydrogenase (CODH) biosynthesis and that for
an Ech-like hydrogenase, a membrane-bound NiFe hydrogenase that has similarity
to NADHquinone oxidoreductases (complex I), were annotated. Table 1.5 shows
the expression levels of these two operons. Clearly the CODH operon (DVU2286
2293) is more highly expressed under the culturing conditions chosen than the cata-
lytic subunit of CODH (DVU2098/2099) or the Ech operon (DVU04290434).
Curiously these operons are not conserved in the genomes of the other SRB
sequenced to date. Thus, the critical nature of the cytoplasmic hydrogenases in the
pathways of sulfate respiration have yet to be established.
Interestingly the genome sequences have also shown that the periplasmic hydroge-
nases are more redundant than previously thought. At least four isozymes have been
identified by sequence analysis, adding one to the number biochemically determined.
All appeared to be expressed at reasonable levels when D. vulgaris was cultured on
lactate plus sulfate: Fe-only hydrogenase (DVU1769/1770) at log2 of 13.8 1.4; NiFe
isozyme-1 (DVU1921/1922), 13.1 1.7; NiFe isozyme-2 (DVU2525/2526), 14.4
1.0; and NiFeSe hydrogenase (DVU1917/1918), 11.5 1.5. The observation that
8 J.D. Wall et al.
Table 1.5 Expression levels of putative cytoplasmic hydrogenase complexes in

CODH hydrogenase operon
DVU2286 cooM 12.4 0.7
DVU2287 cooK 12.1 0.5
DVU2288 cooL 12.0 0.5
DVU2289 cooX 11.9 0.5
DVU2290 cooU 11.6 0.7
DVU2291 cooH 11.8 0.6
DVU2292 hypA 10.8 0.6
DVU2293 cooF 11.0 0.8
DVU2098 cooS 13.1 1.2
DVU2099 cooC-2 15.2 0.8a
Ech hydrogenase operon
DVU0429 echF 15.0 1.1
DVU0430 echE ND
DVU0431 echD 14.8 0.9
DVU0432 echC 14.5 0.8
DVU0433 echB 15.0 1.0
DVU0434 echA 14.0 0.8
ND insufficient useable data for calculation.
a
Fifty-four of 173 data points were not significant.
transcripts for the NiFeSe enzyme were most abundant was unexpected since the
biochemical and mutational data indicated that the Fe-only hydrogenase accounts for
most of the hydrogenase activity of the periplasm (Pohorelic et al. 2002).
1.6 Transmembrane Electron-Conducting Complexes
To complete the circuit of electrons in the hydrogen cycling model, there must be
mechanisms that conduct the electrons across the cytoplasmic membrane from the
periplasmically oxidized hydrogen to sulfate. In addition, growth on hydrogen or
formate would also require a transmembrane conduit for the electrons generated
from the periplasmic oxidation of these substrates. The high molecular mass cyto-
chrome c complex (Table 1.2) that has a hexadecaheme cytochrome facing the
periplasm was the first such transmembrane complex (TMC) described (Rossi et al.
1993). It was proposed to function for electron transfer from periplasmically
oxidized hydrogen, regardless of the origin of the hydrogen, to the cytoplasm for
sulfate reduction (Voordouw 2000). Deletions of this operon, however, demonstrate
that this TMC is not essential for this activity (Dolla et al. 2000) and the expression
levels of the genes were unexpectedly low (Table 1.2).
Much progress has been achieved in the molecular analysis of additional TMCs
of the SRB and sulfur-oxidizing bacteria (Matias et al. 2005; Pereira et al. 2006;
Table 1.6). A three-subunit conserved complex, Qmo (DVU08480850), is encoded
promoter distal in the same operon as the genes for APS reductase. The location
of these genes suggests a possible role in providing electrons for APS reduction.
A six-gene operon (DVU12861291) coding for another apparent TMC suggested
to provide electrons to DsrAB, the bisulfite reductase, has also been identified. This
operon includes a type II tetraheme cytochrome c3, DsrJ. Recently, the isolation and
characterization of the Tmc complex (DVU02630266) was reported. Although a
role in electron transfer from periplasmic oxidations or from reduced menaqui-
nones to sulfate would seem likely for this complex, no experimental evidence
supported this possibility (Pereira et al. 2006). An additional complex with compo-
nents sharing sequence similarity with heterodisulfide reductase (DVU23992405)
and another with similarity to sodium-translocating NADH:quinone oxidore-
ductase complex (DVU27912798, rnf) have been annotated but do not yet have
functions assigned.
Table 1.6 shows that each of these putative TMCs appears to be expressed at
levels similar to those of ribosomal protein genes. The Rnf complex genes were
somewhat less abundantly transcribed but were still more abundant than Hmc
genes. The need for different conduits to supply electrons to APS reductase and
to bisulfite reductase was proposed many years ago (Peck 1993). Thus, two
conduits would be predicted. Why the multiplicity of TMCs, all of which
appear to be synthesized? The information provided by genome sequences has
served to emphasize that our models for energy generation in the SRB are still
inadequate.
1.7 Conclusions
For the first time, we now have a glimpse at the relative transcription of all putative
ORFs in the genome of a sulfate-reducing bacterium, D. vulgaris Hildenborough.
Certainly this information is limited by differential stability of transcripts, stability
of the proteins encoded, and enzyme activity regulation. However, a few surprises
have been observed.
As predicted, genes for the sulfate-reducing enzymes were highly expressed.
The apparent redundancy for other steps in metabolism of substrates, i.e., peri-
plasmic hydrogenases and TMCs, was affirmed by robust expression of the genes
for the redundant systems. In contrast, cytoplasmic hydrogenases (CODH com-
plex and Ech) were eightfold different in expression levels. The much greater
transcription levels of the CODH complex might suggest a more important
role under these experimental conditions. Finally, the low level of expression of
the Hmc complex might suggest that it may be supplementary to other TMCs or
that it may be more important during growth on other substrates or in other cellu-
lar growth phases.
10 J.D. Wall et al.
Table 1.6 Expression levels of genes coding for putative transmembrane complexes in
Type II c3 transmembrane complex
DVU0258 COG-BaeS 14.1 1.0
DVU0259 divK 9.7 1.0
DVU0260 mtrA 11.2 1.0
DVU0261 COG-UspA 11.6 1.0
DVU0262 NA 12.2 0.9
DVU0263 tmcA 12.0 0.9
DVU0264 tmcB 11.4 1.0
DVU0265 tmcC 12.2 1.0
DVU0266 tmcD 11.8 1.0
Qmo transmembrane complex
DVU0848 qmoA 11.0 0.7
DVU0849 qmoB 12.0 0.7
DVU0850 qmoC 12.8 0.7
DVU0851 NA 12.1 0.7
Dsr transmembrane complex
DVU1286 dsrP 11.6 0.9
DVU1287 dsrO 12.2 0.9
DVU1288 dsrJ 12.2 0.9
DVU1289 dsrK 12.6 0.9
DVU1290 dsrM 12.3 0.9
DVU1291 NA 13.2 0.7
Heterodisulfide reductase
DVU2399 NA 11.8 1.0
DVU2400 NA 11.5 1.0
DVU2401 NA 12.1 0.7
DVU2402 hdrA 12.0 0.9
DVU2403 hdrB 11.5 0.9
DVU2404 hdrC 11.2 1.0
DVU2405 eutG 9.1 1.4
Rnf transmembrane complex
DVU2791 dhcA 12.6 1.0
DVU2792 rnfC 13.4 1.0
DVU2793 rnfD 14.1 1.0
DVU2794 rnfG 13.7 1.1
DVU2795 rnfE 14.7 0.9
DVU2796 rnfA 14.4 1.0
DVU2797 rnfB 13.6 0.9
DVU2798 apbE 13.3 1.0
NA not annotated with gene name.
Acknowledgements. This work was part of the Virtual Institute for Microbial Stress and
Survival (http://vimss.lbl.gov) supported by the US Department of Energy, Office of Science,
Office of Biological and Environmental Research, Genomics Program:GTL through contract
DE-AC02-05CH11231 between Lawrence Berkeley National Laboratory and the US Department
of Energy; the DOE Energy Biosciences Program, Office of Basic Energy Sciences, grant
number DE-FG02-87ER13713, and a University of Missouri Life Sciences Post Doctoral
Fellowship to N.C.B.
References
BRENDA (1987) Cologne University Bioinformatics Center. http://www.brenda.uni-koeln.de/

index.php4
Broco M, Rousset M, Oliveira S, Rodrigues-Pousada C (2005) Deletion of flavoredoxin gene
in Desulfovibrio gigas reveals its participation in thiosulfate reduction. FEBS Lett
579:48034807
Dolla A, Pohorelic BK, Voordouw JK, Voordouw G (2000) Deletion of the hmc operon of
Desulfovibrio vulgaris subsp. vulgaris Hildenborough hampers hydrogen metabolism and
low-redox-potential niche establishment. Arch Microbiol 174:143151
Drake HL, Akagi JM (1978) The dissimilatory reduction of bisulfite by Desulfovibrio vulgaris.
J Bacteriol 136:916923
Gorby YA, Lovley DR (1992) Enzymatic uranium precipitation. Environ Sci Technol
26:205207
Hamilton WA (2003) Microbially influenced corrosion as a model system for the study of metal
microbe interactions: a unifying electron transfer hypothesis. Biofouling 19:6576
Hansen TA (1994) NAD(P)-independent lactate dehydrogenase from sulfate-reducing prokaryotes.
Methods Enzymol 243:2123
Hatchikian EC, Chaigneau M, Le Gall J (1976) Analysis of gas production by growing cultures
of three species of sulfate reducing bacteria. In: Schlegel HG, Gottschalk G, Pfennig N (eds)
Microbial production and utilization of gases. Goltze, Gttingen, pp 109118
Heidelberg JF, Seshadri R, Haveman SA, Hemme CL, Paulsen IT, Kolonay JF, Eisen JA, Ward N,
Methe B, Brinkac LM, Daugherty SC, Deboy RT, Dodson RJ, Durkin AS, Madupu R, Nelson
WC, Sullivan SA, Fouts D, Haft DH, Selengut J, Peterson JD, Davidsen TM, Zafar N, Zhou L,
Radune D, Dimitrov G, Hance M, Tran K, Khouri H, Gill J, Utterback TR, Feldblyum TV,
Wall JD, Voordouw G, Fraser CM (2004) The genomic sequence of the anaerobic, sulfate-
reducing bacterium Desulfovibrio vulgaris Hildenborough. Nat Biotechnol 22:554559
Klenk HP, Clayton RA, Tomb JF, White O, Nelson KE, Ketchum KA, Dodson RJ, Gwinn M, Hickey
EK, Peterson JD, Richardson DL, Kerlavage AR, Graham DE, Kyrpides NC, Fleischmann RD,
Quackenbush J, Lee NH, Sutton GG, Gill S, Kirkness EF, Dougherty BA, McKenney K, Adams
MD, Loftus B, Peterson S, Reich CI, McNeil LK, Badger JH, Glodek A, Zhou L, Overbeek R,
Gocayne JD, Weidman JF, McDonald L, Utterback T, Cotton MD, Spriggs T, Artiach P, Kaine
BP, Sykes SM, Sadow PW, DAndrea KP, Bowman C, Fujii C, Garland SA, Mason TM, Olsen
GJ, Fraser CM, Smith HO, Woese CR, Venter JC (1997) The complete genome sequence of the
hyperthermophilic, sulphate-reducing archaeon Archaeoglobus fulgidus. Nature 390:364370
Li X, Zhili H, Zhou J (2005) Selection of optimal oligonucleotide probes for microarrays using
multiple criteria, global alignment and parameter estimation. Nucleic Acids Res 33:61146123
Lovley DR, Phillips EJP, Gorby YA, Landa E (1991) Microbial reduction of uranium. Nature
350:413416
Lupton FS, Conrad R, Zeikus JG (1984) Physiological function of hydrogen metabolism during
growth of sulfidogenic bacteria on organic substrates. J Bacteriol 159:843849
Matias PM, Pereira IAC, Soares CM, Carrondo MA (2005) Sulphate respiration from hydrogen
in Desulfovibrio bacteria: a structural biology overview. Prog Biophys Mol Biol 89:292329
12 J.D. Wall et al.
Mukhopadhyay A, He Z, Alm EJ, Arkin AP, Baidoo E, Borglin SC, Chen W, Hazen TC, He Q,
Holman H-Y, Huang K, Huang R, Joyner DC, Katz N, Keller M, Oeller P, Redding A, Sun J,
Wall J, Wei J, Yang Z, Yen H-C, Zhou J, Keasling JD (2006) Salt stress in Desulfovibrio
vulgaris Hildenborough: an integrated genomics approach. J Bacteriol 188:40684078
Odom JM, Peck HD Jr (1981a) Hydrogen cycling as a general mechanism for energy coupling in
the sulfate reducing bacteria, Desulfovibrio sp. FEMS Microbiol Lett 12:4750
Odom JM, Peck HD Jr (1981b) Localization of dehydrogenases, reductases, and electron transfer
components in the sulfate-reducing bacterium Desulfovibrio gigas. J Bacteriol 147:161169
Peck Jr HD (1993) Bioenergetic strategies of the sulfate-reducing bacteria. In: Odom JM, Singleton
R Jr (eds) The sulfate-reducing bacteria: contemporary perspectives. Springer, New York,
pp 4176
Peck HD Jr, LeGall J (1982) Biochemistry of dissimilatory sulphate reduction. Philos Trans R Soc
Lond Ser B 298:443466
Pereira PM, Teixeira M, Xavier AV, Louro RO, Pereira AC (2006) The TMC complex from
Desulfovibrio vulgaris Hildenborough is involved in transmembrane electron transfer from
periplasmic hydrogen oxidation. Biochemistry 45:1035910367
Pohorelic BK, Voordouw JK, Lojou E, Dolla A, Harder J, Voordouw G (2002) Effects of deletion
of genes encoding Fe-only hydrogenase of Desulfovibrio vulgaris Hildenborough on hydrogen
and lactate metabolism. J Bacteriol 184:679686
Postgate JR (1984) The sulphate reducing bacteria, 2nd edn. Cambridge University Press, Cambridge
Price MN, Huang KH, Alm EJ, Arkin AP (2005) A novel method for accurate operon predictions
in all sequenced prokaryotes. Nucleic Acids Res 33:880892
Rabus R, Ruepp A, Frickey T, Rattei T, Fartmann B, Stark M, Bauer M, Zibat A, Lombardot T,
Becker I, Amann J, Gellner K, Teeling H, Leuschner WD, Glockner FO, Lupas AN, Amann R,
Klenk HP (2004) The genome Desulfotalea psychrophila, a sulfate-reducing bacterium from
permanently cold artic sediments. Environ Microbiol 6:887902
Rossi M, Pollock WBR, Reij MW, Keon RG, Fu R, Voordouw G (1993) The hmc operon of
Desulfovibrio vulgaris subsp. vulgaris Hildenborough encodes a potential transmembrane
redox protein complex. J Bacteriol 175:46994711
Tsuji K, Yagi T (1980) Significance of hydrogen burst from growing cultures of Desulfovibrio
vulgaris Miyazaki and the role of hydrogenase and cytochrome c3 in energy production system.
Arch Microbiol 125:3542
Virtual Institute for Microbial Stress and Survival (2002) Publications. http://vimss.lbl.gov/findings/
publications.php
Voordouw G (1993) Molecular biology of the sulfate-reducing bacteria. In: Odom JM, Singleton R
Jr (eds) The sulfate-reducing bacteria: contemporary perspectives. Springer, New York,
pp 88130
Voordouw G (2000) A universal system for the transport of redox proteins: early roots and latest
developments. Biophys Chem 86:13140
Chapter 2
Living on Sulfate: Three-Dimensional
Structure and Spectroscopy of Adenosine
5-Phosphosulfate Reductase and
Dissimilatory Sulfite Reductase
Gnter Fritz, Alexander Schiffer, Anke Behrens, Thomas Bchert,

Ulrich Ermler, Peter M.H. Kroneck
Abstract The reduction of sulfate to sulfide and the reverse reaction are widespread
biological processes. Hereby, microorganisms play a central role. Plants also
reduce sulfate for the purpose of biosynthesis, and both plants and animals convert
reduced sulfur compounds to sulfate. Sulfate respiration is used for energy conser-
vation by strictly anaerobic bacteria and archaea. The redox equivalents generated
by the oxidation of organic compounds are transferred to sulfate as the terminal
electron acceptor. There are three key enzymes localized in the cytoplasm or at the
cytoplasmic aspect of the inner membrane: ATP sulfurylase (ATPS), adenosine
5-phosphosulfate reductase (APSR), and dissimilatory sulfite reductase (SIR).
Sulfate (S6+) cannot be directly reduced by dihydrogen or organic acids, it has to be
activated to adenosine 5-phosphosulfate (APS) catalyzed by ATPS. The enzyme
APSR (cofactors flavin adenine dinucleotide, [4Fe4S]) catalyzes the conversion of
APS to sulfite (S4+) and AMP, followed by the complex multicomponent enzyme
SIR (cofactors siroheme, [4Fe4S]) which catalyzes the reduction of sulfite (S4+) to
sulfide (S2). In this contribution we present the three-dimensional structures of
APSR from Archaeoglobus fulgidus and of catalytically relevant reaction interme-
diates. In addition, we discuss spectroscopic and structural data of SIR purified
from this organism.
2.1 Introduction
The biogeochemical cycles of the basic elements of life, such as nitrogen, oxygen,
and sulfur, have attracted the interest of many researchers over the past few
decades. Of similar importance, the biochemistry of the transition metals has been
extensively studied because of their functions as cofactors, or as part of cofactors
in enzymes, and as structural elements in proteins. Many processes strictly depend
on transition metal ions and their ability to catalyze multielectron redox and hydro-
lytic transformations (Kroneck 2005). Sulfur can exist in the biosphere in several
oxidation states, such as S6+ in sulfate, S4+ in sulfite, S0 in elemental sulfur, or
13
Springer 2008
14 G. Fritz et al.
S2 in hydrogen sulfide (Amend and Shock 2001). Interconversions of these species

constitute their biogeochemical cycles which are sustained by complex biological
processes, with bacteria playing a prominent role. Numerous studies suggest that
the ability to reduce sulfate was developed early during prokaryotic evolution. As
life may have originated in hot environments (Achenbach-Richter et al. 1987), the
occurrences of sulfate-reducing prokaryotes among hyperthermophilic archaea and
deep-branching thermophilic bacteria indicate an early origin of this process (Dahl
and Trper 2001). Isotopic data suggest that dissimilatory sulfate reduction began
2.8 billion to 3.1 billion years ago (Schidlowski 1983) but acquired global signifi-
cance only after sulfate concentrations had considerably increased in the Precambrian
oceans approximately 2.35 billion years ago (Cameron 1982).
Dissimilatory sulfate reduction operates under strictly anaerobic conditions and
represents an important element within the biogeochemical sulfur cycle (Peck 1959;
LeGall and Fauque 1988; Hansen 1994). The sulfur-oxidizing pathway proceeds in
the opposite direction, starting from sulfide, elemental sulfur, or thiosulfate. Sulfate-
reducing bacteria, such as Desulfovibrio sp., use sulfate as the terminal electron
acceptor, and hydrogen sulfide is formed as a final product. They derive energy
from the dissimilatory reduction of sulfate by dihydrogen or organic substrates. The
electron transport chain catalyzing this reaction involves periplasmic hydrogenases,
several multi-heme cytochromes, such as cytochrome c3 and nonaheme c, and other
both membrane-bound and cytoplasmic redox enzymes (Matias et al. 2005;
Pires et al. 2006).
In this chapter we will focus on two enzymes involved in sulfate respiration:
(1) adenosine 5-phosphosulfate reductase (APSR), which converts adenosine
5-phosphosulfate (APS) to sulfite and AMP, and (2) sulfite reductase (SIR),
which reduces sulfite to hydrogen sulfide. These multimetal enzymes have been
purified to homogeneity under the exclusion of dioxygen and characterized bio-
chemically and spectroscopically. The structural information on both APSR and
SIR from the thermophilic archeon Archaeoglobus fulgidus (Stetter et al. 1987)
was obtained by X-ray crystallography and electron paramagnetic resonance
(EPR) spectroscopy. Furthermore, catalytically competent intermediates could be
trapped and characterized.
2.2 Adenosine 5-Phosphosulfate Reductase
Sulfate cannot oxidize H2 or organic acids in view of its negative redox potential of
516 mV; thus, it has to be activated to APS at expense of ATP (via ATP sulfury-
lase), which shifts the standard redox potential (APS/AMP+HSO3) to 60 mV
(Thauer et al. 1977). Note that the formation of APS is endergonic and probably
driven by the subsequent cleavage of pyrophosphate. All dissimilatory APSRs
isolated so far contain flavin adenine dinucleotide (FAD) and FeS clusters; they
catalyze the two-electron reduction of APS to sulfite (Lampreia et al. 1994):
2 Three-Dimensional Structure and Spectroscopy of Adenosine 15
APS + 2e AMP + HSO3 ( )

E 0 APS AMP + HSO3 = 60 mV
Hydrolytic cleavage of the SOP moiety in APS yields approximately 80 kJ mol1

(Lipmann 1958), which is among the highest values reported so far for an XOP
bond in a biological molecule. This energy is utilized by APSR in the reductive
transformation of APS to sulfite and AMP.
2.2.1 Molecular Properties of APSR
The molecular parameters of APSR from sulfate-reducing bacteria have been a

matter of debate. Originally, an 2-subunit composition with one FAD and two
[4Fe4S] prosthetic groups ( approximately 70 kDa, approximately 20 kDa) was
proposed (Lampreia et al. 1994), or an 22 complex with one single ironsulfur
cluster/ heteromer (Verhagen et al. 1994). This FeS cluster was thought to con-
sist of more than four iron atoms. Analysis of the genes encoding the -subunit and
the -subunit of APSR from the sulfate-reducing hyperthermophilic A. fulgidus
revealed a putative FAD-binding domain on the -subunit. On the -subunit, the
arrangement of seven cysteine residues suggested the presence of a ([3Fe4S][4Fe4S])
cluster similar to 7Fe ferredoxins (Speich et al. 1994). Recently, a purification pro-
cedure was developed which led to active samples of APSR from four different
microorganisms, Desulfovibrio desulfuricans, D. vulgaris, Thiobacillus denitrifi-
cans, and A. fulgidus. Their UVvis spectra were practically identical, and the ana-
lytical data indicated the presence of one FAD and two [4Fe4S] centers per
heterodimer as depicted in a structural model (Fritz et al. 2000). Purification under
the strict exclusion of dioxygen led to a 10% increase in activity of APSR, and the
absence of dioxygen was crucial for the crystallization of the enzyme. APSR purified
under an inert atmosphere (N2/H2 95%/5%) did not show any EPR signal at g=2.01
characteristic for a [3Fe4S] center. In these preparations the enzyme was in a
partially reduced state according to UVvis and EPR spectra (Bchert et al. 1999).
Dithionite reduction of APSR resulted in a rhombic EPR spectrum with gz, gy and
gz at 2.08, 1.94 and 1.90. One single [4Fe4S] cluster (center I) became reduced
(0.91.0 spin per mole); there were no EPR signals originating from the second FeS
center. The reduction potentials of center I in APSR of D. desulfuricans and of A.
fulgidus were determined by an EPR-monitored titration to be 59 12 and 57
5 mV (pH 7.0), which are close to literature values (Lampreia et al. 1994; Verhagen
et al. 1994). Note that these values are rather high for [4Fe4S] clusters, which usu-
ally have values ranging between 200 and 500 mV (Beinert et al. 1997). Such low
potentials were found for the potentials of FeS center II of APSR from D. desulfuri-
cans and A. fulgidus, 540 15 and 520 10 mV, respectively.
The lineshape and the parameters of the EPR spectra of FeS center I in all four
APS reductases were independent of the ionic strength of the buffer, indicating that
FeS center I is shielded from the solvent. At pH 8, reduction of APSRs with
16 G. Fritz et al.
dithionite gave EPR signals characteristic for magnetically interacting [4Fe4S]

centers, including the so-called g4 signal due to dipolar interaction of two S=1/2
systems (Fritz et al. 2002a). An increase in ionic strength clearly influenced the
lineshape of these signals. Since the ionic strength did not influence the EPR spectra
of FeS center I, it was concluded that FeS center II must be affected because of
partial exposure to the solvent. Consequentially, FeS center I was assigned to the
[4Fe4S] cluster in close proximity to FAD (Sect. 2.2.2). Upon addition of both
sulfite and AMP to APSR, EPR signals from FeS center I and a distinct flavin radi-
cal were detected in agreement with the absorption maximum observed at 380 nm
which is typical for the flavosemiquinone radical anion.
The assignment of the high-potential FeS center I to the buried [4Fe4S] cluster
and the low-potential center II to the surface-exposed [4Fe4S] received strong
support from features in the crystal structure of APSR from A. fulgidus (Fritz et al.
2002b). A different number of backbone amides interacted with the sulfur atoms of
the two [4Fe4S] sites. Local dipoles formed by the amide groups stabilized the
additional negative charge upon reduction of FeS center I; thus, the reduction
potential was shifted to a more positive value. Fewer interactions are observed for
FeS cluster II at the surface, leading to a lower reduction potential.
Upon addition of sulfite to APSR of all four organisms, a slight increase around
320 nm was observed indicative for the formation of a sulfite adduct in the N5 posi-
tion of the isoalloxazine ring of the FAD moiety.
Addition of AMP to the APS reductasesulfite complex of the sulfate-reducing
organisms in the absence of dioxygen led to the disappearance of the band at
320 nm and caused a slight increase around 380 nm as well as a decrease around
448 nm. The data are consistent with the decay of the flavin N5-sulfite adduct and
reduction of FAD and reduction of one [4Fe4S] cluster; part of the flavin was only
one-electron-reduced as indicated by the formation of a stable anionic flavin radical
according to EPR spectroscopy. Only the enzyme purified from T. denitrificans did
not form a flavin radical under these conditions. After addition of sulfite and AMP
to APS reductase under exclusion of dioxygen, the FAD and one ironsulfur cluster
became reduced immediately.
The physiological electron donor/acceptor of APS reductase in sulfate reduction
and sulfur oxidation still remains unknown. As electrons generated in the periplasm
must be transferred to APS reductase in the cytoplasm, there must exist an electron
donor shuttling electrons from a transmembrane electron translocating complex to
APS reductase (see Chap. 3 by Pereira). Preliminary experiments with crude
extracts from D. desulfuricans indicated that a low molecular mass component (less
than 1 kDa), perhaps a thiol, might be involved. Thus, the reactivity of APSR with
different thiols was investigated, including glutathione, l-cysteine, 2-mercaptoethanol,
dithiothreitol, 2-mercaptoethylamine and coenzyme M. Among these thiols, dithio-
threitol worked best. 2-Mercaptoethylamine and coenzyme M were much less
active; glutathione, l-cysteine and 2-mercaptoethanol remained without effect
under our experimental conditions. With 10 mM dithiothreitol, the degree of reduc-
tion of APSR from D. desulfuricans was comparable to that achieved with sulfite
and AMP. However, the specific activity of APSR, with dithiothreitol as the
electron donor and APS as the electron acceptor, was much lower compared
with the standard activity assay, with reduced methylviologen and APS (Bchert
et al. 1999).
2.2.2 Three-Dimensional Structure of APSR
Recently, the X-ray structures of APSR from A. fulgidus in several enzymatic states
have been solved, including the structure of the FADsulfite adduct which allowed
a structure-based reaction mechanism to be developed (Fritz et al. 2002b; Schiffer
et al. 2006). The enzyme consists of the -subunit (75 kDa) and the -subunit
(20 kDa) arranged as an 22 heterotetramer (Fig. 2.1). The -subunit harbors the
FAD prosthetic group and can be divided into three domains. This architecture
classifies APSR as member of the fumarate reductase family (Lancaster 2003). The
-subunit consists of a bacterial ferredoxin-type segment with two [4Fe4S] clus-
ters, a three-stranded antiparallel -sheet and a tail with a length of 50 (Fig. 2.2).
The global part of the -subunit is embedded into a broad cleft of the -subunit,
while its long tail wraps around the -subunit (Fig. 2.1).
The reaction of APSR consists of an electron transfer step and the reductive
cleavage of the ester moiety of APS. Two electrons are transferred to the [4Fe4S]
centers and from there via the conserved Trp B48 to the isoalloxazine ring via its
si side; the hydrolytic reaction occurs at the re side of FAD within a 17--long chan-
nel (Fritz et al. 2002b; Schiffer et al. 2006; Fig. 2.1).
2.2.3 Reaction Mechanism of APSR
The reaction catalyzed by APSR comprises a nucleophilic attack of the N5

atom of reduced FAD on the sulfur of APS presumably via an FADAPS inter-
mediate (Massey et al. 1969). This intermediate decays to AMP and the FAD
sulfite adduct, which is subsequently cleaved and sulfite is finally liberated
(Figs. 2.3, 2.4). This mode of action was originally postulated by Michaels et al.
(1970) and has now been experimentally corroborated on the basis of structures
of APSR in catalytically relevant states (Fig. 2.4). In the oxidized state (FADox)
the isoalloxazine moiety of the FAD cofactor exhibits a similarly bent confor-
mation as observed in the structure of the reduced enzyme. In the APS-bound
state (FADoxAPS) the substrate APS is embedded into a 17--long substrate
channel in such a way that the isoalloxazine ring is pushed towards the channel
bottom, thereby producing a compressed enzymesubstrate complex. A clamp
formed by residues Arg A317 and Leu A278 to fix the adenine ring, as well as
the curved APS conformation appear to be key factors to hold APS in a strained
conformation. This energy-rich state becomes relaxed during the attack of APS
on the reduced FAD. A relaxed FADsulfite adduct is observed in the structure
18 G. Fritz et al.
Fig. 2.1 Left: The adenosine 5-phosphosulfate reductase (APSR) heterodimer from
Archaeoglobus fulgidus. The -subunit, which harbors the flavin adenine dinucleotide (FAD; yellow),
is shown in blue; the -subunit, with the two [4Fe4S] clusters, is shown in red. The substrate-binding
channel is illustrated by approximately 35 tightly bound water molecules (green), indicating a
strong electrostatic field favorable for binding charged groups (Fritz et al. 2002b; Schiffer et al.
2006). Right: Active-site channel and electron transfer pathway
Fig. 2.2 Alignment of the FeS-binding domain of the APSR -subunit (red) from A. fulgidus
with the ferredoxin (green) from Clostridium acidiurici; the -subunit has an elongated loop
(magenta) that presumably represents the docking site for the physiological electron donor.
Electron transfer over a distance of about 30 proceeds from the protein surface to FAD via the
two [4Fe4S] clusters and conserved TrpB48 to the C8 methyl group of FAD (Fritz et al. 2002b)
Fig. 2.3 The FADsulfite adduct of APSR from A. fulgidus. The three sulfite oxygens are hydro-
gen-bonded; His A398 and Arg A265 appear to be key residues for substrate binding and catalysis
(Schiffer 2004)
CH3
A N
-
N N H
O B CH3
-
N O H N N O
W 234 H W 234 N H
NH
N 74 N
HN
O APS NH
O N 74
O HN
O
O O -
H 398 O S H2 O
OH 2
R 265
NH +
NH2 HN O
H 398 O
N E 141 NH + O
R 265 NH2 O
H2N
O P HN O
N E 141
H2N O
- R
2e
F C CH 3
CH3
N N -
N O N O
W 234 N N W 234
H
NH O N 74 N NH N 74
HN NH O
O O HN
S O- O
OH 2 H2O H O
OH 2 + H2O
H 398 NH NH2 O OH 2 O
NH + O
R 265 NH2 HN R 265 -
O H 398 O
N H 2N P E 141
O O HN
H 2N O
E 141 R N
-
HSO 3
E CH3 D CH3
- O
N N - O
NH N N
W 234 N W 234 NH
O N 74 N
O
N 74
NH S HN NH HN
O O S O
O O
OH 2 O OH 2 OH 2 O O
+ H2 O O
H3O O H3 O
+
AMP H2 O
- O
NH + HN O O E 141
R 265 NH2 H 398 NH2
+
O
NH
E 141 NH
R 265 - P H 398
N O N
H2 N NH2 O
R
Fig. 2.4 Reaction cycle of APSR. A represents reduced APSR, B APSRadenosine 5-phosphosul-
fate (APS), D APSRAMP, E reduced APSRsulfite and F oxidized APSR; B does not exactly
represent the APSRAPS state, as the latter contains FAD in the oxidized state; the postulated
short-lived state C was modeled (Schiffer et al. 2006)
20 G. Fritz et al.
of the FADsulfite state. Finally, a FADsulfiteAMP1 state could be trapped

with AMP within van der Waals distance. This structure documents how adja-
cent negative charges can be stabilized by the protein, which is crucial for the
back reaction to form APS from AMP and sulfite (Fritz et al. 2002b; Schiffer
et al. 2006).
2.3 Dissimilatory SIR
SIRs are key enzymes for both biosynthetic assimilation of sulfur and dissimilation
of oxyanions, such as sulfate, for energy conservation (LeGall and Fauque 1988).
Found throughout the three major kingdoms of living organisms, many of these
enzymes employ a siroheme that is exchange-coupled with an ironsulfur cluster
(Belinsky 1996; Crane et al. 1995). SIRs catalyze the six-electron reduction of
sulfite to sulfide (Thauer et al. 1977):
HSO3 + 6e + 6H + HS + 3H2 O E 0 ( HSO3 HS ) = 116 mV.
SIR has been described as 22mn multimers with approximately 50 kDa,

approximately 45 kDa, approximately 11 kDa, approximately 8 kDa, and a
total molecular mass of approximately 200 kDa (Steuber and Kroneck 1998).
The enzyme has been isolated from several microorganisms (Lee et al. 1973;
Moura et al. 1988; Lui et al. 1994; Steuber et al. 1995; Dahl and Trper 2001).
The -subunit from Pyrobaculum aerophilium reveals a novel fold consisting of
a -hairpin and an orthogonal helix bundle. A flexible seven-residue C-terminal
arm with a C-terminal cysteine is suggested to be involved in interaction with the
22 tetramer (Cort et al. 2001). In the X-ray structure of the A. fulgidus -subunit,
however, this highly conserved C-terminal arm adopts a well-defined confirma-
tion and the C-terminal cysteine might be a constituent of a redox-active
disulfide bond (Mander et al. 2005). The -subunit from D. vulgaris contains a
winged helix motif, suggesting its participation in DNA-binding (Mizuno et al.
2003). The proposed binding of sulfate or sulfite (Karkhoff-Schweizer et al.
1995) was meanwhile ruled out (Hittel and Voordouw 2000).
Active SIR from A. fulgidus was recently purified under the strict exclusion of
dioxygen (Schiffer 2004). The enzyme consisted of an -subunit (51 kDa) and a
-subunit (45 kDa) arranged as an 22 heterotetramer. Its specific activity (approx-
imately 50 nmol sulfite min1 mg1) was lower than values obtained for crude
extracts of A. fulgidus with reduced methylviologen (Dahl et al. 1994; Dahl and
Trper 2001); note that those initial rates were also found by Schiffer (2004) for the
crude extract of A. fulgidus.
The enzyme from A. fulgidus exhibited a complicated set of EPR resonances at
low and high magnetic field (Schiffer 2004; Fritz et al. 2005) The interpretation of
the high-spin EPR signals and the assignment of the individual resonances to the
metal sites of dissimilatory SIR represents a major challenge. In the oxidized state
there were two types of high-spin signals, with spin S = 5/2 and S = 9/2 (Pierik and
Hagen 1991). The signals with spin S = 5/2 were present in assimilatory as well as
dissimilatory SIRs (Pierik and Hagen 1991; Wolfe et al. 1994), whereas the S = 9/2
signals were only observed in several dissimilatory SIRs, including the enzymes
from D. vulgaris and A. fulgidus. The spin S = 5/2 signal results from the coupled
high-spin siroheme center. As shown by Mssbauer spectroscopy, the siroheme is
in the high-spin state and is strongly exchange coupled to the [4Fe4S]2+ cluster
(Christner et al. 1981). There are exchange and hyperfine interactions between the
heme iron and the ironsulfur cluster (Belinsky 1996).
The iron content of dissimilatory SIR has been a matter of controversy, with ten
to 24 Fe/22nm (Steuber and Kroneck 1998). For the A. fulgidus enzyme 2224
non-heme Fe/22 were reported, indicative for the presence of six [4Fe4S] clusters
(Dahl et al. 1994). Recently, SIR from A. fulgidus was crystallized in the absence
of dioxygen (Schiffer 2004). The green-brown crystals diffracted well below 2.5
and were suitable for X-ray structure analysis, which is currently in progress.
Acknowledgements. Financial support from the Deutsche Forschungsgemeinschaft (P.M.H.K.)

and the Max-Planck-Gesellschaft (U.E.) is gratefully acknowledged.
References
Achenbach-Richter L, Gupta R, Stetter, KO, Woese CR (1987) Were the original eubacteria
thermophiles? Syst Appl Microbiol 9:3439
Amend JP, Shock EL (2001) Energetics of overall metabolic reactions of thermophilic and hyper-
thermophilic Archaea and Bacteria. FEMS Microbiol Rev 25:175243
Beinert H, Holm RH, Mnck E (1997) Iron-sulfur clusters: natures modular, multipurpose
structures. Science 277:653659
Belinsky MI (1996) Exchange model of the {[Fe4S4]-Fe} active site of sulfite reductase. Chem
Phys 201:343356
Bchert T, Fritz G, Kroneck, PMH (1999) Towards the natural electron donor of adenosine-5-
phosphosulfate (APS) reductase from Desulfovibrio desulfuricans Essex. In: Ghisla S,
Kroneck PMH, Macheroux P, Sund H (eds) Flavins and flavoproteins. Rudolf Weber Agency
for Scientific Publications, Berlin, pp 803806
Cameron EM (1982) Sulfate and sulfate reduction in the early Precambrian oceans. Nature
296:145148
Christner JA, Mnck E, Kent TA, Janick PA, Salerno JC, Siegel L M (1984) Exchange coupling
between siroheme and [4Fe-4S] cluster in E. coli sulfite reductase. Mssbauer studies and
coupling models for a 2-electron reduced enzyme state and complexes with sulfide. J Am
Chem Soc 106:67866794
Cort JR, Santhana Mariappan SV, Kim C-Y, Park M S, Peat T S, Waldo G S, Terwilliger T C,
Kennedy MA (2001) Solution structure of Pyrobaculum aerophilium DsrC, an archaeal
homologue of the gamma subunit of dissimilatory sulfite reductase. Eur J Biochem 268:
58425850
Crane BR, Siegel LM, Getzoff ED (1995) Sulfite reductase at 1.6 : evolution and catalysis for
reduction of inorganic anions. Science 270:5967
22 G. Fritz et al.
Dahl C, Trper HG (2001) Sulfite reductase and APS reductase from Archaeoglobus fulgidus.
Dahl C, Speich N, Trper HG (1994) Enzymology and molecular biology of sulfate reduction in
extremely thermophilic archaeon Archaeoglobus fulgidus. Methods Enzymol 243:331352
Fritz G, Bchert T, Huber H, Stetter KO, Kroneck PMH (2000) reductases from archaea and
bacteria are 1:1 alphabeta-heterodimeric iron-sulfur flavoenzymes. High similarity of molecular
properties emphasizes their central role in sulfur metabolism. FEBS Lett 473:6366
Fritz G, Bchert T, Kroneck PMH (2002a) The function of the [4Fe-4S] clusters and FAD in
bacterial and archaeal adenosine 5-phosphosulfate reductases. Evidence for flavin-catalyzed
reduction of adenosine 5-phosphosulfate. J Biol Chem 277:2606626073
Fritz G, Roth A, Schiffer A, Bchert T, Bourenkov G, Bartunik HD, Huber H, Stetter KO,
Kroneck PMH, Ermler U (2002b) Crystal structure of the adenosine 5-phosphosulfate reduct-
ase from the hyperthermophilic Archaeon Archaeoglobus fulgidus at 1.6 resolution. Proc
Natl Acad Sci USA 99:18361841
Fritz G, Einsle O, Rudolf M, Schiffer M, Kroneck PMH (2005) Key bacterial multi-centered metal
enzymes involved in nitrate and sulfate respiration. J Mol Microbiol Biotechnol 10:223233
Hansen TA (1994) Metabolism of sulfate-reducing prokaryotes. Antonie Van Leeuwenhoek
66:165185
Hittel DS, Voordouw G (2000) Overexpression, purification and immunodetection of DsrD from
Desulfovibrio vugaris (Hildenborough). Antonie Van Leeuwenhoek 77:1322
Karkhoff-Schweizer RR, Huber D P, Voordouw G (1995) Conservation of the genes for dissimila-
tory sulfite reductase from Desulfovibrio vulgaris and Archaeoglobus fulgidus allows their
detection by PCR. Appl Environ Microbiol 61:290296
Kroneck PMH (2005) The biogeochemical cycles of the elements and the evolution of life. In:
Sigel A, Sigel H, Sigel RKO (eds) Metal ions in biological systems, vol. 43. Taylor & Francis,
Baton Rouge, pp 17
Lampreia J, Pereira AS, Moura JJG (1994) Adenosine 5-phosphosulfate reductase from sulfate-
reducing bacteria. Methods Enzymol. 243:241260
Lancaster CRD (2003) Wolinella succinogenes quinol:fumarate reductase and its comparison to
E. coli succinate:quinone reductase. FEBS Lett 555:2128
Lee J-P, LeGall J, Peck HD (1973) Isolation of assimilatory- and dissimilatory-type sulfite reduct-
ases from Desulfovibrio vulgaris. J Bacteriol 115:529542
LeGall J, Fauque G (1988) Dissimilatory reduction of sulfur compounds. In: Zehnder AJB (ed)
Biology of anaerobic microorganisms. Wiley, New York, pp 587639
Lipmann F (1958) Biological sulfate activation and transfer: studies on a mechanism of group
activation and its role in biosynthesis are described. Science 128:575580
Lui S M, Soriano A, Cowan JA (1994) Electronic properties of the dissimilatory sulfite reductase
from Desulfovibrio vulgaris (Hildenborough): comparitative studies of optical spectra and rel-
ative reduction potentials for the [Fe4S4]-sirohaem prostetic centers. Biochem J 304:441447
Mander GJ, Weiss MS, Hedderich R, Kahnt J, Ermler U, Warkentin E (2005) X-ray structure of
the -subunit of a dissimilatory sulfite reductase: fixed and flexible C-terminal arms. FEBS
Lett 579:46004604
Massey V, Mller F, Feldberg R, Schuman M, Sullivan PA, Howell LG, Mayhew SG, Matthews RG,
Foust GP (1969) The reactivity of flavoproteins with sulfite. Possible relevance to the problem
of oxygen reactivity. J Biol Chem.244: 39994006
Matias PM, Pereira IAC, Soares CM, Carrondo MA (2005) Sulphate respiration from hydrogen
in Desulfovibrio bacteria: a structural biology overview. Prog Biophys Mol Biol 89:292329
Michaels GB, Davidson JT, Peck HD Jr (1970) A flavin-sulfite adduct as an intermediate in the
reaction catalyzed by adenylyl sulfate reductase from Desulfovibrio vulgaris. Biochem
Biophys Res Commun 39:321328
Mizuno N, Voordouw G, Miki K, Sarai A, Higuchi Y (2003) Crystal structure of dissimilatory
sulfite reductase D (DsrD) protein possible interaction with B- and Z-DNA by Its winged-
helix motif. Structure 11:11331140
Moura I, LeGall J, Lino AR, Peck HD, Fauque G, Xavier AV, DerVartanian DV, Moura JJG,
Huynh BH (1988) Characterisation of two dissimilatory sulfite reductases from the sulfate-
reducing bacteria. Mssbauer and EPR studies. J Am Chem Soc 110:10751082
Peck HD Jr (1959) The ATP-dependent reduction of sulfate with hydrogen in extracts of
Desulfovibrio desulfuricans. Proc Natl Acad Sci USA 45:701708
Pierik AJ, Hagen WR (1991) S = 9/2 EPR signals are evidence against coupling between the siro-
heme and the Fe/S cluster prosthetic groups in Desulfovibrio vulgaris (Hildenborough)
dissimilatory sulfite reductase. Eur J Biochem 195:505516
Pires RH, Venceslau SS, Morais F, Texeira M, Xavier AV, Pereira IAC (2006) Characterization of
the Desulfovibrio desulfuricans ATCC 27774 DsrMKJOP complex a membrane-bound
redox complex involved in the sulfate respiratory pathway. Biochemistry 45:249262
Schidlowski M, Hayes JM, Kaplan IR (1983) Isotopic inferences of ancient biochemistries:
carbon, sulfur, hydrogen, and nitrogen In: Schopf JW (ed) Earths earliest biosphere, its origin
and evolution. Princeton University Press, Princeton, pp 149186
Schiffer A (2004) Structural and functional investigations on multi-site metallo enzymes of the
biological sulfur cycle. Dissertation, Universitt Konstanz
Schiffer A, Fritz G, Kroneck PMH, Ermler U (2006) Reaction mechanism of the iron-sulfur
flavoenzyme adenosine-5-phosphosulfate reductase based on the structural characterization of
different enzymatic states. Biochemistry 45:29602967
Speich N, Dahl C, Heisig P, Klein A, Lottspeich, F, Stetter KO, Trper HG (1994) Adenylylsulphate
reductase from the sulphate-reducing archaeon Archaeoglobus fulgidus: cloning and charac-
terization of the genes and comparison of the enzyme with other iron-sulphur flavoproteins.
Microbiology 140:12731284
Stetter KO, Lauerer G, Thomm M, Neuner A (1987) Isolation of extreme thermophilic sulfate
reducers: Evidence for a novel branch of archaebacteria. Science 236:822824
Steuber J, Kroneck PMH (1998) Desulfoviridin, the dissimilatory sulfite reductase from
Desulfovibrio desulfuricans (Essex): new structural and functional aspects of the membranous
enzyme. Inorg Chim Acta 275276:5257
Steuber J, Arendsen AF, Hagen WR, Kroneck PMH (1995) Molecular properties of the dissimilatory
sulfite reductase from Desulfovibrio desulfuricans (Essex) and comparison with the enzyme
from Desulfovibrio vulgaris (Hildenborough). Eur J Biochem 233:873879
Thauer RK, Jungermann K, Decker K (1977) Energy conservation in chemotrophic anaerobic
bacteria. Bacteriol Rev 41:100180
Verhagen MFJM, Kooter IM, Wolbert RBG, Hagen WR (1994) On the iron-sulfur cluster of
adenosine phosphosulfate reductase from Desulfovibrio vulgaris (Hildenborough). Eur J
Biochem 221:831837
Wolfe BM, Lui SM, Cowan JA (1994) Desulfoviridin, a multimeric-dissimilatory sulfite reductase
from Desulfovibrio vulgaris (Hildenborough). Eur J Biochem 223:7989
Chapter 3
Respiratory Membrane Complexes
of Desulfovibrio
Ins A. Cardoso Pereira
Abstract Despite many years of research the process of sulfate respiration is still
not fully understood. The mechanisms and components associated with energy
conservation have not been clearly identified, and the electron donors to the cyto-
plasmic adenosine 5-phosphosulfate (APS) and sulfite reductases are not known.
Recently, considerable progress has been achieved through genome analysis and
other biochemical and genetic studies. This review presents our current knowl-
edge of transmembrane redox complexes of Desulfovibrio spp. that are proposed
to play a role in the respiratory electron transfer chain. Two of these complexes,
Qmo and Dsr, are apparently conserved in all sulfate reducers, pointing to an
essential role in sulfate respiration, most likely as electron donors to the APS and
sulfite reductases, respectively. In contrast, the Hmc, 9Hc and Tmc complexes
are only present in Desulfovibrio organisms, suggesting a role in alternative
pathways. The presence of the latter complexes correlates with the large pool
of periplasmic cytochromes c found in Desulfovibrio spp., which act as electron
donors to the complexes upon periplasmic oxidation of hydrogen or formate.
Future studies are required to establish the exact function of all the complexes
discussed, namely, their electron donors and acceptors and their involvement in
energy-conserving mechanisms.
3.1 Introduction
It has long been recognized that dissimilatory sulfate reduction is associated with
oxidative phosphorylation, and is a true respiratory process (Peck 1960).
However, despite many years of research into sulfate-reducing bacteria (SRB) it
has still not been clearly established how the electron transport chain is associ-
ated with the generation of a proton-motive force. The terminal reductases (APS
reductase and sulfite reductase) are cytoplasmic and so are not directly involved
in proton translocation. A typical complex I or bc1 complex is not present. Several
important intervenients in the electron transport chain have not been identified,
such as the electron donors to APS and sulfite reductases, or the electron acceptor
of the lactate dehydrogenase. It is also not clear what is the role of important
24
Springer 2008
3 Respiratory Membrane Complexes of Desulfovibrio 25
electron carriers such as NAD(P)H (Kremer and Hansen 1989) and menaquinone
(Collins and Widdel 1986).
Several bioenergetic mechanisms have been proposed to explain energy conser-
vation in SRB. Odom and Peck (1981) proposed a mechanism of hydrogen cycling
for Desulfovibrio vulgaris growing in lactate/sulfate. In this mechanism electrons
from lactate oxidation are transferred to a cytoplasmic hydrogenase that produces H2.
This diffuses to the periplasm, where its reoxidation generates electrons that are
shuttled across the membrane for the cytoplasmic reduction of sulfate, leaving
protons in the periplasm that generate a pH gradient. This mechanism seems not to
be applicable to all SRB since genome analysis shows that a cytoplasmic hydroge-
nase is absent in several organisms. In addition, H2 formation from lactate oxida-
tion to pyruvate is energetically very unfavorable, suggesting that other mechanisms
are operative. More recent evidence indicates that cycling of other reduced inter-
mediates, like CO or formate, may also function in Desulfovibrio (Voordouw 2002;
Heidelberg et al. 2004). Genome analysis points to the existence of differences in
energy metabolism between different SRB, e.g., D. vulgaris versus Desulfotalea
psychrophila (Pereira et al. 2007). The former has a much higher number of peri-
plasmic cytochromes, hydrogenases and formate dehydrogenases, suggesting that
in D. vulgaris cycling of reduced intermediates may play a more important role
than in Dt. psychrophila. Chemiosmotic processes in which energy conservation is
achieved by a membrane-bound electron transport chain that transfers protons to
the periplasm have also been proposed (Wood 1978; Lupton et al. 1984), and are
most probably operative since electron-transport-driven proton translocation has
been demonstrated for several Desulfovibrio spp. (Fitz and Cypionka 1991).
Whatever the mechanisms operating, membrane-associated electron transport is a
requirement. These membrane processes most likely involve menaquinone and may
contribute to energy conservation through standard mechanisms like redox loops
(Jormakka et al. 2003). In recent years, considerable progress has been achieved in
our understanding of membrane-bound redox proteins in SRB, through genetic,
genomic and biochemical studies (Matias et al. 2005; Pereira et al. 2007). These studies
revealed the presence of several transmembrane redox complexes, which are prob-
ably involved in the electron transfer chain. These complexes are unique to sulfur-
metabolizing organisms and contain several novel and interesting proteins, but further
studies are required to establish their precise physiological function.
3.2 Membrane Complexes Conserved in Sulfate Reducers
Analysis of the membrane electron transport complexes present in the complete

genomes of four sulfate-reducers [D. vulgaris Hildenborough (Heidelberg et al.
2004), D. desulfuricans G20 (Joint Genome Initiative 1997), Dt. psychrophila
(Rabus et al. 2004) and Archaeoglobus fulgidus (Klenk et al. 1997)] reveals that
only two such complexes are conserved among the four organisms, suggesting they
may be the only ones essential for sulfate reduction. They are the QmoABC
26 I.A.C. Pereira
complex, which was isolated from D. desulfuricans ATCC 27774 (Pires et al. 2003),
and the DsrMKJOP complex, first isolated from A. fulgidus (Mander et al. 2002) and
more recently also from D. desulfuricans ATCC 27774 (Pires et al. 2006). Indirect
evidence suggests that QmoABC is involved in electron transfer to the APS reductase
and DsrMKJOP is involved in electron transfer to the sulfite reductase (Pires et al.
2003, 2006; Haveman et al. 2004; Dahl et al. 2005; Mussmann et al. 2005).
A striking point regarding the two complexes is that both contain subunits that
are related to subunits of heterodisulfide reductases (Hdr) of methanogens (Fig.
3.1), which catalyze the reduction of the heterodisulfide of two thiol coenzymes
(CoMSH and CoBSH). The heterodisulfide CoMSSCoB is formed in the last
step of methanogenesis and acts as the terminal electron acceptor in the respiratory
chain of these organisms (Hedderich et al. 1999). Its reduction is linked to energy
conservation by generation of a proton-motive force. In hydrogenotrophic metha-
nogens like Methanothermobacter marburgensis the Hdr is soluble and composed
of three subunits, HdrA, HdrB and HdrC. HdrA is a flavo-FeS protein, HdrC is also
an FeS protein, and HdrB contains two five-cysteine motifs that are proposed to
bind also FeS cluster(s) (Hedderich et al. 2005). In methylotrophic methanogens
like Methanosarcina sp. the Hdr is membrane-bound and composed of only two
subunits, HdrD and HdrE. HdrD is a homologue of a hypothetical fusion of the
HdrBC subunits. HdrE is an integral membrane subunit containing two heme
b groups. The catalytic subunits of both Hdrs are HdrB and HdrD, which contain a
catalytic FeS cluster that forms a paramagnetic [4Fe4S]3+ center upon oxidation in
the presence of HSCoM or HSCoB (Hedderich et al. 2005). The two types of
Hdrs have different electron donors. The membrane-bound HdrED enzyme receives
electrons from the membrane cofactor methanophenazine via the cytochrome
b HdrE subunit, whereas the soluble HdrABC enzyme forms a complex with the
F420-non-reducing hydrogenase that catalyzes reduction of the heterodisulfide by H2
(Stojanowic et al. 2003).
3.2.1 The Qmo Complex
The Qmo complex was isolated from the membranes of D. desulfuricans ATCC
27774 (Pires et al. 2003). It is composed of three subunits and contains two hemes
b, two flavin adenine dinucleotide groups and several ironsulfur centers. The
genes encoding these proteins form a putative operon and were named qmoABC
for quinone-interacting membrane-bound oxidoreductase. Homologous genes
are found in the genomes of the sulfate reducers D. vulgaris Hildenborough,
D. desulfuricans G20, Dt. psychrophila, Desulfotomaculum reducens MI-1 (Joint
Genome Initiative 1997) and A. fulgidus. Interestingly, the qmo genes are also
present in sulfur-oxidizing bacteria like the phototrophic Chlorobium tepidum
(Eisen et al. 2002) and Chlorobium chlorochromatii (Joint Genome Initiative 1997),
and the chemotrophic Thiobacillus denitrificans (Beller et al. 2006; Fig. 3.2). In
several of these genomes the qmo genes are found adjacent to the APS reductase
Fig. 3.1 The proteins discussed in the text, as deduced from sequence data. Related subunits are
in similar shades of gray. The putative catalytic [4Fe4S] center in HdrB, HdrD, DsrK, HmcF and
TmcB is depicted in light gray; 4C represents a conserved four-cysteine motif
genes (apsAB), providing evidence of a physiological relationship between these

proteins. In some cases the ATP sulfurylase gene (sat) is also present in the same locus.
QmoA and QmoB are predicted to be cytoplasmic flavo-FeS proteins that are both
related to HdrA (Fig. 3.1), but QmoA is smaller and shows sequence similarity only
to an N-terminal segment of HdrA and QmoB (Pires et al. 2003). The QmoB protein
probably resulted from a gene fusion since the N-terminus region is similar to HdrA
and the C-terminus region is similar to the -subunit of the F420-non-reducing hydro-
genases (MvhD). The QmoC protein seems to be also the result of a gene fusion as the
N-terminus region is hydrophilic and shows similarity to HdrC, including the binding
sites for two [4Fe4S] clusters, whereas the C-terminus is hydrophobic and includes six
transmembrane helices. This membrane-bound domain is similar to HdrE, and the
UVvis spectrum of the QmoABC complex supports the presence of two hemes b.
Thus, QmoC belongs to the family of membrane subunits of respiratory complexes
that bind two hemes b on opposite sides of the bilayer and that are responsible for
electron transfer with the membrane quinones in a process that may be associated with
generation of a proton gradient (Berks et al. 1995). However, QmoC is a unique case
within this family, since it is the first example, to our knowledge, of a protein that con-
tains an additional hydrophilic domain with ironsulfur centers. The sequence of
QmoC strongly suggests that menaquinone is the electron donor to the QmoABC
complex, and the two hemes of QmoC are reduced by menadiol, a menaquinol ana-
logue. Interestingly, in Dm. reducens and T. denitrificans the qmoC gene is not found,
and instead two genes related to hdrC and hdrB are present (Fig. 3.2).
28 I.A.C. Pereira
Fig. 3.2 Representation of the qmo and dsr gene arrangement in several of the organisms discussed.
* the dsrS gene is not found in Thiobacillus denitrificans
The exact function of the QmoABC complex remains to be established; how-

ever, the present evidence points to its involvement in the sulfate respiratory chain,
and more specifically as a probable electron donor to the APS reductase (Pires et al.
2003; Haveman et al. 2004; Mussmann et al. 2005). However, no in vitro electron
transfer could be observed between the isolated QmoABC complex and APS
reductase (Pires et al. 2003). This may have been due to experimental problems or
may indicate that an additional redox partner is involved. The macroscopic redox
potentials of the two hemes b in the Qmo complex were determined to be 20 and
+75 mV. These potentials are in a suitable range to be involved in electron transfer
from menaquinol (70 mV) to APS (E0 APS/SO32 = 60 mV). Thus, the QmoABC
complex provides a link between the menaquinone pool and the cytoplasmic reduc-
tion of sulfate. If oxidation of menaquinol by QmoC occurs at the heme closest to
the positive side of the membrane, protons may be released to the periplasm and
electrons transferred to the negative side of the membrane to QmoAB, with subse-
quent reduction of APS. Hence, electron transfer through QmoABC may lead to
formation of a proton gradient through a redox-loop mechanism, but this has to be
verified experimentally.
3.2.2 The Dsr Complex
The DsrMKJOP complex was isolated from A. fulgidus (Mander et al. 2002) and
also from D. desulfuricans ATCC 27774 (Pires et al. 2006). Sequence analysis
reveals that DsrM is a membrane cytochrome b like HdrE and the C-terminal
domain of QmoC (Fig. 3.1). DsrM is predicted to contain six transmembrane
helices and has four conserved histidines, which are likely candidates to bind two
hemes b. DsrK is predicted to be a cytoplasmic ironsulfur protein that is related
to the catalytic subunit HdrD. DsrK contains only one of the five-cysteine motifs of
HdrD, which are probably involved in binding the [4Fe4S] catalytic center for
disulfide reduction. The DsrJ protein contains three heme c binding sites, and an
N-terminal signal peptide for export to the periplasm. This peptide is not cleaved
off in the corresponding protein of the A. fulgidus complex, and probably serves as
a membrane anchor for the periplasmic cytochrome. The DsrJ sequence shows no
homology to other cytochromes in the databases and so corresponds to a novel
family of cytochromes c. There are not enough histidines in DsrJ for all the hemes
to have bishistidine ligation that is usually found in multiheme cytochromes. The
Dsr O protein is a periplasmic FeS protein that belongs to the family of ferredoxin-
like subunits found in several respiratory enzymes. Dsr O includes a typical signal
peptide for translocation to the periplasm. DsrP is an integral membrane protein
predicted to contain ten transmembrane helices. It is related to the membrane
subunit of Escherichia coli hydrogenase-2 (HybB), which acts as a menaquinone
reductase, and to a whole family of membrane subunits of respiratory enzymes.
The operon coding for the Dsr complex is present in the genomes of all sulfate-
reducing organisms sequenced to date. In two bacteria that reduce sulfite, but not
sulfate, Moorella thermoacetica and Desulfitobacterium hafniense, the Dsr complex
is encoded in the same locus as the dsrAB genes coding for the two subunits of the
dissimilatory sulfite reductase (Fig. 3.2). Strikingly, both the sulfite reductase and
the Dsr complex are also found in organisms that oxidize reduced sulfur com-
pounds like the phototrophs Allochromatium vinosum and C. tepidum, or the chem-
otroph T. denitrificans (Sander et al. 2006). In these organisms the dsrAB and
dsrMKJOP are also part of the same gene cluster that includes other conserved dsr
genes like dsrC and dsrN. It was in A. vinosum that the dsrMK genes were first
identified as belonging to the same gene cluster as dsrAB (and were thus named
also dsr for dissimilatory sulfite reductase), and these genes were shown to be
obligatory for sulfur oxidation (Pott and Dahl 1998).
Spectroscopic characterization of the D. desulfuricans Dsr complex confirms the
presence of a [4Fe4S]3+ cluster (Pires et al. 2006), with similar characteristics to the
one reported in A. fulgidus, and analogous to that observed in Hdrs where it acts as
the catalytic site (Hedderich et al. 2005). On the basis of sequence and spectro-
scopic data, the hemes b in DsrM are proposed to be bishistidine-ligated, whereas
the three hemes c in DsrJ have each different coordination with one bishistidine,
one histidine/methionine and another a very unusual histidine/cysteine coordina-
tion (Pires et al. 2006). There are very few precedents for cysteine coordination in
30 I.A.C. Pereira
hemes c, which include the SoxAX cytochrome (Bamford et al. 2002) that is
involved in thiosulfate oxidation (Friedrich et al. 2005), the triheme PufC cyto-
chrome of the photosynthetic reaction center (Alric et al. 2004), and possibly a new
green cytochrome from Halochromatium salexigens (Van Driessche et al. 2006).
The histidine/cysteine coordinated heme cannot be fully reduced. Treatment with
DMNH2 led only to approximately40% reduction of the hemes.
There is considerable evidence to indicate that the Dsr complex is part of the
same metabolic pathway as the sulfite reductase: all prokaryotic genomes that con-
tain a DsrAB dissimilatory sulfite reductase contain also a DsrMKJOP complex; in
A. vinosum the DsrKJO proteins associate with the DsrABC proteins (Dahl et al.
2005), and the genes for the sulfite reductase and Dsr complex are coordinately
regulated by sulfide (Pott and Dahl 1998; Dahl et al. 2005). In this organism the
proteins encoded by the dsr genes were shown to be essential for oxidation of intra-
cellular stored sulfur (Pott and Dahl 1998; Dahl et al. 2005; Sander et al. 2006).
However, the precise physiological role of the Dsr complex has still not been
established. Sequence analysis suggests that electron transfer may occur in the
periplasm, in the membrane and in the cytoplasm. The unique nature of DsrJ pre-
vents any hints as to its physiological function. It is not an electron acceptor for the
hydrogenase/type I cytochrome c3 (TpIc3) couple, and its heme coordination is
suggestive of a specialized role, possibly catalytic. The cytoplasmic DsrK protein
is most probably a catalytic subunit, given its similarity to the catalytic subunit
HdrD of Hdrs. This suggests that DsrK may be involved in catalyzing a thiol/
disulfide type of redox chemistry.
3.3 Membrane Complexes Found Only in Desulfovibrio spp.
Desulfovibrio spp. are characterized by a high content of periplasmic or membrane-

associated cytochromes c (Matias et al. 2005; Pereira and Xavier 2005). Of these,
the tetraheme TpIc3 is very abundant, suggesting an important role in energy
metabolism. It is thus somewhat surprising that neither Dt. psychrophila nor
A. fulgidus have a TpIc3 (and indeed have very few cytochromes c), showing that it
is not essential for sulfate reduction. In Desulfovibrio spp. the pool of periplasmic
cytochromes c act as electron acceptors for periplasmic hydrogenases and formate
dehydrogenases (Heidelberg et al. 2004; Elantak et al. 2005; Matias et al. 2005).
In these organisms the two classes of proteins lack the cytochrome b membrane
subunit typically present in such enzymes (Pereira et al. 2007). Dt. psychrophila
and A. fulgidus have hydrogenases and formate dehydrogenases with a cytochrome
b membrane subunit, which transfers electrons to the menaquinone pool. In
Desulfovibrio organisms the electrons resulting from periplasmic hydrogen and
formate oxidation are likely to be transferred from the pool of cytochromes c3 to
one of several transmembrane complexes found only in Desulfovibrio spp., which
include a cytochrome c subunit also belonging to the cytochrome c3 family (Matias
et al. 2005; Fig. 3.1). The cytochrome c subunits of three of these complexes,
HmcA, 9HcA and TmcA, have been isolated and characterized, and recently the
Tmc complex was also isolated. It is still not clear whether these three complexes
transfer electrons to the menaquinone pool and/or directly to the cytoplasm for
reduction of sulfate. The subunits of the Hmc, 9Hc and Tmc complexes have a
strong sequence similarity between them.
3.3.1 The Hmc and 9Hc Complexes
The first transmembrane complex to be recognized in Desulfovibrio spp. was the

Hmc complex of D. vulgaris (Rossi et al. 1993). Sequence analysis indicates that
this complex has a subunit composition strikingly similar to the Dsr complex in
terms of the type of subunits present: a cytoplasmic FeS protein related to HdrD,
two integral membrane proteins, a periplasmic ferredoxin-like protein and a
periplasmic cytochrome c (Fig. 3.1). This suggests that both complexes have
related functions, but the actual sequence identity between subunits is very low. The
multiheme cytochrome c subunit is the most dissimilar since it is a large, 65-kDa,
sixteen-heme cytochrome in Hmc and a small three-heme cytochrome of 15 kDa in
Dsr. The HmcA cytochrome is a poor electron acceptor for the periplasmic hydro-
genases, but its reduction rate increases significantly in the presence of TpIc3
(Matias et al. 2005). Several studies indicate that the Hmc complex accepts elec-
trons from periplasmic hydrogen oxidation and a mutant deleted in the hmc operon
grew at a slower rate than the wild type with hydrogen and sulfate (Dolla et al.
2000). Since deletion of the hmc operon does not prevent growth on hydrogen,
there are probably other proteins that can fulfill the same role, and a likely candi-
date is the Tmc complex described in Sect. 3.3.2.
In D. desulfuricans no HmcA has been detected, but a nine-heme cytochrome
(9HcA) that is structurally very similar to the C-terminal domain of HmcA is
present (Matias et al. 2005). This cytochrome is part of a transmembrane redox
complex (9Hc) that lacks the heme b and cytoplasmic FeS subunits of Hmc, but
includes two membrane subunits (9HcC and 9HcD) and a periplasmic FeS subunit
(9HcB) (Saraiva et al. 2001; Fig. 3.1). The 9HcA cytochrome is a much better elec-
tron acceptor for the hydrogenases than HmcA, but its reduction is most probably
also mediated by TpIc3 (Matias et al. 2005).
3.3.2 The Tmc Complex
The TmcABCD complex is the first one of this family to have been isolated (Pereira et
al. 2006). Its cytochrome c subunit (TmcA) had previously been characterized in several
Desulfovibrio spp. and was named type II cytochrome c3 (TpIIc3) since it has several
features that distinguish it from TpIc3 (Matias et al. 2005). The Tmc complex is encoded
in a ten-gene operon present in D. vulgaris Hildenborough and D. desulfuricans G20.
32 I.A.C. Pereira
Four of the genes encode regulatory proteins and one gene encodes a hypothetical
protein. The other four genes, tmcABCD, encode the functional proteins of the com-
plex (Fig. 3.1). The tmcB gene codes for a cytoplasmic FeS protein, homologous to
HmcF, and of the same family as DsrK and HrdD, and includes a binding site for a
putatively catalytic [4Fe4S] center. The tmcC gene encodes a membrane cytochrome
b homologous to HmcE, and of the same family as DsrM and HdrE. The tmcD gene
encodes a tryptophan-rich protein that shows no similarity to any proteins in the
databases. The electron paramagnetic resonance spectrum of the oxidized complex
confirms the presence of the TmcB FeS center with similar characteristics to that of
the D. desulfuricans DsrK and assigned to a [4Fe4S]3+ center.
TpIIc3 (TmcA) is efficiently reduced by the pair hydrogenase/ TpIc3 (Matias
et al. 2005). Reduction of the Tmc complex with H2 and hydrogenase/ TpIc3 led to
almost complete reduction of all the redox centers. This supports the prediction that
the Tmc complex is a transmembrane conduit for electrons resulting from periplasmic
hydrogen oxidation. The hemes of the Tmc complex were not reduced with the
menaquinol analogue DMNH2. Conversely, the reduced Tmc complex could transfer
electrons to DMN, but the reduction rate was similar using only TpIIc3, suggesting
the process may be nonphysiological. These results do not support, but also cannot
discard, the involvement of the menaquinone pool in electron transfer through Tmc.
3.4 Conclusions
We have, nowadays, a lot more pieces of the puzzle of sulfate respiration, but the
whole picture is still far from complete. It seems quite certain that the Qmo and Dsr
complexes will be involved in the electron transfer pathways to the APS reductase
and sulfite reductase, respectively, but the precise mechanism of interaction is not
clear and there may be other molecules involved. Another important conclusion is
that menaquinol is most likely the electron donor to the Qmo complex, which
finally assigns a role for the membrane quinone pool in sulfate respiration. It seems
very plausible that oxidation of menaquinol and electron transfer to APS reductase
by Qmo may lead to a proton gradient across the membrane, but this has to be veri-
fied experimentally. Reduction of menaquinone is likely to occur at least in the first
step of lactate oxidation.
The similar subunit architecture of the Dsr, Hmc and Tmc complexes suggests
their functions may be related. In particular, the similarity between the subunits DsrK,
HmcF and TmcB points to a common (or similar) electron acceptor on the cytoplas-
mic side. This electron acceptor may be a disulfide-containing species that is reduced
to a thiol, which in turn could be an electron donor for the DsrAB sulfite reductase.
This thiol/disulfide could either be a small molecular weight compound, or a thiol
group of a protein. One very likely candidate for this latter case is the DsrC protein
that has two strictly conserved cysteines at the C-terminal in all organisms that have
a dissimilatory sulfite reductase (Cort et al. 2001; Mander et al. 2005). DsrC is present
in all organisms containing a dissimilatory sulfite reductase, irrespective of whether
they are sulfur-compound reducers or oxidizers (Sander et al. 2006). In Desulfovibrio

spp. the DsrAB sulfite reductase forms a stable complex with DsrC (Pierik et al.
1992). DsrC is not cotranscribed with DsrAB and is actually one of the most highly
expressed proteins in D. vulgaris Hildenborough (Haveman et al. 2003).
The similarity between Dsr, Hmc and Tmc complexes does not extend, however,
to the periplasmic cytochrome subunit. As described above, the role of the Hmc and
Tmc complexes is most likely as receptors for electrons resulting from periplasmic
H2, and possibly also formate, oxidation. A proton gradient will be associated with
oxidation of these two compounds as protons are left in the periplasm and electrons
are transferred through the membrane for reduction of sulfate. For now the role of
DsrJ remains as an open question, as does the possibility of electron transfer
between the menaquinone pool and the Dsr, Hmc or Tmc complexes.
In conclusion, despite considerable progress in our understanding of mem-
brane proteins in SRB, many important questions remain, like the precise role of
the membrane complexes, the mechanism(s) of proton translocation and the elec-
tron donors to the terminal reductases. Future research should aim to answer
these questions before we can fully understand the bioenergetics of sulfate
respiration.
Acknowledgements. I would like to thank all my colleagues whose names appear in the
references and in particular Miguel Teixeira and Antnio Xavier, who introduced me to the study
of Desulfovibrio, for many enlightening discussions and for their support over the years. Our work
was funded by Fundao para a Cincia e Tecnologia, MCES, Portugal.
References
Alric J, Tsukatani Y, Yoshida M, Matsuura K, Shimada K, Hienerwadel R, Schoepp-Cothenet B,

Nitschke W, Nagashima KV, Vermeglio A (2004) Structural and functional characterization of
the unusual triheme cytochrome bound to the reaction center of Rhodovulum sulfidophilum. J
Biol Chem 279:2609026097
Bamford VA, Bruno S, Rasmussen T, Appia-Ayme C, Cheesman MR, Berks BC, Hemmings AM
(2002) Structural basis for the oxidation of thiosulfate by a sulfur cycle enzyme. EMBO J
21:55995610
Beller HR, Chain PSG, Letain TE, Chakicherla A, Larimer FW, Richardson PM, Coleman MA,
Wood AP, Kelly DP (2006) The genome sequence of the obligately chemolithoautotrophic,
facultatively anaerobic bacterium Thiobacillus denitfificans. J Bacteriol 188:14731488
Berks BC, Page MD, Richardson DJ, Reilly A, Cavill A, Outen F, Ferguson SJ (1995) Sequence
analysis of subunits of the membrane-bound nitrate reductase from a denitrifying bacterium:
the integral membrane subunit provides a prototype for the dihaem electron-carrying arm of a
redox loop. Mol Microbiol 15:319331
Collins MD, Widdel F (1986) Respiratory quinones of sulfate-reducing and sulfur-reducing
bacteria a systematic investigation. Syst Appl Microbiol 8:818
Cort JR, Mariappan SVS, Kim CY, Park MS, Peat TS, Waldo GS, Terwilliger TC, Kennedy MA
(2001) Solution structure of Pyrobaculum aerophilum DsrC, an archaeal homologue of the
gamma subunit of dissimilatory sulfite reductase. Eur J Biochem 268:58425850
Dahl C, Engels S, Pott-Sperling AS, Schulte A, Sander J, Lubbe Y, Deuster O, Brune DC (2005)
Novel genes of the dsr gene cluster and evidence for close interaction of Dsr proteins during
34 I.A.C. Pereira
sulfur oxidation in the phototrophic sulfur bacterium Allochromatium vinosum. J Bacteriol

187:13921404
Dolla A, Pohorelic BKJ, Voordouw JK, Voordouw G (2000) Deletion of the hmc operon of
Desulfovibrio vulgaris subsp. vulgaris Hildenborough hampers hydrogen metabolism and
low-redox-potential niche establishment. Arch Microbiol 174:143151
Eisen JA, Nelson KE, Paulsen IT, Heidelberg JF, Wu M, Dodson RJ, Deboy R, Gwinn ML, Nelson
WC, Haft DH, Hickey EK, Peterson JD, Durkin AS, Kolonay JL, Yang F, Holt I, Umayam LA,
Mason T, Brenner M, Shea TP, Parksey D, Nierman WC, Feldblyum TV, Hansen CL, Craven
MB, Radune D, Vamathevan J, Khouri H, White O, Gruber TM, Ketchum KA, Venter JC,
Tettelin H, Bryant DA, Fraser CM (2002) The complete genome sequence of Chlorobium
tepidum TLS, a photosynthetic, anaerobic, green-sulfur bacterium. Proc Natl Acad Sci
99:95099514
Elantak L, Dolla A, Durand MC, Bianco P, Guerlesquin F (2005) Role of the tetrahemic subunit
in Desulfovibrio vulgaris Hildenborough formate dehydrogenase. Biochemistry 44:
1482814834
Fitz RM, Cypionka H (1991) Generation of a proton gradient in Desulfovibrio vulgaris. Arch
Microbiol 155:444448
Friedrich CG, Bardischewsky F, Rother D, Quentmeier A, Fischer J (2005) Prokaryotic sulfur
oxidation. Curr Opin Microbiol 8:253259
Haveman SA, Brunelle V, Voordouw JK, Voordouw G, Heidelberg JF, Rabus R (2003) Gene
expression analysis of energy metabolism mutants of Desulfovibrio vulgaris Hildenborough
indicates an important role for alcohol dehydrogenase. J Bacteriol 185:43454353
Haveman SA, Greene EA, Stilwell CP, Voordouw JK, Voordouw G (2004) Physiological and gene
expression analysis of inhibition of Desulfovibrio vulgaris Hildenborough by nitrite.
Hedderich R, Klimmek O, Kroger A, Dirmeier R, Keller M, Stetter KO (1999) Anaerobic
respiration with elemental sulfur and with disulfides. FEMS Microbiol Rev 22:353381
Hedderich R, Hamann N, Bennati M (2005) Heterodisulfide reductase from methanogenic
archaea: a new catalytic role for an iron-sulfur cluster. Biol Chem 386:961970
WC, Sullivan SA, Fouts D, Haft DH, Selengut J, Peterson JD, Davidsen TM, Zafar N, Zhou
LW, Radune D, Dimitrov G, Hance M, Tran K, Khouri H, Gill J, Utterback TR, Feldblyum TV,
Wall JD, Voordouw G, Fraser CM (2004) The genome sequence of the anaerobic, sulfate-reduc-
ing bacterium Desulfovibrio vulgaris Hildenborough. Nat Biotechnol 22:554559
Jormakka M, Byrne B, Iwata S (2003) Protonmotive force generation by a redox loop mechanism.
FEBS Lett 545:2530
Joint Genome Initiative (1997) http://www.jgi.doe.gov. Cited 1 Oct 2006
Klenk HP, Clayton RA, Tomb JF, White O, Nelson KE, Ketchum KA, Dodson RJ, Gwinn M,
Hickey EK, Peterson JD, Richardson DL, Kerlavage AR, Graham DE, Kyrpides NC,
Fleischmann RD, Quackenbush J, Lee NH, Sutton GG, Gill S, Kirkness EF, Dougherty BA,
McKenney K, Adams MD, Loftus B, Venter JC et al. (1997) The complete genome sequence
of the hyperthermophilic, sulphate-reducing archaeon Archaeoglobus fulgidus. Nature
390:364370
Kremer DR, Hansen TA (1989) Demonstration of HOQNO and antimycin-a sensitive coupling of
NADH oxidation and Aps and sulfite reduction in a Marine Desulfovibrio Strain. FEMS
Microbiol Lett 58:4347
Lupton FS, Conrad R, Zeikus JG (1984) Physiological-function of hydrogen metabolism during
growth of sulfidogenic bacteria on organic substrates. J Bacteriol 159:843849
Mander GJ, Duin EC, Linder D, Stetter KO, Hedderich R (2002) Purification and characterization
of a membrane-bound enzyme complex from the sulfate-reducing archaeon Archaeoglobus
fulgidus related to heterodisulfide reductase from methanogenic archaea. Eur J Biochem
269:18951904
Mander GJ, Weiss MS, Hedderich R, Kahnt J, Ermler U, Warkentin E (2005) X-ray structure of
the gamma-submit of a dissimilatory sulfite reductase: fixed and flexible C-terminal arms.
FEBS Lett 579:46004604
Matias PM, Pereira IA, Soares CM, Carrondo MA (2005) Sulphate respiration from hydrogen in
Desulfovibrio bacteria: a structural biology overview. Prog Biophys Mol Biol 89:292329
Mussmann M, Richter M, Lombardot T, Meyerdierks A, Kuever J, Kube M, Glockner FO, Amann R
(2005) Clustered genes related to sulfate respiration in uncultured prokaryotes support the
theory of their concomitant horizontal transfer. J Bacteriol 187:71267137
Odom JM, Peck HD Jr (1981) Hydrogen cycling as a general mechanism for energy coupling in
the sulfate-reducing bacteria, Desulfovibrio sp. FEMS Microbiol Lett 12:4750
Peck HD (1960) Evidence for oxidative phosphorylation during the reduction of sulfate with
hydrogen by Desulfovibrio desulfuricans. J Biol Chem 235:27342738
Pereira IAC, Xavier AV (2005) Multi-heme c cytochromes and enzymes. In: King RB (ed)
Encyclopedia of inorganic chemistry, vol 5, 2nd edn. Wiley, New York, pp 33603376
Pereira IAC, Haveman SA, Voordouw G (2007) Biochemical, genetic and genomic characteriza-
tion of anaerobic electron transport pathways in sulphate-reducing delta-proteobacteria. In:
Barton LL, Hamilton WA (eds) Sulphate-reducing bacteria: environmental and engineered
systems. Cambridge University Press, Cambridge (in press)
Pereira PM, Teixeira M, Xavier AV, Louro RO, Pereira IA (2006) The Tmc complex from
Desulfovibrio vulgaris Hildenborough is involved in transmembrane electron transfer
from periplasmic hydrogen oxidation. Biochemistry 45:1035910367
Pierik AJ, Duyvis MG, van Helvoort JM, Wolbert RB, Hagen WR (1992) The third subunit of
desulfoviridin-type dissimilatory sulfite reductases. Eur J Biochem 205:111115
Pires RH, Lourenco AI, Morais F, Teixeira M, Xavier AV, Saraiva LM, Pereira IA (2003) A novel
membrane-bound respiratory complex from Desulfovibrio desulfuricans ATCC 27774.
Biochim Biophys Acta 1605:6782
Pires RH, Venceslau SS, Morais F, Teixeira M, Xavier AV, Pereira IAC (2006) Characterization
of the Desulfovibrio desulfuricans ATCC 27774 DsrMKJOP complex a membrane-bound
redox complex involved in sulfate respiration. Biochemistry 45:249262
Pott AS, Dahl C (1998) Sirohaem sulfite reductase and other proteins encoded by genes at the dsr
locus of Chromatium vinosum are involved in the oxidation of intracellular sulfur. Microbiology
144:18811894
Becker I, Amann J, Gellner K, Teeling H, Leuschner WD, Glockner FO, Lupas AN, Amann R,
Klenk HP (2004) The genome of Desulfotalea psychrophila, a sulfate-reducing bacterium
from permanently cold Arctic sediments. Environ Microbiol 6:887902
Rossi M, Pollock WB, Reij MW, Keon RG, Fu R, Voordouw G (1993) The hmc operon of
Desulfovibrio vulgaris subsp. vulgaris Hildenborough encodes a potential transmembrane
redox protein complex. J Bacteriol 175:46994711
Sander J, Engels-Schwarzlose S, Dahl C (2006) Importance of the DsrMKJOP complex for sulfur
oxidation in Allochromatium vinosum and phylogenetic analysis of related complexes in other
prokaryotes. Arch Microbiol 186:357366
Saraiva LM, da Costa PN, Conte C, Xavier AV, LeGall J (2001) In the facultative sulphate/nitrate
reducer Desulfovibrio desulfuricans ATCC 27774, the nine-haem cytochrome c is part of a
membrane-bound redox complex mainly expressed in sulphate-grown cells. Biochim Biophys
Acta 1520:6370
Stojanowic A, Mander GJ, Duin EC, Hedderich R (2003) Physiological role of the F-420-non-
reducing hydrogenase (Mvh) from Methanothermobacter marburgensis. Arch Microbiol
180:194203
Van Driessche G, Devreese B, Fitch JC, Meyer TE, Cusanovich MA, Van Beeumen JJ (2006)
GHP, a new c-type green heme protein from Halochromatium salexigens and other
proteobacteria. FEBS J 273:28012811
Wood PM (1978) Chemiosmotic model for sulfate respiration. FEBS Lett 95:1218
Chapter 4
Biochemical and Evolutionary Aspects of
Eukaryotes That Inhabit Sulfidic Environments
Ursula Theissen, William Martin
Abstract Various eukaryotes inhabit environments that harbor high concentrations

of sulfide, which is a potent inhibitor of complex IV in the mitochondrial respiratory
chain. They must therefore posses means by which they can detoxify sulfide, or use
alternative electron routes that circumvent oxygen as the terminal acceptor, or
both. The biochemical mechanisms through which eukaryotes deal with sulfide are
beginning to come into focus, with sulfide:quinone oxidoreductase and the energy
metabolism germane to anaerobic mitochondria standing in the foreground. This
chapter briefly covers current progress in understanding the biochemistry of sulfide
detoxification and utilization by eukaryotes. In light of newer views of ocean geo-
chemistry (Canfield oceans), both the anaerobic biochemistry of mitochondria
and their capacity to deal with sulfide are most easily interpreted as evolutionary
holdovers from the anoxic and sulfidic phase of ocean history between about 2.3 billion
and about 0.6 billion years ago.
4.1 Introduction
For prokaryotes, it is well known that sulfide is a rich and widely used energy
source. Many bacteria can survive with sulfide as their only electron source
(Kelly et al. 1997; Brune 1995). In eubacteria, sulfide is usually oxidized by
the enzymes flavocytochrome c (van Beeumen et al. 1991) and sulfide:quinone
oxidoreductase (SQR). Bacterial SQR has been characterized in detail (Reinartz
et al. 1998; Griesbeck et al. 2002) and an enzymatic mechanism has been pro-
posed (Griesbeck et al. 2002). The enzyme catalyzes the transfer of electrons
from sulfide to quinones as their entry point into the photosynthetic or respira-
tory membrane.
But prokaryotes are not the only inhabitants of sulfidic environments; many
eukaryotes inhabit sulfidic environments as well. Traditionally, sulfide is viewed as
an environmental toxin for eukaryotes, rather than as an energy source, because
sulfide is a potent toxin that has long been known to inhibit complex IV of the
36
Springer 2008
4 Biochemical and Evolutionary Aspects of Eukaryotes 37
mitochondrial respiratory chain (Nicholls 1975). The purpose of this chapter is to

point out some examples of animals that inhabit sulfide-rich environments, to
summarize the means with which they deal with sulfide, and to consider some of
the evolutionary implications of sulfide metabolism in eukaryotes. The physiology
of animals that inhabit sulfidic environments has been reviewed by Grieshaber and
Vlkel (1998).
4.2 Animals in Sulfidic Environments
Animals can be confronted with sulfide from biological or geological sources.

Some animals, like the giant tubeworm Riftia pachyptila, live at deep-sea hydro-
thermal vents, the exhalate of which can contain more than 0.3 mM sulfide (van
Dover 2000). A far more common, and less spectacular, kind of sulfidic environ-
ment inhabited by animals is marine sediment. The sulfide in marine sediments
is generally of biogenic origin, stemming mainly from sulfate-reducing prokaryotes
(Trper 1984). Various invertebrates, for example, the lugworm Arenicola
marina or the ribbed mussel Geukensia demissa, inhabit such sulfidic marine
sediments, where sulfide concentrations can reach 3 M to 2 mM (Fenchel and
Riedl 1970; Vlkel and Grieshaber 1992; Vlkel et al. 1995), in some cases up to
8 mM (Lee et al. 1996).
Animals that live in sulfidic environments must possess strategies to avoid
sulfide-poisoning and/or or to utilize sulfide outright. Several evolutionary strate-
gies are apparent. Many species of mussels have opted for what might be the
simplest solution to short-term sulfide exposure: they simply close their shells.
Animals lacking protective shells, like worms, must find different solutions,
because biological membranes are permeable for sulfide (Beerman 1924). The
oligochaete Tubificoides benedii precipitates FeS in the outer mucus layer, which
leads to the black color of the animals (Dubilier et al. 1995). Other animals harbor
ectosymbiotic bacteria on their body surface that afford protection by oxidizing
sulfide as an energy source. This has been described for various worms, including
the priapulid Halicryptus spinulosus (Oeschger and Schmaljohann 1988; Oeschger
and Janssen 1991).
Animals that must endure high concentrations or long-term exposure to sulfide
require biochemical means of dealing with sulfide. The giant gutless tubeworm
Riftia pachyptila at hydrothermal vents hosts sulfide-oxidizing bacterial symbionts
(Dubilier et al. 2001). The animals have not only to tolerate sulfide, but they also
have to transport sulfide and oxygen to their symbionts, chemoautotrophs that
supply the wormss reduced carbon, thereby functionally replacing the digestive
tract. Many such worms have specialized blood proteins suited to this lifestyle. For
Riftia, two different extracellular hemoglobins have been described that bind not
only oxygen, but sulfide in addition, simultaneously and reversibly (Arp et al. 1985).
The sulfide binding occurs via conserved cysteine residues and disulfide groups
(Zal et al. 1998; Bailly et al. 2002).
38 U. Theissen, W. Martin
Other invertebrates like the lugworm Arenicola marina and the mussel
Geukensia demissa inhabit sulfidic environments (marine sediments) but they
do not host intracellular endosymbionts. Both species possess their own bio-
chemical means of sulfide oxidation, the main product of which is thiosulfate.
This has been shown for Arenicola (Vlkel and Grieshaber 1992) and Geukensia
(Doeller et al. 2001) as well as for the bivalve Solemya reidi (OBrien and
Vetter 1990).
In the case of the ribbed mussel Geukensia demissa, sulfide oxidation takes
place in the gill mitochondria and is directly linked to ATP synthesis: sulfide-supported
oxygen consumption that matches the energy demand of ciliary beating (Doeller
et al. 2001). These mitochondria are thus chemolithoheterotrophic, since the
electrons for ATP synthesis via the respiratory chain stem from an inorganic
donor. For the lugworm, Vlkel and Grieshaber (1994) showed that the oxidation
of sulfide to thiosulfate takes place in mitochondria. They proposed a model for
the mitochondrial respiratory chain involving an SQR similar to the bacterial
enzyme (Vlkel and Grieshaber 1996, 1997). If sulfide concentrations do not
exceed 30 M, at which concentration mitochondrial cytochrome c oxidase is
inhibited, the electrons from sulfide are thought to be transferred to ubiquinone
by the SQR-like enzyme and then used for oxygen-dependent ATP production
(Vlkel and Grieshaber 1997).
4.3 Sulfide-Oxidizing Enzymes in Eukaryotes
Beyond the well-documented circumstances that sulfide is consumed and thiosulfate

is produced, and that mitochondria are involved, comparatively little is known
about the molecular details or biochemical mechanisms by which eukaryotes
oxidize environmental sulfide. Vande Weghe and Ow (1999) identified a gene
from the fission yeast Schizosaccharomyces pombe in a screen for heavy-metal
tolerance, hmt2. The S. pombe HMT2 gene product showed sequence similarity
to the biochemically characterized SQR from the alphaproteobacterium
Rhodobacter capsulatus (Schtz et al. 1997) and furthermore catalyzed electron
transfer from sulfide to quinone, suggesting that HMT2 is, functionally, an
SQR enzyme. But the affinities of the S. pombe HMT2 protein for both sub-
strates were extremely low (in the millimolar range), raising doubts about the
in vivo function of the S. pombe protein (Griesbeck et al. 2002). Vande Weghe
and Ow (1999) reported that SQR homologues occur in a few animal genomes.
A closer look into the databases revealed SQR homologs in virtually all
sequenced genomes from animals, and many unicellular eukaryotes as well
(Theissen et al. 2003).
The widespread occurrence of SQR-related genes in eukaryotes poses two
questions: What are the products of these eukaryotic genes actually doing? Does
eukaryotic sulfide metabolism make any sense in an evolutionary context?
4.4 The Possible Functions of SQR-Related

Genes in Eukaryotes
On the one hand, the occurrence of SQR homologues in eukaryotes such as

humans or fruit flies, neither of which inhabit sulfidic environments, would
appear as a bit of a puzzle. But on the other hand, it places some older biochemical
findings in a new light. In 1972, rat mitochondria were reported to oxidize sulfide
(Curtis et al. 1972; Bartholomew et al. 1980). Fish mitochondria have also been
shown to oxidize sulfide (Bagarinao and Vetter 1990). More recently, isolated
chicken mitochondria were shown to oxidize sulfide, and the sulfide oxidation
was coupled with ATP synthesis (Yong and Searcy 2001), similar to the case
mentioned earlier for Geukensia (Doeller et al. 2001). Are SQR homologues
doing the job?
As a BLAST search with the S. pombe SQR sequence (accession number NP_
596067) will quickly reveal, SQR homologues occur in genomes of all the eukaryotic
lineages in question. We have isolated a complementary DNA encoding an SQR
homologue from Arenicola marina (Theissen 2006) and expressed it in bakers
yeast, which lacks an SQR homologue in the genome sequence. The Arenicola
enzyme expressed in yeast catalyzes the electron transfer from sulfide to ubiquinone,
and with considerably better substrate affinities than reported for the S. pombe pro-
tein, but it does not produce thiosulfate directly; rather, additional enzymes appear
to be necessary in the pathway that leads to the formation of thiosulfate from sulfide
in the lugworm (Theissen 2006). Reports of enzymatic activity for eukaryotic SQR
homologues other than the initial report on S. pombe (Vande Weghe and Ow 1999)
are apparently lacking in the literature so far.
Sulfide has been discussed as an atypical neuromodulator, in addition to the
gases NO and CO (Baranano et al. 2001); hence, one possible role for SQR in
eukaryotes that do not inhabit sulfidic environments might involve the modulation
of physiological responses. Endogenous sulfide production has been described not only
for marine invertebrates like Arenicola marina and the mussel Tapes philippinarum
(Julian et al. 2002) that deal with high sulfide environmental concentrations daily, but
also for various mammals, which do not (Goodwin et al. 1989; Warenycia et al. 1989;
Savage and Gould 1990).
Starting from l-cysteine, eukaryotes can synthesize endogenous sulfide in at least
four different ways (reviewed in Kamoun 2004). In mitochondria, cysteine ami-
notransferase (EC 2.6.1.3) and 3-mercaptopyruvate sulfurtransferase (EC 2.8.1.2) can
be involved in sulfide production (reviewed in Kamoun 2004). Cysteine
aminotransferase catalyzes the reaction of l-cysteine with a ketoacid (e.g., -ketoglu-
tarate) with formation of 3-mercaptopyruvate and an amino acid (e.g., l-glutamate).
3-Mercaptopyruvate is desulfurated by 3-mercaptopyruvate sulfurtransferase, result-
ing in formation of sulfide and pyruvate (Julian et al. 2002). In the cytosol, sulfide
can be generated by cystathione -synthase (CBS; EC 4.2.1.22). Alongside
endogenous sulfide production in mammals, considerable amounts of sulfide can be
produced by anaerobic sulfate-reducing bacteria in the human colon, posing a chal-

lenge to cells of the intestinal epithelium (MacFarlane et al. 1992).
Such findings suggest that even such animals that are not exposed to environmental
sulfide nonetheless require biochemical means of dealing with sulfide, albeit at
lower concentrations than those experienced by sulfide-exposed marine inverte-
brates. Failure to deal with endogenous sulfide can have dire consequences in
humans. For example, overproduction of sulfide owing to enhanced CBS activity
can exacerbate cognitive effects in Down-syndrome patients (Chadefaux et al.
1985; Kamoun 2001) and insufficient detoxification of sulfide produced in the
human colon can lead to inflammatory diseases and might affect the frequency of
colon cancer (Pitcher et al. 2000). Whether mammalian SQR plays a significant
physiological role in sulfide metabolism, or not, remains to be shown.
4.5 Sulfide and Eukaryotic Evolution
Thoughts about eukaryotic evolution are traditionally couched in the context of

oxygen, and very rarely have anything to do with sulfide whatsoever. However,
that view needs to change for reasons that we will briefly outline here. For several
decades, the popular opinion about eukaryote evolution has been that the earliest
eukaryotes were anaerobic, fermenting amoebae, that the origin of mitochondria
corresponded to the origin of the aerobic, respiring lifestyle among eukaryotes, and
that ATP yield from glucose was the prime advantage conferred by the mitochondrial
endosymbiont. This notion is as old as all modern formulations of endosymbiotic
theory. Margulis (p. 229 in Sagan 1967), for example, wrote: The anaerobic break-
down of glucose to pyruvate along the EmbdenMeyerhof pathway occurred in the
soluble cytoplasm under the direction of the host genome.The greater amounts
of energy available after the incorporation of the mitochondrion resulted in large
cells with amoebiod and cyclotic movement. That view meshed well with the
geological view that was emerging at about the same time, namely, that the origin
of eukaryotes (and their mitochondria) corresponded temporally and causally to the
global rise in atmospheric oxygen levels about two billion years ago. Quoting
Margulis again (p. 225 in Sagan 1967): The subsequent evolution of aerobic
metabolism in prokaryotes to form aerobic bacteria (protoflagella and protomito-
chondria) presumably occurred during the transition to the oxidizing atmosphere.
Seen from todays standpoint, quite a few things have changed since 1967, both
from the biological and the geological perspective. (Curiously, however, the
popular opinion about the context of mitochondrial origin has remained largely
unchanged.) From the biological perspective it is now clear that eukaryotic anaer-
obes are not restricted to any kind of early branching lineages. Rather, anaerobic
eukaryotes occur across many independent lineages spanning the breadth and depth
of eukaryote evolution (Embley and Martin 2006). It is furthermore clear that those
eukaryotic lineages that were once thought to be the most primitive and were also
thought to lack mitochondria have mitochondria after all, but anaerobic forms of
the organelle called hydrogenosomes and mitosomes (Mller 1993, 2003; Tovar
et al. 1999, 2003; van der Giezen and Tovar 2005; van der Giezen et al. 2005;
Embley and Martin 2006; Martin and Mller 2007).
From the geological perspective, things have changed since 1967 as well,
perhaps even more dramatically. The view that the appearance of oxygen in the
atmosphere at about two billion years before the present corresponded to some sort
of oxygen catastrophe for the entire planet is no longer current among geologists
studying the history of oxygen on Earth. Instead, a newer model is now current that
is often designated as intermediate oxidation state or Canfield ocean (Canfield
1998; Canfield et al. 2000; Anbar and Knoll 2002; Shen et al. 2003; Poulton et al.
2004; Arnold et al. 2004; Brocks et al. 2005). Summarized briefly, this newer view
of Canfield oceans suggests that during the time from the appearance of oxygen in
the atmosphere at about 2.3 billion years ago up until about 0.6 billion years ago
marine sulfate reduction was globally widespread in the oceans, leading to anoxic
and highly sulfidic water below the photic zone. The most recent evidence to
support this view comes from geological findings that suggest the appearance of
the Ediacara fauna (the earliest metazoan fossils) to correspond with the completion
of ocean oxygenation and the end of anoxic and sulfidic (Canfield) oceans about
550 million years ago (Fike et al. 2006; Canfield et al. 2007).
That newer view of oxygen history on Earth (Canfield oceans) has major and
far-reaching consequences for our understanding of early eukaryote evolution,
although the community of biologists is not awakening to this realization as rapidly
as it probably should. How might Canfield oceans affect the views of biologists
concerning the course of evolution during the last about two billion years? Three
main points are of importance.
First, and perhaps foremost in the context of this volume, the biochemistry of
sulfur metabolism would move to center stage for understanding ecosystems and
their inhabitants during the last two billion years of evolution. Put another way, the
global significance and chemical impact of sulfate reducers (and other microbes
that depend on redox reactions involving sulfur) and their main end product
sulfide in marine environments could be seen on a level comparable to that
traditionally attached to oxygen production by cyanobacteria. Biologists have
always made a big fuss about the difference between anaerobic and aerobic habitats
in evolution; if we trust the geologists (as we probably should) it would appear that
the difference between sulfidic and non-sulfidic habitats during Earth history might
be just as big. Atmospheric oxygen is one thing, but the brunt of evolution during
the time from 2.3 billion to 0.6 billion years ago was going on in the oceans, not in
the atmosphere. Biologists would probably do well to let the message of Canfield
oceans sink into their thinking about biochemistry and evolution during that time.
Second, with evidence continuing to pour in about Canfield oceans (Canfield
1998; Canfield et al. 2000; Anbar and Knoll 2002; Shen et al. 2003; Poulton et al.
2004; Arnold et al. 2004; Brocks et al. 2005), geologists are telling us in no uncer-
tain terms that the oceans were anoxic and sulfidic during the time from 2.3 billion
to 0.6 billion years ago, the time during which the major eukaryotic lineages were
emerging and diversifying. Hence, the widespread occurrence of SQR among
eukaryotes should hardly be surprising, because eukaryotes were born and raised
in sulfidic marine environments (Theissen et al. 2003; Martin et al. 2003; Embley
and Martin 2006). Eukaryotic sulfide metabolism (also among eukaryotes that do
not inhabit sulfidic environments today) is thus easily seen as a holdover from
anaerobic and sulfidic times and is easily understood in that context.
Third, the widespread occurrence of the anaerobic lifestyle among eukaryotes
would no longer need to be seen as a secondary adaptation, but as a direct holdover
from the not-too-distant anaerobic (and/or microaerophilic) past of the eukaryotic
lineage. Indeed, biologists still tend to equate the concept of possessing mitochon-
dria with oxygen. But mitochondria now appear to be ubiquitous among all
eukaryotes, including the anaerobic forms (Embley and Martin 2006; Tovar et al.
2006), which do not require oxygen. The mitochondria of eukaryotic anaerobes fall
into basically three types. The first type are those that possess quinones and pro-
duce ATP via anaerobic respirations such as the succinate-producing mitochondria
of many worms (van Hellemond et al. 1995; Tielens et al. 2002; Tielens and van
Hellemond 2007), and the denitrifying mitochondria of some benthic forams
(Risgaard-Petersen et al. 2006) or the succinate-producing mitochondria of many
marine invertebrates (Grieshaber and Vlkel 1998). Notably, eukaryotes that pro-
duce succinate as a major end product very often contain rhodoquinone as well
because it is needed for the fumarate reductase reaction (Tielens et al. 2002; van
Hellemond et al. 2003). The second type are hydrogenosomes, which produce ATP
(and molecular hydrogen) but lack cytochromes (Mller 1993, 2003; Martin and
Mller 1998; Embley and Martin 2006; Mller and Martin 2007). The third type
are mitosomes, which apparently do not produce ATP at all but still fulfill some
important biochemical functions for the cell (van der Giezen et al. 2005; van der
Giezen and Tovar 2005). Many biologists would still like to view eukaryotes as
ancestrally oxygen-dependent organisms, but in light of Canfield oceans, it would
seem far more reasonable to view the ability to produce ATP without the help of
molecular oxygen as an attribute that was present in the eukaryote common
ancestor.
4.6 Conclusion
Various eukaryotes inhabit sulfidic environments. Some eukaryotes can use sulfide
as an electron donor for ATP synthesis in their mitochondrial respiratory chain. The
enzyme that oxidizes sulfide in eukaryotes, SQR, is a mitochondrial enzyme, but in
contrast to the situation in prokaryotes, the nature of the oxidized sulfur product in
the eukaryotic SQR reaction is not yet known. Geologists are telling us that the
oceans were anaerobic and sulfidic during the time from about 2.3 billion to about
0.6 billion years ago. Eukaryotes arose and underwent their early diversification in
a global ecological setting dominated by anaerobic and sulfidic environments. It is
therefore not surprising to see the biochemical traces of that anaerobic and sulfidic
past preserved in the biochemistry of modern eukaryotic groups.
References
Anbar AD, Knoll AH (2002) Proterozoic ocean chemistry and evolution: a bioinorganic bridge.
Science 297:11371142
Arnold GL, Anbar AD, Barling J, Lyons TW (2004) Molybdenum isotope evidence for wide-
spread anoxia in mid-Proterozoic oceans. Science 304:8790
Arp AJ, Childress JJ, Fisher CR (1985) Blood gas transport in Riftia pachyptila. Biol Soc Wash
Bull 6:289300
Bagarinao T, Vetter RD (1990) Oxidative detoxification of sulfide by mitochondria of the
California killifish Fundulus parvipinnis and the speckled sanddab Citharichthys stigmaeus.
J Comp Physiol 160B:519527
Bailly X, Jollivet D, Vanin S, Deutsch J, Zal F, Lallier F, Toulmond A (2002) Evolution of the
sulfide-binding function within the globin multigenic family of the deep-sea hydrothermal
vent tubeworm Riftia pachyptila. Mol Biol Evol 19:14211433
Baranano DE, Ferris CD, Snyder SH (2001) Atypical neural messengers. Trends Neurosci 24:99106
Bartholomew TC, Powell GM, Dodgson KS, Curtis CG (1980) Oxidation of sodium sulfide by
rat-liver, lungs and kidney. Biochem Pharmacol 29:24312437
Beerman H (1924) Some physiological actions of hydrogen sulfide. J Exp Zool 41:3343
Brocks JJ, Love GD, Summons RE, Knoll AH, Logan GA, Bowden SA (2005) Biomarker evi-
dence for green and purple sulphur bacteria in a stratified Palaeoproterozoic sea. Nature
437:866870
Brune DC (1995) Sulfur compounds as photosynthetic electron donors. In: Blankenship RE, Madigan
MT, Bauer CE (eds) Anoxygenic photosynthetic bacteria. Kluwer, Dordrecht, pp 847870
Canfield DE (1998) A new model for Proterozoic ocean chemistry. Nature 396:450453
Canfield DE, Habicht KS, Thamdrup B (2000) The Archean sulfur cycle and the early history of
atmospheric oxygen. Science 288:658661
Canfield DE, Poulton SW, Narbonne GM (2007) Late-Neoproterozoic deep-ocean oxygenation
and the rise of animal life. Science 315:9295. doi:10.1126/science.1135013
Chadefaux B, Rethore MO, Raoul O, Ceballos I, Gilgenkrantz S, Allard D (1985) Cystathionine
beta synthase gene dosage effect in trisomy 21. Biochem Biophys Res Commun 128:4044
Curtis CG, Bartholomew TC, Rose FA, Dodgson KS (1972) Detoxification of sodium 35S sulfide
in rat. Biochem Pharmacol 21:23132321
Doeller JE, Grieshaber MK, Kraus DW (2001) Chemolithoheterotrophy in a metazoan tissue:
thiosulfate production matches ATP demand in ciliated mussel gills. J Exp Biol 204:
37553764
Dubilier N, Giere O, Grieshaber MK (1995) Morphological and ecophysiological adaptations of
the marine oligochaete Tubificoides benedii to sulfidic sediments. Am Zool 35:163173
Dubilier N, Mlders N, Ferdelman T, de Beer D, Pernthaler A, Klein M, Wagner M, Ersus C,
Thierman F, Krieger J, Giere O, Amann R (2001) Endosymbiotic sulphate-reducing and
sulphide-oxidizing bacteria in an oligochaete worm. Nature 411:298302
Embley TM, Martin W (2006) Eukaryotic evolution, changes and challenges. Nature 440:623630
Fenchel TM, Riedl (1970) The sulfide system: a new biotic community underneath the oxidized
layer of marine sand bottoms. Mar Biol 7:255268
Fike DA, Grotzinger JP, Pratt LM, Summons RE (2006) Oxidation of the Ediacaran ocean. Nature
444:744747
Goodwin LR, Francom D, Dieken FP, Taylor JD, Warenycia NW, Reiffenstein RJ, Dowling G
(1989) Determination of sulfide in brain tissue by gas dialysis/ion chromatography: post-mortem
studies and two case reports. J Anal Toxicol 13:105109
Griesbeck C, Schtz M, Schdl T, Bathe S, Nausch L, Mederer N, Vielreicher M, Hauska G
(2002) Mechanism of sulfide-quinone reductase investigated using site-directed mutagenesis
and sulfur analysis. Biochemistry 41:1155211565
Grieshaber MK, Vlkel S (1998) Animal adaptations for tolerance and exploitation of poisonous
sulfide. Annu Rev Physiol 60:3053
Julian D, Statile JL, Wohlgemuth SE, Arp AJ (2002) Enzymatic hydrogen sulfide production in
marine invertebrate tissues. Comp Biochem Physiol A 133:105115
Kamoun P (2001) Mental retardation in Down syndrome: a hydrogen sulfide hypothesis. Med
Hypotheses 57:389392
Kamoun P (2004) Endogenous production of hydrogen sulfide in mammals. Amino Acids
26:243254
Kelly DP, Shergill JK, Lu WP, Wood AP (1997) Oxidative metabolism of inorganic sulfur com-
pounds by bacteria. Antonie Van Leeuwenhoek 71:95107
Lee RW, Kraus D, Doeller JE (1996) Sulfide-stimulation of oxygen consumption rate and cyto-
chrome reduction in gills of the estuarine mussel Geukensia demissa. Biol Bull 191:421430
MacFarlane GT, Gibson GR, Cummings JH (1992) Comparison of fermentation reactions in dif-
ferent regions of the human colon. J Appl Bacteriol 72:5764
Martin W, Mller M (1998) The hydrogen hypothesis for the first eukaryote. Nature 392:3741
Martin W, Mller M (eds) (2007) Origin of mitochondria and hydrogenosomes. Springer, Heidelberg
Martin W, Rotte C, Hoffmeister M, Theissen U, Gelius-Dietrich G, Ahr S, Henze K (2003) Early
cell evolution, eukaryotes, anoxia, sulfide, oxygen, fungi first (?), and a tree of genomes revis-
ited. IUBMB Life 55:193204
Mller M (1993) The hydrogenosome. J Gen Microbiol 139:28792889
Mller M (2003) Energy metabolism. Part I: anaerobic protozoa. In: Marr J (ed) Molecular
medical parasitology. Academic, London, pp 125139
Nicholls P (1975) The effect of sulfide on cytochrome aa3. Isosteric and allosteric shifts of the
reduced?-peak. Biochim Biophys Acta 396:2435
OBrien J, Vetter RD (1990) Production of thiosulfate during sulphide oxidation by mitochondria
of the symbiont-containing bivalve Solemya reidi. J Exp Biol 149:133148
Oeschger R, Janssen HH (1991) Histological studies on Halicryptus spinulosus (Priapulida) with
regard to environmental hydrogen sulfide resistance. Hydrobiologia 222:112
Oeschger R, Schmaljohann R (1988) Association of various types of epibacteria with Halicryptus
spinulosus (Priapulida). Mar Ecol Prog Ser 48:285293
Pitcher MCL, Beatty ER, Cummings JH (2000) The contribution of sulphate reducing bacteria and
5-aminosalicylic acid to faecal sulphide in patients with ulcerative colitis. Gut 46:6472
Poulton SW, Fralick PW, Canfield DE (2004) The transition to a sulphidic ocean 1.84 billion
years ago. Nature 431:173177
Reinartz M, Tschpe T, Brser T, Trper HG, Dahl C (1998) Sulfide oxidation in the phototrophic
bacterium Chromatium vinosum. Arch Microbiol 170:5968
Risgaard-Petersen N, Langezaal AM, Ingvardsen S, Schmid MC, Jetten MS, Op den Camp HJ,
Derksen JW, Pina-Ochoa E, Eriksson SP, Nielsen LP, Revsbech NP, Cedhagen T, van der
Zwaan GJ (2006) Evidence for complete denitrification in a benthic foraminifer. Nature
443:9396
Sagan L (1967) On the origin of mitosing cells. J Theor Biol 14:225274
Savage JC, Gould DH (1990) Determination of sulfides in brain tissue and rumen fluid by ion-inter-
action reversed-phase high-performance liquid chromatography. J Chromatogr 526:540545
Schtz M, Shahak Y, Padan E, Hauska G (1997) Sulfide quinone reductase from Rhodobacter
capsulatus. J Biol Chem 272:98909894
Shen Y, Knoll AH, Walter MR (2003) Evidence for low sulphate and anoxia in a mid-Proterozoic
marine basin. Nature 423:632635
Theissen U (2006) Die Sulfid:Chinon Oxidoreduktase des Wattwurms Arenicola marina:
Funktion, Mechanismus und Evolution. Dissertation, University of Dsseldorf
Theissen U, Hoffmeister M, Grieshaber M, Martin W (2003) Single eubacterial origin of
eukaryotic Sulfide:quinone oxidoreductase, a mitochondrial enzyme conserved from the
early evolution of eukaryotes during anoxic and sulfidic times. Mol Biol Evol 20:15641574
Tielens AGM, van Hellemond JJ (2007) Anaerobic mitochondria: properties and origins. In:
Martin W, Mller M (eds) Origin of mitochondria and hydrogenosomes. Springer, Heidelberg,
pp 85103
Tielens AGM, Rotte C, van Hellemond JJ, Martin W (2002) Mitochondria as we dont know them.
Trends Biochem Sci 27:564572
Tovar J, Fischer A, Clark CG (1999) The mitosome, a novel organelle related to mitochondria in
the amitochondrial parasite Entamoeba histolytica. Mol Microbiol 32:10131021
Tovar J, Len-Avila G, Snchez LB, Sutak R, Tachezy J, van der Giezen M, Hernndez M, Mller M,
Lucocq JM (2003) Mitochondrial remnant organelles of Giardia function in iron-sulphur
protein maturation. Nature 426:172176
Trper HG (1984) Microorganisms and the sulfur cycle: In: Mller A, Krebs B (eds) Sulfur: Its
significance for chemistry, for the geo-, bio-, and cosmosphere and technology. Studies in
inorganic chemistry. Elsevier, Amsterdam, pp 351365
van der Giezen M, Tovar J (2005) Degenerate mitochondria. EMBO Rep 6:525530
van der Giezen M, Tovar J, Clark CG (2005) Mitochondrion-derived organelles in protists and
fungi. Int Rev Cytol 244:175225
van Dover CL (2000) The ecology of deep-sea hydrothermal vents. Princeton University Press,
Princeton
van Hellemond JJ, Klockiewicz M, Gaasenbeek CPH, Roos MH, Tielens AGM (1995)
Rhodoquinon and complex II of the electron transport chain in anaerobically functioning
eukaryotes. J Biol Chem 270:3106531070
van Hellemond JJ, van der Klei A, van Weelden SW, Tielens AG (2003) Biochemical and evolu-
tionary aspects of anaerobically functioning mitochondria. Philos Trans R Soc Lond B Biol
Sci 358:205213 (not cited in the text!)
van Beeumen JJ, Demol H, Samyn B, Bartsch RG, Meyer TE, Dolata MM, Cusanovich MA
(1991) Covalent structure of the diheme cytochrome subunit and amino-terminal sequence of
the flavoprotein subunit of flavocytochrome c from Chromatium vinosum. J Biol Chem
266:1292112931
Vande Weghe JG, Ow DW (1999) A fission yeast gene for mitochondrial sulfide oxidation. J Biol
Chem 274:1325013257
Vlkel S, Grieshaber MK (1992) Mechanisms of sulfide tolerance in the peanut worm Sipunculus
nudus (Sipunculida) and in the lugworm Arenicola marina (Polychaeta). J Comp Physiol B
162:469477
Vlkel S, Grieshaber MK (1994) Oxygen-dependent sulfide detoxification in the lugworm
Arenicola marina. Mar Biol 118:137147
Vlkel S, Grieshaber MK (1996) Mitochondrial sulfide oxidation in Arenicola marina: Evidence
for alternative electron pathways. Eur J Biochem 235:231237
Vlkel S, Grieshaber MK (1997) Sulphide oxidation and oxidative phosphorylation in the mito-
chondria of the lugworm Arenicola marina. J Exp Biol 200:8392
Vlkel S, Hauschild K, Grieshaber MK (1995) Sulfide stress and tolerance in the lugworm
Arenicola marina during low tide. Mar Ecol Prog Ser 122:205215
Warenycia MW, Goodwin LR, Benishin CG, Reiffenstein RJ, Francom DM, Taylor JD, Dicken FP
(1989) Acute hydrogen sulfide poisoning: demonstration of selective uptake of sulfide by the
brainstem by measurement of brain sulfide levels. Biochem Pharmacol 38:973981
Yong R, Searcy DG (2001) Sulfide oxidation coupled to ATP synthesis in chicken liver mitochon-
dria. Comp Biochem Physiol B 129:129137
Zal F, Leize E, Lallier FH, Toulmond A, Van Dorsselaer A, Childress JJ (1998) S-sulfohemoglobin
and disulfide exchange: the mechanisms of sulfide binding by Riftia pachyptila hemoglobins.
Proc Nat Acad Sci 95:89979002
Chapter 5
Evolution and Ecology of Microbes
Dissimilating Sulfur Compounds: Insights
from Siroheme Sulfite Reductases
Alexander Loy, Stephan Duller, Michael Wagner
Abstract Sulfur microorganisms have been thriving on Earth since the dawn of
life and are still of central importance for the functioning of modern ecosystems.
Here, we summarize the current perception of the evolution of dissimilatory siro-
heme sulfite reductases (DSRs), antique key enzymes in the energy metabolism of
sulfur microbes. We further give recent examples of the diversity and ecology of
uncultured sulfur-dissimilating microorganisms; unprecedented insights that were
only made possible by exploiting DSR-encoding genes as molecular markers in
environmental surveys.
5.1 Introduction
Some of the first microorganisms under the anoxic, reduced atmosphere of the
primordial Earth gained energy for growth and maintenance of cellular processes
by dissimilating sulfur compounds (Canfield and Raiswell 1999; Huston and Logan
2004). Today, phylogenetically distinct bacteria and archaea still have the unifying
ability to employ sulfur compounds as either electron donors or acceptors for
energy-generating redox reactions. In the environment they are thus pivotal for the
biogeochemical cycling of sulfur, but also of carbon, as sulfur oxidation/reduction
in these microbes is coupled to carbon dioxide assimilation or heterotrophic break-
down of organic matter. A presumably ancient group of enzymes, that could have
played a fundamental role in mediating biological conversions of sulfur compounds
from the very first appearance of these microorganisms on Earth, are sulfite reduct-
ases (SRs), catalyzing the six-electron reduction of sulfite to sulfide. The electron
transfer is mediated by a metallocofactor that is composed of a metalloporphyrin,
the so-called siroheme, bound to an ironsulfur [Fe4S4] cluster in the active redox
centers of the protein (Crane and Getzoff 1996; Crane et al. 1995; Siegel et al.
1978). Apart from the evolutionarily related ammonia-forming assimilatory nitrite
reductases, SRs are the only proteins known to contain siroheme, a reduced iron
tetrahydroporphyrin of the isobacteriochlorin class, or siroamide, an amidated
siroheme variant, as a prosthetic group (Matthews et al. 1995). Because siroheme
is vital for the catalytic activity of the enzyme, a typical siroheme[Fe4S4] binding
46
Springer 2008
5 Evolution and Ecology of Microbes Dissimilating Sulfur Compounds 47
motif of four cysteines Cys-X5CysXn-Cys-X3Cys is thus highly conserved in all

SRs. Two main categories of SRs can be distinguished on the basis of their phylog-
eny: cellular function and structure (Crane and Getzoff 1996). Assimilatory SRs
produce sulfide for incorporation in biomass and are present in many microorganisms,
algae, fungi, and plants. In contrast, dissimilatory SRs (DSRs) are restricted to spe-
cialized groups of bacteria and archaea, allowing them to gain energy by reducing/
oxidizing sulfur compounds. While DSRs are essential in the energy metabolism of
all sulfate/sulfite-reducing microorganisms (SRMs) that anaerobically respire
sulfite and/or sulfate and thus catalyze a unique step in the reductive part of the
sulfur cycle (Rabus et al. 2000), some but not all phototrophic and chemotrophic
microorganisms that oxidize reduced sulfur species also possess DSRs.
5.2 Evolution of Dissimilatory Sulfite Reductases
5.2.1 Sulfate/Sulfite-Reducing Microorganisms
A typical sulfate/sulfite-reducing pathway, in which a dissimilatory (bi)sulfite

reductase catalyzes the energy-yielding final step, is common to all SRMs investi-
gated so far. This DSR consists of a heterotetramer core with an 22 quaternary
protein structure. The - and -subunits are encoded by the adjoining genes dsrA
and dsrB, respectively, which most likely arose by duplication of an ancient dsr
gene and are thus paralogous (Dahl et al. 1993, Karkhoff-Schweizer et al. 1995,
Molitor et al. 1998; Fig. 5.1).
Initial phylogenetic analysis of only a few described SRMs showed that the
tree based on their DsrAB sequences was largely congruent with the 16S ribos-
omal RNA (rRNA) based tree, indicating that in the course of SRM evolution,
dsrAB was mainly inherited via vertical transmission from the parent organism
to its progeny (Wagner et al. 1998). However, already with this limited dataset,
inconsistencies in the branching pattern and phylogenetic distances in DsrAB
and 16S rRNA gene-based trees were indicative of possible lateral gene transfer
(LGT) events of dsrAB (Larsen et al. 1999; Wagner et al. 1998). In order to
investigate this further, subsequent studies have significantly extended the dsrAB
dataset by including sequences from several representatives of all known major
lineages of SRMs, namely, the bacterial phyla Proteobacteria, Firmicutes,
Thermodesulfobacteria, and Nitrospira, and the archaeal phyla Crenarchaeota
and Euryarchaeota (Friedrich 2002; Klein et al. 2001; Molitor et al. 1998;
Zverlov et al. 2005). Phylogenetic data, such as characteristic insertions/dele-
tions within dsrAB sequences, collectively confirmed the initial view that verti-
cal transmission was the main evolutionary process responsible for the
distribution of dsrAB among todays microorganisms and that a few defined
SRM groups most likely received their dsrAB by LGT. Those SRMs with an
unusual (i.e., laterally acquired) dsrAB include members of the genus
Fig. 5.1 Phylogeny of assimilatory and dissimilatory siroheme sulfite reductases. The unrooted
trees are based on a short, highly conserved sequence stretch surrounding the siroheme-binding
site. It is noteworthy that the resolution of the trees is limited because only 89 amino acid positions
were considered for treeing. a The schematic maximum-likelihood tree (ProtML-Molphy) shows
the affiliation of ungrouped sequences and was edited with iTOL (http://itol.embl.de/). b Polytomic
nodes in the consensus tree connect branches for which a relative order could not be determined
unambiguously by applying maximum-parsimony and maximum-likelihood (ProtML-Molphy,
ProML-Phylip) treeing methods. Parsimony bootstrap values (100 resamplings) are shown for all
nodes. Dashed lines indicate enzymes for which a dissimilatory function remains to be proven.
SOBs sulfur-compound-oxidizing bacteria, SRMs sulfate/sulfite-reducing microorganisms
Thermodesulfobacterium and some low-G+C Gram-positive bacteria of the

phylum Firmicutes, namely, Moorella thermoacetica, Ammonifex degensii, and
Desulfotomaculum subcluster Ib (including Sporotomaculum hydroxybenzoi-
cum), Ic, Id, and Ie bacteria. The dsrAB donor(s) for these bacteria probably
originated from the Deltaproteobacteria; Desulfobacterium anilini and the
related SRM strain mXyS1 are vertical descendants of a common deltaproteo-
bacterial ancestor that might have functioned as a dsrAB donor for Firmicutes
with a laterally acquired dsrAB (Zverlov et al. 2005; Fig. 5.2). Another postu-
lated case involved an ancient horizontal transfer of dsrAB even across the
boundaries of the domains of life; members of the euryarchaeotal genus
Archaeoglobus contain bacterial versions of dsrAB. The only other archaea
known to carry dsrAB are members of the sulfite-reducing, crenarchaeotal genus
Pyrobaculum (Fitz-Gibbon et al. 2002; Molitor et al. 1998), representing the
deepest branch in the DsrAB tree (Fig. 5.2).
The observation of dsrAB LGTs among major lineages of SRMs inevitably
evokes two important questions. What are the mechanisms underlying these
LGTs and what is the competitive advantage conferred by the acquisition of a
xenologous dsrAB? Although definite answers to these questions are not known,
some hypothetical ones are shortly discussed. An initial theory that a whole gene
set for sulfate reduction could be tightly clustered in the genomes of SRMs and
thus mobilized as a metabolic island (Klein et al. 2001) was rapidly rejected
because of the following findings. Although apsA (encoding the -subunit of the
adenosine 5-phosphosulfate reductase, another key enzyme in the dissimilatory
sulfate reduction pathway) was also subject to a few LGTs among SRMs, apsA
and dsrAB LGT patterns did not match (Friedrich 2002). The notion that genes
with key functions in sulfate reduction, such as aps and dsr, are not physically
linked and thus not cotransferred was confirmed by sequencing the genomes of
Archaeoglobus fulgidus (Klenk et al. 1997), Desulfovibrio vulgaris (Heidelberg
et al. 2004), and Desulfotalea psychrophila (Rabus et al. 2004). However, the
metabolic island theory was recently resurrected by the unexpected discovery
of a whole cluster of genes for sulfate reduction on genomic fragments from yet
uncultured and unidentified marine microorganisms (Mussmann et al. 2005).
This showed that LGT of dsrAB might be mediated by (a combination of)
different processes; e.g., the xenologous displacement (Koonin et al. 2001) of
preexisting dsrAB with a foreign version from another phylogenetically distant
SRM and/or by acquisition of a complete gene set for sulfate reduction by a non-
SRM, the latter resulting in an obvious metabolic advantage for the recipient.
The mechanistic bases of dsrAB LGT, i.e., if conjugation, transformation, or
transduction play a role, are unknown. However, given that SRMs can be
infected by phages (Rapp and Wall 1987; Walker et al. 2006), it is conceivable
that transduction is/was mediating the evolution of sulfate/sulfite respiration, in
analogy to cyanophages that are important vectors for reshuffling and exchange
of photosynthetic genes in and among Prochlorococcus and Synechococcus
species (Lindell et al. 2004; Zeidner et al. 2005).
50 A. Loy et al.
Fig. 5.2 Phylogeny of DsrAB sulfite reductase families. The consensus tree was constructed on
the basis of 367 amino acid alignment positions using the neighbor-joining method with the Kimura
model of amino acid substitution. Polytomic nodes connect branches for which a relative order
could not be determined unambiguously by applying distance-matrix, maximum-parsimony, and
maximum-likelihood treeing methods. Parsimony bootstrap values (100 resamplings) are indicated
for highly supported (values greater than 75%) branches. Selected environmental dsrAB clones
from Loy et al. (2004), Mussmann et al. (2005), Sabehi et al. (2005), and Venter et al. (2004) are
shown in bold. The bar indicates estimated sequence divergence. LA-dsrAB microorganisms with
a laterally acquired dsrAB (Klein et al. 2001; Zverlov et al. 2005)
5.2.2 DsrAB-Containing Syntrophs: Former

Sulfate/Sulfite-Reducing Microorganisms?
An unexpected peculiarity of some but not all dsrAB-containing and dsrAB-

expressing bacteria such as Pelotomaculum and Sporotomaculum is that they are
not able to grow with sulfite and/or sulfate as electron acceptors although they are
closely affiliated with SRMs (Brauman et al. 1998; Imachi et al. 2006). It is possi-
ble that some of these dsrAB-carrying non-SRMs use organosulfonates as electron
acceptors for anaerobic respiration instead, e.g., Bilophila wadsworthia degrades
taurine to sulfite, the actual substrate for its DSR (Cook et al. 1998). However, a
whole range of organosulfonates did not support growth of the spore-forming, low-
G+C bacteria of the genus Pelotomaculum (Imachi et al. 2006). It thus remains a
mystery why some bacteria are endued with actively expressed dsrAB genes, but
cannot utilize sulfate, sulfite, and/or organosulfonates for anaerobic respiration.
One might speculate that these microbes were formerly SRMs but have lost this
trait owing to the necessity to cope with a low-sulfate/sulfite, methanogenic envi-
ronment (Imachi et al. 2006). Hence, the presence of dsrAB in these bacteria, which
often live in close association with hydrogen-consuming microorganisms for the
syntrophic oxidation of substrates, would be a genetic remnant and thus indicative
of an ancient sulfate/sulfite-respiring potential. This theory receives some support
from physiological data on other syntrophs such as members of the deltaproteobac-
terial genus Syntrophobacter, which are also frequently encountered in methanogenic
environments (Loy et al. 2004; Lueders et al. 2004), but still have retained their
sulfate-reducing capability (Harmsen et al. 1998; Wallrabenstein et al. 1994).
Syntrophic dsrAB-containing non-SRMs, syntrophic SRMs, and authentic SRMs
are phylogenetically intermingled, indicating an evolutionary connection between
the lifestyles of SRMs and syntrophs. An alternative explanation is that the actual
substrate for the DSR in syntrophic bacteria has not yet been identified. Genomic
and metagenomic analyses of Pelotomaculum species are under way and might
provide some answers to this riddle.
5.2.3 Sulfur-Oxidizing Bacteria
Apart from microorganisms that do or did gain energy by reducing sulfate/sulfite,

it has long been known that also some phototrophic and chemotrophic sulfur-
compound-oxidizing bacteria (SOBs) contain a reversely operating DsrAB-type
DSR (rDSR; Schedel and Trper 1979). However, we only recently began to
understand the actual physiological function of this enzyme and how widespread
it is among SOBs; rDSRs occur in members of the Alphaproteobacteria,
Betaproteobacteria, and Gammaproteobacteria and of the phylum Chlorobi
(green sulfur bacteria) (Dahl et al. 1999, Sabehi et al. 2005; Fig. 5.2). On the
basis of mutagenesis studies of Allochromatium vinosum, it is evident that rDSR
in these SOBs is essential for the oxidation, and thus mobilization, of intracellularly
stored sulfur or polysulfides (obligate intermediates during oxidation of sulfide
and thiosulfate). Direct oxidation of external sulfide is not mediated by DsrAB
but via alternative pathways (Dahl et al. 2005). Currently available phenotypic
and genotypic data suggest that the presence of rDSR is associated with the capa-
bility of forming intracellular sulfur globules and lack of soxCD, genes encoding
a sulfur dehydrogenase (Friedrich et al. 2005).
52 A. Loy et al.
5.2.4 The Root and Major Branches of the DsrAB Tree
The DsrAB tree is subdivided into three distinct, well-supported branches that
represent three different DsrAB protein families (Molitor et al. 1998): (1) bacterial
DsrAB from microbes with the (former) capability to reduce sulfite, sulfate, and/or
organosulfonates (including the archaeon Archaeoglobus with a laterally acquired
bacterial dsrAB), (2) bacterial DsrAB from SOBs, and (3) the archaeal DsrAB from
sulfite-reducing Pyrobaculum species (Fig. 5.2). Revealing whether DsrAB in
Pyrobaculum is truly of archaeal nature will require the discovery and phylogenetic
analysis of additional sequences from this branch. Candidate archaeal species that
showed faint growth with sulfate and/or sulfite as the terminal electron acceptor but
that have thus far been overlooked as potential SRMs have been proposed (Dhillon
et al. 2005). The genome of one of these species, Caldivirga maquilingensis, is
currently being sequenced.
As mentioned already, dsrA and dsrB arose by a gene duplication event and thus
allow one to determine the root of the DsrAB tree by so-called paralogous rooting
(Klein et al. 2001; Fig. 5.1). Pyrobaculum occupies the deepest position in the
nearly bilaterally symmetrical DsrA and DsrB branches, indicating that, if dsrAB in
Pyrobaculum are archaeal, the duplication of an ancestral dsr gene preceded the
diversification of the domains Archaea and Bacteria and that the ancestral DsrAB
functioned in the reductive direction (Molitor et al. 1998). The latter finding is
supported by biogeochemical data suggesting that the rise of sulfite respiration took
place early in Earths genesis, possibly even before the evolution of sulfate respira-
tion (Skyring and Donnelly 1982). In the further course of evolution, the bacterial
DsrAB version presumably underwent a functional split, leading to maintenance of
the ancestral, sulfite-reducing enzyme type in SRMs and the first appearance of a
new reversely operating DSR in SOBs.
5.2.5 Other Non-DsrAB Dissimilatory Sulfite Reductases
Besides DsrAB, two further siroheme SRs, AsrABC and Fsr, are presumably
involved in dissimilatory processes. AsrABC is best studied in Salmonella enterica
serovar Typhimurium and is encoded by a functional operon consisting of three genes
asrA, asrB, and asrC (Clark and Barrett 1987; Huang and Barrett 1990, 1991). The
deduced amino acid sequences of asrA and asrC contain conserved cysteine residues
that are characteristic for [Fe4S4]ferredoxin binding domains (Huang and Barrett
1991). An additional siroheme[Fe4S4]-binding motif is only present in the -subunit
AsrC, which is homologous to DsrA and DsrB (Dhillon et al. 2005). In S. enterica
serovar Typhimurium, asr genes are part of a larger set of genes, which act in concert
to facilitate a very specific metabolism, i.e., B12-dependent anaerobic growth by oxi-
dizing ethanolamine or 1,2-propanediol with tetrathionate as an electron acceptor
(Price-Carter et al. 2001). Interestingly, this assemblage of many genes, including
genes for three sulfur compound reducing enzyme systems (ttr, phs, and asr),
1,2-propanediol catabolism (pdu), and de novo synthesis of B12 (cbi), is absent in the
genomes of closely related Escherichia coli strains. Thus, B12-dependent anaerobic
degradation of small molecules, driven by reduction of sulfur compounds, is a
characteristic trait of Salmonella species. It has been suggested that this metabolism
evolved during the divergence of Salmonella from E. coli by a combination of acqui-
sition of novel genes and loss of ancestral genes. Because tetrathionate can be
reduced by many other enteric bacteria (Barrett and Clark 1987), it seems likely that
ttr, phs, and asr genes are ancestral and hence were evolutionary eradicated from the
genomes of E. coli strains (Price-Carter et al. 2001). Clearly, additional studies are
necessary to shed further light on this hypothesis. For example, the distribution of asr
genes among different microbial taxa is not well known. A BLASTp search against
all 623 microbial genome sequences (3 October 2006) revealed that asrABC genes
are present in the gammaproteobacterial genera Salmonella and Photobacterium and
in the low-G+C Gram-positive bacteria Clostridium (Harrison et al. 1984; Laishley
et al. 1984), Thermoanaerobacterium, and Moorella (Fig. 5.1).
A new type of SR, Fsr, was recently discovered in the methanogenic
archaeon Methanocaldococcus jannaschii (Johnson and Mukhopadhyay 2005).
Although sulfite can be inhibitory to methanogens (Balderston and Payne
1976), other methanogens such as M. jannaschii, a strictly hydrogenotrophic,
thermophilic microorganism, not only tolerate but even grow with sulfite as the
sole source of sulfur (Daniels et al. 1986; Rothe and Thomm 2000). Fsr is an
unusual, chimeric protein, with the N-terminal half being an H2F420 dehydroge-
nase and the C-terminal half being a siroheme SR, which might have been
generated by fusion of a laterally acquired DSR gene and a fqoF or fpoF gene,
coding for an H2F420 dehydrogenase subunit (Johnson and Mukhopadhyay
2005). The physiological role of Fsr appears to be detoxification of sulfite
rather than sulfite-reduction-based energy production, as M. jannaschii could
thus far not been grown with acetate (as the sole carbon source), hydrogen, and
sulfite. In the phylogenetic tree, the siroheme-binding sequence range of Fsr
forms a monophyletic group with AsrC and other SRs (e.g., from Moorella and
Clostridium), whose functions are unknown. However, with the exception of
the DsrA and DsrB branches (see above), the direction of evolution, i.e., the
root of the tree, cannot be inferred from Fig. 5.1.
5.3 Molecular Insights into the Ecology

of DsrAB-Employing Microorganisms
5.3.1 PCR-Based Surveys
The fortunate state that SRMs and SOBs are clearly separated in the bacterial part
of the DsrAB tree (Fig. 5.2) and that the course of DsrAB evolution within these two
functional guilds largely paralleled their 16S rRNA evolution makes dsrAB an ideal
54 A. Loy et al.
molecular marker for determinative and ecological studies. Specific primer sets
were developed (and are continuously updated upon the availability of new complete
dsrAB sequences) together with protocols for amplification, cloning, and comparative
sequence analysis of a large dsrAB fragment from almost all SRMs (excluding
Pyrobaculum species) (Wagner et al. 2005). Two fascinating discoveries were made
after application of these primer sets for cultivation-independent recovery of dsrAB
sequences from a wide variety of ecosystems and geographic regions (Dhillon et al.
2003; Leloup et al. 2007; Loy et al. 2004). Firstly, many environmental dsrAB
sequences are not affiliated with known SRMs but occupy basal positions in the SRM
branch of the DsrAB tree (Fig. 5.2); hence, these novel dsrAB variants might derive
from microbes that either are members of recognized major taxa not yet known to
contain SRMs or represent yet unknown microbial classes or phyla. Secondly, dsrAB
richness in many habitats is dominated by these novel sequence types. This observa-
tion, and the possibility that these environmental surveys underestimated the number
and diversity of yet uncultured SRMs (because the primers are based on only a few
complete dsrAB sequences deriving mainly from cultured SRMs), indicates that these
previously unrecognized SRMs are of significant ecological importance. Although
the current set of cultivated SRMs already is an assemblage from diverse microbial
phyla, we now understand that they only constitute the tip of the iceberg of the nat-
ural SRM diversity. However, only the dsrAB sequences are known from these novel
SRMs. Even the DsrAB-based phylogeny may not reflect the phylogeny of the SRMs
carrying these novel dsrAB owing to the blurring effect of possible LGT events.
Our knowledge of the distribution of rDSR among described SOBs, a prerequisite
for sound analysis and interpretation of environmentally retrieved dsrAB sequences
from this functional guild, is scant. An initial step towards closing this gap in our
knowledge was made by developing specific dsrAB-targeted primer sets for SOBs
(Duller and Loy, unpublished data). SOBs from different taxonomic groups are cur-
rently being screened for the presence of dsrAB. Furthermore, the applicability of
these primers for environmental surveys of dsrAB-carrying SOBs is also being tested
with samples that contain microbial communities of varying complexity.
With the exception of dsrAB from Pyrobaculum, virtually nothing is known
about the phylogenetic breadth or environmental diversity of dsrAB sequences
belonging to the archaeal DsrAB family. However, it is noteworthy that
Pyrobaculum aerophilum has two different dsrAB copies in its genome (Fig. 5.2).
The presence of multiple copies of a functional gene in one strain is not uncommon
and might extend an organisms ability to cope with varying environmental
conditions, given that the different gene versions code for functional enzymes with
slightly different characteristics (Tchawa Yimga et al. 2003).
5.3.2 Metagenomics
Recent studies emphasize the potential of metagenomics (DeLong 2002;

Handelsman 2004) for environmental analysis of uncultured sulfur organisms
without the need for PCR or cultivation. Fosmid libraries containing several
thousand clones with insert sizes ranging from 32 to 44 kb were created from
sediment samples from an intertidal sand flat in the Wadden Sea and from the
deep ocean at the Hydrate Ridge (Mussmann et al. 2005). Detailed sequence
analysis of three selected fosmids (1) supported earlier theories about possible
mechanisms of gene flow among SRMs, owing to the presence of genomic
islands for sulfate reduction (see above) and (2) enabled extensions of the current
model of sulfur-based energy metabolism.
In another study, bacterial artificial chromosome (BAC) libraries with insert
sizes averaging 80 kb were established from surface waters of the Mediterranean
Sea and the Red Sea in order to better comprehend the genetic variability and physi-
ological capabilities of proteorhodopsin-containing microorganisms (Sabehi et al.
2005). These microbes gain energy with help of a light-driven proton pump, the
membrane-spanning proteorhodopsin, and are one of the most abundant microbial
guilds on Earth. Surprisingly, one of the 11 proteorhodopsin gene-carrying BAC
clones that were completely sequenced also contained a whole reverse DSR
operon with high sequence similarity to, and identical arrangement of, dsr genes as
in A. vinosum. Additionally, the dsrAB sequence from this Mediterranean BAC
clone clustered tightly with nine dsrAB sequences from a large, shotgun-library-
based environmental sequencing project of the Sargasso Sea (Venter et al. 2004;
Fig. 5.2), suggesting ubiquity of the respective microbes in the photic zones of the
oceans. Some anoxygenic phototrophs gain energy by complete oxidation of dime-
thyl sulfide to sulfate (Jonkers et al. 1999) and possess a reverse DSR (Dahl et al.
1999). In contrast to hydrogen sulfide, which is spontaneously oxidized under oxic
conditions, dimethyl sulfide is the most important volatile biogenic sulfur
compound in ocean waters and thus is mainly responsible for the transfer of
marine-derived sulfur to the air (Lovelock et al. 1972). Once released into the
atmosphere, dimethyl sulfide is chemically oxidized to acidic aerosol sulfates,
which serve as condensation nuclei for cloud formation and thus increase the
absorption and scattering of incoming sunlight (cloud albedo) over the remote
oceans. However, up to about 90% of the dimethyl sulfide is biologically oxidized
by marine microorganisms before it can diffuse into the atmosphere (Kiene and
Bates 1990). It has been speculated that these novel proteorhodopsin- and reverse-
DSR-exploiting microbes are part of this important microbial community that
regulates the sea-to-air flux of dimethyl sulfide (Sabehi et al. 2005).
5.4 Conclusions
Genetic, biochemical, and phylogenetic studies have raised our awareness of

novel aspects of functional properties of DSRs and the evolutionary flow of their
coding genes among diverse microbial lineages. This knowledge gain was
extended by ecological investigations that showed an unforeseen diversity of
DSR genes in the environment. However, revealing the identities and (eco) physiolo-
gies of these previously hidden microbes with novel DSR gene variants remains
a challenge. In this respect, the genetically modifiable SOBs A. vinosum (Pott and
56 A. Loy et al.
Dahl 1998) and Chlorobaculum (Chlorobium) tepidum (Frigaard and Bryant

2001) could become valuable model systems for heterologous expression analy-
sis in order to prove the functions of environmental dsrAB and for revealing the
theoretical phylogenetic and functional boundaries of lateral dsrAB transfers
among different microbial taxa.
Acknowledgements. We acknowledge support from the Fonds zur Frderung der wissenschaftli-
chen Forschung (project P18836-B17) to A.L. and the bmb+f (project 01 LC 0021A-TP2 in the
framework of the BIOLOG II program) to M.W. Kasper Kjeldsen and Mike Taylor are acknowl-
edged for valuable comments on the manuscript. We are indebted to Dave Stahl and Michael
Friedrich for long-term collaboration on the ecology/evolution of SRMs and we thank Michael
Klein, Vladimir Zverlov, Natuschka Lee, Doris Steger, Stephanie Freder, Ivan Barisic, Sebastian
Lcker, and Christian Baranyi, who have contributed in many ways to our work on microbes of
the sulfur cycle.
References
Balderston WL, Payne WJ (1976) Inhibition of methanogenesis in salt marsh sediments and
whole-cell suspensions of methanogenic bacteria by nitrogen oxides. Appl Environ Microbiol
32:264269
Barrett EL, Clark MA (1987) Tetrathionate reduction and production of hydrogen sulfide from
thiosulfate. Microbiol Rev 51:192205
Brauman A, Muller JA, Garcia JL, Brune A, Schink B (1998) Fermentative degradation of
3-hydroxybenzoate in pure culture by a novel strictly anaerobic bacterium, Sporotomaculum
hydroxybenzoicum gen. nov., sp. nov. Int J Syst Bacteriol 48:215221
Canfield DE, Raiswell R (1999) The evolution of the sulfur cycle. Am J Sci 299:697723
Clark MA, Barrett EL (1987) The phs gene and hydrogen sulfide production by Salmonella typh-
imurium. J Bacteriol 169:23912397
Cook AM, Laue H, Junker F (1998) Microbial desulfonation. FEMS Microbiol Rev 22:399419
Crane BR, Getzoff ED (1996) The relationship between structure and function for the sulfite
reductases. Curr Opin Struct Biol 6:744756
Crane BR, Siegel LM, Getzoff ED (1995) Sulfite reductase structure at 1.6 : evolution and
catalysis for reduction of inorganic anions. Science 270:5967
Dahl C, Kredich NM, Deutzmann R, Trper HG (1993) Dissimilatory sulphite reductase from
Archaeoglobus fulgidus: physico-chemical properties of the enzyme and cloning, sequencing
and analysis of the reductase genes. J Gen Microbiol 139:18171828
Dahl C, Rkhely G, Pott-Sperling AS, Fodor B, Takcs M, Tth A, Kraeling M, Gyrfi K, Kovcs
A, Tusz J, Kovcs KL (1999) Genes involved in hydrogen and sulfur metabolism in
phototrophic sulfur bacteria. FEMS Microbiol Lett 180:317324
Dahl C, Engels S, Pott-Sperling AS, Schulte A, Sander J, Lbbe Y, Deuster O, Brune DC (2005)
187:13921404
Daniels L, Belay N, Rajagopal BS (1986) Assimilatory reduction of sulfate and sulfite by metha-
nogenic bacteria. Appl Environ Microbiol 51:703709
DeLong EF (2002) Microbial population genomics and ecology. Curr Opin Microbiol 5:520524
Dhillon A, Teske A, Dillon J, Stahl DA, Sogin ML (2003) Molecular characterization of sulfate-
reducing bacteria in the Guaymas Basin. Appl Environ Microbiol 69:27652772
Dhillon A, Goswami S, Riley M, Teske A, Sogin M (2005) Domain evolution and functional
diversification of sulfite reductases. Astrobiology 5:1829
Fitz-Gibbon ST, Ladner H, Kim UJ, Stetter KO, Simon MI, Miller JH (2002) Genome sequence of the
hyperthermophilic crenarchaeon Pyrobaculum aerophilum. Proc Natl Acad Sci USA 99:984989
Friedrich MW (2002) Phylogenetic analysis reveals multiple lateral transfers of adenosine-
5-phosphosulfate reductase genes among sulfate-reducing microorganisms. J Bacteriol
184:278289
Frigaard NU, Bryant DA (2001) Chromosomal gene inactivation in the green sulfur bacterium
Chlorobium tepidum by natural transformation. Appl Environ Microbiol 67:25382544
Handelsman J (2004) Metagenomics: application of genomics to uncultured microorganisms.
Microbiol Mol Biol Rev 68:669685
Harmsen HJM, Van Kuijk BLM, Plugge CM, Akkermans ADL, De Vos WM, Stams AJM (1998)
Syntrophobacter fumaroxidans sp. nov., a syntrophic propionate-degrading sulfate-reducing
bacterium. Int J Syst Bacteriol 48:13831387
Harrison G, Curle C, Laishley EJ (1984) Purification and characterization of an inducible dissimi-
latory type sulfite reductase from Clostridium pasteurianum. Arch Microbiol 138:7278
WC, Sullivan SA, Fouts D, Haft DH, Selengut J, Peterson JD, Davidsen TM, Zafar N, Zhou
L, Radune D, Dimitrov G, Hance M, Tran K, Khouri H, Gill J, Utterback TR, Feldblyum TV,
Wall JD, Voordouw G, Fraser CM (2004) The genome sequence of the anaerobic, sulfate-
reducing bacterium Desulfovibrio vulgaris Hildenborough. Nat Biotechnol 22:554559
Huang CJ, Barrett EL (1990) Identification and cloning of genes involved in anaerobic sulfite
reduction by Salmonella typhimurium. J Bacteriol 172:41004102
Huang CJ, Barrett EL (1991) Sequence analysis and expression of the Salmonella typhimurium
asr operon encoding production of hydrogen sulfide from sulfite. J Bacteriol 173:15441553
Huston DL, Logan GA (2004) Barite, BIFs and bugs: evidence for the evolution of the Earths
early hydrosphere. Earth Planet Sci Lett 220:4155
Imachi H, Sekiguchi Y, Kamagata Y, Loy A, Qiu YL, Hugenholtz P, Kimura N, Wagner M, Ohashi
A, Harada H (2006) Non-sulfate-reducing, syntrophic bacteria affiliated with Desulfotomaculum
cluster I are widely distributed in methanogenic environments. Appl Environ Microbiol
72:20802091
iTOL (2006) Interactive tree of life. http://itol.embl.de/
Johnson EF, Mukhopadhyay B (2005) A new type of sulfite reductase, a novel coenzyme F420-
dependent enzyme, from the methanarchaeon Methanocaldococcus jannaschii. J Biol Chem
280:3877638786
Jonkers HM, Jansen M, Van der Maarel MJ, Van Gemerden H (1999) Aerobic turnover of dime-
thyl sulfide by the anoxygenic phototrophic bacterium Thiocapsa roseopersicina. Arch
Karkhoff-Schweizer RR, Huber DP, Voordouw G (1995) Conservation of the genes for dissimila-
tory sulfite reductase from Desulfovibrio vulgaris and Archaeoglobus fulgidus allows their
detection by PCR. Appl Environ Microbiol 61:290296
Kiene RP, Bates TS (1990) Biological removal of dimethylsulphide from sea water. Nature
345:702705
Klein M, Friedrich M, Roger AJ, Hugenholtz P, Fishbain S, Abicht H, Blackall LL, Stahl DA,
Wagner M (2001) Multiple lateral transfers of dissimilatory sulfite reductase genes between
major lineages of sulfate-reducing prokaryotes. J Bacteriol 183:60286035
Klenk H-P, Clayton RA, Tomb J-F, White O, Nelson KE, Ketchum KA, Dodson RJ, Gwinn M,
McKenney K, Adams MD, Loftus B, Peterson S, Reich CI, McNeil LK, Badger JH, Glodek
A, Zhou L, Overbeek R, Gocayne JD, Weidman JF, McDonald L, Utterback T, Cotton MD,
Spriggs T, Artiach P, Kaine BP, Sykes SM, Sadow PW, DAndrea KP, Bowman C, Fujii C,
Garland SA, Mason TM, Olsen GJ, Fraser CM, Smith HO, Woese CR, Venter JC (1997) The
58 A. Loy et al.
complete genome sequence of the hyperthermophilic, sulphate-reducing archaeon

Archaeoglobus fulgidus. Nature 390:364370
Koonin EV, Makarova KS, Aravind L (2001) Horizontal gene transfer in prokaryotes: quantifica-
tion and classification. Annu Rev Microbiol 55:709742
Laishley EJ, Tyler MG, Krouse HR (1984) Sulfur isotope fractionation during SO32 reduction by
different clostridial species. Can J Microbiol 30:841844
Larsen O, Lien T, Birkeland NK (1999) Dissimilatory sulfite reductase from Archaeoglobus pro-
fundus and Desulfotomaculum thermocisternum: phylogenetic and structural implications
from gene sequences. Extremophiles 3:6370
Leloup J, Loy A, Knab NJ, Borowski C, Wagner M, Jrgensen BB (2007) Diversity and abun-
dance of sulfate-reducing microorganisms in the sulfate and methane zones of a marine sedi-
ment, Black Sea. Environ Microbiol 9:131142
Lindell D, Sullivan MB, Johnson ZI, Tolonen AC, Rohwer F, Chisholm SW (2004) Transfer of
photosynthesis genes to and from Prochlorococcus viruses. Proc Natl Acad Sci USA
101:1101311018
Lovelock JE, Maggs RJ, Rasmussen RA (1972) Atmospheric dimethyl sulfide and the natural
sulfur cycle. Nature 237:452453
Loy A, Ksel K, Lehner A, Drake HL, Wagner M (2004) Microarray and functional gene analyses
of sulfate-reducing prokaryotes in low sulfate, acidic fens reveal co-occurence of recognized
genera and novel lineages. Appl Environ Microbiol 70:69987009
Lueders T, Pommerenke B, Friedrich MW (2004) Stable-isotope probing of microorganisms thriv-
ing at thermodynamic limits: syntrophic propionate oxidation in flooded soil. Appl Environ
Matthews JC, Timkovich R, Liu MY, Le Gall J (1995) Siroamide: a prosthetic group isolated from
sulfite reductases in the genus Desulfovibrio. Biochemistry 34:52485251
Molitor M, Dahl C, Molitor I, Schfer U, Speich N, Huber R, Deutzmann R, Trper HG (1998)
A dissimilatory sirohaem-sulfite-reductase-type protein from the hyperthermophilic archaeon
Pyrobaculum islandicum. Microbiology 144:529541
Mussmann M, Richter M, Lombardot T, Meyerdierks A, Kuever J, Kube M, Glockner FO, Amann
R (2005) Clustered genes related to sulfate respiration in uncultured prokaryotes support the
144:18811894
Price-Carter M, Tingey J, Bobik TA, Roth JR (2001) The alternative electron acceptor tetrathion-
ate supports B12-dependent anaerobic growth of Salmonella enterica serovar Typhimurium on
ethanolamine or 1,2-propanediol. J Bacteriol 183:24632475
Rabus R, Hansen T, Widdel F (2000) Dissimilatory sulfate- and sulfur-reducing prokaryotes. In:
Dworkin M, Falkow S, Rosenberg E, Schleifer K-H, Stackebrandt E (eds) The prokaryotes: an
evolving electronic resource for the microbiological community. Springer, New York
Becker I, Amann J, Gellner K, Teeling H, Leuschner WD, Glckner FO, Lupas AN, Amann R,
Klenk HP (2004) The genome of Desulfotalea psychrophila, a sulfate-reducing bacterium
Rapp BJ, Wall JD (1987) Genetic transfer in Desulfovibrio desulfuricans. Proc Natl Acad Sci
USA 84:91289130
Rothe O, Thomm M (2000) A simplified method for the cultivation of extreme anaerobic Archaea
based on the use of sodium sulfite as reducing agent. Extremophiles 4:247252
Sabehi G, Loy A, Jung KH, Partha R, Spudich JL, Isaacson T, Hirschberg J, Wagner M, Beja O
(2005) New insights into metabolic properties of marine bacteria encoding proteorhodopsins.
PLoS Biol 3:e273
Schedel M, Trper HG (1979) Purification of Thiobacillus denitrificans siroheme sulfite reductase
and investigation of some molecular and catalytic properties. Biochim Biophys Acta
568:454467
Siegel LM, Murphy MJ, Kamin H (1978) Siroheme: methods of isolation and characterization.
Skyring GW, Donnelly TH (1982) Precambrian sulfur isotopes and a possible role for sulfite in
the evolution of biological sulfate reduction. Precambrian Res 17:4161
Tchawa Yimga M, Dunfield PF, Ricke P, Heyer J, Liesack W (2003) Wide distribution of a novel
pmoA-like gene copy among type II methanotrophs, and its expression in Methylocystis strain
SC2. Appl Environ Microbiol 69:55935602
Venter JC, Remington K, Heidelberg JF, Halpern AL, Rusch D, Eisen JA, Wu D, Paulsen I, Nelson
KE, Nelson W, Fouts DE, Levy S, Knap AH, Lomas MW, Nealson K, White O, Peterson J,
Hoffman J, Parsons R, Baden-Tillson H, Pfannkoch C, Rogers YH, Smith HO (2004)
Environmental genome shotgun sequencing of the Sargasso Sea. Science 304:6674
Wagner M, Roger AJ, Flax JL, Brusseau GA, Stahl DA (1998) Phylogeny of dissimilatory sulfite
reductases supports an early origin of sulfate respiration. J Bacteriol 180:29752982
Wagner M, Loy A, Klein M, Lee N, Ramsing NB, Stahl DA, Friedrich MW (2005) Functional
marker genes for identification of sulfate-reducing prokaryotes. Methods Enzymol
397:469489
Walker CB, Stolyar SS, Pinel N, Yen H-CB, He Z., Zhou J, Wall JD, Stahl DA (2006) Recovery
of temperate Desulfovibrio vulgaris bacteriophage using a novel host strain. Environ Microbiol
8:19501959
Wallrabenstein C, Hauschild E, Schink B (1994) Pure culture and cytological properties of
Syntrophobacter wolinii. FEMS Microbiol Lett 123:249254
Zeidner G, Bielawski JP, Shmoish M, Scanlan DJ, Sabehi G, Beja O (2005) Potential photosyn-
thesis gene recombination between Prochlorococcus and Synechococcus via viral intermediates.
Environ Microbiol 7:15051513
Zverlov V, Klein M, Lcker S, Friedrich MW, Kellermann J, Stahl DA, Loy A, Wagner M (2005)
Lateral gene transfer of dissimilatory (bi)sulfite reductase revisited. J Bacteriol
187:22032208
Chapter 6
Genomic and Evolutionary Perspectives on
Sulfur Metabolism in Green Sulfur Bacteria
Niels-Ulrik Frigaard, Donald A. Bryant
Abstract Green sulfur bacteria (GSB) are anaerobic photoautotrophs that oxi-
dize sulfide, elemental sulfur, thiosulfate, ferrous iron, and hydrogen for growth.
We present here an analysis of the distribution and evolution of enzymes
involved in oxidation of sulfur compounds in GSB based on genome sequence
data from 12 strains. Sulfide:quinone reductase (SQR) is found in all strains.
Chlorobium ferrooxidans, which cannot grow on sulfide but grows on Fe2+, has
apparently lost all genes involved in oxidation of sulfur compounds other than
sqr. Instead, this organism possesses genes involved in assimilatory sulfate
reduction, a trait that is unusual in GSB. The dissimilatory sulfite reductase
(Dsr) enzyme system, which appears to be involved in elemental sulfur utiliza-
tion, is found in all sulfide-utilizing strains except Chloroherpeton thalassium.
The absence of Dsr enzymes in this early diverging GSB, in combination with
phylogenetic analyses, suggests that the Dsr system in GSB could be a recent
acquisition, which was obtained by lateral gene transfer in part from sulfide-
oxidizing bacteria and in part from sulfate-reducing bacteria. All thiosulfate-uti-
lizing GSB strains have an identical sox gene cluster. The soxCD genes, which
are found in certain other thiosulfate-utilizing organisms like Paracoccus pan-
totrophus, are absent from GSB. Flavocytochrome c, adenosine 5-phosphosul-
fate reductase, ATP-sulfurylase, the Qmo complex, and other enzymes related to
the utilization of sulfur compounds are found in some, but not all sulfide-utiliz-
ing strains. Even though different GSB strains superficially exhibit a similar
sulfur oxidation phenotype, this may be caused by different combinations of
enzymes. Thus, genome analyses have revealed that GSB have greater diversity
in sulfur metabolism than previously suspected.
6.1 Introduction
Inorganic sulfur metabolism in prokaryotic organisms is a complex topic owing to

the complex chemistry of sulfur and the multitude of enzymes that have evolved to
catalyze its chemistry. Nonetheless, the ability to use inorganic sulfur compounds
for growth is widespread among very different prokaryotes of both archaeal and
60
Springer 2008
6 Genomic and Evolutionary Perspectives on Sulfur Metabolism 61
bacterial affiliation. The phototrophic sulfur bacteria oxidize reduced inorganic sulfur
compounds for photosynthetic CO2 fixation and growth under anaerobic conditions.
Despite decades of research on the enzymology and genetics of sulfur-compound
oxidation in these bacteria, much remains to be learned about the biochemistry and
evolution of this essential part of their metabolism. Recently, genome sequence
information has become available for 12 strains of phototrophic green sulfur bacteria
(GSB). This information holds the key to substantial advances in understanding the
inorganic sulfur metabolism of these bacteria, which constitute an interesting model
not only for bacterial oxidation of sulfur compounds but also for the evolution of a
complex metabolic network in different strains of a closely related group of bacteria.
Based on genome sequence information, this chapter discusses the distribution and
possible functions of known and putative enzymes metabolizing sulfur compounds
as well as the evolution of these enzymes and the metabolic networks that they con-
stitute in GSB. Other recent publications on this subject are also available (Hanson
and Tabita 2001, 2003; Frigaard and Bryant 2008).
6.1.1 Green Sulfur Bacteria
Phototrophic sulfur bacteria thrive either in planktonic or benthic forms in aquatic,

anoxic environments where sulfide and light coincide. In these environments,
sulfide is often produced by sulfate-reducing bacteria, with which phototrophic
sulfur bacteria sometimes form more or less stable multicellular aggregates, but
the sulfide may also be of geological or anthropogenic origin. They are divided
into the GSB and the purple sulfur bacteria (PSB), both of which have been stud-
ied in pure cultures for about 100 years. The GSB, which owe their name and color
to their pigmentation by bacteriochlorophyll c, d, and e, comprise the family
Chlorobiaceae and represent the only cultivated members of the phylum Chlorobi
(Garrity and Holt 2001). The PSB owe their name and color to their pigmentation
by bacteriochlorophyll a and b and various carotenoids, and all are members of the
class Gammaproteobacteria, which is a major subdivision of the highly diverse
phylum Proteobacteria (Imhoff et al. 2005). Sulfide and thiosulfate can also sup-
port photosynthetic growth of some purple nonsulfur bacteria, which belong to the
class Alphaproteobacteria, but these compounds are usually not the preferred
substrates for growth of these bacteria (Imhoff et al. 2005). The ecology of GSB
and that of PSB are to some extent similar (van Gemerden and Mas 1995) and
their oxidative sulfur metabolism probably shares many characteristics (Brune
1989, 1995). However, most other aspects of their physiology and evolution are
rather different. For example, GSB are generally much less physiologically versa-
tile than PSB.
GSB are commonly found in sulfide-rich freshwater and estuarine environments
and are rare in marine environments (Overmann 2000; Garrity and Holt 2001).
They typically occur in the water column, in sediments, or within microbial mats,
and may occasionally be found as dense accumulations in planktonic forms in
62 N.-U. Frigaard, D.A. Bryant
stratified lakes or in benthic forms as microbial mats in sulfide-rich springs.

Recently characterized unusual habitats include the anoxic zone 100 m below the
surface of the Black Sea (Overmann et al. 1992; Manske et al. 2005), deep-sea
hydrothermal vents in the Pacific Ocean (Beatty et al. 2005), and the microbial
mats of Octopus and Mushroom Springs in Yellowstone National Park (Ward et al.
1998). GSB are obligately anaerobic and obligately photoautotrophic (Overmann
2000; Garrity and Holt 2001). All characterized GSB strains use the reductive (also
called reverse) tricarboxylic acid cycle for CO2 fixation. In addition, most GSB can
assimilate a small number of simple, organic compounds such as acetate, but only
in the presence of CO2 and a photosynthetic electron donor. Most strains use elec-
trons derived from oxidation of sulfide, but some strains can also oxidize elemental
sulfur, thiosulfate, H2, and Fe2+ (Sect. 6.2). All GSB characterized to date have
unique light-harvesting organelles known as chlorosomes, which allow highly effi-
cient capture of light energy (Frigaard and Bryant 2006). This ability provides a
substantial competitive advantage over PSB at low light intensities.
6.1.2 Genome Sequencing Projects of Green Sulfur Bacteria
Twelve strains of GSB have been selected for genome sequencing (Fig. 6.1). The
genome of one of the best characterized strains, Chlorobaculum tepidum TLS (pre-
viously known as Chlorobium tepidum TLS), was sequenced and annotated in 2002
by The Institute for Genome Research (Eisen et al. 2002). Other strains are cur-
rently at various stages of genome sequencing and annotation at the Joint Genome
Institute and in the laboratories of Donald A. Bryant, Stephan C. Schuster (both of
The Pennsylvania State University, USA), and Jrg Overmann (Ludwig-
Maximilians-Universitt, Germany). Currently, genome sequence data are publicly
available for ten GSB strains and can be accessed and analyzed on the Web sites of
the Joint Genome Institute (2007a) and the National Center for Biotechnology
Information (2007). These ten genomes were sequenced by a traditional shotgun
cloning approach. The draft genomes of Chlorobaculum parvum NCIMB 8327d
(= DSMZ 263T; previously known as Chlorobium vibrioforme subsp. thiosulfat-
ophilum; the d indicates that this strain contains bacteriochlorophyll d) and
Chloroherpeton thalassium ATCC 35110T have recently been determined by pyro-
sequencing (D.A. Bryant and S.C. Schuster, unpublished results), and PCR-based
methods are currently being used for gap closure. The information in this chapter
is based on the genome sequence data available for the ten GSB strains currently
present in the NCBI GenBank and has been supplemented with information from
the two unfinished sequences. Recent reviews on the physiological and metabolic
inferences derived from genome sequence data of GSB are available (Frigaard
et al. 2003, 2006; Frigaard and Bryant 2004; 2008). At present, the only available
genome sequence information for PSB is that for the halophilic Halorhodospira
halophila SL1 (Joint Genome Institute (2007b)).
Fig. 6.1 Neighbor-joining phylogenetic tree of the 16S ribosomal RNA gene of selected strains
of green sulfur bacteria (GSB). Common strain designations are shown. Asterisks indicate the
demonstrated ability to grow on thiosulfate. Strains for which genome sequence data are available
are marked in bold. Sequence accession numbers are either from the JGI data base or from
GenBank. The tree is based on 1,115 nucleotide positions and was made with MEGA version 3.1
(Kumar et al. 2004). Bootstrap values in percent are shown for 1,000 replications. Except for the
position of Chlorobium chlorochromatii CaD3, whose position is not resolved, minimum-evolution
and maximum-parsimony analyses support the topology of this tree
6.2 Compounds Oxidized by Green Sulfur Bacteria
In addition to being obligately phototrophic, GSB are obligately lithoautotrophic, which

means they can only grow by oxidizing inorganic compounds and reducing CO2 in the
presence of light. Inorganic compounds known to be oxidized by GSB include sulfide
(S2), elemental sulfur (S0), polysulfides (Sn2), thiosulfate (S2O32), tetrathionate
(S4O62), hydrogen (H2), and ferrous iron (Fe2+) (Brune 1995; Heising et al. 1999;
Garrity and Holt 2001). Most strains can oxidize sulfide and H2, while oxidation of thio-
sulfate and Fe2+ is less commonly encountered in cultivated strains. GSB are not known
to oxidize sulfite (SO32) for growth, although it probably is an intracellular intermediate
in the oxidation of other sulfur compounds (Sect. 6.3.5). Some of the sulfur compounds
utilized by the genome-sequenced strains are shown in Table 6.1.
GSB have a high affinity for sulfide, and this is usually the preferred substrate
even if other sulfur substrates are available. Sulfide is usually initially only incom-
pletely oxidized to elemental sulfur, which is deposited extracellularly as highly
refractive sulfur globules. These sulfur globules are usually but not always oxidized
completely to sulfate when the sulfide has been consumed.
Several strains of GSB are capable of growth on thiosulfate (see the overview
in Imhoff 2003). Such strains are often preferred for laboratory work because
thiosulfate can conveniently be included in high concentrations in liquid or solid

growth media and does not inhibit growth. Some of these strains have been
shown to photochemically disproportionate elemental sulfur into sulfide and thio-
sulfate in the absence of CO2 (strains DSMZ 249, 255, 257, and 263) an ability
not observed in strains that cannot utilize thiosulfate (Trper et al. 1988, Brune
1989). Two thiosulfate-utilizing strains of GSB have also been reported to utilize
tetrathionate (Brune et al. 1989).
Table 6.1 Phenotypes and genotypes of genome-sequenced green sulfur bacteria

Electron donora Genotypeb
Strain S2 S0 S2O32 sqr dsr fcc soy sox apr sat qmo PSRLC3
Chlorobaculum + + + + + + +
parvum
DSMZ 263
Chlorobaculum + + + + + + + + + +
tepidum TLS
Chlorobium + + + + + + + + +
chlorochromatii
CaD3
Chlorobium + + + + + + + + + + +
clathratiforme
DSMZ 5477
Chlorobium +
ferrooxidans
DSMZ 13031
Chlorobium + + + + + + +
limicola
DSMZ 245
Chlorobium + + + + +
luteolum
DSMZ 273
Chlorobium + + + + + + +
phaeobacteroides
DSMZ 266
Chlorobium + + + + + + + +
phaeovibrioides
DSMZ 265
Chloroherpeton + + + + + +
thalassium
ATCC 35110
Prosthecochloris + ND ND + + + + + +
sp. BS1
Prosthecochloris + + + + + + +
aestuarii
DSMZ 271
ND not determined.
a
Garrity and Holt (2001), Heising et al. (1999), Vogl et al. (2006).
b
The following abbreviations designate more than one gene: apr, aprBA; dsr, dsrNCABLEFHT-
MKJOP; fcc, fccAB; sox, soxJXYZAKBW; soy, soyYZ; qmo, qmoABC.
Chlorobium ferrooxidans DSMZ 13031, a GSB strain that oxidizes Fe2+, has
been characterized (Heising et al. 1999). This strain also oxidizes H2 but appears to
have lost the ability to oxidize sulfur compounds because it does not grow on sulfide,
elemental sulfur, or thiosulfate. This phenotype is largely confirmed by the absence
of many genes related to oxidation of sulfur compounds in its genome (Table 6.1). It
is not known how common or important this mode of photoferrotrophy is in nature,
but photosynthetic growth with Fe2+ as an electron donor has also been demonstrated
in some purple bacteria (Widdel et al. 1993; Ehrenreich and Widdel 1994).
6.3 Enzymes Involved in Sulfur-Compound Oxidation
Several enzymes potentially involved in sulfur metabolism can readily be identified

in the genome sequences by sequence homology with enzymes for which functional
information is available (Table 6.1). This information, combined with biochemical
and physiological information about the strains, can be used to produce a putative
scheme for the metabolic reactions in GSB (Fig. 6.2). The enzymes and the meta-
bolic capabilities they may confer are discussed in the following sections.
Fig. 6.2 Overview of the proposed pathways in the oxidative sulfur metabolism of GSB. Not all
GSB strains have all pathways shown here. See text for details. The electron carriers menaquinone
(MK) and cytochrome c (CycA) are reoxidized by the photosynthetic processes that fix CO2.
(Derived from information in Eisen et al. 2002 and Dahl 2008)
6.3.1 Overview of the Putative Sulfur Compound

Oxidation Enzymes
Sulfide:quinone reductase (SQR) and flavocytochrome c probably constitute two

alternative pathways of sulfide oxidation (Sects. 6.3.3, 6.3.4). The dissimilatory
sulfite reductase (Dsr) system probably is involved in the oxidation of sulfur glob-
ules (Sect. 6.3.2). A putative thiol (RSH) is shown in Fig. 6.2 that may be involved
in the oxidation of the sulfur globules (Brune 1989, 1995; Hanson and Tabita
2001). The Sox system is involved in oxidation of thiosulfate and can account for
this activity observed in all strains capable of oxidizing thiosulfate (Sect. 6.3.6).
The adenosine 5-phosphosulfate (APS) reductase (Apr), sulfate adenylyltrans-
ferase (Sat), and quinone-interacting membrane-bound oxidoreductase (Qmo)
complex probably constitute a sulfite-oxidizing system (Sect. 6.3.5). A putative
alternative sulfite-oxidizing system has been identified in the GSB strains that do
not have the Apr, Sat, and Qmo enzymes (Sect. 6.3.5).
Homologs of polysulfide reductase, and heterodisulfide reductase are also found in
GSB (Frigaard and Bryant 2008). However, functions cannot easily be assigned to
these proteins in GSB because (1) they are too distantly related to characterized
enzymes, (2) they are not distributed among the GSB strains in a manner that obviously
correlates with known physiological traits, and (3) the sulfur compound and hydrogen
oxidation properties of the GSB strains can be accounted for by other enzymes.
6.3.2 Dissimilatory Sulfite Reductase
The well-studied PSB, Allochromatium vinosum, contains a gene cluster with

high sequence similarity to the dissimilatory sulfite reductase dsr gene cluster
of sulfate-reducing bacteria (Dahl et al. 2005; Dahl 2008). The dsr gene cluster
in A. vinosum, dsrABEFHCMKLJOPNRS, is essential for the oxidation of intra-
cellular sulfur globules, and thus it is assumed that the Dsr enzyme system in
this organism functions in the oxidative direction to produce sulfite (Pott and
Dahl 1998; Dahl et al. 2005; Sander et al. 2006). GSB contain a very similar
dsr cluster, dsrNCABLEFHTMKJOP, the only difference being the absence of
dsrRS and the presence of dsrT. This cluster is present in all GSB, except Chl.
ferrooxidans and Chp. thalassium, and it most likely encodes the same function
as in A. vinosum.
The absence of dsr genes in Chl. ferrooxidans is consistent with the observation
that this bacterium is incapable of growth on elemental sulfur and sulfide. Because
this strain appears to have ancestors that are sulfur- and sulfide-oxidizing GSB that
contain the dsr genes (Fig. 6.1), it seems highly likely that Chl. ferrooxidans has
lost the dsr genes as a consequence of adapting to growth on Fe2+.
The absence of dsr genes in Chp. thalassium is especially interesting for two
reasons: firstly, this organism is a very early diverging GSB, and, secondly, this
organism grows poorly on elemental sulfur. Like other GSB, Chp. thalassium
grows well on sulfide and forms extracellular sulfur globules as an oxidation
product (Gibson et al. 1984); however, this elemental sulfur is only very slowly
oxidized, and this behavior could be due to the absence of the Dsr system. It is at
present unclear what might constitute an alternative sulfur-oxidizing system in
Chp. thalassium. Such a system might somehow involve the ribulose-1,5-
bisphosphate carboxylase/oxygenase (Rubisco) like protein (RLP), which is
present in all GSB, including Chp. thalassium, and which has been shown to be
involved in growth on elemental sulfur in Cba. tepidum TLS (Hanson and Tabita
2001, 2003). It is an interesting possibility that it might have been the acquisition
of the Dsr-dependent system, which seems to be involved in efficient and com-
plete oxidation of elemental sulfur, that led to the relatively recent, explosive
radiation of the lineages of GSB that are not closely related to Chloroherpeton
(Fig. 6.1).
Phylogenetic analyses of the DsrA protein and other Dsr proteins in GSB
show that these proteins constitute a monophyletic group (Fig. 6.3a). Thus, the
DsrA phylogeny is congruent with the 16S ribosomal RNA phylogeny at least at
the phylum level. However, the dsr genes have experienced lateral gene transfer
(LGT) within the GSB phylum; for example, DsrA from Prosthecochloris aes-
tuarii DSMZ 271 is located within the Chlorobium/Chlorobaculum cluster (Fig.
6.3a). On the basis of further phylogenetic analyses, the cytoplasmic DsrAB
sulfite reductase and other cytoplasmic Dsr proteins in GSB are most closely
related to the Dsr proteins from other sulfide-oxidizing prokaryotes (Sander et
al. 2006). This is in contrast to the subunits of the membrane-bound DsrMKJOP
complex, which are most closely related to the DsrMKJOP proteins from sul-
fate-reducing prokaryotes. In addition, the DsrT protein (unknown function) is
only found in GSB and sulfate-reducing prokaryotes and not in other sulfide-
oxidizers. This suggests that the Dsr system in GSB has an intriguing chimeric
nature.
6.3.3 Sulfide:Quinone Reductase
SQR catalyzes the oxidation of sulfide with a membrane-bound isoprenoid quinone as

the electron acceptor. This enzyme occurs in both chemotrophic and phototrophic
prokaryotes (Griesbeck et al. 2000; Theissen et al. 2003). Membrane-bound SQR
activity has been demonstrated biochemically in GSB, and this enzyme presumably
feeds electrons into the photosynthetic electron transfer chain (Shahak et al. 1992).
The genome sequences of all 12 GSB strains, including Chl. ferrooxidans and Chp.
thalassium, encode either one or two homologs of the biochemically characterized
SQRs from Rhodobacter capsulatus (CAA66112) and Oscillatoria limnetica
(AAF72962) (Fig. 6.3b). The SQR homologs of GSB are flavoproteins with predicted
masses of about 53 kDa, and each contains all three conserved cysteine residues that
are essential for sulfide oxidation in R. capsulatus SQR (Griesbeck et al. 2002).
Fig. 6.3 Neighbor-joining phylogenetic tree of a DsrA and b sulfide:quinone reductase (SQR)
proteins from GSB and other organisms. Bootstrap values in percent are shown for 1,000 rep-
licates. Nodes with less than 50% support in neighbor-joining and minimum-evolution analyses
are collapsed
It is interesting that Chl. ferrooxidans contains an SQR (ZP_01385816), because

this organism cannot grow on sulfide as the sole electron donor (Heising et al.
1999). This organism may benefit from SQR activity as a supplement to its energy
metabolism. Alternatively, it could also use SQR as a protective mechanism to
remove sulfide, which prevents growth when it is present in high concentrations.
6.3.4 Flavocytochrome c
Flavocytochrome c is a periplasmic enzyme consisting of a large sulfide-binding

FccB flavoprotein subunit and a small FccA cytochrome c subunit (Brune 1995).
Except for Chl. ferrooxidans and Chlorobium luteolum DSMZ 273, an fccAB-
encoded flavocytochrome c is found in all GSB strains for which genome sequence
data are available. The flavocytochrome c of GSB consists of a 10-kDa FccA
cytochrome c553 subunit, which binds a single heme, and an approximately 47-kDa
sulfide-binding FccB flavoprotein subunit. FccAB is constitutively expressed in
Chlorobium limicola DSMZ 249 (Vert et al. 2002).
Although flavocytochrome c from various organisms has been shown to oxidize
sulfide and reduce cytochrome c in vitro, the exact function and significance of this
protein in vivo is still not clear. While many sulfide-utilizing organisms produce
flavocytochrome c, some sulfide-utilizing GSB and PSB do not, which clearly
demonstrates that flavocytochrome c is not essential for sulfide oxidation (Brune
1995). Additionally, a mutant of the purple sulfur bacterium A. vinosum DSMZ
180, in which flavocytochrome c has been eliminated genetically, exhibits sulfide
and thiosulfate oxidation rates similar to the wild type (Reinartz et al. 1998). If
indeed the FccAB flavocytochrome c oxidizes sulfide in vivo, both GSB and PSB
apparently have alternative sulfide-oxidizing enzyme systems, possibly SQR (Sect.
6.3.3) and the Dsr system (Sect. 6.3.2), that may be quantitatively more important.
However, it is also possible that flavocytochrome c is advantageous under certain
growth conditions and that such conditions have not yet been identified.
6.3.5 Sulfite Oxidation
Although GSB cannot grow on sulfite as sole sulfur source and electron donor,
sulfite appears to be the product of the Dsr enzyme system (Sect. 6.3.2, Fig. 6.2).
The only known dissimilatory sulfite oxidation enzyme that has homologs with
high sequence similarity in GSB is the Apr-type APS reductase (also called adeno-
sine 5-phosphosulfate reductase). (A CysH-type APS reductase is also found in
two GSB strains, but the gene encoding this enzyme is part of an assimilatory sul-
fate reduction gene cluster; see Sect. 6.4.) The aprAB genes, encoding the Apr
enzyme, are only found in four GSB strains (Table 6.1). In each of the four strains
these genes occur in a cluster with the sat gene encoding an ATP sulfurylase and
the qmoABC genes encoding a heterodisulfide-reductase-like quinone oxidoreduct-

ase known as the Qmo complex that has its substrate-binding site in the cytoplasm
(Pires et al. 2003). In combination, the products of this sataprBAqmoABC gene
cluster might constitute an enzyme system (SatAprQmo) that in principle could
oxidize sulfite to sulfate, with APS as an intermediate, and reduce a membrane-
bound quinone to quinol (Fig. 6.2). A similar aprBAqmoABC operon with high
sequence similarity occurs in some sulfate-reducing Desulfovibrio-like strains. It is
therefore possible that the sataprBAqmoABC gene cluster in GSB is derived
from sulfate-reducing organisms.
How sulfite is oxidized in GSB that lack the putative SatAprQmo system is not
clear. However, with the sole exception of Cba. parvum DSMZ 263 (Table 6.1), an
unusual homolog of polysulfide reductase, denoted polysulfide-reductase-like com-
plex 3 (PSRLC3), is present in all Dsr-containing GSB that lack the SatAprQmo
system (Frigaard and Bryant 2008). PSRLC3 is a membrane-bound molybdopterin-
binding enzyme that has a quinone oxidoreductase domain in the membrane and,
unlike other common polysulfide reductases, its substrate-binding site in the cyto-
plasm. Many known and putative sulfite oxidoreductases are molybdopterin-
binding, oxotransferase enzymes and the PSRLC3 complex may constitute such an
enzyme in GSB. Phylogenetic analyses show that the PSRLC3 complex in Chp.
thalassium is basal to all other PSRLC3 complexes in GSB. This suggests that (1)
the PSRLC3 complex was present in an early ancestor of GSB and (2) GSB lacking
a PSRLC3 complex may have lost it by gene elimination, perhaps as a consequence
of acquiring the SatAprQmo system.
6.3.6 Thiosulfate Oxidation by the Sox System
In the chemolithoautotrophic alphaproteobacterium Paracoccus pantotrophus, seven

sox genes (soxXYZABCD) constitute a complete thiosulfate-oxidizing enzyme system
(Friedrich et al. 2001, 2005). In this system, thiosulfate is bound to the SoxYZ carrier
by an oxidation reaction catalyzed by SoxAX and further processed by the hydrolase
SoxB and the oxidase SoxCD to regenerate the SoxYZ carrier and liberate two sulfate
molecules per thiosulfate molecule. In GSB the sox cluster, orf1015soxXYZA
orf1020soxBW, is conserved in the genomes of five strains (Table 6.1). This includes
all four thiosulfate-utilizing strains and one strain (CaD3) that has not been reported
to grow on thiosulfate. Because of the organizational conservation and the congruent
phylogeny of the genes in this cluster, the genes CT1015 and CT1020 are likely
involved in the Sox system. Thus, these two genes are now denoted as soxJ and soxK,
respectively. The soxCD genes, which are essential components of the Sox system in
P. pantotrophus, do not occur in the genome sequences of GSB. Instead, owing to the
conservation of the soxJ/orf1015 and soxK/orf1020 genes in the GSB sox gene clus-
ter, the process(es) in GSB that regenerates the SoxYZ complex probably involves
the SoxJ and SoxK proteins. No other easily identifiable thiosulfate oxidation
enzymes (such as rhodaneses) are encoded in the GSB genome sequences in a
manner that obviously matches the pattern of thiosulfate utilization. Thus, thiosulfate
utilization in GSB can almost certainly be attributed to the Sox system. A similar
SoxCD-independent, thiosulfate-oxidizing Sox system is present in the PSB A. vinosum
(see Chap. 9 by Grimm et al.). SoxY (J. van Beeumen, personal communication) and
the SoxYZ complex (B.C. Berks, personal communication) from GSB have recently
been crystallized and their structures determined.
6.3.7 A Novel Complex: SoyYZ
The heterodimeric SoxYZ complex carries sulfur substrates on a conserved

cysteine residue in the SoxY subunit (Sect. 6.3.6; Quentmeier and Friedrich 2001).
The soxYZ gene cluster has been duplicated in five GSB and is here denoted soyYZ
(Table 6.1). A signal sequence at the amino termini of the SoyY sequences suggests
that, like SoxYZ, SoyYZ is a periplasmic complex. Neither SoxZ nor SoyZ has a
signal sequence, and both are probably transferred across the cytoplasmic mem-
brane as part of complexes with SoxY or SoyY, respectively. In all GSB that have
soyYZ, these genes are located immediately upstream of the fccAB genes in an
apparent operon; therefore, it is attractive to propose that SoyY and SoyZ form a
complex in the periplasm that carries a sulfur substrate and that this complex reacts
with the periplasmic FccAB flavocytochrome c. However, not all GSB that encode
fccAB also encode soyYZ. The presence of soyYZ does not correlate with thiosulfate
utilization and the substrate SoyYZ may carry is unclear.
In most organisms of all taxonomic affiliations that have SoxY, the sulfur-
substrate-carrying cysteine residue of SoxY is located at the C-terminus within the
motif GGC(G12)COOH. SoyY differs from all known SoxY proteins by having a
C-terminus in which the putative sulfur-substrate-binding cysteine is the terminal
residue. The proximity of the C-terminal carboxyl group and the thiol group of the
substrate-carrying cysteine residue in SoyY is likely to affect the chemistry at this
site in a manner that does not occur in SoxY. If this is the case, this might explain
the evolution of this particular motif in SoyY. In GSB the conserved motif in SoyY
is VXAQACCOOH. The soyY gene has only been found in one organism other
than GSB: the anaerobic, sulfide-oxidizing, chemoautotrophic Alkalilimnicola
ehrlichei MLHE-1, which based on ribosomal RNA phylogeny is closely related to
PSB of the Ectothiorhodospiraceae family.
6.4 Assimilatory Sulfur Metabolism
It is often stated in the literature that GSB cannot perform assimilatory sulfate
reduction (Lippert and Pfennig 1969). Nevertheless, a recently isolated strain, Chl.
ferrooxidans DSMZ 13031, grows with sulfate as the sole sulfur source and cannot
utilize sulfide, thiosulfate, or elemental sulfur for growth (Heising et al. 1999).
In agreement with this observation, the Chl. ferrooxidans genome encodes a single
gene cluster that includes the assimilatory sulfate reduction genes cysIHDNCG and
the sulfate permease genes cysPTWA, which are transcribed in opposite directions.
These assimilatory sulfate reduction genes share a high degree of sequence similar-
ity with those from the clostridia Clostridium thermocellum and Desulfitobacterium
hafniense. However, sequence analyses show that the APS reductase encoded by
cysH in Chl. ferrooxidans is related to the plant-type enzyme that uses APS and not
3-phosphoadenosine 5-phosphosulfate (PAPS) as a substrate. An identical cys
gene cluster is observed in Chl. luteolum DSMZ 273, but not in any other GSB
genome. This raises the possibility that Chl. luteolum DSMZ 273 also is capable of
assimilatory sulfate reduction and growth in the absence of reduced sulfur com-
pounds using alternative electron donors.
6.5 Possible Phage-Mediated Lateral Gene Transfer
Mobile genetics elements are poorly characterized in GSB. Transposases and other
insertion elements are found in the genome sequences, and a few plasmids have also
been identified (Mndez-Alvarez et al. 1994). However, no phage has yet been
characterized that infects GSB, but there is no reason to believe that such a phage
does not exist. In fact, the presence of RNA-directed DNA polymerases and inte-
grases in several GSB genomes in different genetic clusters is a strong indication
that viral infections of GSB do occur. Phage can potentially cause lateral exchange
of host genes between successive hosts and are thus interesting from an evolu-
tionary point of view. One such example may be found in an 11,000-bp island in
Chl. phaeovibrioides DSMZ 265 (Frigaard and Bryant 2008). This island contains
the Chlorobium-type sox cluster with eight genes, in addition to a transposase, an
integrase, and an RNA-directed DNA polymerase. Genes that are unrelated to sulfur
metabolism surround this thiosulfate utilization island. It is possible that this
island is a remnant structure derived from an RNA viral genome. The sox cluster
could have been transferred into the viral genome by a transposase in a previous host
and then integrated laterally into the genome of strain DSMZ 265.
6.6 Conclusions
On the basis of genome sequence analyses, some conclusions can be made about
the sulfur compound oxidation enzymes in GSB. SQR (encoded by sqr) is the only
known sulfur-oxidizing enzyme that is found in all GSB strains. Most strains utiliz-
ing sulfide or elemental sulfur contain the dissimilatory sulfite reductase dsrNCA-
BLEFHTMKJOP genes. Although Chp. thalassium appears to have a (probably
less efficient) alternative enzyme system for elemental sulfur oxidation, the dsr
genes appear to be involved in elemental sulfur utilization in all other GSB strains.
All thiosulfate-utilizing strains have an identical sox gene cluster (soxJXYZAKBW).

The soxCD genes found in certain other thiosulfate-utilizing organisms like P. pan-
totrophus are absent from GSB. A putative complex denoted SoyYZ, related to the
thiosulfate-binding SoxYZ complex, could be involved in the processing of an as
yet unidentified sulfur compound in GSB. Genes encoding flavocytochrome c
(fccAB), APS reductase (aprAB), ATP sulfurylase (sat), and a heterodisulfide
reductase homolog (qmoABC) were found in some, but not all strains. Given the
patchy distribution of these and other enzymes among the strains, it seems likely
that different enzymes perform some sulfur-oxidation activities, such as sulfite oxi-
dation, in different strains of GSB. The Fe2+-oxidizing Chl. ferrooxidans, which
cannot grow on sulfide, has no genes obviously involved in sulfur utilization other
than sqr, but contains a full complement of genes involved in assimilatory sulfate
reduction (cysIHDNCG), a trait that is not widely distributed among the GSB.
Analyses indicate that, although the phylogenies of some enzymes (e.g., the
DsrA protein; Fig. 6.3a) are congruent with the organismal phylogeny at the phy-
lum level, the phylogenies of other enzymes are not (e.g., the SQR protein; Fig.
6.3b). Some enzyme systems that are only present in some strains exhibit a phyl-
ogeny incongruent with the cellular core phylogeny (e.g., the Sox and the SatApr
Qmo systems), and thus these systems appear to result from LGT rather than gene
elimination. In the case of the sox gene cluster, evidence for phage-mediated LGT
was identified.
As a final point, although the GSB are closely related, with the exception of
Chp. thalassium, genomic analyses clearly show that gene elimination and LGT
substantially influence the distribution of sulfur-metabolism genes both within the
Chlorobi and among prokaryotes from other phyla. These observations illustrate
the dynamic structures of prokaryotic genomes and in addition demonstrate that
even organisms that superficially appear to be very closely related on the basis of
their cellular core machinery nevertheless can have unexpected differences in
physiology and life style.
Acknowledgements. N.-U.F gratefully acknowledges support from the Danish Natural Science
Research Council (grant 21040463). D.A.B. gratefully acknowledges support for genomics
studies of GSB from the United States Department of Energy (grant DE-FG0294ER20137) and
the National Science Foundation (grant MCB-0523100).
References
Beatty JT, Overmann J, Lince MT, Manske AK, Lang AS, Blankenship RE, Van Dover CL,
Martinson TA, Plumley FG (2005) An obligately photosynthetic bacterial anaerobe from a
deep-sea hydrothermal vent. Proc Natl Acad Sci USA 102:93069310
Brune DC (1989) Sulfur oxidation by phototrophic bacteria. Biochim Biophys Acta
975:189221
Brune DC (1995) Sulfur compounds as photosynthetic electron donors. In: Blankenship RE,
Madigan MT, Bauer CE (eds) Anoxygenic photosynthetic bacteria. Kluwer, Dordrecht,
pp 847870
Dahl C (2008) Inorganic sulfur compounds as electron donors in purple sulfur bacteria. In:
Govindjee (series ed) Advances in photosynthesis and respiration, vol 27, Hell R, Dahl C,
Knaff DB, Leustek T (eds) Sulfur metabolism in phototrophic organisms. Springer, New York
(in press)
187:13921404
tepidum TLS, a photosynthetic, anaerobic, green-sulfur bacterium. Proc Natl Acad Sci USA
99:95099514
Ehrenreich A, Widdel F (1994) Anaerobic oxidation of ferrous iron by purple bacteria, a new type
of phototrophic metabolism. Appl Environ Microbiol 60:45174526
Friedrich CG, Rother D, Bardischewsky F, Quentmeier A, Fischer J (2001) Oxidation of reduced
inorganic sulfur compounds by bacteria: emergence of a common mechanism? Appl Environm
Frigaard N-U, Bryant DA (2004) Seeing green bacteria in a new light: genomics-enabled studies
of the photosynthetic apparatus in green sulfur bacteria and filamentous anoxygenic pho-
totrophic bacteria. Arch Microbiol 182: 265276
Frigaard N-U, Bryant DA (2006) Chlorosomes: Antenna organelles in photosynthetic green bacteria.
In: Shively JM (ed) Complex intracellular structures in prokaryotes. Springer, Berlin, pp 79114
Frigaard N-U, Bryant DA (2008) Genomics insights into the sulfur metabolism of phototrophic
green sulfur bacteria. In: Govindjee (series ed) Advances in photosynthesis and respiration, vol
27, Hell R, Dahl C, Knaff D, Leustek T (eds) Sulfur metabolism in phototrophic organisms.
Springer, New York (in press)
Frigaard N-U, Gomez Maqueo Chew A, Li H, Maresca JA, Bryant DA (2003) Chlorobium tepi-
dum: Insights into the structure, physiology, and metabolism of a green sulfur bacterium
derived from the complete genome sequence. Photosynth Res 78:93117
Frigaard N-U, Gomez Maqueo Chew A, Maresca JA, Bryant DA (2006) Bacteriochlorophyll bio-
synthesis in green bacteria. In: Grimm B, Porra R, Rdiger W, Scheer H (eds) Advances in
photosynthesis and respiration, vol 25. Springer, Dordrecht, pp 201221
Garrity GM, Holt JG (2001) Phylum BXI. Chlorobi phy. nov. In: Boone DR, Castenholz RW (eds)
Bergeys manual of systematic bacteriology, vol 1, 2nd edn. Springer, New York, pp 601623
Gibson J, Pfennig N, Waterbury JB (1984) Chloroherpeton thalassium gen. nov. et spec. nov., a
non-filamentous, flexing and gliding green sulfur bacterium. Arch Microbiol 138:96101
Griesbeck C, Hauska G, Schtz M (2000) Biological sulfide oxidation: Sulfide-quinone reductase
(SQR), the primary reaction. In: Pandalai SG (ed) Recent research developments in microbiology,
vol 4. Research Signpost, Trivadrum, pp 179203
(2002) Mechanism of sulfide-quinone reductase investigated using site-directed mutagenesis
and sulfur analysis. Biochemistry 41:1155211565
Hanson TE, Tabita FR (2001) A ribulose-1,5-bisphosphate carboxylase/oxygenase (Rubisco)-like
protein from Chlorobium tepidum that is involved with sulfur metabolism and the response to
oxidative stress. Proc Natl Acad Sci USA 98:43974402
Hanson TE, Tabita FR (2003) Insights into the stress response and sulfur metabolism revealed by
proteome analysis of a Chlorobium tepidum mutant lacking the Rubisco-like protein.
Photosynth Res 78:231248
Heising S, Richter L, Ludwig W, Schink B (1999) Chlorobium ferrooxidans sp. nov., a pho-
totrophic green sulfur bacterium that oxidizes ferrous iron in coculture with a Geospirillum
sp. strain. Arch Microbiol 172:116124
Imhoff JF (2003) Phylogenetic taxonomy of the family Chlorobiaceae on the basis of 16S rRNA
and fmo (Fenna-Matthews-Olson protein) gene sequences. Intl J Syst Evol Microbiol
53:941951
Imhoff JF, Hiraishi A, Sling J (2005) Anoxygenic phototrophic purple bacteria. In: Brenner DJ,
Krieg NR, Staley JT (eds) Bergeys manual of systematic bacteriology, vol 2, part A, 2nd edn.
Springer, New York, pp 119132
Joint Genome Institute (2007a) Integrated microbial genomes. http://img.jgi.doe.gov. Cited 15 Jan
2007
Joint Genome Institute (2007b) Microbial genomics. http://genome.jgi-psf.org/mic_cur1.html.
Cited 15 Jan 2007
Kumar S, Tamura K, Nei M (2004) MEGA3: integrated software for molecular evolutionary
genetics analysis and sequence alignment. Brief Bioinformatics 5:150163
Lippert KD, Pfennig N (1969) Die Verwertung von molekularem Wasserstoff durch Chlorobium
thiosulfatophilum. Arch Microbiol 65:2947
Manske AK, Glaeser J, Kuypers MAM, Overmann J (2005) Physiology and phylogeny of green
sulfur bacteria forming a monospecific phototrophic assemblage at a depth of 100 meters in
the Black Sea. Appl Environ Microbiol 71:80498060
Mndez-Alvarez S, Pavn V, Esteve I, Guerrero R, Gaju N (1994) Transformation of Chlorobium
limicola by a plasmid that confers the ability to utilize thiosulfate. J Bacteriol
176:73957397
National Center for Biotechnology Information (2007) Genomic biology. http://www.ncbi.nlm.
nih.gov/Genomes. Cited 15 Jan 2007
Overmann J (2000) The family Chlorobiaceae. In: Dworkin M, Falkow S, Rosenberg E, Schleifer
K-H, Stackebrandt E (eds) The prokaryotes: an evolving electronic resource for the microbio-
logical community, 3rd edn, release 3.1. Springer, New York
Overmann J, Cypionka H, Pfennig N (1992) An extremely low-light-adapted phototrophic sulfur
bacterium from the Black Sea. Limnol Oceanogr 37:150155
Pires RH, Loureno AI, Morais F, Teixeira M, Xavier AV, Saraiva LM, Pereira IAC (2003)
A novel membrane-bound respiratory complex from Desulfovibrio desulfuricans ATCC
27774. Biochim Biophys Acta 1605:6782
144:18811894
Quentmeier A, Friedrich CG (2001) The cysteine residue of the SoxY protein as the active site of
protein-bound sulfur oxidation of Paracoccus pantotrophus GB17. FEBS Lett 503:168172
Reinartz M, Tschpe J, Brser T, Trper HG, Dahl C (1998) Sulfide oxidation in the phototrophic
sulfur bacterium Chromatium vinosum. Arch Microbiol 170:5968
Shahak Y, Arieli B, Padan E, Hauska G (1992) Sulfide quinone reductase (SQR) activity in
Chlorobium. FEBS Lett 299:127130
Theissen U, Hoffmeister M, Grieshaber M, Martin W (2003) Single eubacterial origin of eukaryo-
tic sulfide:quinone oxidoreductase, a mitochondrial enzyme conserved from the early evolu-
tion of eukaryotes during anoxic and sulfidic times. Molec Biol Evol 20:15641574
Trper HG, Lorenz C, Schedel M, Steinmetz M (1988) Metabolism of thiosulfate in Chlorobium.
In: Olson JM, Ormerod JG, Amesz J, Stackebrandt E, Trper HG (eds) Green photosynthetic
bacteria. Plenum, New York, pp 189200
van Gemerden H, Mas J (1995) Ecology of phototrophic sulfur bacteria. In: Blankenship RE, Madigan
Vert F, Kostanjevecki V, De Smet L, Meyer TE, Cusanovich MA, Van Beeumen JJ (2002)
Identification of a thiosulfate utilization gene cluster from the green phototrophic bacterium
Chlorobium limicola. Biochemistry 41:29322945
Vogl K, Glaeser J, Pfannes KR, Wanner G, Overmann J (2006) Chlorobium chlorochromatii sp.
nov., a symbiotic green sulfur bacterium isolated from the phototrophic consortium
Chlorochromatium aggregatum. Arch Microbiol 185:363372
Ward DM, Ferris MJ, Nold SC, Bateson MM (1998) A natural view of microbial biodiversity
within hot spring cyanobacterial mat communities. Microbiol Mol Biol Rev 62:13531370
Widdel F, Schnell S, Heising S, Ehrenreich A, Assmus B, Schink B (1993) Ferrous iron oxidation
by anoxygenic phototrophic bacteria. Nature 362:834836
Chapter 7
Differential-Expression Proteomics
for the Study of Sulfur Metabolism
in the Chemolithoautotrophic
Acidithiobacillus ferrooxidans
Lissette Valenzuela, An Chi, Simn Beard, Jeffrey Shabanowitz,

Donald F. Hunt, Carlos A. Jerez
Abstract Acidithiobacillus ferrooxidans obtains its energy from the oxidation of

ferrous iron, elemental sulfur, or partially oxidized sulfur compounds. The ability of
this microorganism to solubilize metal sulfides is successfully applied in biomining
operations. Genomic, metagenomic, and high-throughput proteomic studies of the global
regulatory responses that biomining microorganisms use to adapt to their changing
environment are just beginning to emerge. To further study some of the components
involved in sulfur metabolism, differential expression proteomics of total periplasmic
proteins was done by high-resolution LTQ FT ion trap mass spectrometry. Of 216 pro-
teins found in the periplasm, several of them changed their levels of synthesis during
growth of A. ferrooxidans ATCC 23270 in thiosulfate, elemental sulfur and ferrous
iron. Thirty-four percent of them corresponded to unknown proteins. Forty-one pro-
teins were exclusively present in sulfur-grown cells and 14 in thiosulfate-grown cells.
The putative genes coding for these proteins were localized in the available genomic
sequence of A. ferrooxidans ATCC 23270. The genomic context around several of
these genes suggests their involvement in sulfur metabolism and possibly in sulfur
oxidation and formation of FeS clusters. Many of the periplasmic proteins changing
their expression during growth in sulfur compounds may have important roles yet to
be described in the sulfur metabolism of this acidophilic microorganism. This knowl-
edge will eventually help to improve mineral bioleaching processes.
7.1 Introduction
Acidithiobacillus ferrooxidans is a chemolithoauthotrophic bacterium that obtains

its energy from the oxidation of ferrous iron, elemental sulfur or partially oxidized
sulfur compounds (Olson et al. 2003; Harrison 1984; Lundgren 1980; Suzuki 2001;
Rawlings 2002).
Effective tools for the study of A. ferrooxidans genetics and physiology are not
in widespread use and, despite considerable effort, an understanding of its unusual
physiology remains at a rudimentary level. An efficient and reproducible tech-
nique for DNA transfer is still missing (Valenzuela et al. 2006; Rawlings 2005).
77
Springer 2008
78 L. Valenzuela et al.
A. ferrooxidans was the first biomining microorganism to have its genome entirely
sequenced and the annotation of all its genes has recently been made available
(J. Craig Ventner Institute 2007). This information has been very useful to many
researchers to look for the genome-wide candidate genes for important metabolic
pathways and several important physiological functions and to predict for the func-
tions of many new genes. The main focus of research has been the energy metabo-
lism which is directly responsible of bioleaching. Some researchers used
chromosome walking to find genes involved in sulfur and iron metabolisms
(Valenzuela et al. 2006; Rawlings 2005). Genomic, metagenomic, and high-
throughput proteomic studies of the global regulatory responses that biomining
microorganisms use to adapt to their changing environment are just beginning to
emerge (Valenzuela et al. 2006). In this chapter, we will concentrate specifically on
proteomic analysis of A. ferrooxidans to better understand its sulfur metabolism.
This knowledge together with that obtained in other bioleaching microorganisms
will allow future improvements in industrial bioleaching processes.
7.2 Sulfur Metabolism in A. ferrooxidans
The aerobic oxidation of elemental sulfur by A. ferrooxidans and other microor-

ganisms is carried out by a sulfur dioxygenase (Rohwerder et al. 2003; Rohwerder
and Sand 2003; Silver and Lundgren 1968a; Mller et al. 2004; Sugio et al.
1987). Recently, thiosulfate has been postulated as a key compound in the oxida-
tion of the sulfur moiety of pyrite (Schippers and Sand 1999). Iron(III) ions are
exclusively the oxidizing agents for the dissolution. Thiosulfate would be conse-
quently degraded in a cyclic process to sulfate, with elemental sulfur being a side
product. This explains why only Fe(II) ion oxidizing bacteria are capable of oxi-
dizing these metal sulfides (Schippers and Sand 1999). All reactions comprising
this oxidation have been shown to occur chemically (Sand et al. 1995, 2001).
However, sulfur-compound-oxidizing enzymes such as the tetrathionate hydro-
lase of A. ferrooxidans, A. thiooxidans, or the former Thiobacillus acidophilus
(now renamed Acidiphilium acidophilum) may also be involved in the process
(De Jong et al. 1997; Kelly et al. 1997; Suzuki 1999; Friedrich et al. 2001).
In addition, enzymes for thiosulfate or sulfite oxidation of A. ferrooxidans or
A. thiooxidans may successfully compete with the chemical reactions with
iron(III) ions as an oxidizing agent (Schippers and Sand 1999).
7.3 Proteomics of A. ferrooxidans Grown in Sulfur Compounds
Two-dimensional polyacrylamide gel electrophoresis (2D PAGE) in combination

with mass spectrometry is currently the most widely used technology for compara-
tive bacterial proteomic analysis (Gygi et al. 2000). A set of A. ferrooxidans ATCC
7 Differential-Expression Proteomics for the Study of Sulfur Metabolism 79
19859 proteins differentially expressed when grown in metal sulfides, thiosulfate,

elemental sulfur, and ferrous iron media were characterized by using 2D PAGE
(Ramirez et al. 2004). N-terminal amino acid sequencing and tandem mass spec-
trometry analysis of these proteins allowed the identification and localization of
their corresponding genes in the available genomic sequence of A. ferrooxidans
ATCC 23270. The genomic context around several of these genes suggests their
involvement in the energy metabolism of A. ferrooxidans. Two groups of proteins
could be distinguished. Proteins highly upregulated by growth on sulfur compounds
but downregulated by growth on ferrous iron are particularly interesting, such as a
44-kDa outer-membrane protein, an exported 21-kDa putative thiosulfate sulfur
transferase protein, a 33-kDa putative thiosulfate/sulfate binding protein, and a
45-kDa putative capsule polysaccharide export protein (WcbC). Polysaccharides
may play a role in the adherence capability of bacteria to solid surfaces. In this
regard, it is known that most leaching bacteria grow attached to the surface of the
solid substrates such as elemental sulfur and metal sulfides.
On the other hand, A. ferrooxidans proteins that are downregulated when grow-
ing on sulfur but upregulated when growing on ferrous iron were also analyzed by
2D PAGE (Ramirez et al. 2004). These include rusticyanin, a cytochrome c552, a
putative phosphate binding protein (PstS), the small and large subunits of ribulose-1,
5-bisphosphate carboxylase/oxygenase (Rubisco), and a 30-kDa putative CbbQ
protein, amongst others. These results suggested a separation of the iron and sulfur
utilization pathways. Rusticyanin in addition of being highly expressed on ferrous
iron was also newly synthesized as determined by metabolic labeling, although
at lower levels during growth on sulfur compounds and iron-free metal sulfides
(Ramirez et al. 2004). These results were in agreement with those of Yarzabal et al.
(2004). The capacity of A. ferrooxidans to oxidize thiosulfate and tetrathionate was
found to be inhibited by the presence of ferrous iron (Das et al. 1993). However,
during the growth of A. ferrooxidans on iron-containing metal sulfides, such as
pyrite and chalcopyrite, we found elevated expression of proteins involved in both
ferrous iron and sulfur compound utilization, indicating that the two energy-
generating pathways are simultaneously induced depending on the type and the
concentration of the available oxidizable substrates (Ramirez et al. 2004). In agree-
ment with these results, it was previously suggested that A. ferrooxidans can simul-
taneously utilize both ferrous iron and elemental sulfur as energy sources (Espejo
and Romero 1987).
7.4 Thiosulfate Sulfur Transferases from A. ferrooxidans
Rhodanese activity has been previously reported in A. ferrooxidans (Tabita et al.

1969). This enzyme is a thiosulfate:cyanide sulfur transferase (TST), which breaks
the SS bond present in thiosulfate, generating sulfur and sulfite. Other enzymes may
also participate in the mechanism proposed by Schippers and Sand (1999), such as
the thiosulfate-oxidizing enzyme of A. ferrooxidans (Silver and Lundgren 1968b).
Fig 7.1 Possible cellular localization and in vitro thiosulfate:cyanide sulfur transferase activities
of rhodanese-like proteins from Acidithiobacillus ferrooxidans. The genes coding for proteins
P14, P15, P16, P16.2, and P21 were cloned and expressed in Escherichia coli and the thiosulfate:
cyanide sulfur transferase activities were determined. Activities were in the low (+) to high (+++)
ranges or were absent (). The presence of signal peptide in P21 is also indicated (tilde)
Recently, new rhodanese-like proteins were identified, the expression of which

is regulated depending on the growth substrate and is probably related to sulfur
metabolism and/or oxidation (Acosta et al. 2005). Eight nucleotide sequences
containing a single rhodanese domain are present in the genome of A. ferrooxidans
ATCC 23270 (Fig. 7.1): p11, p14, p14.3, p15, p16, p16.2, p21, and p28 (Acosta
et al. 2005). Amino acid sequence comparisons of all eight proteins allowed us to
identify the potential catalytic cysteine residues and other highly conserved rhoda-
nese family features. The genomic contexts of some of the rhodanese-like genes
suggested their implication in sulfur oxidation and metabolism, formation of FeS
clusters and detoxification mechanisms. Several of the putative rhodanese genes
were successfully isolated, cloned, and overexpressed in Escherichia coli and
their TST and 3-mercaptopyruvate:cyanide sulfur transferase (MST) activities
were determined. On the basis of their sulfur transferase activities and structural
comparisons of catalytic sites and electrostatic potentials between homology-
modeled A. ferrooxidans rhodaneses and the reported crystal structures of E. coli
GlpE (TST) and SseA (MST) proteins, two of the rhodanese-like proteins, P15
and P16.2, could clearly be defined as TSTs, and P14 and P16 could possibly
correspond to MSTs (Acosta et al. 2005). Nevertheless, several of the eight A.
ferrooxidans rhodanese-like proteins may have some different functional activi-
ties yet to be discovered.
The comparison of messenger RNA (mRNA) abundance of some of the genes
involved in sulfur metabolism in A. ferrooxidans grown on different oxidizable
substrates was started by performing a preliminary pilot DNA macroarray formed
Fig. 7.2 Analysis of cotranscription of modA2 and doxDA2 genes of A. ferrooxidans by reverse-
transcription (RT) PCR. A Map of the positions of the genes in the genomic context of p21.
Arrows indicate locations of primers used for the RT and PCR reactions. B Agarose gel electro-
phoresis of RT-PCR products. The primers used in each reaction are numbered. The RT reaction
was carried out on 3 g of total RNA obtained from thiosulfate-grown cells of A. ferrooxidans. RT
reactions with (+ lanes) and without ( lanes) the Moloney murine leukemia virus reverse tran-
scriptase enzyme were carried out in order to exclude amplification due to genomic DNA con-
tamination. A control with genomic DNA was also included (C). Sizes of DNA markers are shown
on the left of each gel. Expected sizes (in base pairs) for the corresponding RT-PCR products are
given at the bottom of the gels
with 70 different genes (Acosta et al. 2005). As already mentioned, the gene p21
codes for a putative thiosulfate sulfur transferase protein and all putative genes
upstream of it have been found to form a cluster (Ramirez et al. 2002, 2004). They
were all highly expressed in cells grown on sulfur compared with the levels seen on
ferrous iron (Acosta et al. 2005). This clearly supports our previous proposal based
on proteomic analysis that the rhodanese-like gene p21 forms part of a group of genes
related with sulfur oxidation (Ramirez et al. 2002). In addition, the DNA macroarray
results obtained for p21 were validated by our previous reverse-transcription PCR
studies, indicating the induced expression of p21 by growth on sulfur compounds
both at the transcriptional (Acosta et al. 2005) and at the translational levels (Ramirez
et al. 2002, 2004).
Unlike cytoplasmic rhodaneses, P21 was located in the periphery of A. ferrooxidans
cells (Fig. 7.1) and was regulated depending on the oxidizable substrate. If P21 and
some of the proteins coded by its adjacent genes (Fig. 7.2a) are involved in thiosulfate
metabolism, one should expect an increased expression of these proteins when the
cells are grown on pyrite, thiosulfate, or sulfur, as we have observed by proteomics
(Ramirez et al. 2002). However, we could not detect an in vitro TST activity for puri-
fied P21 (Ramirez et al. 2004). Protein P21 may not be a periplasmic rhodanese
enzyme but rather part of a possible complex in charge of thiosulfate oxidation.
This putative complex could be different from the Sox model proposed for sulfur
oxidation in many bacteria (Friedrich 1998) since we did not find any sox-like
genes in the genome of A. ferrooxidans (Ramirez et al. 2004).
7.5 Other Proteins Involved in Sulfur Metabolism
Genes modA1 and modA2 form a part of the genomic context of p21 (Fig. 7.2a).
Proteins ModA1 and ModA2 may be a part of the ATP-binding cassette (ABC)
superfamily of transporters (Self et al. 2001). These transport systems are involved
in both uptake and efflux and have different substrate specificities. Structural mod-
eling of ModA proteins with crystal structures of known similar proteins strongly
suggests a conserved functional mechanism for the transport of thiosulfate/sulfate
or molybdate in A. ferrooxidans. Previously, we found that modA1 is cotranscribed
with p21 and a putative thiosulfate:quinone oxidoreductase (TQR) (doxDA1)
(Ramirez et al. 2004). modA1 is upregulated both in sulfur and thiosulfate, whereas
the mRNA of ModA2 is expressed only in cells grown on thiosulfate and it is not
cotranscribed with doxDA2 (Fig. 7.2b).
By expression proteomics (Fig. 7.3) it is clear that ModA2 is synthesized in
much higher amounts in cells grown on thiosulfate. On the other hand, this protein
appears entirely repressed in cells grown on sulfur or iron. The almost absent
expression of ModA1 and ModA2 on ferrous iron suggests that these putative
Fig. 7.3 ModA2 is synthesized in higher levels in thiosulfate-grown A. ferrooxidans. A Total

proteins from A. ferrooxidans ATCC 23270 grown in thiosulfate were separated by 2D sodium
dodecyl sulfatepolyacrylamide gel electrophoresis (PAGE). A rectangular section containing
protein ModA2 is indicated. B The sections where ModA2 migrates in 2D PAGE are compared
for the total proteins of A. ferrooxidans grown in thiosulfate (T), ferrous iron (Fe), or elemental
sulfur (S). The proteins were separated by 2D nonequilibrium pH gel electrophoresis with a pH
gradient between 3.0 (right side of the gel) and 10.0 (left side of the gel). Spots were detected by
silver staining and analyzed by Delta 2D version 3.3 software. The positions of spots corresponding
to ModA2 are indicated by circles. Molecular mass standards (in kilodaltons) are seen on the left
of the gels. Numbers on the right of the sections indicate the relative intensity volumes of the
circled spots with respect to all detected spots in the corresponding gels
Table 7.1 Real-time PCR quantitation of the levels of

gene expression of doxDA1 and doxDA2 in cells
grown in different oxidizable substrates
Ratio doxDA1 doxDA2
Sulfur/Fe(II) 2.5 3.8
Thiosulfate/Fe(II) 6.7 3.3
Pyrite/Fe(II) 10.8 6.7
Unpublished results. The experiments were performed
twice with both independent total RNA and comple-
mentary DNA preparations. Values shown are the
average of these experiments.
transporters may function more likely as periplasmic thiosulfate or molybdate bind-

ing proteins in this acidophile.
A. ferrooxidans harbors duplicated doxDA genes (Fig. 7.2a) that are homologous
to the genes encoding TQR in A. ambivalens, an enzyme oxidizing thiosulfate with
tetrathionate as a product and ferricyanide or decylubiquinone as electron acceptors
(Mller et al. 2004). As pointed out very recently by Friedrich et al. (2005), this
gene duplication points to a yet undemonstrated significance in thiosulfate metabo-
lism in A. ferrooxidans. doxDA1 and doxDA2 are in the same genomic context of
p21 (Fig. 7.2) and as seen in Table 7.1, their expression determined by real-time
PCR is enhanced several fold in cells grown on elemental sulfur or on thiosulfate
compared with the expression levels of cells grown on ferrous iron. Similar behavior
was seen by DNA macroarray analysis in the case of doxDA1 (Acosta et al. 2005).
The upregulated expression of the doxDA genes may be of particular relevance
during the degradation of the mineral pyrite, where much higher levels of transcription
are seen (Table 7.1). In agreement with these results, thiosulfate is an important
intermediate during pyrite oxidation (Schippers and Sand 1999). These results
support the significance of the two doxDA genes in thiosulfate metabolism in
A. ferrooxidans and are in favor of the idea that acidophilic bacteria oxidize sulfur
compounds by a system different from the Sox enzyme system.
7.6 High-Throughput Proteomics of Periplasmic Proteins

Induced by Growth of A. ferrooxidans on Sulfur Compounds
Several of the proteins involved in sulfur and iron oxidation have been described as
forming part of the periplasm of A. ferrooxidans; therefore, to further study some
of the components involved in sulfur metabolism, differential proteomic analysis of
total periplasmic proteins was performed using high-resolution LTQ FT ion trap
mass spectrometry.
We identified 216 proteins in the periplasm of A. ferrooxidans ATCC 23270,

several of them changing their levels of synthesis when growing the bacterium on
thiosulfate, elemental sulfur, and ferrous iron media (Fig. 7.4). Thirty-four percent
of them corresponded to unknown proteins. Forty-one proteins were exclusively
present in sulfur-grown cells and 14 in thiosulfate-grown cells (unpublished results).
The putative genes coding for all the proteins were localized in the available
genomic sequence of A. ferrooxidans ATCC 23270. The genomic context around
several of these genes suggests their involvement in sulfur metabolism and
possibly in sulfur oxidation and formation of FeS clusters. Some of the periplas-
mic proteins were upregulated by growth on sulfur compounds, for example, the
exported P 21, the ModA1 and ModA2 putative transporters, and a sulfide:quinone
oxidoreductase.
Fig. 7.4 Distribution of periplasmic proteins from

cells of A. ferrooxidans ATCC 23270 grown in different
oxidizable substrates
Fig. 7.5 A speculative working model for sulfur oxidation in A. ferrooxidans. Proteins resulting
from our studies are shaded. (Data taken and adapted from Rawlings 2005 and Rohwerder and
Sand 2003)
A summary of our findings and those from other researchers are put together in
a speculative working model for sulfur-compound oxidation in A. ferrooxidans
(Fig. 7.5). Many of the proteins changing their expression during growth in sulfur
compounds may have important roles yet to be described in the sulfur metabolism
of this acidophilic microorganism.
7.7 Conclusions
Proteomics is a powerful tool for the study of differential expression of

microorganisms such as A. ferrooxidans.
Acidophilic chemolithoautotrophs such as A. ferrooxidans do not possess a Sox
system, and therefore should oxidize thiosulfate by other means.
During the oxidation of a metal sulfide such as pyrite, the two doxDA genes of
A. ferrooxidans are upregulated, in agreement with the proposed thiosulfate
mechanism for pyrite oxidation.
A. ferrooxidans has a thiosulfate dehydrogenase activity. It remains to be
demonstrated whether this activity belongs to the DoxDA proteins.
References
Acosta M, Beard S, Ponce J, Vera M, Mobarec JC, Jerez CA (2005) Identification of putative
sulfurtransferase genes in the extremophilic Acidithiobacillus ferrooxidans ATCC 23270
genome: structural and functional characterization of the proteins. OMICS 9:1328
Das A, Mishra AK, Roy P (1993) Inhibition of thiosulfate and tetrathionate oxidation by ferrous
iron in Thiobacillus ferrooxidans. FEMS Microbiol Lett 112:6772
De Jong GAH, Hazeu W, Bos P, Kuenen G (1997) Polythionate degradation by tetrathionate
hydrolase of Thiobacillus ferrooxidans. Microbiology 143:499504
Espejo RT, Romero P (1987) Growth of Thiobacillus ferrooxidans on elemental sulfur. Appl
Environ Microbiol 19071912
Friedrich CG (1998) Physiology and genetics of sulfur-oxidizing bacteria. Adv Microb Physiol
39:235289
inorganic sulfur compounds by bacteria: emergence of a common mechanism? Appl Environ
Gygi SP, Corthals GL, Zhang Y, Rochon Y, Aebersold R (2000) Evaluation of two-dimensional
gel electrophoresis-based proteome analysis technology. Proc Natl Acad Sci USA 97:
93909395
Harrison AP (1984) The acidophilic Thiobacilli and other acidophilic bacteria that share their
habitat. Annu Rev Microbiol 38:26592
J. Craig Ventner Institute (2007) The new JCVI. http://www.tigr.org. Cited 16 Jan 2007
Kelly DP, Shergill JK, Lu W-P, Wood AP (1997) Oxidative metabolism of inorganic sulfur
compounds by bacteria. Antonie Van Leeuwenhoek 71: 95107
Lundgren DG (1980) Ore leaching by bacteria. Annu Rev Microbiol 34:263283
Mller FH, Bandeiras TM, Urich T, Teixeira M, Gomes CM, Kletzin A (2004) Coupling of the
pathway of sulphur oxidation to dioxygen reduction: characterization of a novel membrane-
bound thiosulphate:quinone oxidoreductase. Mol Microbiol 53:11471160
Olson GJ, Brierley JA, Brierley CL (2003) Bioleaching review part B: progress in bioleaching:
applications of microbial processes by the minerals industries. Appl Microbiol Biotechnol
63:249257
Ramirez P, Toledo H, Guiliani N, Jerez CA (2002) An exported rhodanese-like protein is induced
during growth of Acidithiobacillus ferrooxidans in metal sulfides and different sulfur com-
pounds. Appl Environ Microbiol 68:18371845
Ramirez P, Guiliani N, Valenzuela L, Beard S, Jerez CA (2004) Differential protein expression
during growth of Acidithiobacillus ferrooxidans on ferrous iron, sulfur compounds, or metal
sulfides. Appl Environ Microbiol 70:44914498
Rawlings DE (2002) Heavy metal mining using microbes. Annu Rev Microbiol 56:6591
Rawlings DE (2005) Characteristics and adaptability of iron- and sulfur-oxidizing microorganisms
used for the recovery of metals from minerals and their concentrates. Microb Cell Fact 4:13
Rohwerder T, Sand W (2003) The sulfane sulfur of persulfides is the actual substrate of the sulfur-
oxidizing enzymes from Acidithiobacillus and Acidiphilium spp. Microbiology149:16991709
Rohwerder T, Gehrke T, Kinzler K, Sand W (2003) Bioleaching review part A: progress in
bioleaching: fundamentals and mechanisms of bacterial metal sulfide oxidation. Appl Microbiol
Biotechnol 63:239248
Sand W, Gehrke T, Hallmann R, Schippers A (1995) Sulfur chemistry, biofilm, and the (in)direct
attack mechanism a critical evaluation of bacterial leaching. Appl Microbiol Biotechnol
43:961966
Sand W, Gehrke T, Jozsa PG, Schippers A (2001) (Bio)chemistry of bacterial leaching-direct vs.
indirect bioleaching. Hydrometallurgy 59:159175
Schippers A, Sand W (1999) Bacterial leaching of metal sulfides proceeds by two indirect mecha-
nisms via thiosulfate or via polysulfides and sulfur. Appl Environ Microbiol 65:319321
Self WT, Grunden AM, Hasona A, Shanmugam, KT (2001) Molybdate transport. Res Microbiol
152:311321
Silver M, Lundgren DG (1968a) Sulfur-oxidizing enzyme of Ferrobacillus ferrooxidans
(Thiobacillus ferrooxidans). Can J Biochem 46:457461
Silver M, Lundgren DG (1968b) The thiosulfate-oxidizing enzyme of Ferrobacillus ferrooxidans
Sugio T, Mizunashi W, Inagaki K, Tano T (1987) Purification and some properties of sulfur:ferric
ion oxidoreductase from Thiobacillus ferrooxidans. J. Bacteriol 169:49164922
Suzuki I (1999) Oxidation of inorganic sulfur compounds: chemical and enzymatic reactions. Can
J Microbiol 45:97105
Suzuki I (2001) Microbial leaching of metals from sulfide minerals. Biotechnol Adv 19:
119132
Tabita R, Silver M, Lundgren DG (1969) The rhodanese enzyme of Ferrobacillus ferrooxidans
Valenzuela L, Chi A, Beard S, Orell A, Guiliani N, Shabanowitz J, Hunt DF, Jerez CA (2006)
Genomics, metagenomics and proteomics in biomining microorganisms. Biotechnol Adv
24:197211
Yarzabal A, Appia-Ayme C, Ratouchniak J, Bonnefoy V (2004) Regulation of the expression of
the Acidithiobacillus ferrooxidans rus operon encoding two cytochromes c, a cytochrome
oxidase and rusticyanin. Microbiology 150:21132123
Chapter 8
Sulfur and Light? History and Thiology
of the Phototrophic Sulfur Bacteria
Hans G. Trper
Abstract This chapter describes how our present knowledge of sulfur metabolism
of phototrophic sulfur bacteria accumulated through several major steps of
experimental progress. Among these are the following: discovery of microbial cells
with conspicuous inclusions (purple bacteria and colorless ones); verification that
such inclusions consist of sulfur; detection of phototactic behavior in purple bacte-
ria; enrichment cultures (Winogradsky columns), consequent detection of sulfide
requirement and of the liquid stage of sulfur inclusions, globules; identification
of the pigments as bacteriochlorin (now chlorophylls) and bacterioerythrin (now
carotenoids); discovery of Chlorobium, a green bacterium that deposits sulfur glob-
ules outside its cells; discovery of photosynthesis in purple non-sulfur bacteria;
postulation of a photosynthetic metabolism in purple sulfur bacteria by combina-
tion of photosynthesis and chemosynthesis; evidence that the red (purple) and
green sulfur bacteria perform anaerobic photosynthesis (carbon dioxide fixation)
dependent on the oxidation of reduced sulfur compounds; discovery of
Ectothiorhodospira, a halophilic phototrophic purple sulfur bacterium that deposits
sulfur globules outside its cells; systematic development of enrichment and pure
culture media and techniques for the Chromatiaceae and Chlorobiaceae; the first
specific studies on sulfur metabolism in purple bacteria on whole cells and crude
extracts; purification of adenosine 5-phosphosulfate reductase and reverse siroheme
sulfite reductase; finding and purification of sulfide quinone reductase; discovery
and characterization of the dsr gene cluster in Allochromatium vinosum; isolation
and gene sequence of the sulfur globule encoating periplasmic proteins in
Chromatiaceae; determination of the inner structure of the sulfur globules by X-ray
absorption near-edge spectroscopy in phototrophic and chemotrophic sulfur
bacteria; characterization of sulfite oxidoreductase; characterization of the thiosul-
fate-oxidizing (Sox) multienzyme complex in Allochromatium vinosum.
87
Springer 2008
88 H.G. Trper
8.1 Introduction
The chemical element sulfur (L. sulfur or sulfur, Gr. thion) as a material that natu-
rally occurs in volcanic areas, in calcite and gypsum deposits and at certain
beaches has been known to humans since prehistoric times and as it burns with
an unusual blue flame and a stinging smell has been ascribed magic powers as
well as connections to devilish underground spirits and hell. Only since 1870 has
it been known that considerable amounts of elemental sulfur may occur in living
beings as well.
8.2 Discovery of Sulfur-Oxidizing Microorganisms
Since 1786 (Mller 1786, in Denmark) several microorganisms (then considered

either animals or plants, animalcula or infusoria) have been described to contain
intracellular conspicuous light-refringent granules of unknown composition.
Christian Gottfried Ehrenberg (1838) described the discovery of the first purple
sulfur bacteria Monas okenii and Ophidomonas jenensis in 1836 with the fol-
lowing words (here translated into English):
In 1836, 18 September, the day of the constitutional opening of the 14th meet-
ing of the German Association of Naturalists (in Jena), founded by (Ludwig)
Oken, during an excursion that I had undertaken with Professor Weisse I found
somewhat below the church in Ziegenhain this beautifully red monad in consid-
erable amounts in a small basin of the village creek. It formed handwide red
patches, and between its legions in considerable numbers Ophidomonas jenensis
occurred, a new genus of armored monads, together with Euglena viridis and
Spirogyra. During favorable development this form can easily cause a very inten-
sive blood color in the stagnant water. The animalcules collected on 18 September
in a little bottle, which I demonstrated to the zoological section of the naturalists
convention in Jena, still existed in small numbers on 11 December in Berlin, and
while I am writing this I have them right beside me together with Ophidomonas
alive under the microscope.
After Ehrenbergs description of the two organisms was published in 1838, in the
following years further forms were described. The Italian botanist Trevisan (1842)
described the large filamentous Beggiatoa with similar conspicuous inclusions
occurring in the sulfur springs of the Euganean Hills west of Padua. The Swiss bota-
nist Maximilian Perty (1852) introduced the genus name Chromatium and described
the species Chromatium vinosum, C. weissei, C. violascens and C. erubescens, how-
ever, without making attempts to isolate them or to study their nutritional physi-
ology. He just called them pigment bacteria. The British biologist Ray
Lankester (1873) described the reddish microbial mats on the mud surface in
ponds as Bacterium rubescens and for the first time classified them as Bacteria;
he called the characteristic red pigment bacteriopurpurin. The Danish botanist
8 Sulfur and Light? History and Thiology of the Phototrophic Sulfur Bacteria 89
Eugen Warming (1875) consequently coined the term purple bacteria for these
organisms and added further new species. The first scientist who started to
experiment beyond the predominantly simple descriptions was Theodor W.
Engelmann (1883), who discovered and described in 1883 the light-excitation
reactions of purple bacteria. Working with Bacterium photometricum (probably
Allochromatium vinosum) crude cultures, he found that these bacteria assemble
in the light in the so-called Engelmanns light trap, thus finding a behavior simi-
lar to that of green algae, where he had found a correlation between phototaxis
and photosynthesis before (Drews 2005). At that time to biologists oxygen pro-
duction in the light meant photosynthesis. Although he believed in it, Engelmann
was not able to prove oxygen production in purple bacteria in a convincing
manner.
8.3 Identification of Conspicuous Inclusions as Sulfur
It had also been recognized that some of the microbes containing the conspicuous
inclusions were purple or red in color, while others were not (Table 8.1). The
Swiss botanists Cramer and Meyer-Ahrens in 1870 (Mller 1870) and Ferdinand
Cohn (1872, 1875), the German botanist founder of bacterial systematics, in 1872
and in 1875 were the first who unequivocally proved that such cellular inclusions
of Beggiatoa and of the colored microbes, respectively, consisted of elemental
sulfur. Cramer and Meyer-Ahrens spoke of granules and considered their
strong refraction as a sign for their solid state, while Cohn spoke of granules and
crystals. Cohn considered the occurrence of elemental sulfur as a singular phe-
nomenon in the plant world and although studies of environmental samples in
the laboratory by himself as well as by Sergei N. Winogradsky (1887) showed
that these organisms depended on sulfide in their aquatic medium both hypotheses
brought forward for this dependence could not really be proven: on one hand it
was believed that these organisms reduced sulfate under production of hydrogen
sulfide and elemental sulfur; on the other it was thought that they oxidized hydro-
gen sulfide to elemental sulfur. Winogradsky was the first to realize the nonsolid
nature of the sulfur inclusions and called them globules. He observed that only
in dead cells the sulfur transformed into crystals. Much later (Winogradsky 1949)
he wrote that it was easy to prove that the inclusions in Beggiatoa cells are not in
a solid stage: En effect il est facile de demontrer quelles sont de consistance
sirupeuse en chauffent des filaments bourrs de soufre a 70 dans un peu deau
toute les inclusions qui remplissent la cellule des filaments ne tardent pas alors
confluer en formant une seule grosse goutte par cellule. As a consequence one
should not speak of granules but of droplets or better globules of sulfur.
Actually it is the global shape that causes the strong light refringence, as one can
easily observe by comparing them with microscopically small air or gas bubbles.
Winogradsky considered the metabolism of purple sulfur bacteria as a type
of chemolithotrophy like in Beggiatoa, where he had observed the oxidation of
90 H.G. Trper
Table 8.1 Genera of conspicuous or prominent colored and colorless sulfur bacteria. (List
incomplete after 1900. Data from Buchanan and Gibbons 1974)
1786 Animalcula infusoria O.F. Mller = Monas (muelleri) 1875 Warming = Thiovulum
muelleri (=majus) 1913 Hinze (CSB)
1833 Micraloa Ktzing = Lamprocystis 1886 Schroeter (PSB)
1838 Ophidomonas Ehrenberg = Thiospirillum 1888 Winogradsky (PSB)
1838 Monas (okenii) Ehrenberg = Chromatium 1852 Perty (PSB)
1842 Beggiatoa Trevisan = Beggiatoa (CSB)
1865 Beggiatoa (nivea) Rabenhorst = Thiothrix 1888 Winogradsky (CSB)
1887 Spirillum (rubrum) Esmarch = Rhodospirillum 1907 Molisch (PB)
1888 Thiocystis Winogradsky (PSB)
1888 Amoebobacter Winogradsky (PSB)
1888 Thiodictyon Winogradsky (PSB)
1888 Thiothece Winogradsky (PSB)
1888 Thiopedia Winogradsky (PSB)
1888 Thiocapsa Winogradsky (PSB)
1888 Rhabdochromatium Winogradsky (PSB)
1888 Thiopolycoccus Winogradsky (PSB)
1888 Thiosarcina Winogradsky (PSB)
1893 Achromatium (oxaliferum) Schewiakoff (CSB)
1902 (Thiobacillus thioparus) Nathanson; 1904 Thiobacillus Beijerinck (CSB)
1905 Thiospirillum winogradskyi Omelianski = Thiospira 1914 Visloukh (CSB)
1906 Chlorobium Nadson (green PSB)
1907 Thioploca Lauterborn (CSB)
1912 Bacterium (bovista) Molisch = Thiobacterium 1924 Janke (CSB)
1915 Achromatium (mobile) Lauterborn = Macromonas 1924 Utermhl, Koppe
1936 Ectothiorhodospira Pelsh (PSB)
1999 Thiomargarita Schulz et al. (CSB)
Entries given as year of description, name of organism, name of author and subsequent
nomenclatural changes.
CSB chemolithotrophic sulfur bacterium, PSB phototrophic sulfur bacterium, PB phototrophic
non-sulfur bacterium.
hydrogen sulfide to sulfur and sulfate in the dark under consumption of molecular
oxygen. He realized that in purple bacteria this process apparently occurred
under anaerobic conditions, and concluded that the necessary oxygen would
apparently be provided by a light-dependent splitting of water. In principle,
however, he considered the metabolism of colorless bacteria and that of purple
sulfur bacteria to be identical. By 1904 it had become clear through the work
of Nadson (1903) and Arcichowskij (1904) that the pigments of the purple bac-
teria consisted of two different types, a green component, then called bacterio-
chlorin (today called chlorophylls) and a red component, then called bacterioerythrin
(today called carotenoids).
Still these days, large-cell bacteria with conspicuous sulfur globule inclusions
may be newly found, as the discovery of the rather huge bacterium Thiomargarita
by Schulz et al. (1999) showed.
8.4 Enrichment Cultures First Taxonomy and the Question

of Photosynthesis
An important step forward was the introduction of enrichment cultures in glass

cylinders; these were only much later called Winogradsky columns after their
inventor. These allowed the maintenance of enrichment cultures of colored sulfur
bacteria in the laboratory over longer periods of time.
After Erwin von Esmarch (1887) had established the first pure culture of a
Spirillum rubrum, Hans Molisch performed a large study on purple bacteria,
included this organism in his studies and renamed it Rhodospirillum. Molisch
applied varieties of Winogradsky colums to enrich so-called non-sulfur purple bac-
teria as well. He discovered, isolated and described a larger number of new genera
and species of purple bacteria, however only of purple non-sulfur bacteria (that he
called Athiorhodaceae), which as he found out assimilate organic substances in
the light. He had detected a new type of photosynthesis which we now call photoor-
ganoheterotrophy, and ascribed the two pigments roles similar to those of the chlo-
rophyll and the carotenoids in carbon dioxide assimilation of green plants (Molisch
1907). But Molisch never studied purple sulfur bacteria!
Summarizing his studies on sulfur bacteria, Winogradsky (1887) produced a
paper of 73 pages reporting numerous from our present view simple little experi-
ments and observations, which were helpful to develop a first view on these organ-
isms. In this paper he also proposed a first taxonomic scheme for the sulfur bacteria
(Sulfobacteria), which he simply subdivided into colorless and red sulfobacteria, the
former group represented by the two genera Beggiatoa and Thiothrix, the latter by
the 11 genera Thiocystis, Lamprocystis, Amoebobacter, Thiopolycoccus, Thiodictyon,
Thiothece, Thiocapsa, Thiopedia, Chromatium, Rhabdochromatium and
Thiospirillum. The differentiating criteria were, as was the tradition of botanists and
zoologists, only morphological characters.
Johannes Buder (1919) confirmed Engelmanns theories that the purple bacteria
perform a photosynthetic metabolism. He argued as follows. The dependence of
these organisms upon organic substances or upon hydrogen sulfide cannot be
denied. Living under anaerobic conditions, they are forced to produce the required
oxygen themselves. The possession of photosynthesis pigments thus makes possible
that an oxidation of organic substances is also possible under anaerobic conditions.
Buder made clear that Engelmann, Winogradsky and Molisch had been working
with different organisms. He realized that purple bacteria represented a new type of
metabolism: They assimilate carbon dioxide or organic compounds anaerobically in
the light. He thus united Engelmanns photosynthesis and Winogradskys
chemosynthesis.
The definite proof for the phototrophic nature of the purple and green sulfur
bacteria, however, as well as the similarities with and the differences from plant
photosynthesis came in 1931 from the Dutch American microbiologist Cornelis B.
van Niel, working in Pacific Grove, CA, USA. Using numerous illuminated cul-
tures in closed bottles under the absence of oxygen, van Niel in long-term
92 H.G. Trper
experiments established the stoichiometric relationship between hydrogen sulfide

oxidation, sulfur and sulfate formation, disappearance of carbonate from the medium
and increase in cellular carbon. He realized the surprising similarity with the material
turnover in green plant photosynthesis and established the generalized stoichiometric
photosynthesis equation:
CO2 + 2H2 A light

CH2 O + H2 O + 2 A
For the phototrophic sulfur bacteria the reducing agent H2A is hydrogen sulfide (or
sulfur or thiosulfate) and for the photosynthesis of plants it is water, with the oxida-
tion products molecular sulfur (and/or sulfate) and molecular oxygen, respectively.
Using water labeled with the oxygen isotope 18O, Ruben et al. (1941) proved that
in plant photosynthesis the O2 formed is indeed derived from water. The similarities
between the purple sulfur and non-sulfur bacteria were further strengthened by the
finding that species of both groups were able to use molecular hydrogen as the H2A
of the general photosynthesis equation (Roelofsen 1934; Gaffron 1935).
8.5 Pure Cultures of Phototrophic Sulfur Bacteria at Last!
Until the 1960s most of the research on anaerobic phototrophic bacteria aimed at
elucidating the mechanisms of photosynthesis, carbon dioxide fixation, pigment
synthesis, phototaxis, dark metabolism and diversity of metabolic physiology were
done with the more easily cultivable purple non-sulfur bacteria Rhodospirillum
rubrum, Rhodobacter capsulatus and Rhodobacter sphaeroides, or with more or
less alleged pure cultures of Allochromatium (then Chromatium) vinosum. Since
1953 also the green sulfur bacteria (genus Chlorobium) had been cultivated in pure
cultures and studied thoroughly by Helge Larsen (1953).
The great breakthrough came when Norbert Pfennig in the early 1960s first
improved Winogradsky columns (by quantification of ingredients in the mud, and
preincubation in the dark), from which he step by step deduced an optimal growth
medium for red and green phototrophic sulfur bacteria, Pfennigs medium (consisting
of three or four separately sterilized solutions), and introduced the technique of feed-
ing with neutralized sulfide (this was necessary because sulfide tolerance turned out
to be limited even in these sulfide-requiring bacteria), screw-capped bottles (the
Pfennig bottle) instead of the glass-stoppered ones of van Niel, which easily got
contaminated, agar shake dilution series, capillary isolation of single large cells, regu-
lar routine controls for contaminating sulfate-reducing and heterotrophic anaerobic
bacteria and many other tricks. The large-cell purple sulfur bacteria described by
Ehrenberg, Perty, Cohn and Winogradsky became cultivable and Pfennigs medium
also turned out to be optimal for the green sulfur bacteria. So Pfennig isolated one
after the other of these old literature bacteria, and proved, how careful the old sci-
entists had perceived and studied these organisms (Pfennig and Trper 1989).
Besides the feeding technique the secret of his medium was that he based it on a
delicate bicarbonate buffering system and that the large-cell species as well as many
green bacteria depended upon the availability of vitamin B12, which in the Winogradsky
columns had been provided by the anaerobic prokaryotes present in the mud (Schlegel
and Pfennig 1961; Pfennig 1961, 1962; Trper 1970).
Practically all the following important research on phototrophic sulfur bacteria
owes homage to Pfennig. By inventing the methods for handling these fastidious
bacteria he practically opened a barn door for us, like Robert Hungate and Ralph
Wolfe did for the strict anaerobic rumen bacteria and methanogens, Roger Stanier
for the Cyanobacteria, Roger Whittenbury for the methylotrophs, Karl Stetter for
the hyperthermophiles and Fritz Widdel for the sulfate reducers.
I am personally glad that I was an eyewitness and coworker during these fruitful
years of Pfennig in Gttingen. After that I spent most of my research life until
retirement working together with many of my students to elucidate the sulfur
metabolism of purple (including Ectothiorhodospira) and green phototrophic sulfur
bacteria even by taking detours or side interests such as assimilatory pathways in
yeasts, sulfur and non-sulfur purple bacteria as well as dissimilatory sulfur metabo-
lism in the anaerobic Thiobacillus denitrificans and in sulfate reducers like
Desulfovibrio and extreme thermophilic Archaeoglobus and Pyrobaculum species.
Often the choice of another model organism helped to overcome a dead end in our
research on phototrophic sulfur bacteria. In those years the progresses in sulfur
metabolism research no matter whether on oxidative or reductive pathways were
due to rapid developments in enzymology, protein chemistry and radioactive labe-
ling techniques besides more reliable analytical chemistry of sulfur compounds.
8.6 The Age of Enzymology and Isotope Labeling
The book The Biochemistry of Inorganic Compounds of Sulfur by Roy and Trudinger
(1970) presented the state of the art at that time. The bibliography in that book reveals
that until it appeared, there existed several highly active groups working on dissimila-
tory sulfur metabolism in thiobacilli (sensu lato) or sulfate-reducing bacteria, while
only a few fighters had taken on the phototrophic sulfur bacteria. Some leading labo-
ratories on sulfur metabolism in thiobacilli then were those of M.I.H. Aleem, J.P.
Aubert, W.P. Hempfling, D.P. Kelly, H. Lees, M. Okizumi, W. Ostrowski, H.D.
Peck, S.C. Rittenberg, R.L. Starkey, I. Suzuki, P.A. Trudinger, W.W. Umbreit and
W. Vishniac and on dissimilatory sulfur metabolism in sulfate reducers were those of
J.B. Adams, J.M. Akagi, L.L. Campbell, C. Furusaka, M. Ishimoto, J. Le Gall, H.D.
Peck and J.R. Postgate. The few fighters on phototrophic sulfur bacteria came from
only three laboratories: Arnold Smith from June Lascelles laboratory, Thiele and I
from Pfennigs laboratory and Harry D. Peck, who had a big laboratory behind him and
great experience in the two other fields mentioned above.
I entered this field with a minor part of my doctoral thesis, repeating the long-term
stoichiometry experiments of van Niel (1931) with a pure culture of Chromatium
okenii in the form of short-term ones in a special vessel using radioactively labeled
94 H.G. Trper
carbon and sulfur compounds. Besides exactly confirming the stoichiometric

turnover relationships with this technique, I found that light-dependent anaerobic
oxidation of sulfide by Chromatium okenii depended upon the presence of car-
bon dioxide, indicating that a close coupling existed between sulfide oxidation
and photosynthetic carbon dioxide fixation (Trper 1964a, b; Trper and
Schlegel 1964).
We further proved by radioactive labeling that in the light thiosulfate in
Thiocapsa roseopersicina and Allochromatium (then Chromatium) vinosum is split
into the sulfane group, forming intracellular sulfur globules, and the sulfone group
that appears in the medium as sulfate (Trper and Pfennig 1966), confirming results
of Arnold Smith and June Lascelles with Allochromatium vinosum (Smith 1965,
1966; Smith and Lascelles 1966). Smith (1965) further isolated an enzyme from
that organism which catalyzes the oxidation of thiosulfate to tetrathionate.
After a first hint given by Peck (1961), Thiele (1966, 1968) in Gttingen and
Peck (1966) in Athens, GA, USA, independently reported the first finding of an
adenosine 5-phosphosulfate (APS) reductase activity in Chromatium and Thiocapsa
strains. Thiele quit the field of sulfur research soon after 1966, and I took the
chance to work in Pecks laboratory for a few weeks, during which we found APS
reductase activities in several green and purple phototrophic sulfur bacteria, but not,
however, in any of the non-sulfur purple bacteria tested. It also became clear that
this enzyme activity was originally membrane-bound and leached off at different
degrees in different organisms. It was practically impossible to get it into the clear
supernatant from Allochromatium vinosum strain D, while it easily leached off in
Thiocapsa roseopersicina (Trper and Peck 1970). The latter enzyme was purified
(Trper and Rogers 1971); the corresponding one from Chlorobium limicola
followed (Kirchhoff and Trper 1974).
We were facing an oxidative chain from sulfide to sulfate with at least sulfur,
sulfite and APS as intermediates. In addition, thiosulfate as a substrate was either split
into sulfur and (end-product) sulfate or oxidized to tetrathionate. All electrons set free
during these oxidations would go into photosynthesis, as the stoichiometry studies
had proven before.
Our aims then were clear but far away, and the way to get there was foggy. In
those years cells were defined as bags full of enzymes, although it was hard to
imagine that all these different proteins in addition to all intermediates, educts and
products, harmless as well as poisonous ones, should coexist in one compartment.
We spoke of soluble enzymes when they came out in the supernatant after ultracen-
trifugation and of particle-, membrane- or chromatophore-bound enzymes, when
they resided in the pellet. The former could be tested by combined optical enzyme
tests, the latter only with complicated substrate or product determinations including
radioactive labeling methods.
As we were concerned with oxidation reactions we had to find suitable electron
acceptors for testing before we could find the genuine ones.
As soon as the outer membrane and the periplasmic space of Gram-negative bacteria
had been discovered and the concept of the periplasm had been developed
(Mitchell 1961) and became known, we suddenly had to deal with two compartments
within which our reactions could take place in vivo. In addition we had to answer
the topological questions at which side of the cellular membrane such enzymes were
situated that we had localized in the insoluble fraction. This meant that the ultrastruc-
ture of the bacteria that are capable of storing sulfur globules inside their cells
became very important. Several studies proposed that the so-called chromatophores
in Chromatiaceae that obviously carried the photosynthetic pigments are invagina-
tions of the cellular membrane, thus forming a large interconnected intracellular
vesicular system, the so-called intracytoplasmic membrane system, the inside of
which topologically was part of the periplasm (Remsen 1978).
Working with cell extracts under normal, i.e., aerobic laboratory conditions, brought
up the question whether we would not measure competing reactions with oxygen. As
sulfide and sulfite are strong reductants they would easily interfere with many meta-
bolic redox reactions in the cell besides being undoubtedly toxic to metal-containing
and other enzymes. This meant that we also had to look for possible organic carrier
molecules that would mask such aggressive features. Another important question was
that of the exact chemical nature of the conspicuous sulfur globules. How could water-
insoluble elemental sulfur participate in biochemical reactions?
The first leap forward from that situation was the purification of a reverse siroheme-
containing sulfite reductase from both Thiobacillus denitrificans and Allochromatium
vinosum in 1979 by Michael Schedel (Schedel et al 1979; Schedel and Trper 1979).
Besides that we worked intensively on c cytochromes, flavocytochromes, ironsulfur
proteins, ATP sulfurylases and ADP sulfurylases. In 1986 Dan Brune (Brune and
Trper 1986) found the first evidence for the possible participation of sulfide quinone
reductase as a possible first step in sulfide oxidation. This line was successfully
pursued later by Hauska and Shahak (Shahak et al. 1999).
8.7 Advent of Molecular Genetics
As soon as we had learned the basic methods of molecular biology, Christiane Dahl
and her group brought Schedels results to a new high by discovering the dsr gene
cluster (Pott and Dahl 1998, Dahl et al. 2005; Chap. 9 by Grimm et al.). Kobchai
Pattaragulwanit in our laboratory developed a new method to use gene technology in
Allochromatium vinosum and thus came up with the then sensational finding that the
sulfur globules in Chromatiaceae are contained in envelopes consisting of two to
three types of proteins, the sulfur globule proteins (SGP), which are free of sulfhydryl
groups. In their genes they revealed, however, typical leader sequences characterizing
them as periplasmic proteins (Pattaragulwanit et al. 1998). Thus, the sulfur globules
are stored periplasmically in the intracytoplasmic membrane vesicles!
With cinematographic techniques we proved with fixed living cells that the intracel-
lular sulfur globules do not originate at the cellular membrane to be moved to the inner
part of the cell as postulated by Remsen (1978) but that they are formed at any
place in the cell (Herrmann 1984; Herrmann and Trper, unpublished data), which is
further proof for the continuity of the intracytoplasmic membrane system.
96 H.G. Trper
After a long period of cooperation with the inorganic chemist Ralf Steudel,
Berlin, on the nature of the sulfur in the globules (Steudel 1989), finally Alexander
Prange, for the first time employing X-ray absorption near-edge spectroscopy (at the
cyclotron of the University of Bonn Physics Department), succeeded in determining
the status of this sulfur in living cells of phototrophic as well as chemolithotrophic
sulfur bacteria (Prange et al. 2002; Chap. 20 by Prange).
Ulrike Kappler working with the enzyme sulfite oxidoreductase (sulfite dehy-
drogenase) as the alternative to the APS pathway had to switch from Allochromatium
vinosum to Starkeya (formerly Thiobacillus) novella before she succeeded (Kappler
et al. 2000, 2001; Chap. 13 by Kappler).
On the basis of path-breaking studies by Don Kellys group on aerobic thiobacilli
(Lu and Kelly 1983ac, 1988), Cornelius Friedrich and his coworkers studied the peri-
plasmic thiosulfate-oxidizing multienzyme (Sox) system in Paracoccus panthotrophus
(Friedrich et al. 2001, 2005, Chap. 12), a system that also exists in Allochromatium
vinosum and Thiocapsa sp., as was found by Dahls group (Hensen et al. 2006). In
Chromatiaceae this multienzyme system is probably not involved in sulfide
oxidation.
I leave the explanation of the present status of the art in this field to the next
generation. I am very happy and thankful that we have come so far through many
frustrating but also many highly exciting periods of work. As far as the work was
done in our laboratory I thank my ingenious diploma and doctoral students, post-
docs and coworkers, and those who will continue. I apologize for not having been
able to mention the merits of all the other colleagues who had and have their share
in the progress of sulfur metabolism research. To do that I would have needed about
some 20 lecture hours.
8.8 Further Reading
For further reading I recommend the proceedings of international scientific symposia,

workshops and meetings on microbial sulfur metabolism and related fields of the
last 30 years:
1974: Meeting on the sulfur cycle, Wageningen, The Netherlands (van Egeraat
and Huntjens 1975)
1979: Symposium on biology of sulfur, London, UK (Ciba Foundation
1980)
1979: Low molecular weight sulfur containing natural products, Rome, Italy
(Cavallini et al. 1980)
1979: SCOPE/UNEP workshop on the global biogeochemical sulfur cycle,
Pushchino, Russia (Ivanov and Freney 1983)
1980: Biology of inorganic nitrogen and sulfur, Bochum, Germany (Bothe and
Trebst 1981)
1982: Sulfur bacteria, London, UK (Postgate and Kelly 1982)
1983: Sulfur, its significance for chemistry, for the geo-, bio- and cosmosphere
and technology, Bielefeld, Germany (Mller and Krebs 1984)
1984: Evolution of the global biogeochemical sulfur cycle, Tallinn, Estonia

(Brimblecombe and Lein 1989)
1988: The nitrogen and sulfur cycles, Southampton, UK (Cole and Ferguson
1988)
1995: First international symposium on DMSP and related sulfonium com-
pounds, Mobile, AL, USA (Kiene et al. 1996)
1996: Processes and structures in marine methane and sulfide biotopes,
Winterscheid, Germany (Grieshaber and Fischer 1996)
1998: The biological sulfur cycle: environmental science and technology,
Wageningen, The Netherlands (Lens and Hulshoff Pol 1998)
Of the meetings listed above, only that of the Royal Society in London in 1982 had
the scope of the International Symposium in Mnster in 2006.
Acknowledgements. I am personally glad and proud that I have been an eyewitness and cow-
orker during the fruitful years of Norbert Pfennig in Gttingen in the so-called Sulfur Department.
Since then, the fascination of phototrophic bacteria has never left my mind. Without the marve-
lous book by Schlegel (1999) on the history of microbiology I would not have been able to write
this chapter. I thank him wholeheartedly!
References
Arcichowskij V (1904) Zur Frage ber das Bakteriopurpurin. Bull Jard Bot St Petersbourg 4:97
Bothe H, Trebst A (eds) (1981) Biology of inorganic nitrogen and sulfur. Springer, Berlin
Brimblecombe P, Lein AY (1989) SCOPE 39: Evolution of the global biogeochemical sulphur
cycle. Wiley, Chichester
Brune DC, Trper HG (1986) Noncyclic electron transport in chromatophores from photolitho-
trophically grown Rhodobacter sulfidophilus. Arch Microbiol 145:295301
Buchanan RE, Gibbons NE (eds) (1974) Bergeys manual of determinative bacteriology, 8th edn.
Williams and Wilkins, Baltimore
Buder J (1919) Zur Bakteriologie des Bakteriopurpurins und der Purpurbakterien. Jahrb Wiss Bot
58:525628
Cavallini D, Gaull DG, Zappia V (eds) (1980) Natural sulfur compounds. Plenum, New York
Ciba Foundation (1980) Ciba Foundation symposium 72, new series. Excerpta Medica,
Amsterdam
Cohn F (1872) Untersuchungen ber Bakterien. Beitr Biol Pflanzen 1 2:127224
Cohn F (1875) Untersuchungen ber Bakterien. Beitr Biol Pflanzen 1 3:141207
Cole JA, Ferguson SJ (eds) (1988) 42nd symposium of the SGM. Cambridge University Press,
Cambridge
Dahl C, Engels S, Pott-Sperling A, Schulte A,Sander J, Lbbe Y, Deuster O, Brune DC (2005)
Novel genes of the dsr gene cluster and evidence for close interaction of dsr proteins during
sulfur oxidation in the phototrophic sulphur bacterium Allochromatium vinosum. J Bacteriol
187:13921404
Drews G (2005) Contributions of Theodor Wilhelm Engelmann on phototaxis, chemotaxis and
photosynthesis. Photosynth Res 83:2534
Ehrenberg CG (1838) Die Infusionsthierchen als vollkommene Organismen, ein Blick in das
tiefere organische Leben der Natur. Voss, Leipzig
Engelmann TW (1883) Bacterium photometricum. Pflgers Arch Ges Physiol 30:95124
98 H.G. Trper

inorganic sulphur compounds by bacteria: emergence of a common mechanism? Appl Environ
Gaffron H (1935) ber den Stoffwechsel der Purpurbakterien. Biochem Z 275:301319
Grieshaber MK, Fischer U (eds) (1996) Abstracts and program of the workshop on processes and
structures in marine methane and sulfide biotopes. Shaker, Aachen
Hensen D, Sperling D, Trper HG, Brune DC, Dahl C (2006) Thiosulfate oxidation in the
phototrophic sulfur bacterium Allochromatium vinosum. Mol Microbiol 62:794810
Herrmann R (1984) Photographische Dokumentation der Bildung von Schwefeleinschlssen in
Zellen von Chromatien. Diploma thesis, Fachhochschule Kln, Cologne
Ivanov MI, Freney JR (eds) (1983) SCOPE 19: The biogeochemical sulphur cycle. Wiley,
Chichester
Kappler U, Bennett B, Rethmeier J, Schwarz G, Deutzmann R, McEwan AG, Dahl C (2000)
Sulfite:cytochrome c oxidoreductase from Thiobacillus novellus purification, characteriza-
tion, and molecular biology of a heterodimeric member of the sulfite oxidase family. J Biol
Chem 275:1320213212
Kappler U, Friedrich CG, Trper HG, Dahl C (2001) Evidence for two pathways of thiosulfate
oxidation in Starkeya novella (formerly Thiobacillus novellus). Arch Microbiol 175:102111
Kiene RP, Visscher PT, Keller MD, Kirst GO (eds) (1996) Biological and environmental chemis-
try of DMSP and related sulfonium compounds. Plenum, New York
Kirchhoff J, Trper HG (1974) Adenosine 5-phosphosulfate reductase of Chlorobium limicola.
Lankester ER (1873) On a peach-coloured bacterium Bacterium rubescens. Q J Microsc Sci
13:408425
Larsen H (1954) On the microbiology and biochemistry of the photosynthetic green sulfur bacte-
ria. K Nor Vidensk Selsk Skr 1
Lens PNL, Hulshoff Pol LW (eds) (1998) Biodegradation 9:157318
Lu WP, Kelly DP (1983a) Thiosulphate oxidation, electron transport and phosphorylation in a
cell-free system from Thiobacillus A2. J Gen Microbiol 129:16611671
Lu WP, Kelly DP (1983b) Partial purification and resolutution of a thiosulphate-oxidizing system
from Thiobacillus A2. J Gen Microbiol 129:16731681
Lu WP, Kelly DP (1983c) Purification and some properties of two principal enzymes of the thio-
sulphate-oxidizing multi-enzyme system from Thiobacillus A2. J Gen Microbiol
129:35493564
Lu WP, Kelly DP (1988) Cellular location and partial purification of the thiosulphate-oxidizing
enzyme and trithionate hydrolase from Thiobacillus tepidarius. J Gen Microbiol
134:877885
Mitchell P (1961) Approaches to the analysis of specific membrane transport. In: Goodwin TW,
Lundberg O (eds) Biological structure and function, vol 2. Academic, New York, pp 581603
Molisch H (1907) Die Purpurbakterien nach neuen Untersuchungen. Fischer, Jena
Mller A, Krebs B (eds) (1984) Studies in inorganic chemistry 5. Elsevier, Amsterdam
Mller C (1870) Chemisch-Physikalische Beschreibung der Thermen von Baden in der Schweiz
(Canton Aargau). Zehnder, Baden
Mller OF (1786) Animalcula infusoria fluviatila et marina. Havniae, Copenhagen
Nadson GA (1903) Observations sur les bactries pourpres. Bull Jard Bot St Petersbourg 3:109
Pattaragulwanit K, Brune DC, Trper HG, Dahl C (1998) Molecular genetic evidence for extracy-
toplasmic localization of sulphur globules in Chromatium vinosum. Arch Microbiol
169:434444
Peck HD (1961) Enzymatic basis for assimilatory and dissimilatory sulfate reduction. J Bacteriol
82:933939
Peck HD (1966) Some evolutionary aspects of inorganic sulfur metabolism. In: Lecture series on
theoretical and applied aspects of modern microbiology. University of Maryland, College
Park, pp 122
Perty M (1852) Zur Kenntnis kleinster Lebensformen. Jent and Reinert, Bern
Pfennig N (1961) Eine vollsynthetische Nhrlsung zur selektiven Anreicherung einiger
Schwefelpurpurbakterien. Naturwissenschaften 48:136
Pfennig N (1962) ber die Kultur von Chromatium okenii. Vortr Gesamtgeb Bot 1:8485
Pfennig N, Trper HG (1989) Section 18. Anoxygenic phototrophic bacteria. In: Staley JT, Bryant
MP, Pfennig N, Holt JG (eds) Bergeys manual of systematic bacteriology, vol 3. Williams and
Wilkins, Baltimore, pp 16351709
Postgate JR, Kelly DP (eds) (1982) Philos Trans R Soc Lond Ser B 298:431602
locus of Chromatium vinosum are involved in the oxidation of intracellular sulphur.
Prange A, Chauvistr R, Modrow H, Hormes J, Trper HG, Dahl C (2002) Quantitative speciation
of sulphur in bacterial sulphur globules: X-ray absorption spectroscopy reveals at least three
different species of sulphur. Microbiology 148:267276
Remsen CC (1978) Comparative subcellular architecture of photosynthetic bacteria. In Clayton
RK, Sistrom WR (eds) The photosynthetic bacteria. Plenum, New York, pp 3160
Roelofsen PA (1934) On the metabolism of the purple sulfur bacteria. Proc K Ned Acad Wet
37:660668
Roy AB, Trudinger PA (1970) The biochemistry of inorganic compounds of sulphur. Cambridge
University Press, Cambridge
Ruben S, Randall M, Kamen M, Hyde JL (1941) Heavy oxygen (O18) as a tracer in the study of
photosynthesis. J Am Chem Soc 63:877878
and investigation of some of its molecular and catalytic properties. Biochim Biophys Acta
568:454467
Schedel M, Vanselow M, Trper HG (1979) Siroheme sulfite reductase isolated from Chromatium
vinosum. Arch Microbiol 121:2936
Schlegel HG (1999) Geschichte der Mikrobiologie. Leopoldina, Halle
Schlegel HG, Pfennig N (1961) Die Anreicherungskultur einiger Schwefelpurpurbakterien. Arch
Mikrobiol 38:139
Schulz HN, Brinkhoff T, Ferdelman TG, Henndez Marin M, Teske A, Jrgensen BB (1999). Dense
population of a giant sulfur bacterium in Namibian shelf sediments. Science 284:493495
Shahak J, Schtz M, Bronstein M, Griesbeck C, Hauska G, Padan E (1999)
Sulfide-dependent anoxygenic photosynthesis in prokaryotes: sulfide-quinone reductase (SQR),
the initial step. In: Peschek GA, Lffelhardt W, Schmetterer G (eds) The phototrophic prokary-
otes, Kluwer/Plenum, New York, pp 217228
Smith A (1965) The discriminative oxidation of the sulphur atoms of thiosulphate by a photosyn-
thetic sulphur bacterium Chromatium strain D. Biochem J 94:27P
Smith A (1966) The role of tetrathionate in the oxidation of thiosulphate by Chromatium sp. strain
D. J Gen Microbiol 42:371380
Smith A, Lascelles J (1966) Thiosulphate metabolism and rhodanese in Chromatium sp. strain D.
J Gen Microbiol 42:257270
Steudel R (1989) On the nature of the elemental sulfur (S) produced by sulfur-oxidizing bacteria
a model for S globules. In: Schlegel HG, Bowien B (eds) Biology of autotrophic bacteria
Science Technology, Madison, pp 289303
Thiele HH (1966) Wachstumsphysiologische Untersuchungen an Thiorhodaceae:
Wasserstoffdonatoren und Sulfatreduktion. Doctoral thesis, University of Gttingen
Thiele HH (1968) Sulphur metabolism in Thiorhodaceae. V. Enzymes of sulphur metabolism in
Thiocapsa floridana and Chromatium species. Antonie Van Leeuwenhoek J Microbiol Serol
34:350361
Trevisan FS (1842) Prospetto della flora euganea. Coi tipi del seminario, Padua, pp 5657
Trper HG (1964a) CO2-Fixierung und Intermedirstoffwechsel bei Chromatium okenii Perty.
Arch Mikrobiol 49:2350
100 H.G. Trper
Trper HG (1964b) Sulphur metabolism in Thiorhodaceae. II. Stoichiometric relationship of CO2

fixation to oxidation of hydrogen sulphide and intracellular sulphur in Chromatium okenii.
Antonie Van Leeuwenhoek J Microbiol Serol 30:385394
Trper HG (1970) Culture and isolation of phototrophic sulfur bacteria from the marine environ-
ment. Helgol Wiss Meeresunters 20:616
Trper HG, Peck HD (1970) Formation of adenosine 5-phosphosulfate in phototrophic bacteria.
Trper HG, Pfennig N (1966) Sulphur metabolism in Thiorhodaceae. III. Storage and turnover of
thiosulphate sulphur in Thiocapsa floridana and Chromatium species. Antonie Van
Leeuwenhoek J Microbiol Serol 32:261276
Trper HG, Rogers LA (1971) Purification and properties of adenosine 5-phosphosulfate reduct-
ase from the phototrophic sulfur bacterium Thiocapsa roseopersicina. J Bacteriol
108:11121121
Trper HG, Schlegel HG (1964) Sulphur metabolism in Thiorhodaceae. I. Quantitative measure-
ments on growing cells of Chromatium okenii. Antonie Van Leeuwenhoek J Microbiol Serol
30:225238
van Egeraat AWSM, Huntjens JLM (eds) (1975) Plant Soil 43:1228
van Niel CB (1931) On the morphology and physiologyof the purple and green sulphur bacteria.
von Esmarch E (1887) ber die Reinkultur eines Spirillum. Zbl Bakteriol I 1:225230
Warming E (1875) Om nogle ved Danmarks kyster levende bakterier. Vidensk Medd Dansk
Naturhist Foren Khobenhavn 306324
Winogradsky SN (1887) ber Schwefelbacterien. Bot Ztg 45:489507, 513523, 529539,
545559, 569576, 585594, 606610
Winogradsky SN (1949) Microbiologie du sol. Problmes et mthodes. Cinquantes ans de recher-
ches. Masson, Paris
Chapter 9
Thiosulfate and Sulfur Oxidation in Purple
Sulfur Bacteria
Frauke Grimm, Bettina Franz, Christiane Dahl
Frauke Grimm and Bettina Franz contributed equally to this work.
Abstract In chemotrophic and phototrophic sulfur oxidizers that do not form sulfur
deposits a periplasmic thiosulfate-oxidizing multienzyme complex (Sox complex) has
been described to be responsible for formation of sulfate from thiosulfate. In the anoxy-
genic phototrophic sulfur bacterium Allochromatium vinosum intracellular sulfur glob-
ules are an obligate intermediate during the oxidation of thiosulfate to sulfate. Despite
this fundamental difference A. vinosum possesses five sox genes in two independent loci
(soxBXA and soxYZ) encoding proteins related to components of the Sox complex from
Paracoccus pantotrophus. Three sox-encoded proteins were purified from A. vinosum:
the heterodimeric c-type cytochrome SoxXA, the monomeric SoxB and the het-
erodimeric thiosulfate-binding protein SoxYZ. Gene inactivation and complementation
studies proved that these proteins are essential for thiosulfate oxidation to sulfate. The
intermediary formation of sulfur globules in A. vinosum appears to be related to the lack
of soxCD genes, the products of which are proposed to oxidize SoxY-bound sulfane
sulfur. In their absence the latter is instead transferred to growing sulfur globules. The
oxidation of the stored sulfur is completely dependent on the proteins encoded in the
dsr operon. The dissimilatory sulfite reductase (DsrAB) interacts with membrane-
bound as well as soluble Dsr proteins. From membranes the protein is copurified with
the transmembrane electron-transporting complex DsrMKJOP. Furthermore, the solu-
ble cytoplasmic proteins DsrC and DsrEFH are found in the same fraction, indicating
an interaction of DsrC and DsrEFH with the reverse sulfite reductase. From the soluble
fraction DsrAB is copurified with DsrL, a homodimeric ironsulfur flavoprotein with
NADH:acceptor oxidoreductase activity. The observed interactions of Dsr proteins
serve as a basis for an improved model of sulfur oxidation in purple sulfur bacteria.
9.1 Introduction
Thiosulfate (S2O32) plays an important role in the natural sulfur cycle, especially
in freshwater sediments (Jrgensen 1990; Sorokin et al. 1999; Podgorsek and
Imhoff 1999). It is a rather stable and environmentally abundant sulfur compound
101
Springer 2008
102 F. Grimm et al.
of intermediate oxidation state. Generally, there are three different ways of bacterial
utilization of thiosulfate: oxidation with tetrathionate or sulfate as the end prod-
uct, reduction to products like hydrogen sulfide, and disproportionation to sulfur
and sulfite or hydrogen sulfide and sulfate (Jrgensen 1990). In this review, we
concentrate on the oxidation of thiosulfate. In many organisms like some
Pseudomonas and Halomonas species (Sorokin et al. 1999; Podgorsek and
Imhoff 1999), tetrathionate is the end product of thiosulfate oxidation. It is
formed by oxidative condensation of two thiosulfate anions catalyzed by thiosul-
fate dehydrogenase (EC 1.8.2.2; thiosulfate:acceptor oxidoreductase). More
widespread than the formation of tetrathionate is the complete oxidation of thio-
sulfate to sulfate. Two different pathways appear to exist: In numerous faculta-
tively chemolithotrophic or photolithotrophic organisms like P. pantotrophus or
Rhodovulum sulfidophilum both sulfur atoms of thiosulfate are oxidized to sulfate
without the appearance of sulfur deposits as intermediates (Appia-Ayme et al.
2001; Friedrich et al. 2001, 2005), whereas in phototrophic purple sulfur bacteria
and many chemotrophic sulfur oxidizers like magnetotactic bacteria, Beggiatoa
sp. or Thiothrix the formation of conspicuous globules of polymeric, water-insol-
uble sulfur appears to be an important step during thiosulfate oxidation (Nelson
and Castenholz 1981; Dahl 1999; Howarth et al. 1999; Dahl and Prange 2006;
Hensen et al. 2006; Williams et al. 2006). Table 9.1 shows selected organisms
forming sulfur globules from thiosulfate.
The oxidation of sulfur deposits is one of the least understood steps of sulfur metab-
olism. The immense diversity of sulfur-forming prokaryotes is reflected by the facts
that the site of sulfur deposition (intracellular or extracellular) as well as the chemical
nature of the deposited sulfur can vary (Table 9.1). Universal biochemical mechanisms
may therefore not exist (Brune 1995a; Prange et al. 2002; Dahl et al. 2002; Friedrich et
al. 2005; Dahl and Prange 2006). One enzyme for which an involvement in the degra-
dation of stored sulfur was suggested is dissimilatory sulfite reductase. In the chemo-
lithotrophic sulfur oxidizer Thiobacillus denitrificans and the photolithotrophic sulfur
oxidizer A. vinosum (formerly Chromatium vinosum; Imhoff et al. 1998) this enzyme
is assumed to be operating in reverse, performing the six-electron oxidation from
sulfide to sulfite (Schedel et al. 1979; Schedel and Trper 1979).
The genetically accessible Gammaproteobacterium A. vinosum, an anoxygenic
purple sulfur bacterium of the family Chromatiaceae, utilizes reduced sulfur com-
pounds like sulfide, thiosulfate and sulfur as electron donors for reductive carbon
dioxide fixation during photolithoautotrophic growth (Brune 1995a). A. vinosum
employs two different pathways for the oxidation of thiosulfate: oxidation to
tetrathionate or complete oxidation to sulfate (Hensen et al. 2006). The formation
of intracellular sulfur globules from sulfide and thiosulfate is obligatory en route to
sulfate (Pott and Dahl 1998). The globules are located in the periplasm
(Pattaragulwanit et al. 1998) and are surrounded by an envelope consisting of three
different proteins, SgpA, SgpB and SgpC (Brune 1995a). The sulfur inside is
present as sulfur chains probably carrying so far unidentified organic residues at
one or at both ends (Prange et al. 2002). The reverse sulfite reductase is encoded by
9 Thiosulfate and Sulfur Oxidation in Purple Sulfur Bacteria 103
Table 9.1 Selected organisms forming sulfur globules from thiosulfate

Organism/group Comments Systematic affiliation References
Chemotrophic sulfur bacteria
Magnetotactic coccus Intracellular and Alphaproteobacteria (Williams et al. 2006)
MC-1 extracellular
Thermothrix Extracellular Betaproteobacteria (Odintsova et al. 1996)
Beggiatoa sp. Intracellular, Gammaproteobacteria (Nelson and Castenholz
periplasmic, 1981; Prange et al.
cyclo-octasulfur 2002)
Thiothrix Intracellular, Gammaproteobacteria (Odintsova et al. 1993;
periplasmic Howarth et al. 1999)
Thioalkalivibrio Extracellular, some Gammaproteobacteria (Sorokin et al. 2001)
strains intracel-
lular in
periplasm
Phototrophic bacteria
Chlorobaculum Extracellular Chlorobi (Steinmetz and Fischer
parvum DSM 263a 1982)
Chromatiaceae Intracellular, Gammaproteobacteria (Smith and Lascelles
periplasmic, 1966; Brune 1989;
organic Prange et al. 2002)
polysulfanes
a
Formerly Chlorobium vibrioforme subsp. thiosulfatophilum (Imhoff 2003).
the first two genes of a large cluster dsrABEFHCMKLJOPNRS, which is essential

for the oxidation of sulfur deposited in sulfur globules to the final product sulfate
(Pott and Dahl 1998; Dahl et al. 2005).
In this review we concentrate on A. vinosum and summarize the current knowl-
edge of the pathways of thiosulfate and intracellular sulfur oxidation to sulfate.
Since the second product of thiosulfate oxidation, tetrathionate, is not further
utilized by A. vinosum, this pathway is not described in detail and the interested
reader is referred to the data presented by Hensen et al. (2006).
9.2 Oxidation of Thiosulfate in A. vinosum
A wealth of biochemical and molecular genetic information on thiosulfate oxidation is

available about organisms that do not form sulfur deposits during thiosulfate utilization.
The groups of Don Kelly and Cornelius Friedrich found and characterized a periplasmic
thiosulfate-oxidizing multienzyme complex (Sox) in P. versutus (Lu et al. 1985) and
P. pantotrophus (Rother et al. 2001; Friedrich et al. 2001). In P. pantotrophus the Sox
complex is essential for thiosulfate oxidation in vivo and catalyzes reduction of
104 F. Grimm et al.
cytochrome c coupled to the oxidation of thiosulfate, sulfide, sulfite and elemental

sulfur in vitro. The proposed mechanism for thiosulfate oxidation requires four dif-
ferent proteins: SoxB, SoxXA, SoxYZ and SoxCD (Friedrich et al. 2001). The het-
erodimeric SoxYZ has been identified as the substrate-binding molecule of the
complex (Quentmeier and Friedrich 2001). SoxXA is a heterodimeric heme enzyme
that is reduced while oxidatively coupling the sulfur compound to SoxYZ. The mono-
meric, manganese-containing SoxB has been proposed to act as a sulfate thiol este-
rase or sulfate thiol hydrolase and is responsible for hydrolytic cleavage of a sulfate
group from the bound sulfur substrate. SoxCD oxidizes the remaining sulfane sulfur,
acting as a sulfur dehydrogenase. Further action of SoxB releases a second sulfate
molecule and thereby restores SoxYZ. Further details about properties and regulation
of the system are given in Chap. 12 by Friedrich et al.
Much less is known about thiosulfate oxidation involving the intermediate deposi-
tion of sulfur either inside or outside the cells. As mentioned in Sect. 9.1, sulfur globules
are formed as an obligatory intermediate during the oxidation of thiosulfate to sulfate
in A. vinosum (Pott and Dahl 1998). Studies with radioactively labeled thiosulfate in
purple sulfur bacteria demonstrated that the sulfane and the sulfone sulfur atoms of thio-
sulfate are oxidized by different pathways (Smith and Lascelles 1966; Trper and
Pfennig 1966). Only the sulfane sulfur accumulates in sulfur globules before further
oxidation, whereas the sulfone sulfur is rapidly converted into sulfate. Thus, the initial
step of thiosulfate oxidation is a cleavage of the molecule. In the past, the detection of
thiosulfate-reducing enzyme activities (i.e., rhodaneses and thiosulfate reductases) in
phototrophic and chemotrophic sulfur bacteria led to the assumption that thiosulfate
would be cleaved into sulfate and sulfide in the presence of suitable thiol acceptors like
glutathione and dihydrolipoic acid, and that the H2S formed during the proposed reac-
tion would be immediately oxidized to stored sulfur (Brune 1989, 1995a; Dahl 1999).
However, genetic proof for this assumption was missing and it was recognized quite
early that rhodanese as well as thiosulfate reductase occur in a wide range of organisms
not able to metabolize thiosulfate (Brune 1989). During the past several years clusters
of sox genes were identified in thiosulfate-oxidizing green sulfur bacteria (Petri et al.
2001; Eisen et al. 2002; Vert et al. 2002; Frigaard and Bryant 2008; see also Chap. 6
by Frigaard and Bryant). Therefore, an essential role of rhodanese or thiosulfate reduct-
ase during the initial steps of thiosulfate oxidation in sulfur-storing bacteria appeared
increasingly unlikely. Now, sox genes have also been identified in A. vinosum, proteins
essential for thiosulfate oxidation have been purified and mutational analysis has shown
that the earlier models of thiosulfate oxidation via intermediate sulfur formation have to
be completely revised (Hensen et al. 2006).
9.2.1 sox Genes in A. vinosum
Nucleotide sequence analysis revealed that, unlike the situation in P. pantotrophus, the
genes soxXAB and soxYZ are located in two independent gene regions in A. vinosum
and genes coding for SoxCD are not present in the organism (Hensen et al. 2006).
The genes soxB and soxXA are transcribed divergently. The two sequenced DNA
fragments include ten further open reading frames. Especially notable is the rhd
gene, the product of which contains a conserved domain typical for rhodaneses,
enzymes responsible for sulfur group transfer that are found in
all three domains of life. Since the sulfane sulfur is transferred to the sulfur glob-
ules, a rhodanese could well be part of a thiosulfate oxidizing pathway (Hensen
et al. 2006).
9.2.2 Sox Proteins in A. vinosum
Three periplasmic proteins SoxXA, SoxB and SoxYZ were identified in and puri-
fied from A. vinosum (Hensen et al. 2006). Except SoxZ, all are predicted to be
synthesized as precursors carrying signal peptides. A Sec-dependent transport is
postulated for SoxXA. The protein was purified as a heterodimer (SoxX 11 kDa,
SoxA 29 kDa). Covalently bound heme is present in both subunits. A. vinosum
SoxA is predicted to bind one heme like the protein from Starkeya novella
(Kappler et al. 2004), while two heme binding sites are present in P. pantotrophus
and R. sulfidophilum SoxA (Friedrich et al. 2000; Bamford et al. 2002). Since the
structural analysis of R. sulfidophilum SoxXA revealed that the additional amino-
terminal SoxA heme is at too great a distance from the other hemes to allow effi-
cient electron tunneling (Bamford et al. 2002), it is not clear whether the different
heme contents of SoxA cause different functions. Although SoxXA was purified
under aerobic, nonreducing conditions, the UVvis spectrum was that of a typical
reduced c550-type cytochrome. All other SoxXA proteins described so far have
been isolated in the oxidized state (Friedrich et al. 2000; Cheesman et al. 2001;
Kappler et al. 2004). SoxA is expressed at a low constitutive level in the absence
of thiosulfate and its formation is strongly increased in the presence of thiosulfate
(Hensen et al. 2006).
SoxB was isolated as a monomeric protein (62 kDa) from A. vinosum and
processing and transport by the Tat pathway was experimentally verified
(Hensen et al. 2006). This implies transport as a mature, folded protein proba-
bly containing a cofactor. SoxB of P. pantotrophus contains two manganese
atoms per monomer (Friedrich et al. 2000) and this is probably also the case in
A. vinosum.
SoxYZ was purified as a heterodimer (SoxY 12.7 kDa, SoxZ 11.2 kDa) (Hensen
et al. 2006). Experimental evidence was obtained for a covalent attachment of
thiosulfate to a strictly conserved cysteine at the carboxy terminus of SoxY. This is
in accordance with the suggestion of Quentmeier and Friedrich (2001), who proposed
SoxY as the substrate-binding molecule in the Sox complex of P. pantotrophus. For
SoxY a Tat-dependent transport is predicted, and SoxZ is very likely cotrans-
ported with SoxY, as has also been proposed for P. pantotrophus SoxZ (Friedrich
et al. 2001).
106 F. Grimm et al.
9.2.3 Inactivation and Complementation of sox Genes

in A. vinosum
On the basis of different mutants, the importance of the sox genes and the encoded
proteins for thiosulfate oxidation in A. vinosum was determined. The A. vinosum
mutants DsoxX, DsoxB, DsoxBX and DsoxY entirely lacked sulfate production from
thiosulfate. In contrast, the inactivation of ORF9/rhd had no detectable effect on
thiosulfate utilization (Hensen et al. 2006). Therefore, the products of the latter
genes do not seem to play a vital role in the oxidation of thiosulfate to sulfate under
the experimental conditions chosen, while the proteins SoxXABYZ are absolutely
essential. The A. vinosum DsoxX and DsoxY mutants were complemented in trans.
Thiosulfate oxidation was completely restored to the wild-type phenotype in the
complemented mutant DsoxX and sulfate was again the major product. In the com-
plemented DsoxY mutant the thiosulfate oxidation rate was still significantly lower
than in the wild type, but the principal capability to oxidize thiosulfate to sulfate
was clearly reestablished. In summary, the complementation experiments verified
that the observed lack of sulfate formation from thiosulfate was indeed caused by
inactivation of sox genes (Hensen et al. 2006).
On the basis of these results and the model suggested by Friedrich et al. (2001)
for non-sulfur-storing bacteria, a model for thiosulfate oxidation in sulfur-storing
organisms is proposed (Fig. 9.1): the initial oxidation and covalent binding of
S
- S2O32-
SgpA SoxY
SgpC SgpB
SoxX
Periplasm SgpB SgpC SoxZ
SgpA SgpA
SoxA
SSnS SgpC SO SgpB 2 e-
O
SgpB SgpC -
-O S O-
S S
SgpA SgpA
SgpC SgpB S S
SoxY SoxY
Flavocytochrome c
SoxZ SoxZ
+
2 H + 2 e-
SoxB H2O
SO42-
2 HS SS
+
2 H + 2 e-
Sulfide: QH2
quinone
oxido-
reductase Q
+
2H
Cytoplasm
Fig. 9.1 Model for the oxidation of sulfide and thiosulfate to intracellularly stored sulfur in
Allochromatium vinosum. A sulfur globule is represented with its envelope consisting of the three
proteins SgpA, SgpB and SgpC (Brune 1995b; Pattaragulwanit et al. 1998).
thiosulfate to SoxYZ is catalyzed by SoxXA and sulfate is then hydrolytically

released by SoxB. Owing to the lack of the sulfur dehydrogenase SoxCD, the
sulfane sulfur atom still hooked up to SoxY cannot be directly further oxidized in
organisms like A. vinosum. Probably, the sulfur is instead transferred to growing
sulfur globules. Such a suggestion is feasible as the sulfur globules in A. vinosum
and in many if not all other organisms forming intracellular sulfur deposits reside
in the bacterial periplasm (Pattaragulwanit et al. 1998; Dahl and Prange 2006) and
therefore in the same cellular compartment as the Sox proteins. How the transfer of
SoxY-bound sulfur to the sulfur globules is achieved is currently unclear as the lack
of the potential sulfur transferase encoded by the rhd gene did not lead to a detect-
able phenotype. Possibly, other sulfur transferases present in the cells function as a
backup system (Hensen et al. 2006).
9.3 Oxidation of Stored Sulfur in A. vinosum
In A. vinosum the oxidation of thiosulfate and that of sulfide merge at the level of
stored sulfur. During sulfide oxidation, sulfur stored in sulfur globules is the first
macroscopically and microscopically observable product (Dahl and Prange 2006).
The mechanism by which the periplasmically stored sulfur is made available to the
cytoplasmic sulfite reductase is unclear. In sulfate-reducing bacteria dissimilatory
sulfite reductase catalyzes the six-electron reduction of sulfite to sulfide. It has
therefore been proposed that the stored sulfur has to be reductively activated to the
oxidation state of sulfide in A. vinosum in order to serve as a substrate for sulfite
reductase operating in reverse (Schedel et al. 1979). The importance of the dsr gene
region for the oxidation of stored sulfur has been shown by interposon mutagenesis
(Pott and Dahl 1998; Dahl et al. 2005).
9.3.1 The dsr Operon and Proteins Encoded Therein
The reverse dissimilatory sulfite reductase (DsrAB) of A. vinosum is encoded

together with 13 other proteins in the dsr operon, dsrABEFHCMKLJOPNRS (Pott
and Dahl 1998; Dahl et al. 2005). The dsrAB gene products form the cytoplasmic
22-structured sulfite reductase, which is closely related to the dissimilatory sulfite
reductases of sulfate-reducing prokaryotes (Hipp et al. 1997). The prosthetic group
of DsrAB is siroamide[Fe4S4], with siroamide being an amidated form of the clas-
sic siroheme. The dsrN encoded protein resembles cobyrinic acid a,c-diamide syn-
thases and catalyzes the glutamine-dependent amidation of siroheme. A DdsrN
mutant showed a reduced sulfur oxidation rate. A. vinosum is apparently able to
incorporate siroheme instead of siroamide into sulfite reductase, thereby retaining
some function of the enzyme (Lbbe et al. 2006). The dsrEFH genes are located
adjacent to dsrAB. The products of these three genes show significant similarity to
108 F. Grimm et al.
each other and form a single tight 75-kDa complex with an 222 structure (Dahl
et al. 2005). DsrC is a small soluble cytoplasmic protein with a highly conserved
C-terminus including two conserved cysteine residues. Proteins closely related to
DsrEFH and DsrC have recently been shown to act as parts of a sulfur relay system
involved in thiouridine biosynthesis at transfer RNA wobble positions in Escherichia
coli (Numata et al. 2006; Ikeuchi et al. 2006). The dsrM-encoded protein is
predicted to be a membrane-bound b-type cytochrome and shows similarities to a
subunit of heterodisulfide reductases from methanogenic archaea (Sander et al.
2006). The cytoplasmic ironsulfur protein DsrK exhibits relevant similarity to the
catalytic subunit of heterodisulfide reductases. DsrP is another integral membrane
protein. The periplasmic proteins DsrJ and DsrO are a triheme c-type cytochrome
and an ironsulfur protein, respectively. DsrKJO were copurified from membranes,
pointing at the presence of a transmembrane electron-transporting complex consist-
ing of DsrMKJOP (Dahl et al. 2005). Individual in frame deletions of the dsrMK-
JOP genes led to the complete inability of the mutants to oxidize stored sulfur
(Sander et al. 2006). DsrL is a cytoplasmic ironsulfur flavoprotein with NADH:
acceptor oxidoreductase activity (Y. Lbbe and C. Dahl, unpublished data). In
frame deletion of dsrL completely abolished the oxidation of stored sulfur (Lbbe
et al. 2006). DsrR and DsrS are soluble cytoplasmic proteins of unknown function.
The dsr genes, with the exception of the constitutively expressed dsrC, are
expressed and the encoded proteins are formed at a low basic level even in the
absence of sulfur compounds. An increased production of all Dsr proteins is
induced by sulfide and/or stored sulfur (Dahl et al. 2005).
9.3.2 Distribution of dsr Genes in Organisms with Dissimilatory

Sulfur Metabolism and Phylogenetic Analysis
Dissimilatory sulfite reductase and other Dsr proteins occur in sulfate-reducing

prokaryotes, where sulfite reductase catalyzes the reduction of sulfite to sulfide
as the final step of sulfate reduction, as well as in sulfur-oxidizing prokaryotes,
in which the sulfite reductase works in the reverse direction (Hipp et al. 1997;
Dahl et al. 2005; Sander et al. 2006). When the occurrences of the various dsr
genes in sulfur-oxidizing and sulfate-reducing prokaryotes are compared, it
becomes apparent that certain genes, dsrABCNMKJOP, represent a core unit,
whereas other dsr genes are specific for either sulfur-oxidizing or sulfate-reducing
prokaryotes (Table 9.2). The gene dsrD appears to be typical for sulfate/sulfite-
reducing prokaryotes, whereas the genes dsrEFH and dsrL appear to be restricted
to the sulfur oxidizers.
Phylogenetic analysis of Dsr proteins yielded two separate clusters consisting of
proteins from sulfate reducers, on the one hand, and of proteins from sulfur oxidiz-
ers, on the other (Sander et al. 2006). Astonishingly, the DsrMKJOP proteins of the
members of the green sulfur bacteria (phylum Chlorobi) do not cluster with the
proteins of other sulfur oxidizers but affiliate with the sulfate/sulfite-reducing
Table 9.2 Occurrence of dsr genes in selected sulfur-oxidizing and sulfate/sulfite-reducing prokaryotes
dsr genes
Organisms AB C D EFH L N MKJOP R S T GeneBank accession numbers
SULFUR OXIDIZERS
Alphaproteobacteria
Magnetospirillum magneto- + + + + + + ? NZ_AAAP01003833, NZ_
tacticum AAAP01003703, NZ_AAAP01003586
Magnetococcus sp. + + + + + + NZ_AAAN02000064, NZ_AAA02000091,
NZ_AAAN03000009
Betaproteobacteria
Thiobacillus denitrificans + + + + + + + + NZ_AAFH01000005, NC_007404
Gammaproteobacteria
Allochromatium vinosum + + + + + + + + U84760
Alkalilimnicola ehrlichii + + + + + + NZ_AALK01000002
Halorhodospira halophila + + + + + + NZ_AAOQ01000001
Chlorobi
Chlorobaculum tepidum + + + + + + + NC_002932
Chlorobium phaeobacteroides + + + + + + + NZ_AAIB01000016 (DSM266),
NZ_AAIB01000004 (DSM 266),
9 Thiosulfate and Sulfur Oxidation in Purple Sulfur Bacteria
NZ_AAIC01000113 (BS1),
NZ_AAIC01000057 (BS1)
Chlorobium limicola + + + + + + + NZ_AAHJ01000040
Chlorobium clathratiforme + + + + + + + NZ_AAIK01000042
Prostecochloris aestuarii + + + + + + + NZ_AAIJ01000019, NZ_AAIJ01000014
Prostecochloris vibrioformis + + + + + + + NZ_AAJD01000006
SULFATE/SULFITE
REDUCERS
Deltaproteobacteria
Desulfovibrio vulgaris + + + + + + NC_002937
Desulfovibrio desulfuricans + + + + + + NC_007519, AJ249777, CP000112
109
(continued)
110
Table 9.2 ( continued)

dsr genes
Organisms AB C D EFH L N MKJOP R S T GeneBank accession numbers
Desufotalea psychrophila + + + + + + CR522870
Syntrophobacter fumaroxidans + + + + + + NZ_AAJF01000004, NZ_AAJF01000055
Clostridia
Moorella thermoacetica + + + + + + NZ_AADT03000031, NZ_
AADT02000020, NC_AADT03000011
Desulfitobacterium hafniense + + + + + NZ_AAAW03000053
Euryarchaeota
Archaeoglobus fulgidus + + + + + + NC_000917
F. Grimm et al.
prokaryotes. This phenomenon suggests a horizontal gene transfer, which is also

supported by the presence of dsrT (or ORF9; Mussmann et al. 2005) in the green
sulfur bacteria, a gene otherwise only found in sulfate/sulfite-reducing prokaryotes
(Sander et al. 2006).
9.3.3 Model of the Sulfur Oxidation Pathway in A. vinosum
The periplasmic flavin adenine dinucleotide containing flavocytochrome c and the

membrane-bound sulfide:quinone oxidoreductase (SQR) have long been suspected
to be responsible for sulfide oxidation in A. vinosum (Brune 1995a; Fig. 9.1).
Flavocytochrome c deficient mutants of A. vinosum (Reinartz et al. 1998) showed
no impact on sulfide oxidation rates, indicating SQR to be the main sulfide-oxidizing
enzyme in this organism. The primary in vitro product of the SQR reaction is solu-
ble polysulfide (Griesbeck et al. 2002). Polysulfides were also detected as the pri-
mary product of sulfide oxidation by whole cells of A. vinosum (Prange et al. 2004).
The initial product of sulfide oxidation released from the enzyme is probably
disulfide. Polysulfide anions of different chain lengths are in equilibrium with each
other (Griesbeck et al. 2002). Longer-chained polysulfides are spontaneously
formed from the initial disulfide by disproportionation. It is currently unknown how
polysulfides are converted into sulfur globules containing organic polysulfanes. In
A. vinosum, sulfide and thiosulfate oxidation merge at the level of stored sulfur
(Fig. 9.1), which is an obligate intermediate during the oxidation of both compounds
(Pott and Dahl 1998; Prange et al. 2004; Hensen et al. 2006).
As outlined already, the only gene region known so far to be essential for the
oxidation of stored sulfur is the dsr operon. Since the proteins encoded at the dsr
locus are either cytoplasmic or membrane-bound and cannot act directly on the
extracytoplasmic sulfur globules (Fig. 9.2), it is proposed that the sulfur is reduc-
tively activated, transported to and further oxidized in the cytoplasm (Pott and Dahl
1998; Dahl et al. 2005). DsrL exhibits NADH:acceptor oxidoreductase activity
(Y. Lbbe and C. Dahl, unpublished data) . Interestingly, the protein carries a
thioredoxin motif CysXXCys immediately preceding the carboxy-terminal iron
sulfur cluster binding sites. This indicates a potential disulfide reductase activity
which we could not yet prove experimentally. Still, the possibility exists that DsrL
uses NADH as electron donor for reduction of a disulfidic or persulfidic compound.
Thus, it is possible that DsrL is involved in the reductive release of sulfide from a
carrier molecule probably an organic perthiol that may transport sulfur from the
periplasmic sulfur globules to the cytoplasm, where it is further metabolized by Dsr
proteins (Dahl et al. 2005). Glutathione amide is a likely candidate for carrying
sulfur from the periplasm to the cytoplasm. This derivative of glutathione has been
found to be largely converted into its perthiolic form when A. vinosum is grown
photoautotrophically on sulfide (Bartsch et al. 1996). Recently, transporters have
been characterized in E. coli mediating export (Pittman et al. 2005) and import
(Suzuki et al. 2005) of glutathione. Shuttling of glutathione amide between cytoplasm
112 F. Grimm et al.
SgpA
SgpC SgpB
SgpB SgpC Periplasm
SgpA SgpA
Sulfite: acceptor oxidoreductase
SgpC SO SgpB
2 e-
SgpB SgpC
HSO3- SO42-
SgpA
SgpC SgpB
SgpA 2 H2O
DsrO
RSSH RSH 4 [FeS] DsrJ
APS reductase
3 Heme c
Heme bL QH2
Q
DsrM DsrP ? AprM
Q
HemebH QH2
+
DsrK AprB 2H
HS SH
HS SH 2[Fe4S4]
RSSH RSH 2 [Fe4S4] DsrC SO42-

DsrC AprA
NADH + ATP
NAD
FAD
HS- 3 H2O
HS S S SH PPi
DsrL DsrL DsrC DsrC
Sat
1 [FeS] 1 [FeS] AMP APS
FAD FAD HSO3-
2 [Fe4S4] 2 [Fe4S4] 3 x 2 e+ 6 H+
ATP sulfurylase
SH SH DsrA DsrA
siroamide- siroamide-
2[Fe4S4] 2 [Fe4S4]
DsrE DsrE
Sulfite reductase
DsrH DsrH DsrB DsrB Cytoplasm
HS SH
[Fe4S4] [Fe4S4]
DsrF DsrF
Fig. 9.2 Model for the oxidation of intracellularly stored sulfur to sulfate in A. vinosum and
involvement of the proteins of the dsr locus. The scheme is based on sequence analysis of the
encoding genes and on biochemical information where available. The potential presence of sulfite:
acceptor oxidoreductase is inferred from the fact that adenosine 5-phosphosulfate (APS) reduct-
ase is not essential (Dahl 1996). However, neither the protein nor the corresponding genes could
yet be proven in A. vinosum.
and periplasm in purple sulfur bacteria like A. vinosum therefore also appears fea-
sible. DsrL, being an essential protein for sulfur oxidation, is copurified with the
sulfite reductase (Y. Lbbe and C. Dahl, unpublished data). Sulfide released from
the perthiol could therefore be directly passed to dsrAB-encoded sulfite reductase,
thereby reducing losses caused by evaporation of gaseous H2S. The DsrMKJOP
membrane complex is copurified with sulfite reductase, DsrEFH and DsrC (Dahl
et al. 2005). This is in accordance with the suggestion that DsrMKJOP from dis-
similatory sulfate reducers transfer electrons to sulfite reductase (Pires et al. 2006;
see also Chap. 3 by Pereira). Taken together these observations indicate that sulfite
reductase specifically interacts with the soluble protein DsrL on one hand and with
membrane-bound Dsr proteins and DsrEFHC on the other hand. Electrons released
from the oxidation of sulfide by sulfite reductase may be fed into photosynthetic
electron transport via DsrC and DsrMKJOP, which would be analogous to the
pathway postulated for sulfate reducers, operating in the reverse direction. DsrM
could operate as a quinone reductase, DsrP as a quinol oxidase and finally the
c-type cytochrome DsrJ would be reduced (Dahl et al. 2005). From here, electrons
could be transferred to high-potential iron protein (HiPIP), the primary electron
donor to the photosynthetic reaction center (Vermeglio et al. 2002). The function
of DsrEFH remains unclear, but as it occurs exclusively in sulfur oxidizers and
shows some interaction with DsrC, it may be important for the pathway to operate
in the sulfide oxidizing direction. In the final step, sulfite is oxidized to sulfate,
either directly by a postulated sulfite:acceptor oxidoreductase or via the nonessential
enzymes adenosine 5-phosphosulfate reductase and ATP sulfurylase (Dahl 1996;
Fig. 9.2).
9.4 Conclusions
In the purple sulfur bacterium A. vinosum, Sox and Dsr proteins have been estab-
lished to be absolutely essential for the oxidation of thiosulfate and stored sulfur,
respectively. Clusters of sox and dsr genes have also been identified in the only
distantly related green sulfur bacteria as well as in other sulfur-storing phototrophic
and chemotrophic sulfur oxidizers. This suggests that the mechanisms of thiosul-
fate oxidation via sulfur deposition and of the oxidation of deposited sulfur are
evolutionary highly conserved and that studies in A. vinosum can contribute to the
elucidation of sulfur oxidation pathways in other sulfur-storing bacteria.
Acknowledgements. We thank Birgitt Httig for excellent technical assistance and acknowl-
edge financial support from the Deutsche Forschungsgemeinschaft (grants Da 351/3-3 and
351/3-4 and Da 351/4-1 and 351/4-2). We also thank Hans G. Trper for ongoing interest and
support.
References
Appia-Ayme C, Little PJ, Matsumoto Y, Leech AP, Berks BC (2001) Cytochrome complex essen-
tial for photosynthetic oxidation of both thiosulfate and sulfide in Rhodovulum sulfidophilum.
21:55995610
Bartsch RG, Newton GL, Sherrill C, Fahey RC (1996) Glutathione amide and its perthiol in
anaerobic sulfur bacteria. J Bacteriol 178:47424746
Brune DC (1989) Sulfur oxidation by phototrophic bacteria. Biochim Biophys Acta 975:189221
Brune DC (1995a) Sulfur compounds as photosynthetic electron donors. In: Blankenship RE, Madigan
Brune DC (1995b) Isolation and characterization of sulfur globule proteins from Chromatium
vinosum and Thiocapsa roseopersicina. Arch Microbiol 163:391399
Cheesman MR, Little PJ, Berks BC (2001) Novel heme ligation in a c-type cytochrome involved
in thiosulfate oxidation: EPR and MCD of SoxAX from Rhodovulum sulfidophilum.
Biochemistry 40:1056210569
Dahl C (1996) Insertional gene inactivation in a phototrophic sulphur bacterium: APS-reductase-
deficient mutants of Chromatium vinosum. Microbiology 142:33633372
114 F. Grimm et al.
Dahl C (1999) Deposition and oxidation of polymeric sulfur in prokaryotes. In: Steinbchel A (ed)
Biochemical principles and mechanisms of biosynthesis and biodegradation of polymers.
Wiley-VCH, Weinheim, pp 2734
Dahl C, Prange A (2006) Bacterial sulfur globules: occurrence, structure and metabolism. In:
Shively JM (ed) Inclusions in prokaryotes. Springer, Heidelberg, pp 2151
Dahl C, Prange A, Steudel R (2002) Natural polymeric sulfur compounds. In: Steinbchel A (ed)
Miscellaneous biopolymers and biodegradation of synthetic polymers, vol 9. Wiley-VCH,
Weinheim, pp 3562
187:13921404
tepidum TLS a photosynthetic, anaerobic, green-sulfur bacterium. Proc Natl Acad Sci USA
99:95099514
Friedrich CG, Quentmeier A, Bardischewsky F, Rother D, Kraft R, Kostka S, Prinz H (2000)
Novel genes coding for lithotrophic sulfur oxidation of Paracoccus pantotrophus GB17. J
Bacteriol 182:46774687
Frigaard NU, Bryant DA (2008) Genomic insights into the sulfur metabolism of phototrophic
green sulfur bacteria. In: Govindjee (series ed) Advances in photosynthesis and respiration,
vol. 27, Hell R, Dahl C, Knaff DB, Leustek T (eds) Sulfur metabolism in phototrophic organ-
isms. Springer, New York (in press)
(2002) Mechanism of sulfide-quinone oxidoreductase investigated using site-directed muta-
genesis and sulfur analysis. Biochemistry 41:1155211565
Hensen D, Sperling D, Trper HG, Brune DC, Dahl C (2006) Thiosulfate oxidation in the pho-
totrophic sulfur bacterium Allochromatium vinosum. Mol Microbiol 62:794810
Hipp WM, Pott AS, Thum-Schmitz N, Faath I, Dahl C, Trper HG (1997) Towards the phylogeny
of APS reductases and sirohaem sulfite reductases in sulfate-reducing and sulfur-oxidizing
prokaryotes. Microbiology 143:28912902
Howarth R, Unz RF, Seviour EM, Seviour RJ, Blackall LL, Pickup RW, Jones JG, Yaguchi J, Head
IM (1999) Phylogenetic relationships of filamentous sulfur bacteria (Thiothrix spp. and
Eikelboom type 021N bacteria) isolated from wastewater-treatment plants and description of
Thiothrix eikelboomii sp. nov., Thiothrix unzii sp. nov., Thiothrix fructosivorans sp. nov. and
Thiothrix defluvii sp. nov. Int J Syst Bacteriol 49:18171827
Ikeuchi Y, Shigi N, Kato J, Nishimura A, Suzuki T (2006) Mechanistic insights into sulfur relay
by multiple sulfur mediators involved in thiouridine biosynthesis at tRNA wobble positions.
Mol Cell 21:97108
Imhoff JF (2003) Phylogenetic taxonomy of the family Chlorobiaceae on the basis of 16S rRNA and
fmo (Fenna-Matthews-Olson protein) gene sequences. Int J Syst Evol Microbiol 53:941951
Imhoff JF, Sling J, Petri R (1998) Phylogenetic relationships among the Chromatiaceae, their
taxonomic reclassification and description of the new genera Allochromatium, Halochromatium,
Isochromatium, Marichromatium, Thiococcus, Thiohalocapsa, and Thermochromatium. Int
J Syst Bacteriol 48:11291143
Jrgensen BB (1990) The sulfur cycle of freshwater sediments: role of thiosulfate. Limnol
Oceanogr 35:13291342
Kappler U, Aguey-Zinsou K-F, Hanson GR, Bernhardt PV, McEwan AG (2004) Cytochrome c551
from Starkeya novella: characterization, spectroscopic properties, and phylogeny of a diheme
protein of the SoxAX family. J Biol Chem 279:62526260
Lu W-P, Swoboda EP, Kelly DP (1985) Properties of the thiosulfate-oxidizing multi-enzyme
system from Thiobacillus versutus. Biochim Biophys Acta 828:116122
Lbbe YJ, Youn H-S, Timkovich R, Dahl C (2006) Siro(haem)amide in Allochromatium vinosum
and relevance of DsrL and DsrN, a homolog of cobyrinic acid a,c diamide synthase for sulfur
oxidation. FEMS Microbiol Lett 261:194202
Mussmann M, Richter M, Lombardot T, Meyerdierks A, Kuever J, Kube M, Glckner FO, Amann
R (2005) Clustered genes related to sulfate respiration in uncultured prokaryotes support the
theory of their concomittant horizontal transfer. J Bacteriol 187:71267137
Nelson DC, Castenholz RW (1981) Use of reduced sulfur compounds by Beggiatoa sp. J Bacteriol
147:140154
Numata T, Fukai S, Ikeuchi Y, Suzuki T, Nureki O (2006) Structural basis for sulfur relay to RNA
mediated by heterohexameric TusBCD complex. Structure 14:357366
Odintsova EV, Wood AP, Kelly DP (1993) Chemolithoautotrophic growth of Thiotrix ramosa.
Odintsova EV, Jannasch H, Mamone JA, Langworthy TA (1996) Thermothrix azorensis sp. nov.,
an obligately chemolithoautotrophic, sulfur-oxidizing, thermophilic bacterium. Int J Syst
Bacteriol 46:422428.
Pattaragulwanit K, Brune DC, Trper HG, Dahl C (1998) Molecular genetic evidence for extracy-
toplasmic localization of sulfur globules in Chromatium vinosum. Arch Microbiol
169:434444
Petri R, Podgorsek L, Imhoff JF (2001) Phylogeny and distribution of the soxB gene among
thiosulfate-oxidizing bacteria. FEMS Microbiol Lett 197:171178
Pires RH, Venceslau SS, Morais F, Teixeira M, Xavier AV, Pereira IAC (2006) Characterization
of the Desulfovibrio desulfuricans ATCC 27774 DsrMKJOP complex a membrane-bound
Pittman MS, Robinson HC, Poole RK (2005) A bacterial glutathione transporter (Escherichia coli
CydDC) exports reductant to the periplasm. J Biol Chem 280:3225432261
Podgorsek L, Imhoff JF (1999) Tetrathionate production by sulfur oxidizing bacteria and the role
of tetrathionate in the sulfur cycle of Baltic Sea sediments. Aquat Microb Ecol 17:255265
Pott AS, Dahl C (1998) Sirohaem-sulfite reductase and other proteins encoded in the dsr locus of
Chromatium vinosum are involved in the oxidation of intracellular sulfur. Microbiology
144:18811894
Prange A, Chauvistre R, Modrow H, Hormes J, Trper HG, Dahl C (2002) Quantitative speciation
of sulfur in bacterial sulfur globules: X-ray absorption spectroscopy reveals at least three dif-
ferent speciations of sulfur. Microbiology 148:267276
Prange A, Engelhardt H, Trper HG, Dahl C (2004) The role of the sulfur globule proteins of
Allochromatium vinosum: mutagenesis of the sulfur globule protein genes and expression stud-
ies by real-time RT PCR. Arch Microbiol 182:165174
Reinartz M, Tschpe T, Brser T, Trper HG, Dahl C (1998) Sulfide oxidation in the phototrophic
bacterium Chromatium vinosum. Arch Microbiol 170:5968
Rother D, Heinrich HJ, Quentmeier A, Bardischewsky F, Friedrich CG (2001) Novel genes of the
sox gene cluster, mutagenesis of the flavoprotein SoxF, and evidence for a general sulfur-oxi-
dizing system in Paracoccus pantotrophus GB17. J Bacteriol 183:44994508
and investigation of some molecular and catalytic properties. Biochim Biophys Acta
568:454467
116 F. Grimm et al.
Schedel M, Vanselow M, Trper HG (1979) Siroheme sulfite reductase from Chromatium vino-
sum. Purification and investigation of some of its molecular and catalytic properties. Arch
Microbiol 121:2936
Smith AJ, Lascelles J (1966) Thiosulphate metabolism and rhodanese in Chromatium sp. strain D.
J Gen Microbiol 42:357370
Sorokin DY, Lysenko AM, Mityushina LL, Tourova TP, Jones BE, Rainey FA, Robertson LA,
Kuenen GJ (2001) Thioalkalimicrobium aerophilum gen. nov., sp. nov. and Thioalkalimicrobium
sibericum sp. nov., and Thioalkalivibrio versutus gen. nov., sp. nov., Thioalkalivibrio nitratis
sp. nov. and Thioalkalivibrio denitrificans sp. nov., novel obligately alkaliphilic and obligately
chemolithoautotrophic sulfur-oxidizing bacteria from soda lakes. Int J Syst Evol Microbiol
51:565580
Sorokin DY, Teske A, Robertson LA, Kuenen JG (1999) Anaerobic oxidation of thiosulfate to
tetrathionate by obligately heterotrophic bacteria, belonging to the Pseudomonas stutzeri
group. FEMS Microbiol Ecol 30:113123
Steinmetz MA, Fischer U (1982) Cytochromes of the green sulfur bacterium Chlorobium vibrio-
forme f. thiosulfatophilum. Purification, characterization and sulfur metabolism. Arch
Microbiol 19:26
Suzuki H, Koyanagi T, Izuka S, Onishi A, Kumagai H (2005) The yliA, -B, -C, and -D genes of
Escherichia coli K-12 encode a novel glutathione importer with an ATP-binding cassette.
Trper HG, Pfennig N (1966) Sulphur metabolism in Thiorhodaceae. III. Storage and turnover of
thiosulphate sulphur in Thiocapsa floridana and Chromatium species. Antonie Van
Leeuwenhoek Int J Gen Mol Microbiol 32:261276
Vermeglio A, Li J, Schoepp-Cothenet B, Pratt N, Knaff DB (2002) The role of high-potential iron
protein and cytochrome c(8) as alternative electron donors to the reaction center of Chromatium
vinosum. Biochemistry 41:88688875
Vert F, Kostanjevecki V, de Smet L, Meyer TE, Cusanovich MA, van Beeumen JJ (2002)
Identification of a thiosulfate utilization gene cluster from the green phototrophic bacterium
Chlorobium limicola. Biochemistry 41:29322945
Williams TJ, Zhang CL, Scott JH, Bazylinski DA (2006) Evidence for autotrophy via the reverse
tricarboxylic acid cycle in the marine magnetotactic coccus strain MC-1. Appl Environ
Chapter 10
Sulfur Oxidation in Chlorobium tepidum
(syn. Chlorobaculum tepidum):
Genetic and Proteomic Analyses
Leong-Keat Chan, Rachael Morgan-Kiss, Thomas E. Hanson
Abstract Chlorobium tepidum (syn. Chlorobaculum tepidum) has become the

model system of choice for understanding the unique biological attributes of
the green sulfur bacteria, the Chlorobiaceae. This chapter describes how genome
sequence enabled genetic and proteomic approaches are being applied to under-
stand pathways of anaerobic sulfur oxidation in C. tepidum. Reduced sulfur com-
pounds are the sole source of exogenous reductant that C. tepidum utilizes to
drive all anabolic pathways necessary for cellular growth, including carbon and
nitrogen fixation. The stoichiometries of sulfur-compound conversions in batch
cultures confirm that sulfide oxidation occurs via extracellular elemental sulfur.
No intermediate is apparent for the oxidation of thiosulfate to sulfate, but thiosul-
fate oxidation appears to be stimulated when cells are grown autotrophically.
Mutation of predicted sulfur oxidation genes leads to pleiotropic phenotypes that
appear to affect the organization of photopigments in cells, suggesting that sulfur
oxidation and light harvesting are tightly integrated processes in C. tepidum. In
concert with genetic approaches, proteomics coupled with subcellular fractiona-
tion is being used to identify proteins that are potentially involved in the oxidation
of extracellular elemental sulfur. Observations on the next generation of genetic
techniques to augment those that currently exist in C. tepidum and to extend
proteomic observations are presented throughout.
10.1 Introduction
10.1.1 Background
Anaerobic sulfur oxidation is an important, but poorly understood aspect of the glo-
bal sulfur cycle. This chapter will detail our recent efforts at identifying the relevant
genes encoding enzymes of anaerobic sulfur oxidation in the green sulfur bacterium
Chlorobium tepidum (syn. Chlorobaculum tepidum; Imhoff 2003). To this end, we
have taken two complementary approaches to this goal: genome-directed genetic
analysis of predicted sulfur oxidation genes and proteomic analyses to identify
117
Springer 2008
118 L.-K. Chan et al.
proteins with appropriate subcellular localization and properties to participate in the

oxidation of elemental sulfur, which is accumulated extracellularly.
C. tepidum oxidizes reduced sulfur compounds to provide all necessary reducing
equivalents during growth. This includes ATP and proton motive force generation
via photosynthetic electron transport (Brune 1995), dinitrogen fixation via nitroge-
nase in the absence of combined nitrogen (Wahlund and Madigan 1993), carbon
dioxide fixation via the reductive tricarboxylic acid cycle (Buchanan and Arnon
1990), and all other anabolic processes requiring NADH, NADPH, ferredoxin (Fd),
and other redox mediators. Reduced Fd is the initial product of photosynthetic
electron transport and is utilized directly for carbon dioxide and nitrogen fixation.
Reduced Fd can also be converted to NADPH via a recently described Fd:NADP+
oxidoreductase (Seo and Sakurai 2002). Light harvesting for photosynthetic elec-
tron transport is accomplished via the unique antenna structure of the chlorosome
and energy transfer is accomplished by a type I reaction center (Frigaard and
Bryant 2004).
Much is known about how C. tepidum generates intracellular redox mediators
downstream of the reaction center. Much less is known about the input side of the
photosynthetic electron transport chain, where more limited information has been
gleaned from biochemical studies of other green sulfur bacteria. C. tepidum is an
obvious model system to pursue this question as it is genetically amenable (Frigaard
and Bryant 2001; Hanson and Tabita 2001) and the genome has been sequenced
and annotated (Eisen et al. 2002).
10.1.2 Sulfur-Compound Dynamics in C. tepidum

Batch Cultures
C. tepidum is capable of using multiple forms of reduced sulfur with differing

redox potentials to feed into the photosynthetic electron transport chain. In a
typical batch culture medium, both sulfide and thiosulfate are provided to the
cells (Wahlund et al. 1991). Under these conditions, sulfide is oxidized first and
elemental sulfur accumulates as extracellular sulfur globules. Elemental sulfur
accumulation is stoichiometric relative to the sulfide consumed, with small
amounts of thiosulfate produced in some experiments (L.K. Chan, R.M.
Morgan-Kiss, T.S. Weber, and T.E. Hanson, unpublished results). It has been
reported that sulfide is required for the growth of C. tepidum (Wahlund et al.
1991); however, recent experiments in our laboratory and others have shown
that this is not the case and that C. tepidum can be grown with thiosulfate and
elemental sulfur as electron donors in the absence of exogenously provided
sulfide (data not shown).
The oxidation of elemental sulfur commences only after sulfide has been
depleted to undetectable levels in the medium. This poses an interesting challenge
to C. tepidum and other organisms that use elemental sulfur as an electron donor.
Elemental sulfur is sparingly soluble (less than 5 g l1) and so to extract reducing
10 Sulfur Oxidation in Chlorobium tepidum (syn. Chlorobaculum tepidum) 119
Table 10.1 Thiosulfate consumption and sulfate production in triplicate batch cultures of
Chlorobium tepidum WT2321 under autotrophic and mixotrophic conditions
Culture S2O32 consumed (mM) SO42 produceda (mM) SO42/S2O32
Mixotrophic 2.0 4.7 2.4
Autotrophic 9.0 18.9 2.1
a
Values were corrected for sulfate produced from elemental sulfur by subtracting the maximal
amount of elemental sulfur observed from the total sulfate produced.
equivalents from this material C. tepidum must have a specific mechanism for
accessing and mobilizing it. The problem faced by C. tepidum conceptually resem-
bles the problem of organisms that utilize insoluble materials as electron acceptors
under anaerobic conditions. These organisms usually either directly attach to the
substrate to facilitate reduction or utilize extracellular redox mediators to reduce
surfaces at a distance (Lies et al. 2005). In both cases, it appears that outer mem-
brane associated cytochromes are important for delivering reducing equivalents to
the cell surface. Conceptually, it seems reasonable that a similar mechanism work-
ing in reverse could participate in the oxidation of extracellular elemental sulfur
(Sect. 10.3).
Thiosulfate oxidation commences after the onset of elemental sulfur oxidation.
The extent of thiosulfate oxidation appears to be controlled by the demand for
reducing equivalents. This is obvious when comparing mixotrophic (carbon diox-
ide and acetate as the carbon source) and autotrophic (carbon dioxide as the sole
carbon source) growth. As shown in Table 10.1, mixotrophic batch cultures con-
sumed 4.5-fold less thiosulfate than autotrophic cultures. Biomass yields under
autotrophic and mixotrophic conditions are essentially identical (data not shown),
suggesting that C. tepidum is able to integrate signals for redox demand and adjust
its sulfur-oxidizing capability appropriately.
Sulfate is the final product of anaerobic sulfur oxidation produced by C. tepidum.
It begins to accumulate when elemental sulfur oxidation commences, but no sooner.
This observation, along with the 1:1 stoichiometry of elemental sulfur production
from sulfide (L.K. Chan, R.M. Morgan-Kiss, T.S. Weber, and T.E. Hanson, unpub-
lished results), suggests that sulfide cannot be oxidized to sulfate without elemental
sulfur as an intermediate. Sulfate is produced proportionately to the amount of ele-
mental sulfur and thiosulfate oxidized. The stoichiometry of sulfate production con-
forms to the expected value of 2:1, though this is somewhat elevated in mixotrophic
cultures (Table 10.1), perhaps suggesting that cultures with lower demand for
reducing equivalents store some form of sulfur as a hedge against lean times.
The stoichiometry described above further suggests that C. tepidum fully oxidizes
thiosulfate without any significant accumulation of side products. This is noteworthy
as C. tepidum, like all other green sulfur bacteria and some purple sulfur bacteria,
lacks genes encoding the SoxCD sulfur dehydrogenase that is required for the
complete oxidation of thiosulfate in the Paracoccus pantotrophus sulfur-oxidizing
system studied by Friedrich et al. (2001).
10.2 Genetic Analyses
10.2.1 Organization of Genes Encoding Putative Sulfur

Oxidation Functions
Examination of the complete, annotated genome sequence of C. tepidum led to the

observation that a number of the genes encoding presumptive sulfur oxidation enzymes
are tightly clustered on the genome. The three largest clusters are outlined in Table 10.2.
Together, these three gene clusters contain 52 kilobases of DNA encoding 68 proteins,
which correspond to 2.4% of the total genome and 3.0% of the total number of protein
coding genes in C. tepidum. As noted in Chap. 6 by Frigaard and Bryant, these genes
are generally well conserved in other green sulfur bacterial genomes.
10.2.2 Mutations Affecting Sulfur Oxidation Have Secondary

Effects on Light Harvesting
The tight clustering of many sulfur oxidation genes led us to use an in vitro trans-
position mutagenesis approach (Hayes 2003) to isolate insertion mutations in
defined regions of the C. tepidum genome. We have used a transposon, TnOGm,
derived from the plasposon pTnModOGm to generate mutant strains in a number
of the C. tepidum sulfur islands, including those listed in Table 10.2. Interestingly,
many of the isolated mutants display altered coloration relative to the wild-type
parental strain that apparently results from shifts in the major in vivo absorption
band associated with bacteriochlorophyll c in the chlorosome (Table 10.3). Most of
these strains display moderate to severe defects in growth and sulfur oxidation.
One particular mutant in Sulfur Island-I, C5, carries a TnOGm insertion that has
replaced approximately 5 kb of DNA encoding two subunits of the Hdr/Qmo complex,
an Sqr homolog and several hypothetical proteins (Chan et al. 2007). This strain dis-
plays a blueshifted max for chlorosomal bacteriochlorophyll c at 750 nm relative to the
755 nm maximum of the wild type. In this strain, as in all TnOGm strains containing
Table 10.2 C. tepidum genomic sulfur islands

No. of base Putative sul-
Name Genes pairs fur oxidation Hypothetical Gene Products
SI-I CT0841- 26,636 18 18 Dsr, ApsBA, Sat,
CT0877 Hdr/Qmo, Sqr
SI-II CT2238- 12,505 11 3 Dsr, siroheme
CT2252 biosynthesis
SI-Sox CT1009- 13,567 8 10 Sox complex (no
CT1027 SoxCD)
Totals 52,708 37 31
Table 10.3 Shifts in the major chlorosome absorption peak in C. tepidum strains carrying
TnOGm insertions in Sulfur Island I
TnOGm site Genes No. of strains In vivo maxa (nm) MeOH maxa (nm)
None (WT2321) 755 669
SI-I-2 CT0854-CT0869 2 750 669
2 760 669
SI-I-3 CT0869-CT0877 5 747 669
7 750 669
9 752 669
a
Measurements are the means of three independent cultures. The standard deviation in all
triplicates was 2 nm or less.
similar shifts examined to date, the absorption maximum difference disappears when
pigments are extracted into methanol (Table 10.3), indicating that the shift is due to
the arrangement of the bacteriochlorophyll molecules in the chlorosome rather than
the bulk properties of the molecules themselves. Strain C5 was also found to display
many other properties suggesting that this strain is fundamentally compromised in the
way that it deals with incident light energy, including increases in baseplate bacterio-
chlorophyll a fluorescence yield, which may indicate poor energy transfer efficiency
between the chlorosome and reaction center (Wang et al. 1990; Melo et al. 2000).
None of the genes deleted in this mutant are obvious candidates for chlorosome struc-
tural proteins, so it seems that the effect of this mutation is likely indirect.
Strain C5 is also severely compromised for growth (Chan et al. 2007) and sulfur
oxidation (L.K. Chan, R.M. Morgan-Kiss, T.S. Weber, and T.E. Hanson, unpublished
results); therefore, we propose that the observed alterations in light harvesting and
antenna function are a secondary result of defects in sulfur oxidation pathways. Similar
effects were seen in a strain of C. tepidum lacking the ribulose-1,5-bisphosphate car-
boxylase/oxygenase (Rubisco) like protein encoded by CT1772 that was also defective
in thiosulfate oxidation (Hanson and Tabita 2001). The major difference between these
two strains is that the Rubisco-like protein mutant had a decrease in the level of bacteri-
ochlorophyll per protein, while strain C5 displays no such defect. These results suggest
that C. tepidum regulates the function of its antenna apparatus in response to the availa-
bility of reductant, as has been reported by others working with chlorosomes in
Chloroflexus aurantiacus and C. tepidum (Wang et al. 1990; Melo et al. 2000). Details
regarding the nature of such a signal and the mechanism by which it is transmitted are
currently unclear, but the fact that mutants compromised in sulfur oxidation affect this
regulation indicates a sulfur oxidation intermediate may be involved.
10.2.3 Additional Genetic Techniques Are Needed
While the ability to make targeted gene disruptions in C. tepidum has been and will
continue to be valuable for understanding the biology of this organism, additional
techniques will be required to enable more sophisticated and subtle experimental
manipulations of the genome. Two primary areas are viewed as likely to yield the
greatest benefit. The first is epitope tagging where a gene is modified in place on the
C. tepidum chromosome to produce a variant protein that can be recognized by tag-
specific antibodies. Methods for appending a hexahistidine tag to specific gene prod-
ucts by manipulating the genes in place on the chromosome were recently reported
for Escherichia coli (Morgan-Kiss and Cronan 2004). This allows the detection and
quantification of protein expression without the need for generating a specific anti-
body for each protein of interest. In addition, the hexahistidine tag allows purification
of the tagged protein by chromatographic methods (Morgan-Kiss and Cronan 2004).
This enables the detection and analysis of protein complexes and can greatly facilitate
biochemical analysis of the gene product.
The second is to develop a chromosomal expression system with a regulated
promoter. Currently, no plasmids exist for the complementation of mutants in
C. tepidum. Failing the development of these systems, one can envision comple-
mentation of mutants by ectopic copies of genes under a regulated promoter. This
is one route to the expression of site-directed mutants in genes of interest in a null
mutant background to enable a more detailed understanding of structurefunction
relationships in the original physiological background.
We are currently attempting to adapt protocols for both techniques for use in
C. tepidum.
10.3 Proteomic Analysis
10.3.1 Why Proteomics?
As noted in Table 10.2, a large number of genes, almost 50% of the total, in the
Sulfur Islands of C. tepidum encode hypothetical proteins with unknown functions
relative to sulfur oxidation. Thus, their biological functions have yet to be discov-
ered despite prior investigations into the biochemistry of sulfur oxidation. While
the genetic approach described in Sect. 10.2 will yield some information as to what
genes are important, understanding what proteins are expressed during growth on
particular sulfur compounds and what their location is in the cell (or outside of it)
can provide a second line of evidence for the involvement of particular gene prod-
ucts in the oxidation of particular sulfur compounds. This will be particularly true
if some sulfur oxidation genes are essential to C. tepidums viability.
10.3.2 Proteomic Analysis of Subcellular Fractions
Specific hypotheses amenable to proteomic analyses can be posed regarding par-

ticularly problematic aspects of anaerobic sulfur oxidation like the oxidation of
extracellular elemental sulfur. One such hypothesis that can be directly tested more
easily by proteomics than by current genetic techniques in C. tepidum is whether or
Table 10.4 Proteins identified from inner- and outer-membrane fractions of C. tepidum
No. of MS/MS peptidesa
CT no. Annotation Total membrane Outer membrane
CT2144 Outer surface protein 12 16
CT1499 FmoA, Bchla binding protein 21 16
CT0893 Hypothetical protein 10 15
CT1353 OmpA family protein 5 9
CT1447 Serine protease 7 9
CT0254 OmpH, outer-surface protein 7 4
CT0641 PscD, reaction center protein 2 0
CT0638 Peptidoglycan-associated lipoprotein 5 0
CT2033 ATP synthase F1, subunit 8 0
a
Liquid chromatographytandem mass spectrometry (MS/MS) was performed on proteolytically
digested bands from 1D sodium dodecyl sulfate polyacrylamide gel electrophoresis gels of
the indicated fractions.
not C. tepidum contains electron transfer proteins associated with the outer
membrane or cell surface. This would provide a clear mechanism for transferring
reducing equivalents either directly to elemental sulfur to liberate sulfide and
polysulfides, or to a redox-shuttling compound. Outer-membrane proteins have
been implicated in sulfur oxidation in Thiobacillus ferroxidans and in mediating the
reduction of extracellular electron acceptors in Shewanella oneidensis MR-1
(Buonfiglio et al. 1993; Lies et al. 2005).
A method previously used to enrich outer-membrane proteins from the Gram-negative
marine bacterium Hyphomonas jannaschiana (Shen et al. 1989) was applied to C.
tepidum and was found to reliably provide fractionation of chlorosome-depleted mem-
branes. This method relies on the selective solubilization of inner-membrane proteins
by nonionic detergents. Proteins have been identified from C. tepidum total (inner and
outer) and outer-membrane fractions (Nonidet P-40 insoluble fraction) by standard tan-
dem mass spectrometry methods. The results indicate that the fractionation protocol has
specifically enriched outer-membrane proteins from a total membrane preparation
(Table 10.4). This is clearly seen when the distribution of known inner-membrane pro-
teins like the F1 ATPase subunit and PscD reaction center subunit are examined.
Peptides for these proteins were found solely in the total membrane fraction, and not in
the outer-membrane fraction. The enrichment of predicted outer-surface and outer-
membrane proteins like CT2033 and OmpA was observed in terms of the number of
peptides detected in the outer-membrane fraction.
The exception to this rule is the FmoA protein, which was found to be abundant
in both fractions. The FmoA protein is a peripherally membrane associated protein
in C. tepidum that mediates the association between the inner membrane and the
chlorosome (Imhoff 2003). This result indicates that the outer-membrane fraction
prepared by selective detergents probably enriches both outer-membrane proteins
and peripheral-membrane proteins together. The localization of CT1087, sulfide:
quinone oxidoreductase that is likely a peripheral-membrane protein (Shahak et al.

1992), to this outer-membrane fraction also supports this notion (L.K. Chan, R.M.
Morgan-Kiss, T.S. Weber, and T.E. Hanson, unpublished results).
The outer-membrane and peripheral-membrane fraction has been examined by
heme staining after electrophoresis and three hemoproteins were revealed (Chan
et al. 2007). These are currently being identified (R.M. Morgan-Kiss, L.K. Chan,
and T.E. Hanson, unpublished results). Given the caveat regarding the composition
of the outer-membrane fraction, these hemoproteins may be either peripheral-
membrane proteins or outer-membrane proteins. The C. tepidum genome encodes
26 potential hemoproteins that contain a consensus CXXCH heme attachment
motif. Fifteen of these are associated with functional annotations and represent
expected hemoproteins of the Dsr complex, Sox system, and others. The remain-
ing 11 potential hemoproteins are not functionally annotated. If any of these are
found in the outer-membrane fraction, then the next obvious step will be to deter-
mine their subcellular localization by employing the classic microbial genetic
technique of alkaline phospatase translational fusions. These fusions are only
active when they are transported to the cell exterior and can be used in concert
with -galactosidase fusion proteins to assess protein topology (Haardt and
Bremer 1996). Construction of these fusions in the C. tepidum chromosome will
further develop the capabilities for genetics in this system by demonstrating the
use of translational reporter fusions.
10.4 Conclusions
The combination of both genetic and proteomic techniques to address hypotheses

proposed on the basis of genomic sequences is a hallmark of the postgenomics era.
Both approaches have their strengths and weaknesses and have only recently been
applied to C. tepidum. The development and application of improved genetic
techniques and higher-throughput proteomic techniques will continue to improve the
ability to perform more sophisticated and subtle experiments in these fascinating
organisms. With the accumulation of additional green sulfur bacterial genomes (see
Chap. 6 by Frigaard and Bryant), the next large challenge in the field will be to extend
the techniques to these additional strains so that unique genetic features of each can be
assessed in their proper biological context. In addition, the extension of the understanding
developed through these studies in laboratory cultures must be extended to environmental
populations to truly understand how the green sulfur bacteria function in situ.
Acknowledgements. The authors would like to thank Joy Lawani, Jessica Martin, Egle Burbaite,
Tim Weber, and Michele Madorma for excellent technical assistance over the course of the project
and Ann OBrien for protein identification. This project was supported by grants from the National
Science Foundation (MCB-0447649 to T.E.H. and Delaware EPSCoR grant EPS-0447610 through
the Delaware Biotechnology Institute), and utilized common instrumentation facilities provided in
part by the National Institutes of Health (P20-RR116472-04 from the IDeA Networks of
Biomedical Research Excellence program of the National Center for Research Resources).
References
Brune DC (1995) Sulfur compounds as photosynthetic electron donors. In Bauer CE (ed)

Anoxygenic photosynthetic bacteria. Kluwer, Amsterdam, pp 847870
Buchanan BB, Arnon DI (1990) A reverse Krebs cycle in photosynthesis: consensus at last.
Photosynth Res 24:4753
Buonfiglio V, Polidoro M, Flora L, Citro G, Valenti P, Orsi N (1993) Identification of two outer
membraneproteins involved in the oxidation of sulphur compounds in Thiobacillus ferrooxi-
dans. International Symposium on Advances on Biohydrometallurgy: Microbiol Appl
11:4350
Chan LK, Morgan-Kiss R, Hanson TE (2007) Genetic and proteomic studies of sulfur oxidation
in Chlorobium tepidum. In: Hell R, Dahl C, Leustek T, Knaff D (eds) Sulfur in phototrophic
organisms. Springer, New York (in press)
tepidum TLS, a photosynthetic, anaerobic, green-sulfur bacterium. Proc Natl Aacad Sci USA
99:95099514
Frigaard NU, Bryant DA (2001) Chromosomal gene inactivation in the green sulfur bacterium
Chlorobium tepidum by natural transformation. Appl Environ Microbiol 67:25382544
Frigaard NU, Bryant DA (2004) Seeing green bacteria in a new light: Genomics-enabled studies
of the photosynthetic apparatus in green sulfur bacteria and filamentous anoxygenic
phototrophic bacteria. Arch Microbiol 182:265276
Haardt M, Bremer E (1996) Use of phoA and lacZ fusions to study the membrane topology of
ProW, a component of the osmoregulated ProU transport system of Escherichia coli.
Hanson TE, Tabita FR (2001) A ribulose-1,5-bisphosphate carboxylase/oxygenase (Rubisco)-like
protein from Chlorobium tepidum that is involved with sulfur metabolism and the response to
oxidative stress. Proc Natl Acad Sci USA 98:43974402
Hayes F (2003) Transposon-based strategies for microbial functional genomics and proteomics.
Annu Rev Genet 37:329
Imhoff JF (2003) Phylogenetic taxonomy of the family Chlorobiaceae on the basis of 16s rRNA
and fmo (Fenna-Matthews-Olson protein) gene sequences. Int J Syst Evol Microbiol
53:941951
Lies DP, Hernandez ME, Kappler A, Mielke RE, Gralnick JA, Newman DK (2005) Shewanella
oneidensis MR-1 uses overlapping pathways for iron reduction at a distance and by direct
contact under conditions relevant for biofilms. Appl Environ Microbiol 71:44144426
Melo TB, Frigaard NU, Matsuura K, Razi Naqvi K (2000) Electronic energy transfer involving
carotenoid pigments in chlorosomes of two green bacteria: Chlorobium tepidum and
Chloroflexus aurantiacus. Spectrochim Acta A Mol Biomol Spectrosc 56:20012010
Morgan-Kiss RM, Cronan JE (2004) The Escherichia coli fadK (ydiD) gene encodes an anerobi-
cally regulated short chain acyl-CoA synthetase. J Biol Chem 279:3732437333
Seo D, Sakurai H (2002) Purification and characterization of ferredoxin-NAD(P)(+) reductase
from the green sulfur bacterium Chlorobium tepidum. Biochim Biophys Acta 1597:123132
Shahak Y, Arieli B, Padan E, Hauska G (1992) Sulfide quinone reductase (SQR) activity in
Chlorobium. FEBS Lett 299:127130
Shen N, Dagasan L, Sledjeski D, Weiner RM (1989) Major outer membrane proteins unique to
reproductive cells of Hyphomonas jannaschiana. J Bacteriol 171:22262228
Wahlund TM, Madigan MT (1993) Nitrogen fixation by the thermophilic green sulfur bacterium
Chlorobium tepidum. J Bacteriol 175:474478
Wahlund TM, Woese CR, Castenholz RW, Madigan MT (1991) A thermophilic green sulfur bac-
terium from New Zealand hot springs, Chlorobium tepidum sp. nov. Archiv Microbiol
156:8190
Wang J, Brune DC, Blankenship RE (1990) Effects of oxidants and reductants on the efficiency
of excitation transfer in green photosynthetic bacteria. Biochim Biophys Acta 1015:457463
Chapter 11
Structural Insights into Component SoxY
of the Thiosulfate-Oxidizing Multienzyme
System of Chlorobaculum thiosulfatiphilum
Jan Stout, Lina De Smet, Bjorn Vergauwen, Savvas Savvides,

Jozef Van Beeumen
Abstract We discuss the crystal structure of component SoxY of the SoxYZ com-
plex that is known to play a key role in the sulfur-oxidizing multienzyme system of
the green sulfur bacterium Chlorobaculum thiosulfatiphilum. The protein appears
to be structurally similar to a monomeric immunoglobulin-like protein that oli-
gomerizes into a tetramer via conserved contact regions between the monomers.
The tetramer is a dimer of dimers and exhibits one large hydrophobic contact
region in each dimer, and two small hydrophilic interface patches between the dim-
ers. At the tetramer interface patch, two conserved redox-active C-terminal
cysteines form an intersubunit disulfide bridge. Depending on the redox state of
the cysteines, the tetramer is in equilibrium with the dimers, each one of which is
a candidate to covalently bind a thiosulfate molecule by means of a thioldisulfide
exchange reaction with the interprotein disulfide bonds. The significant conservation
level of the interfaces, the specific interactions between the subunits in the tetramer,
and the dimertetramer equilibrium suggest that these SoxY oligomers are biologi-
cally relevant. A possible role for these protomers in the mechanism of the Sox-
system is proposed.
11.1 Introduction
The oxidation of thiosulfate proceeds in Eubacteria and Archaea mainly via two
distinct pathways: (1) the tetrathionate pathway, which is mainly restricted to
acidophilic thiobacilli and oxidizes thiosulfate to sulfate via tetrathionate as an
intermediate (Kelly et al. 1997), and (2) the ubiquitous periplasmic sulfur oxidizing
(Sox) pathway found in neutrophilic, respiratory, and phototrophic Proteobacteria
and in green sulfur bacteria (Friedrich et al. 2001, Friedrich 2005). The best
characterized Sox systems to date are those of Paracoccus versutus (Lu and Kelly
1983; Lu et al. 1985; Lu 1986) and Paracoccus pantotrophus (Friedrich et al.
2000; Rother et al. 2001), where the combination of four components of the latter
organism, SoxYZ, SoxAX, SoxB, and SoxCD, results in an active system exhibit-
ing a full thiosulfate oxidizing activity and generates eight electrons and two sulfate
molecules as end products (Friedrich et al. 2000). Although the involvement of
127
Springer 2008
128 J. Stout et al.
these components, and, more recently, of other Sox components such as SoxW and
SoxV (Bardischewsky and Friedrich 2001; Appia-Ayme and Berks 2002;
Bardischewsky et al. 2006a), SoxR and SoxS (Rother et al. 2005), SoxT (Lahiri et
al. 2006), and SoxF (Bardischewsky et al. 2006b), in the Sox system has been
established, their specific role, with the exception of that of SoxYZ, remains hard
to determine.
SoxYZ has the capacity to bind reduced sulfur substrates via a thioether or a
thioester bond at a conserved C-terminal cysteine of the SoxY subunit (Quentmeier
and Friedrich 2001), and it is thought to present this activated sulfur substrate
molecule to the other oxidizing Sox enzymes. Although this widely accepted view
is supported by clear evidence (Quentmeier and Friedrich 2001), it remains
unknown, however, how this sulfur substrate binding step exactly occurs. In the
reaction model proposed by Friedrich et al. (2001), the SoxAX cytochrome c is
proposed to mediate the binding, and an appropriate reaction mechanism based on
the crystal structure of SoxAX for this binding event was postulated by Bamford
et al. (2002). In this model, SoxAX transfers a sulfur substrate molecule, covalently
bound to a conserved cysteine residue that also functions as the sixth axial ligand
of a heme prosthetic group, to the sulfhydryl group of a strictly conserved cysteine
residue located in a highly conserved C-terminal sequence motif of SoxY. An alter-
native mechanism was also proposed in which SoxY binds the sulfur substrate via
an intersubunit disulfide bridge formed with a second SoxY molecule (Quentmeier
et al. 2003). This hypothesis was based on the observation that SoxYZ exhibits
redox activity by means of its conserved cysteine residue (Quentmeier et al. 2003)
and is further supported by recent results indicating that SoxYZ, having SoxY
disulfide-bridged subunits and SoxYpersulfide subunits, increases the thiosulfate-
oxidizing activity of the system (A. Quentmeier, P Jannig, and C.G. Friedrich, personal
communication). We here present the crystal structure of a standalone tetrameric
SoxY from the green sulfur bacterium Chlorobium limicola f. sp. thiosulfatophilum
DSM 249T, recently renamed Chlorobaculum thiosulfatiphilum (Imhoff 2003).
The structure reveals specific and well-conserved contact interfaces between the
subunits which are linked via an intersubunit disulfide bond, offering a first detailed
picture of the structural basis of its redox activity.
11.2 SoxY Structure
11.2.1 Overall Structure
The SoxY crystal structure reveals an -protein consisting of an N-terminal -helix

and a -sandwich domain (Fig. 11.1). The two SoxY subunits present in the asym-
metric unit interact extensively with each other and form an extended -sandwich
structure with six and eight -strands in the upper and the lower layer, respectively
(Fig. 11.1d, e). On their turn, two of these SoxY dimers interact with each other via
11 Structural Insights into Component SoxY 129
Fig. 11.1 a The SoxY tetramer of Chlorobaculum thiosulfatiphilum viewed from the side, show-
ing dimer A and dimer A, which are related via a crystallographic twofold axis (dashed line with
filled ellipsoid on top). Tetramerization occurs by means of two small interface patches at the
distal ends of the oligomer and is indicated by two dashed ellipsoids. b Sodium dodecyl sulfate
polyacrylamide gel electrophoresis of crystallized SoxY (lanes 1 and 2) and purified SoxY (lanes 3
and 4). Lanes 1 and 3 are protein samples boiled at 95C in Laemmli buffer. Lanes 2 and 4 are
protein samples treated with b-mercaptoethanol before boiling. c SoxY monomer showing the
secondary structure elements. The N-terminal -helix (S1F20) is connected via a loop to a
-sandwich domain (I31G122) having seven antiparallel -strands: -strands a (I31K34),
b (A43T51), c (N58T63), c (M70L77), e (P82M90), f (E94A102), and g (K105T116). The
nomenclature of the -strands was taken from Bork et al. (1994). The red -strands (a, b, and e)
form -sheet I. The blue -strands (c, c c, f, and g) constitute -sheet II. d, e The SoxY dimer
orientated with both -sheets in the plane of and orthogonal to the page, respectively. The two
monomers are shown in different colors
a twofold crystallographic axis and assemble into a tetramer (Fig. 11.1a). This
agrees well with analytical gel filtration experiments showing that recombinant
SoxY is present in solution as a 52-kDa tetramer (Stout et al. 2006).
Sodium dodecyl sulfate (SDS) polyacrylamide gel electrophoresis revealed that
the tetramer is composed of dimer subunits which are covalently linked via a
disulfide bridge (Fig. 11.1b), a structural property that was also demonstrated for
SoxY of P. pantotrophus (Quentmeier et al. 2003). Interestingly, when solubilized,
crystalline SoxY migrates in a SDS polyacrylamide gel like untreated recombinant
SoxY does, i.e., as dimer subunits linked via a disulfide bridge. This is quite
130 J. Stout et al.
remarkable since the thiol-reducing reagent dithiothreitol (DTT) is absolutely

required for successful crystallization (Stout et al. 2006) and is expected to reduce
solvent-exposed disulfide bonds at millimolar concentrations (10 mM), even under
the acidic crystal growth conditions (pH 4.0). Being the only cysteine residue in the
mature SoxY, C120 has logically been assigned to be the site of covalent linkage.
Consequently, a total of two disulfide bridges are formed in the tetramer.
Unfortunately, no detailed structure for the disulfide bridge region could be
determined, owing to the very poor quality of the electron densities of the C120
residues. This is an indication for a high structural flexibility of this region which
in all likelihood is correlated with the high glycine content of the GGCG(G) motif.
Nevertheless, the molecular assembly of the tetramer makes it feasible to deduce
the position of the disulfide bridges, as will be discussed later.
11.2.2 SoxY Monomer
The 13-kDa SoxY monomer consists of an N-terminal -helix (S1F20) packed

against an immunoglobulin (Ig) like domain (I31I117). Both structural elements
are connected by a loop (G21T30) covering one side of the -sandwich domain
(Fig. 11.1c). Variations are known to exist in the 3D topology of Ig domains, on the
basis of which distinct Ig sets have even been defined (Williams and Barclay 1988;
Bork et al. 1994; Harpaz and Chothia 1994). The Ig-like fold of SoxY is an s-type
Ig domain on the basis of the classification scheme of Bork et al. (1994).
11.2.3 SoxY Dimer
The two monomers within the crystal asymmetric unit use one edge of their Ig-like
domains to form a -sandwich dimer constituted of two extended -sheets: -sheet
I consists of strands a, b, and e of subunits A and B, and -sheet II is composed of
strands c, c, f, and g of both subunits. Both -sheets in SoxY exhibit a continuous
hydrogen-bonding network (Fig. 11.1d, e). This is in contrast to canonical extended
-sandwiches which have one continuous (sheet II) and one discontinuous (sheet I)
-sheet, separated in the middle by water molecules (Richardson and Richardson
2002). In the dimer interface of SoxY, however, a total of 12 direct intersubunit
hydrogen bonds are present between the main chains of -strands c and e and their
equivalent -strands of the opposite monomer.
The structural and physicochemical characteristics of the SoxY dimer interface
are similar to those of other stable proteinprotein interfaces in known protein
complexes. The dimer interface buries a surface of 1,522 2, which agrees well
with canonical proteinprotein interfaces (1,2002,000 2) (Lo Conte et al. 1999;
Bahadur et al. 2004), and is mainly hydrophobic (70.5%). It has a substantial frac-
tion (40%) of completely buried interface atoms (Stout et al. 2007), a value in
accordance with the observed values corresponding to completely shielded inter-

face atoms (3436%) (Lo Conte et al. 1999; Bahadur et al. 2003). The burial of the
hydrophobic surface is assumed to be an important feature of stable proteinprotein
interfaces (Young et al. 1994; Jones and Thornton 1996; Tsai et al. 1997). In the
case of SoxY, the major part of the hydrophobic residues is located at the centre of
the dimer interface. These residues are flanked at the bottom and the top by the two
layers of intersubunit hydrogen bonds (12 in total) that are part of the -ladder net-
work of the two extended -sheets described above. Hydrogen bonds, being formed
by complementary electrostatic, polar, or charged-atom groups of both proteins, are
thought to impose some degree of specificity in proteinprotein binding events (Xu
et al. 1997a, b).
Several studies reported that interface residues are, on average, more conserved
than solvent-exposed surface residues (Valdar and Thornton 2001a; Caffrey et al.
2004; Mintseris and Weng 2005). In the case of SoxY, bootstrap analyses demon-
strated that the core of the dimer interface, which consists of ten residues that
have at least 95% of their total surfaces buried at the interface, is significantly
conserved (Stout et al. 2007).
11.2.4 SoxY Tetramer
In contrast to the dimer interface, contacts that stabilize the tetramer occur at the
distal ends of the tetramer by virtue of two small interface patches, separated by a
solvent-filled space between the two SoxY dimers (Fig. 11.1a). The buried surface
area per tetramer interface patch is 630 2 and the tetramer interface is less polar
(nonpolar area fraction of 57.1 %) than the dimer interface (Stout et al. 2007). Two
strongly conserved protein regions of both SoxY subunits make up the tetramer
interface: the ab loop connecting -strands a and b (P36G42) and the beginning
of -strand b (A43P46), R111I117 of -strand g, and the equivalent regions of
the symmetry-related subunit. The center of the interface is composed of the strictly
conserved peptide sequences E37E40, exhibiting a -like conformation, of the ab
loop of both subunits. These short peptides are orientated in an antiparallel way but
do not interact via direct intersubunit backbone hydrogen bonds. Instead, six
ordered water molecules and two ordered chloride ions intercalate between the
peptides and set up a hydrogen-bonding network with both subunits. The chloride
ions appear to be an artifact of crystal cryo-cooling at 100 K, as they are replaced
by water molecules in a room-temperature crystal structure. Owing to the high level
of hydration, and in contrast to the dimer interface, only a small fraction (6.7%) of
the tetramer interface atoms are fully buried.
Both sides of one tetramer interface patch are defined by two strictly conserved
salt bridges (3.8 ) between E40 and K114 of the symmetry-related subunit (Fig. 11.2).
A second glutamate residue, E37, of the symmetry-related subunit with respect to
E40, is positioned with its carboxyl group between the carboxyl group of E37, and
NZ of K114 and is within hydrogen-bonding distance of OE2 of E40 (2.8 ). This
132 J. Stout et al.
Fig. 11.2 The tetramer interface patch, made up by the subunits A and A, with the two con-
served salt bridges (dashed lines) between E40K114 and E40K114 defining both sides of the
patch. E37 and E37 are within hydrogen-bonding distance of E40 and E40, respectively. K114,
K114, and the differently colored K92, K92, R89, and R89 are potential candidates to interact
with a modified, sulfur substrate bound cysteine
type of electrostatic interaction, where a third charged residue interacts with a pair
of salt bridge forming residues, occurs frequently at proteinprotein interfaces (Xu
et al. 1997a).
By analogy to the dimer interface, a bootstrap procedure to assess the aver-
age conservation of the tetramer interface revealed that about 0.001% of ran-
domly picked sets of surface residues (leaving out the dimer interface residues)
have at least the same (or higher) level of conservation, indicating that the
tetramer interface patches are highly conserved regions of the protein surface
(Stout et al. 2007).
11.2.5 Location of the Disulfide Bridges and the Potential

Sulfur Binding Site
Although the reactive C-terminal cysteines could not be modeled, the position of
the disulfide bridges can be deduced on the basis of the quaternary structure of SoxY
which brings two symmetry-related C120 residues in close proximity to one
another to allow formation of a disulfide bridge. The last modeled residues of both
-strands g, I117 and I117, protrude from the tetramer interface into the solvent
(Fig. 11.2). It is therefore likely that the intersubunit disulfide bond between the
C120 residues is also exposed to the solvent, residing at the top of each tetramer
interface patch.
Several residues which are located on, or in the vicinity of, the tetramer interface
patch may electrostatically stabilize the resulting adduct after sulfur substrate binding.
The conserved residues K114 and K114, which are also involved in salt-bridge for-
mation as described in Sect. 11.2.4, and the conserved residues R89 and R89 are
good candidates to interact with the negatively charged S-thiocysteinesulfonate

(Fig. 11.2). The nonconserved residues K92 and K92 are also potential candidates
for such an interaction.
11.3 Discussion
To date, most biochemical data have been generated using the SoxYZ heterodimer
(Friedrich et al. 2000; Rother et al. 2001; Quentmeier and Friedrich 2001;
Quentmeier et al. 2003), which is generally believed to be a stable, obligate complex
that is only active in the form of a heterodimer. A combination of structural and bio-
chemical results, however, suggests that the individual proteins, apart from forming
a SoxYZ complex, may also exist on their own. First, a recently determined SoxZ
structure, showing an apparent dimer, indicates that also this protein can be stable
without being associated with another protein (pdb 1v8h, unpublished data). Second,
the SoxY structure discussed here exhibits specific intersubunit interactions, which
argues against the notion that SoxYZ is an obligate complex in which both
components depend on each other for their folding and structural integrity.
A first strong argument supporting the biological relevance of the SoxY
oligomers is the statistically significant conservation levels of the dimer and
tetramer interfaces indicated by bootstrap analyses. These are in agreement with
several studies which state that proteinprotein interfaces are more conserved than
solvent-exposed protein surfaces (Valdar and Thornton 2001a; Caffrey et al. 2004;
Mintseris and Weng 2005) and biologically irrelevant crystal contact points (Valdar
and Thornton 2001b). The high conservation level of the tetramer interface can be
explained by the fact that these residues are in the vicinity of the sulfur substrate
binding cysteine, and therefore can also be involved in transient interactions with
other components of the Sox system. Hence, the evolutionary pressure on these
residues may be higher than that on other surface residues since their combination
is likely to be optimal for interactions with the different Sox enzymes.
In addition, a number of biochemical data and structural and physicochemical
analyses of the interfaces convincingly support the relevance of SoxY oligomers.
Analytical gel filtration demonstrated that reduced SoxY is eluted as a dimer, indi-
cating that the dimer on its own constitutes a stable protomer (Stout et al. 2007).
Furthermore, three major properties of the dimer interface strongly argue for the
SoxY dimer as a specific proteinprotein complex. Firstly, a considerable surface
area (1,522 2) is buried at this dimer interface, which, secondly, has a distinct
central core of hydrophobic residues. Several studies (Miller 1989; Tsai et al. 1997;
Bahadur et al. 2003) have emphasized the presence of such a hydrophobic center in
medium and large proteinprotein interfaces, strengthening the basic paradigm that
hydrophobicity is a major stabilizing factor in proteinprotein association (Chothia
and Janin 1975). Thirdly, the considerable number of hydrogen bonds (12 in total)
argues for a biologically relevant proteinprotein interface. These hydrogen bonds
not only make a considerable energetic contribution to proteinprotein binding, but
also enforce binding specificity due to the electrostatic complementarity of the
134 J. Stout et al.
hydrogen-bond donor and acceptor groups (Xu et al. 1997a, b). All hydrogen bonds
at the dimer interface are part of the hydrogen-bond ladder pattern across the two
extended -sheets. This arrangement of two rows of hydrogen-bond donor and
acceptor groups is specific for -sandwiches and likely restricts potential binding
partners to proteins having a similar arrangement of donor and acceptor groups.
In contrast to the extended dimer interface, the two dimers within the tetramer
interact with each other via two small interface patches at the top and the bottom of
the tetramer. These contact interfaces can be interpreted as biologically irrelevant
crystal contacts on the basis of their limited buried surface area (Bahadur et al.
2004), and their high level of hydration (Rodier et al. 2005). This interpretation,
however, would be in contrast to their high conservation levels. In fact, a small
interface for tetramerization can make it easier for dimertetramer transitions to
occur. Such transitions were observed for P. pantotrophus SoxYZ (Quentmeier
et al. 2003) and for C. thiosulfatiphilum SoxY (Stout et al. 2007) to be dependent
on the oxidation state of the C-terminal cysteine and the molecule adhered to this
residue. It has been stated that a considerable number of proteins can change their
activity by forming weak transient oligomers (Nooren and Thornton 2003). It was
reasoned that these proteins can easily stabilize or weaken their complexes by cre-
ating or breaking a limited number of interactions at small interfaces, thus making
a dynamic response to a change in environment or to a covalent modification
possible. In the case of SoxY, the interface reveals a number of hydrogen bonds and
four conserved salt bridges which were shown by the gel filtration experiments to
be of less importance for the integrity of the tetramer than the disulfide bridges
(Stout et al. 2007). Indeed, treatment of SoxY with different reducing agents such
as DTT, glutathione, sulfide, and sulfite, where the last three sulfur compounds
remain covalently linked to the C-terminal cysteine, resulted in the dissociation of
the SoxY tetramer into dimers. An exception to this behavior, however, was
observed for a SoxYthiosulfate adduct which remains a tetramer after the covalent
addition of this substrate. These observations are suggestive for the possibility that
the chemical nature of the adhered molecule plays a key role in the stabilization of
the tetramer interface. The second feature, the creation of an intersubunit disulfide
bond, can also be explained by the specific electrostatic properties of the tetramer
interface. Such an event can only take place when two dimers approach each other
and interact in a way that eventually leads to the closure of this bond. The formation
of the conserved intersubunit salt bridges at the tetramer interface may well be of
major importance in this event. Salt bridges have been shown to play an important
role in the rate of association in proteinprotein interactions (Vijayakumar et al.
1998; Selzer et al. 2000; Selzer and Schreiber 1999, 2001) and it is likely that the
constitutent residues orientate and enhance two SoxY dimers for tetramerization.
However, from an energetic point of view, the contribution of the salt bridges is not
sufficient to make a permanent tetrameric assembly. This is based on the observa-
tion that reduced SoxY is present as dimers in solution, and the SoxY tetramer will
thus redissociate into dimers if the intersubunit disulfide bonds are not formed. This
dissociation must proceed rapidly because of the limited buried surface area of the
tetramer and its high hydration level.
Fig. 11.3 One tetramer interface region of SoxY viewed from the top. The proposed -turns are
drawn as black curved lines. The yellow SS motif represents the proposed disulfide bridge
Although we could not model the C-terminal region, GGCGG, and the
concomitant disulfide bond, we can propose a potential conformation for these resi-
dues. Double glycine motifs, on a regular basis, occur in -turns as the second and
third residues of this structural element (Chou and Fasman 1979). We therefore
suggest that the I117C120 sequence has a -turn conformation. On the basis of
these assumptions, we propose that the -turns reside at the top of the tetramer
interface regions, each forming a hook-like structure, in such a way that their
fourth residues, C120 residues, are in sufficiently close proximity to form a
disulfide bridge. According to crystallographic rules, this should be symmetrical
and centered on a crystallographic twofold axis (Fig. 11.3). As mentioned above,
these bridges should in principle be accessible for other compounds which act on
this bond.
Although biochemical studies have so far not revealed a role for SoxY and SoxZ
homoprotomers, the SoxY and SoxZ structures are suggestive that an exchange
from SoxYZ to the individual constituents may be part of the reaction mechanism.
A generally accepted issue is that SoxZ, lacking a signal peptide, needs SoxY for
translocation to the periplasm (Friedrich et al. 2000), which implies that SoxYZ
heterodimers need to be formed in the cytoplasm. Both proteins can be transported
to the periplasm by the Tat mechanism, a system capable of translocating folded
proteins (Berks et al. 2003). In the periplasm, these heterodimers get involved in
thiosulfate oxidation, which may occur via two possible scenarios. In one scenario,
SoxYZ exchanges subunits before sulfur substrate binding. The SoxY dimers,
being redox-active, then assemble into SoxY tetramers with closure of the
intersubunit disulfide bridges, an event that is likely mediated via a thiol oxidore-
ductase system. Both SoxY protomers comprise the active sulfur substrate binding
species on which the other sulfur-oxidizing Sox enzymes act. The drawback in this
model is that it disregards SoxZ, which is believed to coordinate the sulfur substrate
molecules (Quentmeier and Friedrich 2001). The second scenario does take SoxZ
into account. SoxZ would mediate the interaction between the heterodimer and the
136 J. Stout et al.
proteins responsible for disulfide bond closure and sulfur substrate addition,
respectively, meanwhile protecting the disulfide bridge and the covalently linked
sulfur substrate. An exchange from SoxYZ heterodimers to SoxY homodimers and
tetramers would then happen after substrate binding, when SoxY offers its bound
sulfur molecules to the other Sox enzymes. In both scenarios, however, one SoxY
protomer offers two covalently bound sulfur molecules instead of one at each
encounter with another component of the sulfur-oxidizing system, which should in
principle be a more efficient way of presenting sulfur substrate molecules to the
Sox system.
Future research needs to address the role of homo-oligomeric SoxY and SoxZ
proteins and, in particular, to investigate the role of the intersubunit disulfide
bridges between SoxY monomers.
References
Appia-Ayme C, Berks BC (2002) SoxV, an orthologue of the CcdA disulfide transporter, is

involved in thiosulfate oxidation in Rhodovulum sulfidophilum and reduces the periplasmic
thioredoxin SoxW. Biochem Biophys Res Commun 296:737741
Bahadur RP, Chakrabarti P, Rodier F, Janin, J (2003) Dissecting subunit interfaces in homodimeric
proteins. Proteins 53:708719
Bahadur RP, Chakrabarti P, Rodier F, Janin J (2004) A dissection of specific and non-specific
protein-protein interfaces. J Mol Biol 336:943955
21:55995610
Bardischewsky F, Friedrich CG (2001) The shxVW locus is essential for oxidation of inorganic
sulfur and molecular hydrogen by Paracoccus pantotrophus GB17: a novel function for litho-
trophy. FEMS Microbiol Lett 202:215220
Bardischewsky F, Quentmeier A, Friedrich, CG (2006a) The flavoprotein SoxF functions in
chemotrophic thiosulfate oxidation of Paracoccus pantotrophus in vivo and in vitro. FEMS
Bardischewsky F, Fischer J, Holler B, Friedrich CG (2006b) SoxV transfers electrons to the peri-
plasm of Paracoccus pantotrophus an essential reaction for chemotrophic sulfur oxidation.
Berks BC, Palmer T, Sargent F (2003) The Tat protein translocation pathway and its role in micro-
bial physiology. Adv Microb Physiol 47:187254
Bork P, Holm L, Sander C (1994) The immunoglobulin fold. Structural classification, sequence
patterns and common core. J Mol Biol 242:309320
Caffrey DR, Somaroo S, Hughes JD, Mintseris J, Huang ES (2004) Are protein-protein interfaces
more conserved in sequence than the rest of the protein surface? Protein Sci 13:190202
Chothia C, Janin J (1975) Principles of protein-protein recognition. Nature 256:705708
Chou PY, Fasman GD (1979) Prediction of beta-turns. Biophys J 26:367383
Friedrich CG, Quentmeier A, Bardischewsky F, Rother D, Kraft R, Kostka S, Prinz H (2000)
Novel genes coding for lithotrophic sulfur oxidation of Paracoccus pantotrophus GB17. J
Harpaz Y, Chothia C (1994) Many of the immunoglobulin superfamily domains in cell adhesion
molecules and surface receptors belong to a new structural set which is close to that containing
variable domains. J Mol Biol 238:528539
53:941951
Jones S, Thornton JM (1996) Principles of protein-protein interactions. Proc Natl Acad Sci USA
93:1320
Lahiri C, Mandal S, Ghosh W, Dam B, Roy P (2006) A novel gene cluster soxSRT is essential for
the chemolithotrophic oxidation of thiosulfate and tetrathionate by Pseudaminobacter sali-
cylatoxidans KCT001. Curr Microbiol 52:267273
Lo Conte L, Chothia C, Janin J (1999) The atomic structure of protein-protein recognition sites.
J Mol Biol 285:21772198
Lu WP (1986) A periplasmic location for the thiosulfate-oxidizing multi-enzyme system from
Thiobacillus versutus. FEMS Microbiol Lett 34:313317
Lu WP, Kelly DP (1983) Purification and some properties of two principal enzymes of the thiosulfate-
oxidizing multi-enzyme system from Thiobacillus A2. J Gen Microbiol 129:35493564
Lu WP, Swoboda BEP, Kelly DP (1985) Properties of the thiosulfate-oxidizing multi-enzyme
system from Thiobacillus versutus. Biochim Biophys Acta 828:116122
Miller S (1989) The structure of interfaces between subunits of dimeric and tetrameric proteins.
Protein Eng 3:7783
Mintseris J, Weng Z (2005) Structure, function, and evolution of transient and obligate protein-
protein interactions. Proc Natl Acad Sci USA 102:1093010935
Nooren IM, Thornton JM (2003) Structural characterisation and functional significance of tran-
sient protein-protein interactions. J Mol Biol 325:9911018
Quentmeier A, Hellwig P, Bardischewsky F, Grelle G, Kraft R, Friedrich CG (2003) Sulfur oxida-
tion in Paracoccus pantotrophus: interaction of the sulfur-binding protein SoxYZ with the
dimanganese SoxB protein. Biochem Biophys Res Commun 312:10111018
Richardson JS, Richardson DC (2002) Natural -sheet proteins use negative design to avoid edge-
to-edge aggregation. Proc Natl Acad Sci USA 99:27542759
Rodier F, Bahadur RP, Chakrabarti P, Janin J (2005) Hydration of protein-protein interfaces.
Proteins 60:3645
Rother D, Henrich HJ, Quentmeier A, Bardischewsky F, Friedrich CG (2001) Novel genes of the
sox gene cluster, mutagenesis of the flavoprotein SoxF, and evidence for a general sulfur-oxidizing
system in Paracoccus pantotrophus GB17. J Bacteriol 183:44994508
Rother D, Orawski G, Bardischewsky F, Friedrich CG (2005) SoxRS-mediated regulation of
chemotrophic sulfur oxidation in Paracoccus pantotrophus. Microbiology 151:17071716
Selzer T, Schreiber G (1999) Predicting the rate enhancement of protein complex formation from
the electrostatic energy of interaction. J Mol Biol 287:409419
Selzer T, Schreiber G (2001) New insights into the mechanism of protein-protein association.
Proteins 45:190198
Selzer T, Albeck S, Schreiber G (2000) Rational design of faster associating and tighter binding
protein complexes. Nat Struct Biol 7:537541
Stout J, De Smet L, Panjikar S, Weiss MS, Savvides SN, Van Beeumen J (2006) Crystallization,
preliminary crystallographic analysis and phasing of the thiosulfate binding protein SoxY from
Chlorobium limicola f. thiosulfatophilum. Acta Crystallogr Sect F 62: 10931096
Stout J, Van Driessche G, Savvides SN, Van Beeumen J (2007) X-ray crystallographic analysis of
the sulfur carrier protein SoxY from Chlorobium limicola f. thiosulfatophilum reveals a tetra-
meric structure. Protein Sci 16:589601
Tsai CJ, Lin SL, Wolfson HJ, Nussinov R (1997) Studies of protein-protein interfaces: a statistical
analysis of the hydrophobic effect. Protein Sci 6:5364
138 J. Stout et al.
Valdar WSJ, Thornton JM (2001a) Protein-protein interfaces: analysis of amino acid conservation
in homodimers. Proteins 42:108124
Valdar WSJ, Thornton JM (2001b) Conservation helps to identify biologically relevant crystal
contacts. J Mol Biol 313:399416
Vijayakumar M, Wong KY, Schreiber G, Fersht AR, Szabo A, Zhou HX (1998) Electrostatic
enhancement of diffusion-controlled protein-protein association: comparison of theory and
experiment on barnase and barstar. J Mol Biol 278:10151024
Williams AF, Barclay AN (1988) The immunoglobulin superfamily domains for cell surface
recognition. Annu Rev Immunol 6:381405
Xu D, Tsai CJ, Nussinov R (1997a) Hydrogen bonds and salt bridges across protein-protein
interfaces. Protein Eng 10:9991012
Xu D, Lin SL, Nussinov R (1997b) Protein binding versus protein folding: the role of hydrophilic
bridges in protein associations. J Mol Biol 265:6884
Young L, Jernigan RL, Covell DG (1994) A role for surface hydrophobicity in protein-protein
recognition. Protein Sci 3:717729
Chapter 12
Redox Control of Chemotrophic Sulfur
Oxidation of Paracoccus pantotrophus
Cornelius G. Friedrich, Armin Quentmeier, Frank Bardischewsky,

Dagmar Rother, Grazyna Orawski, Petra Hellwig, Jrg Fischer
Abstract The reaction cycle of the reconstituted Sox enzyme system of

Paracoccus pantotrophus requires the periplasmic proteins SoxYZ, SoxXA,
SoxB, and SoxCD. The heme enzyme SoxXA covalently binds the sulfur substrate
to the thiol of the single cysteine residue of SoxY located at its carboxy-terminal
end. Bound sulfur is then oxidized to sulfate by a series of reactions. These
involve sulfur dehydrogenase SoxCD which oxidizes the protein-bound sulfane
sulfur to sulfone in a unique six-electron transfer. Bound sulfone is then hydro-
lyzed off by the sulfate thiohydrolase SoxB to regenerate SoxYZ. The flavoprotein
SoxF enhances the rate of sulfur oxidation in vivo as evident from mutant analysis
and we have specified its action in vitro. SoxYZ unlike the other Sox proteins is
inactivated upon reduction. When the Sox system is reconstituted with inactivated
SoxYZ, the thiosulfate-oxidizing activity is drastically decreased. SoxF reverses
this inactivation and may mediate a conformational change of SoxYZ possibly by
a transient interprotein disulfide. The membrane protein SoxV and the thioredoxin
SoxS are essential for chemotrophic growth as evident from homogenote mutants
defective in these proteins. Evidence is presented that both proteins transfer
reductant from the cytoplasm to the periplasm and that SoxYZ is the final target
of this transfer to balance the redox state of the Sox enzyme system or reduce a
SoxY..Y interprotein disulfide.
12.1 The Sulfur-Oxidizing Enzyme System

of Paracoccus pantotrophus
Paracoccus pantotrophus is a facultatively aerobic alphaproteobacterium which grows

heterotrophically with various carbon sources and chemoautotrophically with
thiosulfate under aerobic conditions (Robertson and Kuenen 1983; Ludwig et al.
1993, Rainey et al. 1999). The sulfur-oxidizing (Sox) enzyme system of P. pantotrophus
is encoded by the sox gene cluster, which comprises 15 genes organized in three
transcriptional units soxRS, soxVW, and soxXYZA-H. The soxR gene encodes a
DNA-binding repressor protein of the ArsR family (Rother et al. 2005). Seven sox
139
Springer 2008
140 C.G. Friedrich et al.
genes encode polypeptides which form four periplasmic proteins, designated accord-
ing to the gene nomenclature. These proteins reconstitute the Sox enzyme system in
vitro and oxidize hydrogen sulfide, sulfur, thiosulfate, and sulfite with horse cyto-
chrome c as the final electron acceptor (Eq. 12.1) (reviewed in Friedrich et al. 2005):

S SO3 + 5H2 O + 8Cytc3+ 2SO 4 2 + 8Cytc 2 + + 10H + . (12.1)
The SoxYZ complex is the central protein and interacts with SoxXA, SoxCD, and
SoxB. Sulfur is oxidized when covalently bound to the thiol of the invariant
Cys138 of SoxY (Quentmeier and Friedrich 2001; Fig. 12.1). The heme enzyme
SoxXA, a complex of the monoheme c-type cytochrome SoxX and the diheme
c-type cytochrome SoxA, is proposed to link the sulfur substrate to the thiol of
Cys138 of SoxY (Bamford et al. 2002; Dambe et al. 2005). The molybdoprotein
cytochrome complex SoxCD catalyzes a unique six-electron transfer and oxi-
dizes the outer (sulfane) sulfur atom of cysteinepersulfide of SoxY to the sulfone
oxidation state to yield l-cysteine-S-sulfate and acts as sulfur dehydrogenase.
The dimanganese SoxB protein is a paralog of the zinc-containing 5 nucleoti-
dases and is proposed to hydrolyze off sulfate from the l-cysteine-S-sulfate to
Fig. 12.1 Model of the reaction cycle of thiosulfate oxidation by the Sox enzyme system of
Paracoccus pantotrophus and reactivation of SoxYZ by the flavoprotein SoxF. The capital letters
indicate the respective Sox proteins, the central protein SoxYZ in its active form is indicated in
boldface, and the inactive form is indicated in fine type. The SoxYY interprotein disulfide of the
heterotetrameric SoxYY(Z)2 represents a hypothetical intermediate in the transition of the inac-
tive to active form of SoxYZ catalyzed by SoxF. TCEP tris(2-carboxyethyl)phosphine
12 Redox Control of Chemotrophic Sulfur Oxidation of Paracoccus pantotrophus 141
regenerate SoxYZ acting as sulfate thioesterase (Friedrich et al. 2001; Quentmeier

and Friedrich 2001; Rother et al. 2001; Bardischewsky et al. 2005). Therefore,
the sulfur-binding protein SoxYZ is the central protein of the Sox enzyme system
and reacts with SoxXA, SoxB, and SoxCD (Fig. 12.1). Moreover, SoxYZ is
redox-active although it does not contain a cofactor or metal (Quentmeier et al.
2003), and reduction in vitro by the non-sulfur reductant tris(2-carboxyethyl)
phosphine (TCEP) inactivates SoxYZ (Quentmeier and Friedrich 2001). Genetic
inactivation of SoxB or SoxC disables P. pantotrophus from growing with and
oxidizing thiosulfate and hydrogen sulfide to sulfate (Wodara et al. 1997; Rother
et al. 2001).
The flavoprotein SoxF is not part of the reconstituted Sox enzyme system
and its inactivation does not disable P. pantotrophus from growing chemo-
trophically with thiosulfate (Rother et al. 2001) but decreases the rate of thio-
sulfate oxidation (Bardischewsky et al. 2006b). SoxF is a periplasmic monomer
of 42,797 Da, contains covalently bound flavin adenine dinucleotide (FAD),
and is closely related to the flavoprotein subunits of flavocytochromes c of
phototrophic and chemotrophic sulfur-oxidizing bacteria. SoxF like flavocyto-
chrome c (FCSD) has sulfide dehydrogenase (SDH) activity in vitro with horse
cytochrome c as an electron acceptor to yield two electrons per mole of
hydrogen sulfide (Quentmeier et al. 2004). The SDH activity of FCSDs of vari-
ous sulfur-oxidizing bacteria was taken as evidence for its significance in sulfur
energy metabolism (Kusai and Yamanaka 1973; Kostanjevecki et al. 2000;
reviewed in Cusanovich et al. 1991) while genetic evidence did not support this
view (Reinartz et al. 1998).
The soxVW genes comprise a transcriptional unit, and soxV encodes the
membrane protein SoxV with six channel-forming transmembrane helices and
two cysteine residues facing each other on the inner side and able to form a
protein disulfide for transport of reductant. In this respect SoxV is homologous
to CcdA of P. pantotrophus and other bacteria. CcdA is essential for cyto-
chrome c biogenesis (Bardischewsky and Friedrich 2001). The soxW gene
encodes a periplasmic thioredoxin of 166 amino acids after maturation. The
Sox enzyme system of P. pantotrophus requires the function of SoxV. Although
c-type cytochromes are essential for chemotrophic sulfur oxidation and SoxV
transfers electrons to the periplasm it does not reduce apocytochromes and is
not involved in cytochrome c maturation. Instead, a thioredoxin is the proposed
electron acceptor as the nonessential thioredoxin SoxW is reduced by SoxV
(Bardischewsky et al. 2006a). The crucial periplasmic electron acceptor of
SoxV has not been reported so far.
Here we summarize the current knowledge of the Sox system and present
unpublished results demonstrating that this enzyme system is subject to a subtile
redox control. We present preliminary results on the central protein, SoxYZ, and its
inactivation upon reduction and show that the flavoprotein SoxF reactivates
SoxYZ. Evidence is presented that the redox partner of SoxV is the thioredoxin
SoxS, which in turn may interact with SoxYZ.
12.2 Abundance of the sox Genes in Bacteria
The increasing availability of partial and complete microbial genome sequences has
enabled the detection of sox genes among the Bacteria, while in Archaea these
genes are not present. Besides the P. pantotrophus sox gene cluster, mostly incom-
plete sox clusters were reported from seven bacterial strains in 2001 (Friedrich et al.
2001) and from 17 strains in 2005 (Friedrich et al. 2005). To date (scan closed
August 31, 2006) a total of 38 sox gene clusters are known, most of which are
derived from genomic sequences of chemotrophic and phototrophic bacteria of
different genera (Fig. 12.2).
Among the bacteria listed in Fig. 12.2 several strains are unable to grow
chemotrophically with thiosulfate like, e.g., Ralstonia eutropha. Transfer of the
P. pantotrophus sox structural genes to R. eutropha does not add this physiological
trait (F. Bardischewsky, unpublished data). For other strains like, e.g., Bradyrhizobium
japonicum, Nitrobacter hamburgensis, Polaromonas sp., or Methylobium petrophilum,
chemoautotrophic growth with inorganic sulfur compounds is unknown. On the other
hand enzymes involved in sulfur oxidation have been described from chemotrophic
and phototrophic bacteria, the respective genes of which were not identified (reviewed
in Brune 1989; Kelly et al. 1997; Friedrich 1998).
R. eutropha, Polaromonas sp., and Dechloromonas aromatica (Fig. 12.2), like
the other strains listed in Fig. 12.2, harbor complete sets of sox genes soxYZ,
soxXA, soxB, and soxCD, the products of which are required for thiosulfate oxida-
tion in vitro. These strains, however, lack soxVW and soxRS required for sulfur
oxidation to sulfate in vivo by P. pantotrophus. Among the strains lacking these
genes are the green phototrophic bacteria Chlorobaculum tepidum (formely
Chlorobium tepidum; Imhoff 2003), Chlorobium limicola, and Chlorobium clath-
ratiforme (formerly Pelodictyon phaeoclathratiforme; Imhoff 2003) which grow
well photoautotrophically with thiosulfate. The Chlorobiaceae miss the soxCD
genes, which are also missing in the known sequences of the phototrophic purple
sulfur bacterium Allochromatium vinosum (formerly Chromatium vinosum; Imhoff
et al. 1998). SoxCD is essential for growth of the chemotroph P. pantotrophus
(Wodara et al. 1997). Instead of the soxCD genes, green sulfur bacteria and A. vinosum
harbor the dsr operon which is indispensable for oxidation of stored sulfur (Dahl
Fig. 12.2 Map of the sox gene cluster of P. pantotrophus and sox gene homologs of other bacteria.
Capital letters designate the sox genes of P. pantotrophus. Open reading frames (ORFs) predicting
homologous proteins are indicated by the same color. Pink/violet arrows without frame indicate genes
encoding sulfite dehydrogenases and their cytochromes (see chapter 13). Bright yellow arrows as for
Rod. cap. indicate sulfide-quinone oxidoreductase genes. ORFs not encoding Sox homologous pro-
teins are given in white. Par.pan., P. pantotrophus GB17; Par.den., P. denitrificans 1222; Rhd.sph.,
Rhodobacter sphaeroides; Rhv.sul., Rhodovulum sulfidophilum; Rhp.pal., Rhodopseudomonas palus-
tris; Sul.NAS, Sulfitobacter sp. NAS-14.1; Sul.EE, Sulfitobacter sp. EE-36, Rho.bac., rhodobacterales
bacterium; Ros.nub., Roseovarius nubinhibens; Ros.217, Roseovarius sp. 217; pom., Sil. Silicibacter
pomeroyi; Bra.sp, Bradyrhizobium sp.; Bra.jap., Bradyrhizobium japonicum; Sta.nov., Starkeya
novella; Psb.sal., Pseudaminobacter salicylatoxidans KCT001; Met.ext., Methylobacterium
Fig. 12.2 (continued) extorquens; Tms.den., Thiomicrospira denitrificans; Tms.cru., Thiomicrospira

crunogena; Mel.pet., Methylobium petroleophilum; Ane.deh., Anaeromyxobacter dehalogenans; Pol.
sp, Polaromonas sp.; Dec.aro., Dechloromonas aromatica; Ral.eut., Ralstonia eutropha, Ral.sol.,
Ralstonia solanacearum; Ral.met., Ralstonia metallidurans; Thm.the., Thermus thermophilus; Chl.
tep., Chlorobaculum tepidum; Chl.lim., Chlorobium limicola, Chl.chl., Chlorobium chlorochromatii
CaD3; Pel.pha., Pelodictyon phaeoclathratiforme; Nit.ham., Nitrobacter hamburgensis; Tms.den.,
Thiomicrospira denitrificans; All.vin., Allochromatium vinosum, Aqu.aeo., Aquifex aeolicus; Mgc.
MC1, Magnetococcus MC-1; Mgs.mag., Magnetospirillum magnetotacticum; Alk.ehr., Alkalilimnicola
ehrlichii; Mec.cap., Methylococcus capsulatus; Rod.cap., Rhodobacter capsulatus
et al. 2005). The function of dsrAB encoding dissimilatory sulfite reductase and
the interaction of other Dsr proteins is specified in this volume and suggests their
function in oxidation of inorganic sulfur to sulfate in an alternative route in which
elemental sulfur is an obligate intermediate (see Chap. 9 by Grimm et al.).
Therefore, other enzymes are proposed in other bacteria which oxidize sulfur
compounds to sulfate. These enzymes should be functional equivalents to sulfur
dehydrogenase SoxCD, the membrane protein SoxV, and the thioredoxins SoxW
and SoxS.
In the genomes of most bacteria harboring the sox structural genes two other
genes are located either within the sox gene cluster or separate from it. One gene is
homologous to soxF which encodes the flavoprotein SoxF in P. pantotrophus. The
other gene (soxE) encodes a small c-type cytochrome (Fig. 12.2) which, however,
is not closely related to those of other sources. In A. vinosum these genes, desig-
nated fccAB, encode the flavocytochrome c complex which exhibits SDH activity
in vitro. The function of FCSD, which is located in the periplasm in most strains
but which is membrane-bound in others (Visser et al. 1997), has been a matter of
debate. In vitro, sulfur is the product of FCSD. The periplasmic location of the
enzyme in A. vinosum is in agreement with the periplasmic storage of sulfur in
purple sulfur bacteria (Pattaragulwanit et al. 1998).
12.3 The Physiological Function of the Flavoprotein SoxF
This section is devoted to the physiological function of SoxF in P. pantotrophus. The

in vivo function of SoxF differs from the SDH activity determined in vitro
(Quentmeier et al. 2004). Sulfide is not a free intermediate of thiosulfate oxidation
as evident from the mechanism of sulfur oxidation by the Sox enzyme system
(Friedrich et al. 2001, 2005; Rother et al. 2001). Also, homologs of the soxF gene
are missing in some sulfur-oxidizing bacteria like, e.g., in the genome of
Thiomicrospira denitrificans (Fig. 12.2; see also Chap. 19 by Sievert et al.).
Therefore, SDH activity is probably not functionally important in vivo. Also, SDH
yields only two electrons, while from sulfide the Sox enzyme systems yields four
electrons in vitro and eight electrons in vivo. Strain GBsoxF carries a deletion in
the soxF gene which eliminates 90 amino acids around the FAD-binding site and
which causes complete degradation and inactivation of the SoxF protein. The spe-
cific growth rate of strain GBsoxF with thiosulfate and carbon dioxide is about half
of that of the wild type. Whole cells of strain GBsoxF, however, oxidize thiosulfate
and hydrogen sulfide to sulfate as evident from the requirement of about 2 mol of
oxygen per mole of thiosulfate or sodium sulfide (Bardischewsky et al. 2006b). This
result again suggests a role different from SDH for SoxF in sulfur metabolism.
In the 4465% ammonium sulfate precipitate of cell-free extracts of strain
GBsoxF the specific thiosulfate-oxidation rate is about 50% of that of the same
fraction of the wild type. Addition of purified SoxF to enzyme assays containing the
ammonium sulfate fraction from strain GBsoxF increases the activity to the
wild-type level. However, SoxF is unable to metabolize thiosulfate and sulfide is not
an intermediate of thiosulfate oxidation by the Sox enzyme system; therefore, SoxF
is proposed to enhance the activity of some component of the Sox enzyme system.
However, SoxF added to the Sox enzyme system reconstituted from homogeneous
Sox proteins does not affect the thiosulfate-oxidizing activity (Rother et al. 2001;
Bardischewsky et al. 2006b). Consequently, the action of SoxF must relate to some
condition given in cell-free extracts but not in the reconstituted Sox enzyme system.
Proteins of cell-free extracts are considered to be reduced. The key protein of the
Sox enzyme system, SoxYZ, is sensitive to reduction by the non-sulfur reductant
TCEP. Reduction of SoxYZ does not only inhibit SoxYZ as is plausible from shift-
ing the equilibrium of the reaction (Eq. 12.2) to the left side, but it also inactivates
SoxYZ (Quentmeier and Friedrich 2001):
SoxZY S + S SO3 + 2Cytc3+ SoxZY S S SO3 + 2Cytc 2 + . (12.2)
Preincubation of SoxYZ with 1 mM TCEP prior to reconstitution of the Sox

enzyme system selectively inactivates SoxYZ as evident from the 5090%
decrease in thiosulfate-oxidizing activity of the system. Reduction of the other Sox
proteins prior to reconstitution of the Sox enzyme system does not affect its
thiosulfate-oxidizing activity (A. Quentmeier, unpublished data). Changes in the
redox state of redox-active proteins are often linked to conformational changes
upon change in the charge of metals, heme iron ligation, or protein disulfide for-
mation as is known from cytochromes, thioredoxins, flavodoxin, photosystem I
and II, or regulatory proteins (Takano and Dickerson 1980; Swenson et al. 1999;
Arnesano et al. 2003; Lee et al. 2004; Range et al. 2006; reviewed in Ritz and
Beckwith 2001). Infrared spectroscopy of SoxYZ suggests a conformational
change upon various treatments of the protein (P. Hellwig, unpublished data).
Addition of SoxF of P. pantotrophus to the Sox enzyme system reconstituted
with TCEP-inactivated SoxYZ reactivated the thiosulfate-oxidizing activity of the
whole system with time (A. Quentmeier, unpublished data). When SoxF was
included in the assay cytochrome c was reduced with progressively increasing rate.
This progressive increase suggested an action of SoxF which was linked to the reac-
tion cycle of the reconstituted Sox enzyme system. As the inactivated SoxYZ was
the bottleneck of the overall reaction, very likely this protein is the target of the
reactivating activity of SoxF. This SoxYZ-reactivating reaction represents a novel
reaction for a flavoprotein which we consider physiologically significant (C.G.
Friedrich, A. Quentmeier, D. Rother, F. Bardischewsky, and P. Hellwig, unpub-
lished results). How does the SDH activity of SoxF relate to this function?
SoxF transfers electrons to horse cytochrome c and oxidizes thiols as is evident
from its SDH activity. SoxYZ does not contain a redox center or metal and the
single cysteine of SoxY is, therefore, proposed to donate one thiol for formation of
an interprotein disulfide to form a SoxY..Y homodimer (.. stands for protein
disulfide). Dimers of the subunits of SoxYZ covalently linked by protein disulfides
have been identified by sodium dodecyl sulfate gel electrophoresis (Quentmeier
et al. 2003). Interprotein disulfides like SoxY..Z and SoxZ..Z are not considered to
be crucial for the Sox enzyme system since the single cysteine of P. pantotrophus
SoxZ is not present in SoxZ of several other sources (J. Fischer, unpublished data).
The current model of sulfur oxidation by the Sox enzyme system proposes the free
thiol of SoxYCys138 as a hook for the covalent linkage of the sulfur substrate.
Therefore, the SoxY..Y homodimer is considered to be a transient intermediate
for a conformational rearrangement of SoxY to yield its active form (Fig. 12.1).
This assumption, however, implies the re-reduction of the interprotein disulfide
of SoxY..Y to two SoxY and poses the question from where the electrons for
re-reduction of SoxY..Y originate.
12.4 The Periplasmic Partners of SoxV

for Transfer of Electrons
The reversible formation of protein disulfide bonds is an important means for trans-
port of electrons, protein biogenesis, protein stability, and enzyme catalysis
(reviewed in Fabianek et al. 2000; Ritz and Beckwith 2001). The membrane protein
SoxV is a paralog of CcdA, which is essential for re-reduction of apocytochromes,
and the functional difference is evident from the phylogenetic tree of the related
proteins (Bardischewsky et al. 2006a). The function of SoxV in transfer of reduct-
ant to the periplasm is essential as is evident from disruption of the soxV gene by
an Kmr cassette. Since soxVW comprise a transcriptional unit, the inserted
-cassette in soxV acts in a polar manner on soxW expression, leading to a
SoxW-negative phenotype. Strain GBV is unable to grow chemotrophically with
thiosulfate and the in vivo thiosulfate-dependent oxygen uptake rate is only less
than 10% of that of the wild type. Genetic complementation of the soxV gene
restored the ability for chemotrophic growth, while complementation of the soxW
gene did not (Bardischewsky et al. 2006b). The thiosulfate-oxidizing ability of
strain GBV could also be restored by chemical complementation with dithiothrei-
tol (DTT; Fig. 12.3). Such chemical complementation has similarly been described
for CcdA in Escherichia coli (Sambongi and Ferguson 1994) and is in accordance
with the direction of transfer of electrons by SoxV to the periplasm. This poses the
question of the periplasmic redox partner of SoxV.
The electron donor to reduce SoxV is presumably a cytoplasmic thioredoxin as
the thioredoxin is reduced by SoxV. SoxW, however, is not essential for thiosulfate
oxidation in vivo (Bardischewsky et al. 2006a). Recently, the periplasmic thiore-
doxin SoxS was identified to be essential for thiosulfate oxidation as is evident
from a homogenote mutant strain GBS in which the soxS gene is disrupted by the
-Kmr interposon. This mutant forms about 10% of the specific thiosulfate-oxidizing
activity as compared with the wild type. In this mutant, trans complementation of
soxS restores the wild-type phenotype (G. Orawski, unpublished data). Also, similarly
as in the mutant GBV (Fig. 12.3) the mutation in strain GBS can be complemented
chemically by inclusion of 1 mM DTT to the mineral medium to yield almost the
wild-type level of thiosulfate-oxidizing activity (F. Bardischewsky, unpublished data).
Fig. 12.3 Chemical complementation of thiosulfate-oxidizing ability of the homogenote mutant

P. pantotrophus GBV. Strain GBV was cultivated mixotrophically with thiosulfate. In the
stationary growth phase the culture was split and further aerated at 30C by shaking. One culture
was untreated (open diamonds); the other was treated with 1 mM dithiothreitol final concentration
(closed diamonds). The thiosulfate-dependent oxygen uptake rate of whole cells was determined
at the time intervals indicated. (From Bardischewsky et al. 2006a, with permission)
Fig. 12.4 Model of the route of reductants from the

cytoplasm for redox conditioning of SoxYZ. The letters
indicate the Sox proteins
The chemical complementation of strain GBS transposes the previous question

one step further to: Which is the redox partner of the thioredoxin SoxS?
The reduced and the oxidized state of SoxY can be deduced from the availability
of the free sulfhydryl of Cys138 and its reactivity with the sulfhydryl reactant diso-
dium 4-acetamido-4-maleimidstilbene-2,2-disulfonic acid (AMS; cited in
Bardischewsky et al. 2006a). In freshly prepared cell-free extracts of the wild type,
SoxY appears predominantly in the reduced state since AMS binds to SoxY and
increases the molecular mass by about 500 Da. However, SoxY of strain GBS
cannot be trapped by AMS, suggesting the sulfhydryl of Cys138 SoxY is either
oxidized or present in a conformation in which the sulfhydryl is inaccessible for
chemical modification by AMS (F. Bardischewsky, unpublished data).
With this set of experiments we have obtained evidence for the route of electrons
from the cytoplasm via SoxV to SoxS. SoxYZ possibly requires a conformation
which allows access to the sulfhydryl of Cys138 to which the sulfur substrates and
SoxXA bind. SoxS donates electrons and is essential for chemotrophic growth.
Therefore, we suggest SoxYZ as a redox partner of the thioredoxin SoxS to enable
its transition to the active form (Fig. 12.4). It is not yet clear whether an enzymatic
step is required to catalyze the conformational change which is likely to be linked
to the activation of SoxYZ. Our current work aims at answering this question.
References
Arnesano F, Banci L, Bertini I, Mangani S, Thompsett AR (2003) A redox switch in CopC: an

intriguing copper trafficking protein that binds copper(I) and copper(II) at different sites. Proc
Natl Acad Sci USA 100:38143819
21:55995610
Bardischewsky F, Friedrich CG (2001) Identification of ccdA in Paracoccus pantotrophus GB17:
disruption of ccdA causes complete deficiency in c-type cytochromes. J Bacteriol
183:257263
Bardischewsky F, Quentmeier A, Rother D, Hellwig P, Kostka S, Friedrich CG (2005) Sulfur
dehydrogenase of Paracoccus pantotrophus: the heme-2 domain of the molybdoprotein cyto-
chrome c complex is dispensable for catalytic activity. Biochemistry 44:70247034
Bardischewsky F, Fischer J, Hller B, Friedrich CG (2006a) SoxV transfers electrons to the peri-
plasm of Paracoccus pantotrophus an essential reaction for chemotrophic sulfur oxidation.
Bardischewsky F, Quentmeier A, Friedrich CG (2006b) The flavoprotein SoxF functions in
chemotrophic thiosulfate oxidation of Paracoccus pantotrophus in vivo and in vitro. FEMS
Brune D (1989) Sulfur oxidation by phototrophic bacteria. Biochim Biophys Acta
975:189221
Cusanovich MA, Meyer TE, Bartsch RG (1991) Flavocytochrome c. In: Mller F (ed) Chemistry
and biochemistry of flavoenzymes. CRC, Boca Raton, pp 377399
Dahl C, Engels S, Pott-Sperling AS, Schulte A, Sander A, Lbbe Y, Deuster O, Brune DC (2005)
187:13921404
Dambe T, Quentmeier A, Rother D, Friedrich CG, Scheidig AJ (2005) Structure of the cyto-
chrome complex SoxXA of Paracoccus pantotrophus, a heme enzyme initiating chemotrophic
sulfur oxidation. J Struct Biol 152:229234
Fabianek RA, Hennecke H, Thny-Meyer L (2000) Periplasmic protein thiol:disulfide oxidore-
ductases of Escherichia coli. FEMS Microbiol Rev 24:303316
Friedrich CG (1998) Physiology and genetics of sulfur-oxidizing bacteria. Adv Microbial Physiol
39:235289

53:941951
Imhoff JF, Sling J, Petri R (1998) Phylogenetic relationships among the Chromatiaceae, their
taxonomic reclassification and description of the new genera Allochromatium, Halochromatium,
Isochromatium, Marichromatium, Thiococcus, Thiohalocapsa and Thermochromatium. Int J
Syst Bacteriol 48:11291143
Kelly DP, Shergill LK, Lu W-P, Wood AP (1997) Oxidative metabolism of inorganic sulfur com-
Kostanjevecki V, Brige A, Meyer TE, Cusanovich MA, Guisez Y, van Beeumen JJ (2000)
A membrane-bound flavocytochrome c-sulfide dehydrogenase from the phototrophic purple
sulfur bacterium Ectothiorhodospira vacuolata. J Bacteriol 182:3097103
Kusai A, Yamanaka T (1973) Cytochrome c (553, Chlorobium thiosulfatophilum) is a sulfide-
cytochrome c reductase. FEBS Lett 34:235237
Lee C, Lee SM, Mukhopadhyay P, Kim SJ, Lee SC, Ahn W-S, Yu M-H, Storz G, Ryu SE (2004)
Redox regulation of OxyR requires specific disulfide bond formation involving a rapid kinetic
reaction path. Nat Struct Mol Biol 11:11791185
Ludwig W, Mittenhuber G, Friedrich CG (1993) Cleavage of structural proteins during the assem-
bly of the head of bacteriophage T4. Int J Syst Bacteriol 43:363367
Pattaragulwanit K, Brune DC, Trper HG, Dahl C (1998) Molecular genetic evidence for cytoplas-
mic localization of sulfur globules in Chromatium vinosum. Arch Microbiol 169:434444
Quentmeier A, Friedrich CG (2001) The cysteine residue of the SoxY protein as the active site
of protein-bound sulfur oxidation of Paracoccus pantotrophus GB17. FEBS Lett
503:168172
Quentmeier A, Hellwig P, Bardischewsky F, Grelle G, Kraft R, Friedrich CG (2003) Sulfur oxida-
tion in Paracoccus pantotrophus: interaction of the sulfur-binding protein SoxYZ with the
dimanganese SoxB protein. Biochem Biophys Res Comm 312:10111018
Quentmeier A, Hellwig P, Bardischewsky F, Wichmann R, Friedrich CG (2004) Sulfide dehydro-
genase activity of the monomeric flavoprotein SoxF of Paracoccus pantotrophus. Biochemistry
43:1469614703
Rainey FA, Kelly DP, Stackebrandt E, Burghardt J, Hiraishi A, Katayama Y, Wood AP (1999) A
re-evaluation of the taxonomy of Paracoccus denitrificans and a proposal for the combination
Paracoccus pantotrophus comb. nov. Int J Syst Bacteriol 49:645651
Range K, Ayala I, York D, Barry BA (2006) Normal modes of redox-active tyrosine: conformation
dependence and comparison to experiment. J Phys Chem 110:1097010981
Reinartz M, Tschpe J, Brser T, Trper HG, Dahl C (1998) Sulfide oxidation in the phototrophic
sulfur bacterium Chromatium vinosum. Arch Microbiol 170:5968
Ritz D, Beckwith J (2001) Roles of thiol-redox pathways in bacteria. Annu Rev Microbiol
55:2148
Robertson LA, Kuenen JG (1983) Thiosphaera pantotropha gen. nov. sp. nov.: a facultatively
anaerobic, facultatively autotrophic sulfur bacterium. J Gen Microbiol 129:28472855
Rother D, Henrich H-J, Quentmeier A, Bardischewsky F, Friedrich CG (2001) Novel genes of the
sox gene cluster, mutagenesis of the flavoprotein SoxF, and evidence for a general sulfur-oxi-
dizing system in Paracoccus pantotrophus GB17. J Bacteriol 183:44994508
Rother D, Orawski G, Bardischewsky F, Friedrich CG (2005) SoxRS mediated regulation of
chemotrophic sulfur oxidation in Paracoccus pantotrophus. Microbiology 151:17071716
Sambongi Y, Ferguson SJ (1994) Specific thiol compounds complement deficiency in c-type
cytochrome biogenesis in Escherichia coli carrying a mutation in a membrane-bound disul-
phide isomerise-like protein. FEBS Lett 353:235238
Swenson RP, Kasim M, Bradley LH, Druhan LJ (1999) Role of conformational dynamics and
associated electrostatic and hydrogen bonding interactions in the regulation of redox potentials
in the Clostridium beijerinckii flavodoxin. In: Ghisla S, Kroneck P, Macheroux P, Sund H (eds)
Flavins and flavoproteins. Agency for Scientific Publications, Berlin, pp183186
Takano T, Dickerson RE (1980) Redox conformation changes in refined Tuna cytochrome. Proc
Natl Acad Sci 77:63716375
Visser JM, de Jong GAH, Robertson LA, Kuenen JG (1997) A novel membrane-bound flavocyto-
chrome c sulfide dehydrogenase from the colorless sulfur bacterium Thiobacillus sp. W5. Arch
Wodara C, Bardischewsky F, Friedrich CG (1997) Cloning and characterization of sulfite dehy-
drogenase, two c-type cytochromes, and a flavoprotein of Paracoccus denitrificans GB17:
essential role of sulfite dehydrogenase in lithotrophic sulfur oxidation. J Bacteriol
179:50145023
Chapter 13
Bacterial Sulfite-Oxidizing Enzymes Enzymes
for Chemolithotrophs Only?
Ulrike Kappler
Abstract All known sulfite-oxidizing enzymes that have been studied in molecular
detail belong to the sulfite oxidase family of molybdoenzymes. The first bacterial
enzymes in this family were only characterized in 2000, but by now it has become
clear that bacterial enzymes originating from many different types of bacteria
may actually be the most abundant proteins in this enzyme family. This chapter
provides an overview of sulfite oxidase like bacterial enzymes as well as an anal-
ysis of their phylogeny.
13.1 Introduction Sulfite in the Environment

and in Cell Metabolism
Sulfites form naturally during the decomposition of reduced sulfur compounds such
as thiosulfate, polythionates and sulfonates (Roy and Trudinger 1970), and in the
absence of oxygen, sulfites can persist in the environment (Hayes et al. 2006; Sorokin
1995). Another natural process, namely, sulfur dioxide becoming dissolved in water,
can also lead to the formation of hydrogen sulfite (HSO3) and sulfite (SO32).
The sulfite anion is a relatively strong nucleophile and can therefore be used as a
reducing agent (SO42/HSO32 E = 516 mV; Thauer et al. 1977).
As a result of their reactivity, sulfites and so-called sulfiting agents (sulfur dioxide,
bisulfites, metabisulfite) are a major class of industrial chemicals. They are used in
applications such as leather tanning, paper milling, photography and, probably most
importantly, the food industry. Use of sulfites in food as a conditioning or preserva-
tion agent is very common, and residual sulfites in food can cause severe allergic
reactions in some humans following ingestion (Lester 1995; McEvily et al. 1992).
In all living cells, sulfur-containing compounds play a major role in the form of
coenzymes, amino acids and redox-active molecules. In order to be able to synthesize
these compounds, most cells reduce sulfate to the level of sulfur/sulfide and then
incorporate it into the biomass. In this energy-consuming process, sulfate undergoes
a two-step activation to 3-phosphoadenosine 5-phosphosulfate, from which sulfite
is then released and reduced to the desired state via the reaction of a sulfite reductase
(Carroll et al. 2005; Kappler and Dahl 2001).
151
Springer 2008
152 U. Kappler
However, sulfites are also commonly formed as a result of metabolic processes

that break down sulfonates, sulfur-containing amino acids or reduced inorganic
sulfur compounds. In vertebrates, sulfite generally arises during amino acid break-
down (Griffith 1987). It is then oxidized by a sulfite oxidase (SO) to sulfate, which
can be easily excreted (Rajagopalan 1980). The formation of free sulfite within a
cell can lead to irreversible damage via disruption of protein disulfide bonds and
damage to bases present in the DNA, and therefore an efficient mechanism for
removing sulfite from cells by detoxification, export and/or binding of the compound
in another form has to be present. Accumulating sulfite can also cause systemic
effects such as damage to the central nervous system and increased oxidative stress
(Chamulitrat 1999; Zhang et al. 2004). However, there may be other, regulatory
roles for sulfite in the human body as well: elevated sulfite levels are associated
with inflammatory conditions and host defence (Mitsuhashi et al. 1998, et al. 2005;
Ratthe et al. 2002).
Plants generally encounter sulfite after exposure to atmospheric sulfur dioxide
or as a result of amino acid breakdown. However, the physiological role of the
peroxisomal plant SO, which uses oxygen as its preferred electron acceptor, is still
being investigated (Mendel and Bittner 2006).
In microbial cells sulfite can arise from a variety of reactions, such as dissimilatory
sulfur compound oxidation, amino acid degradation, sulfur assimilation, or follow-
ing sulfonate breakdown. Microbial cells can also become exposed to sulfite
present in the environment, e.g. in the anaerobic regions of the Black Sea (Sorokin
1995). In as far as it has been investigated to date, the common strategy for sulfite
detoxification in Bacteria and Archaea seems to involve an oxidation of sulfite to
sulfate, which can proceed either via a direct formation of sulfate from sulfite, or
via the indirect APS reductase pathway (reviewed in Kappler and Dahl 2001).
Despite the fact that the direct oxidation pathway does not allow for a conservation
of energy via substrate-level phosphorylation, it appears to be the more common of
the two (Kappler and Dahl 2001). The enzymes catalysing the direct oxidation of
sulfite to sulfate will be the focus of the remainder of this chapter.
13.2 Sulfite-Oxidizing Enzymes
At present two types of enzymes catalysing the direct oxidation of sulfite to sulfate
are recognized: SOs (EC 1.8.3.1) and sulfite dehydrogenases (SDH; EC 1.8.2.1).
Both enzymes are metalloproteins that possess a molybdenum-containing redox
centre. Additional redox-active centres (e.g. haem groups) may also be present.
The main difference between the two types of enzymes lies in their ability to
transfer electrons to oxygen: SOs transfer electrons to oxygen, ferricyanide and
sometimes cytochrome c, while SDHs use either or both of the latter two electron
acceptors, but do not transfer electrons to oxygen. Typical SOs are the plant SO
(known electron acceptors oxygen, ferricyanide) (Eilers et al. 2001; Hemann et al.
2005) and the well-studied vertebrate SOs that can use all three electron acceptors
13 Bacterial Sulfite-Oxidizing Enzymes Enzymes for Chemolithotrophs Only? 153
listed above but appear to use a cytochrome c as their natural electron acceptor
(Enemark and Cosper 2002; Rajagopalan 1980).
SDHs have been reported in many bacteria, including the soil bacterium
Starkeya novella (Kappler et al. 2000), Sulfitobacter species (Pukall et al. 1999;
Sorokin 1995), various Thiobacilli, alkanesulfonate-degrading and photosyn-
thetic bacteria (Cook et al. 2006; reviewed in Kappler and Dahl 2001). While for
some bacterial SDHs a monohaem cytochrome c has been established as the natural
electron acceptor (Kappler et al. 2000; Yamanaka et al. 1981), in many cases this
acceptor is not known. In fact, bacterial enzymes can be divided into two groups on
the basis of their preference for either ferricyanide or cytochrome c as an electron
acceptor (Kappler and Dahl 2001).
The preference for either of these electron acceptors might be an important
indicator of the function and/or localization of the respective sulfite-oxidizing
enzymes (SOEs): Enzymes that use a cytochrome c as their natural electron acceptor
clearly have to be located in an extracellular compartment or, if they were membrane
proteins, they would have to possess a periplasmic/extracellular domain that transfers
electrons to cytochrome c.
A preference for ferricyanide as an artificial electron acceptor (reported, e.g., for
the SOEs from Thiobacillus acidophilus and Comamonas acidovorans; Kappler
and Dahl 2001) is also found in the SO from Arabidopsis thaliana that uses oxygen
as its preferred electron acceptor (Mendel and Bittner 2006). Although an extracellular
location cannot be excluded for ferricyanide-dependent SOEs, such a preference
might be indicative of SOEs with an intracellular or membrane location which have
been shown to exist in some chemolithotrophic and alkanesulfonate-degrading
bacteria (Cook et al. 2006; Kappler and Dahl 2001).
It should also be noted that while the plant and animal SOs and the so far characterized
bacterial enzymes are all soluble proteins, both soluble and membrane-bound bacterial
SOEs have been reported (reviewed in Kappler and Dahl 2001).
13.3 Structure and Function of Sulfite-Oxidizing Enzymes
All SOs and SDHs characterized to date belong to the SO family of molybdoenzymes.
This enzyme family comprises the known SOEs and enzymes related to these as
well as the assimilatory nitrate reductases found in plants (Hille 1996). In all
enzymes of the SO family, a molybdenum atom chelated by the dithiolene groups
of a single pyranopterin cofactor (also known as molybdenum cofactor or Moco)
forms the active site (Hille 1996). During catalysis, the Mo centre cycles between
the Mo(VI) and Mo(IV) states (Enemark et al. 2006; Enemark and Cosper 2002;
Hille 1996), and in the oxidized Mo(VI) state the SOE molybdenum site has been
shown to contain two oxo and three sulfur ligands (Fig. 13.1), one of which is from
a conserved cysteine residue. The geometry of the molybdenum site is square
pyramidal (Kisker et al. 1997a) and the equatorial oxo ligand (Fig. 13.1) is directly
involved in the oxidationreduction reaction catalysed by the SOEs. The spectroscopic
154 U. Kappler
Fig. 13.1 The sulfite oxidase active site
and catalytic properties of SOEs have been the subject of many publications and
review articles and the reader is referred to these (Enemark et al. 2006; Enemark
and Cosper 2002; Hille 1996, 2005).
Although the SOEs from vertebrates, plants and bacteria all belong to the same
enzyme family there are significant structural differences between them: while both
the vertebrate and the plant SOs are homodimers containing two Mo sites per
enzyme molecule (Kisker et al. 1997a; Schrader et al. 2003), the bacterial enzyme
from Starkeya novella is a heterodimer with only one Mo site per enzyme molecule
(Kappler et al. 2000; Kappler and Bailey 2005). Both the chicken SO and the
bacterial SDH contain a haem group, but while the chicken SO contains a haem b,
the bacterial SDH contains a haem c located on a second subunit that forms a per-
manent complex with the Mo-containing subunit of the enzyme.
Crystal structures are available for several enzymes of the SO family (Fischer
et al. 2005; Kappler and Bailey 2005; Kisker et al. 1997a; Loschi et al. 2004;
Schrader et al. 2003).
The structure of the chicken SO was the first to be solved and revealed a three-
domain architecture: In this SO a mobile haem b binding domain is connected via a
flexible linker region to a central molybdenum-binding and a C-terminal dimerization
domain which mediates formation of the SO homodimer (Kisker et al. 1997a). Both
the plant SO and the bacterial SDH were shown to also contain two of these domains,
namely the molybdenum-binding and the dimerization domain. In contrast, the YedY
protein (Loschi et al. 2004) lacks both the dimerization and the haem b domain.
Although both the chicken SO and the bacterial SDH contain haem groups, only
the structure of the bacterial enzyme has allowed insights into intramolecular
electron transfer in these SOEs (Kappler and Bailey 2005). Homology modelling
suggests that the electron-transfer competent conformation of the chicken SO is
very similar to that seen in the bacterial SorAB SDH, which highlights the functional
similarity of these two structurally different SOEs.
13.4 Phylogeny of Sulfite-Oxidizing Enzymes
Although SOE activities were first described in bacteria more than 40 years ago and
biochemical evidence for SOEs exists for many bacterial species (reviewed in
Kappler and Dahl 2001), it remained unclear for a long time whether these bacterial
enzymes belonged to the SO family or whether they formed a separate class of

enzymes. While we can still only speculate on whether the latter may be true of
some enzymes still awaiting discovery, it is now clear that many bacterial enzymes
exist that are molybdoproteins and belong of the SO family.
The first bacterial SO-like/SDH proteins to be characterized on both the protein
and the genetic level were the SoxCD sulfur dehydrogenase from Paracoccus
pantotrophus (Quentmeier et al. 2000; Wodara et al. 1997), which is part of a
multienzyme complex and requires the other complex components for activity, and
the SorAB SDH from Starkeya novella (Kappler et al. 2000), which is a true SDH
and functions independently of other proteins.
Since then SDHs have been characterized from Deinococcus radiodurans,
Acidiphilium acidophilum and Campylobacter jejuni (DErrico et al. 2006;
deJong et al. 2000; Myers and Kelly 2005). An unusual enzyme, the YedY
protein from Escherichia coli, which is clearly related to the SO family but is of
unknown function, was also characterized and crystallized (Brokx et al. 2005;
Loschi et al. 2004).
However, with the increasing number of completed genome sequences there is
more and more evidence that genes encoding enzymes belonging to the SO family
are actually widespread in bacteria. We have recently repeated our phylogenetic
analysis of the SO family reported in Kappler and Dahl (2001) 6 years ago: The number
of SOE-related gene sequences in GenBank has now increased to over 500 (from
around 23), and new sequences appear at high frequencies. All sequences consid-
ered in the analysis contain the conserved cysteine residue that serves as a ligand
to the Mo atom in this enzyme family; only residues present in all aligned sequences
were used for the determination of the phylogenetic relationships (Kappler and
Dahl 2001).
A number of entries in the conserved protein domains database (CDD) (Marchler-
Bauer et al. 2005) refer to the Moco-dimerization and different Moco-binding
domains (Table 13.1). All putative proteins considered here contain at least one of
these signature sequences.
On the basis of our analysis, three groups of SOEs can now be clearly distin-
guished (Fig. 13.2), and they have been designated as group 1 pathogen enzymes,
group 2 classic SOEs and nitrate reductases and group 3 enzymes from
Archaea, phototrophic and soil bacteria. The groups have been arbitrarily named on
the basis of either enzyme function or the provenance of sequences. The sequences
identified originate from many different bacterial phyla, including the Proteobacteria,
the green non-sulfur bacteria, the high-GC Gram-positive Bacteria, Bacilli,
Actinomycetes, Myxobacteria and Plantomycetes as well as sequences from both
Euryarchaeota and Crenarchaeota.
Only protein sequences of group 2 SOEs have the two-domain architecture seen
in several crystal structures with a Mo-binding domain and a dimerization domain
found in tandem on one polypeptide (Kappler and Bailey 2005; Kisker et al. 1997a;
Schrader et al. 2003). Both group 1 and group 3 sequences lack the dimerization
domain, and the Mo-binding domain of group 3 is further reduced by the loss of
parts of the N-terminal sequence.
156 U. Kappler
Table 13.1 Sulfite oxidase family related entries in the CDD database (Marchler-Bauer et al. 2005)
CDD numbering NAME
CDD28617 Bact_SorA_Moco
CDD28613 SO_family_Moco_dimer
CDD28614 eukary_SO_Mico
CDD28616 Bact_SoxC_Moco
CDD28609 SO_family Moco
CDD28615 eukary_NR_Moco
CDD28612 arch_bact_SO family
CDD28611 bact_SO_family_Moco
CDD28610 YedY_like_Moco
CDD22936 pfam00174 Oxidoreductases_molybdopterin-binding
CDD23450 pfam03404 Mo-co_dimer
CDD11749 COG2041 Sulfite oxidase and related enzymes
Fig. 13.2 The phylogeny of the sulfite oxidase enzyme family. The three main subfamilies are
indicated by the boxes. Shading is used to accentuate the different major groupings within the
subfamilies. The relative position of sequences of enzymes that are discussed in the text are high-
lighted in bold and labelled. The phylogenetic tree was generated using the neighbour-joining
method. SDH sulfite dehydrogenase
The average deduced molecular mass of the Mo-binding proteins in the different
groups is between 30 and 35 kDa for group 1, 40 and 60 kDa for group 2 (SOEs only)
and 20 and 25 kDa for group 3. Interestingly, with the exception of the E. coli YedY
protein (a group 1 enzyme), all of the well-characterized SOEs are found in group 2,
which therefore has been designated as containing the classic SOEs and nitrate
reductases. To the best of my knowledge no protein belonging to group 3 has been
characterized in any detail so far.
13.5 Diversity of Enzymes Within the Sulfite Oxidase Family
13.5.1 Group 1 SOE-Like Enzymes Originating from

Pathogenic Microorganisms
SOE-like enzymes in group 1 mainly originate from known bacterial pathogens such
as Pseudomonas aeruginosa, pathogenic and nonpathogenic E. coli strains,
Salmonella, Yersinia, Shewanella, Burkholderia and Ralstonia species. Exceptions to
this are sequences from organisms related to the Roseobacter lineage, Rhodobacter
species and Thiobacillus denitrificans, which are also found in this group.
On the basis of the structure of the Mo-binding domain, the group 1 SOEs can
be divided into two subgroups, group 1A, which contains YedY-like enzymes
(average molecular mass of Mo domain around 35 kDa, 300330 amino acids),
and group 1B, which contains SOEs with a Moco domain of around 30 kDa
(approximately 240270 amino acid chain length). Within group 1A, sequence
homologies of around 50% identity (approximately 6668% similarity) are
found. Within group 1B, identity values are around 40% (approximately 60%
similarity), while between the two groups identities fall to around 24% (approxi-
mately 40% similarity).
In an analysis of group 1 sequences for the presence of signal peptides (Signal
P and TatP; Bendtsen et al. 2005; Nielsen et al. 1999) that might indicate an extra-
cellular localization most proteins returned ambiguous results or were predicted to
lack a signal peptide. There were differences even in the predicted location of
closely related sequences such as YedY from E. coli (periplasmic) and from
Yersinia species (no signal peptide predicted). It would then seem that various
cellular locations can be assumed for the proteins in this group of SOEs, none of
which have been predicted to be membrane bound.
In contrast, the genetic context of the group 1 SOE genes is very conserved:
all genes encoding group 1 SOEs are associated with genes encoding con-
served, membrane-bound proteins, which, like the group 1 SOEs themselves,
fall into two categories. Genes encoding group 1A proteins occur together with
genes encoding proteins with six transmembrane helices and belonging to the
UPF0191 category of conserved proteins (Marchler-Bauer et al. 2005). It is
likely that all of these proteins bind haem b similar to YedZ, which is a repre-
sentative of this group and has been described as the second subunit of YedY
(Brokx et al. 2005). In contrast, genes encoding group 1B SOEs are found in
association with genes encoding proteins of the COG4117 thiosulfate reduct-
ase cytochrome b subunit type (Marchler-Bauer et al. 2005) that contain four
conserved transmembrane helices.
It would then appear that all group 1 SOEs interact with a second subunit
which in all cases is a membrane-bound, haem b binding protein which should
allow for transfer of electrons to or from the quinone-pool (see also later and
Fig. 13.4).
158 U. Kappler
13.5.1.1 Group 1A Enzymes: YedY and Related Proteins
The YedY protein from E. coli is the only group 1 enzyme that has been character-
ized to date, and its crystal structure clearly showed that it is a member of the SO
family (Loschi et al. 2004). There are several interesting differences between YedY
and other SO/SDH proteins: A crucial and conserved residue in the other SO
enzymes is Arg55 (SorAB SDH numbering), and in YedY this residue is replaced
by an asparagine residue (Kappler and Bailey 2005; Kisker et al. 1997a; Loschi
et al. 2004).
In addition, instead of a positively charged substrate-binding pocket similar to the
one of the group 2 SOEs (Kappler and Bailey 2005; Kisker et al. 1997a), the YedY
substrate-binding pocket is mainly hydrophobic with several tyrosine (Tyr47, Tyr231)
and tryptophan (Trp223, Trp246) residues. A single charged residue, Glu104, occu-
pies a position close to the Mo centre that is very similar to that of a catalytically
important aspartate residue in bacterial dimethyl sulfoxide and N-oxide reductases
(Loschi et al. 2004). Our analyses showed that both residues that are close to the Mo
active site and those lining the substrate-binding pocket are conserved in the majority
of YedY-like group 1A protein sequences. The sequence around the Mo-binding
Cys102 (YedY numbering) is also highly conserved (Fig. 13.3).
The in vivo function of YedY is unclear at present: YedY lacks sulfite-oxidizing
activity, and may function as a reductase in vivo, as some enzymatic activity has
been obtained with sulfoxides and N-oxides (Loschi et al. 2004). YedY interacts
with the haem b binding membrane protein YedZ (Em7 = 8 mV). YedZ was shown
to interact weakly with wild-type YedY and strongly with a YedY C102S variant
(Brokx et al. 2005). YedZ was also shown to interact with menadiol, which indicates
Fig. 13.3 Conserved amino acid residues around the conserved, Mo-binding cysteine. Strictly
conserved residues (bold), conserved residues (normal type), conserved active-site residues
(boxes), conserved cysteine residue (shaded box)
that the YedYZ-catalysed reaction may be linked to the quinone pool in vivo (Brokx
et al. 2005).
Within group 1A, there exist a number of subgroups, and one of these contains
enzymes from bacteria of the Roseobacter lineage such as Silicibacter pomeroyii,
Roseobacter and Sulfitobacter species (Wagner-Doebler and Biebl 2006). During
the oxidation of organosulfonates by these bacteria sulfite is produced and oxidized
by an as yet uncharacterized SDH with a preference for ferricyanide as an electron
acceptor (data from Silicibacter pomeroyii; Denger et al. 2006). The only SOE-like
protein identified in the Silicibacter genome in addition to this group 1 protein is
related to the SoxC sulfur dehydrogenase which is involved in thiosulfate oxidation
via a multienzyme complex (thiosulfate-oxidizing multienzyme system, TOMES)
and has neither sulfite-oxidizing capacity in the absence of other TOMES proteins
nor an ability to transfer electrons to ferricyanide (Quentmeier et al. 2000). It will
therefore be interesting to determine whether the Roseobacter lineage group of
YedY-like proteins has a role in the degradation of organosulfonates or whether a
novel type of SDH is involved in the process. This may be the case if, as so far pre-
dicted, this ferricyanide-linked SDH is located in the cytoplasm: both the YedY-
like and the SoxC-like SOEs found in Silicibacter pomeroyi are predicted to be
periplasmic proteins (Fig. 13.4).
13.5.1.2 Group 1B 30-kDa Mo-Domain Proteins
The group 1B enzymes differ from the group 1A enzymes not only in their
association with a different type of membrane subunit, but also in the much
lower degree of sequence conservation found around the Mo-binding cysteine
residue (Fig. 13.3). In addition, several of the active-site/substrate binding
pocket residues conserved in group 1A SOEs are not conserved in group 1B SOEs,
and this includes the asparagine residue which in YedY occupies a position
similar to that of the crucial Arg55 (SorAB numbering) found in group 2 SOEs.
The substrate-binding pocket appears to be less hydrophobic than that of YedY,
which could be an indication of a different substrate spectrum/reaction cata-
lysed: only Trp223 is strictly conserved; Tyr231 is present in some cases. It will,
however, be necessary to determine the crystal structure of a group 1B enzyme to
determine whether these observations are meaningful.
Most of the group 1B SOEs (i.e. proteins from Pseudomonas, Burkholderia and
Rhodopseudomonas) appear to be located in the cytoplasm, while a few may have
an N-terminal transmembrane helix. In contrast, for another subgroup containing
sequences from cyanobacterial and myxococcal species the cellular localization is
ambiguous, with most sequences returning either clearly ambiguous prediction
results or being predicted to contain a signal peptide (Fig. 13.4).
The group 1B enzyme originating from Thiobacillus denitrificans is of particular
interest, as a membrane-bound SOE that transferred electrons to ferricyanide has
been purified from this organism (Aminuddin and Nicholas 1974). Experiments with
crude extracts or membrane fractions showed that sulfite oxidation in Thiobacillus
160 U. Kappler
Fig. 13.4 The most frequently encountered enzyme conformations in the three different groups
of sulfite-oxidizing enzymes. Top: Group 1A and group 1B enzymes shown in two possible
conformations, with the Mo-binding subunit either in the cytoplasm or in the periplasm. Bottom:
Group 2 and group 3 enzymes. Group 2 enzymes are shown with the Moco-dimer domain. CSO
chicken liver sulfite oxidase, HSO human sulfite oxidase, PSO plant sulfite oxidase, SorAB SDH
SorAB sulfite dehydrogenase, SoxCD SDH Sox CD sulfur dehydrogenase, TMH transmembrane
helices, c c-type cytochrome, b b-type cytochrome
denitrificans was inhibited by various inhibitors of the electron-transfer chain, indi-

cating that the reaction was coupled to a reduction of the quinone pool. It is possible
that the enzyme described by Aminuddin and Nicholas is identical to the group
1B enzyme considered here.
It is also interesting to note that for some group 1B SOEs originating from
Mycobacterium avium, Rubrobacter xylanophilus and a Nocardioides sp., a gene
fusion between the genes encoding the Mo-binding and the membrane subunit
has occurred.
13.5.2 Group 2: Classic Sulfite-Oxidizing Enzymes

and Nitrate Reductases
A large body of literature is available for the sulfite-oxidizing and nitrate-reducing

proteins that fall into this group. Especially the vertebrate SOs have been covered
extensively, and the reader is referred to several excellent reviews on the topic
(Enemark et al. 2006; Enemark and Cosper 2002; Kisker et al. 1997b; Mendel
2005; Mendel and Bittner 2006).
Three subgroups, group 2A SOs and plant nitrate reductases, group 2B
SoxCD sulfur dehydrogenases and group 2C SorAB SDHs are easily
distinguishable within the group 2 SOEs. There is some variation in the average
predicted molecular masses of the SOEs in the three groups: SOs tend to
have a molecular mass of around 5560 kDa, the nitrate reductases about 95
100 kDa, SoxC-like proteins around 45 kDa and SorA-like proteins about
42 kDa.
All group 2 SOEs show the unique combination of a Moco-dimer domain
and a Moco-binding domain (Fig. 13.4). In several cases, fusions of other
domains to the central Mo and dimerization domains have occurred. This is
most obvious for the plant assimilatory nitrate reductases and the vertebrate
SOs: in the former a C-terminal fusion with a cytochrome b domain (pfam
00173), a flavin adenine dinucleotide binding domain (pfam 00970) and an
NAD-binding domain (pfam 00175) has occurred, while in the case of the SOs
an N-terminal fusion with a cytochrome b encoding gene has occurred. Most of
the group 2 SOEs occur in extracytoplasmic compartments of the cell (Fig.
13.4). There seems to be a lesser degree of conservation in the amino acids sur-
rounding the Mo-binding cysteine residue in this group (Fig. 13.3), and those
residues that are conserved differ markedly from those conserved in the two
other groups of SOEs.
As the major groupings contain proteins that are comparatively well studied and
have already been described in the introductory sections, they will be dealt with
here only briefly.
13.5.2.1 Group 2A: Sulfite Oxidases and Plant Nitrate Reductases
The SOs subgroup contains sequences from most forms of life, including
mammals, birds, amphibians, insects, plants and fungi. Although all characterized
SOs from higher animals are soluble proteins located in the mitochondrial
162 U. Kappler
intermembrane space, the SO-related genes from Xenopus laevis encode a protein
with an N-terminal membrane domain (amino acids 5978) that precedes the haem
b and the Moco-dimer domains.
There are even some bacterial sequences located in this group, which appear to
originate from Mycobacterium bovis, Mycobacterium tuberculosis, Oceanicola
granulosus and Streptomyces nodosus. All four of these sequences lack signal
peptides as well as haem b domains. The two mycobacterial genes are associated
with genes encoding MmpL-type membrane proteins. MmpL proteins are a family
of lipid transporters, some of which have been shown to be involved in the viru-
lence and pathogenesis of mycobacteria (Domenech et al. 2005). It remains to be
established whether these are true SOEs and what function they might have in their
respective source organisms.
13.5.2.2 Group 2B: SoxCD-Like Enzymes Sulfur Dehydrogenases
This group contains protein sequences that are related to the SoxC subunit of the
SoxCD protein that was first described in Paracoccus species (Kelly et al. 1997;
Wodara et al. 1997). SoxCD is a periplasmic (SoxCD)2 heterotetramer (Quentmeier
et al. 2000) in which SoxC contains the Mo redox centre, while SoxD is a c-type
cytochrome containing either one or two haem groups.
SoxCD is part of a TOMES (Friedrich et al. 2005; Kelly et al. 1997), and
enhances the reaction rate of the TOMES complex when assayed together with
other complex components (Quentmeier et al. 2000). The exact nature of the reac-
tion catalysed by the SoxCD protein is unknown, and in isolated form it has a low
affinity for sulfite. SoxCD has been proposed to oxidize a sulfane sulfur atom
bound to a conserved cysteine present in one of the other TOMES components to
a sulfone group (Bardischewsky et al. 2005). A crystal structure for SoxCD is, at
present, not available. Genes encoding SoxC-like SOEs usually occur in TOMES-
encoding sox gene clusters (Friedrich et al. 2005) and this is borne out by our
analyses which show that this is true for most of the group 2B sequences.
The majority of genes encoding SoxC-like SOEs are associated with genes
encoding SoxD cytochrome subunits. These SoxD proteins exist in two different
forms that bind one or two haem groups respectively (Appia-Ayme et al. 2001;
Kappler et al. 2001; Quentmeier et al. 2000). The function of the monohaem or
dihaem forms is unclear as it has recently been shown that a single haem group
is sufficient for SoxCD function within the TOMES (Bardischewsky et al. 2005).
The two SoxD forms differ in their domain structure: both monohaem and dihaem
SoxD proteins contain an N-terminal COG3258(cytochrome c)/2863(cytochrome
c553)-like domain, which is followed by a C-terminal COG3474(cytochrome c556 and
cytochrome c) domain in dihaem SoxD proteins. Consequently, monohaem SoxD
proteins have about 180200 amino acids, while the dihaem forms can have up to
about 400 residues. In addition, the SoxD proteins that have been studied so far
contain a conserved CxxxC motif, which is similar to a motif found in metal-bind-
ing redox proteins such as PrrC and Sco (McEwan et al. 2002) and could indicate
the presence of another redox centre in SoxCD proteins. This CxxxC motif, how-
ever, is absent from a group of monohaem SoxD proteins that are associated with
SoxC-related protein sequences from Burkholderia, Nitrobacter and Bradyrhizobium
species (e.g. accession nos. ZP_00457665, ABA06238, NP_772761). This group of
sequences is also conspicuous for another reason: they are not associated with a sox
gene cluster. It therefore seems possible that they represent a novel type of SoxCD-
related enzyme which may function in a different metabolic context from the
SoxCD sulfur dehydrogenases that are part of a TOMES.
13.5.2.3 Group 2C: SorAB-Like Sulfite Dehydrogenases
The SorAB protein from Starkeya novella was the first true bacterial SOE to be
characterized in sufficient detail to enable its classification as a member of the
SO family (Kappler et al. 2000). It is a true SDH in that it does not transfer
electrons to oxygen, and it was also the first bacterial SOE in group 2 for which
a crystal structure was solved (Kappler and Bailey 2005). SorAB is a periplasmic
heterodimer of a large Moco-dimer domain (40.2 kDa) and a small cytochrome
c subunit (8.8 kDa) The protein as well as some SorAB variants containing
site-directed mutations have been studied in detail (Doonan et al. 2006; Kappler
et al. 2006; Raitsimring et al. 2005).
Proteins related to the SorAB SDH are found in organisms such as Xanthobacter,
Campylobacter, Ralstonia, Rhizobia, Nitrobacter, Kineococcus and Brevibacterium.
Most of the genes encoding SorA-related proteins are associated with a gene encoding
a monohaem cytochrome c; however, in some cases such as Sulfitobacter EE-26
and NAS-14, the sorA-like genes are associated with genes encoding dihaem
cytochromes. There is a high degree of diversity between the haem proteins associated
with the SorA-like SOEs, and sequence homology between these SorB-like
proteins is usually limited to phylogenetically closely related sequences.
So far all characterized members of this enzyme family are true SDHs with a peri-
plasmic location and a haem c binding second subunit. One example of such a protein
is the recently characterized Campylobacter jejuni SDH (Myers and Kelly 2005) that
cross-reacts with anti-Starkeya novella SDH antibodies. The Campylobacter SDH
has been suggested to be involved in survival of Campylobacter under microaero-
bic conditions in the environment or to serve as a mechanism for detoxification of
sulfite (Myers and Kelly 2005).
Again, within the group of SorA-related protein sequences there are exceptions
from the rule, and a whole subgroup of sequences originating from high-GC
Gram-positive bacteria (e.g. Kineococcus, Mycobacterium or Streptomyces) and
some euryarchaeota (Haloarcula, Natronomonas) are not located in the vicinity of
a gene encoding a haem c binding subunit. The sequences contained in this
group also share another characteristic: on average they contain more than 500 amino
acids, and have been predicted to contain three to five transmembrane domains
using the TMHMM program (Moller et al. 2001). These membrane domains appear
to be exclusively located in the N-terminal region of these proteins (amino acids
164 U. Kappler
1200), while the Moco-dimer domains are located between amino acid 250 and
the C-terminus and are predicted to reside on the extracytoplasmic face of the mem-
brane, a characteristic shared by most other group 2 SOEs (Fig. 13.4). It would then
seem that this group of enzymes represents another novel type of SOE-fusion pro-
tein. The N-terminal part of these membrane-bound SOEs contains no conserved
domains, and although there are almost always five transmembrane domains
present, there is no significant degree of conservation between these N-terminal
membrane domains.
13.5.2.4 Other Sulfite-Oxidizing Enzymes in Group 2
A small number of group 2 enzymes do not appear to belong directly to any of the
three major subgroups. These are SOE-related enzymes from various microorganisms,
including Arthrobacter (ZP_00410553), Roseovarious (ZP_00961287),
Sinorhizobium (AAK65805) and Deinococccus radiodurans (AAF12408). This
last enzyme has recently been shown to be a ferricyanide-dependent, molybdenum-
containing SDH by DErrico et al. (2006). The enzyme was constitutively expressed
in Deinococccus, but its exact function remains somewhat unclear and it has been
suggested to be involved in intracellular dissimilatory sulfite oxidation.
In summary it appears as if most of the group 2 SOEs are soluble, mainly extra-
cytoplasmic proteins and are often associated with haem groups, although there are
enzymes (e.g. the plant SOs) that lack additional redox centres. This analysis of the
group has also uncovered some novel enzyme groups such as the bacterial enzymes
in group 2A, the SoxCD-like proteins that are not part of a TOMES and the
haemless, membrane-bound SorA-related enzymes.
13.5.3 Group 3: Sulfite-Oxidizing Enzymes Enzymes from

Archaea, Phototrophic and Soil Bacteria
The third group of SOEs is the one that contains the most reduced Mo-binding
domain: in addition to the absence of the dimerization domain, which leads to a
reduction of the Mo-binding domain to around 30 kDa in the group 1 SOEs, the
domain has been further reduced to an average molecular mass of 2225 kDa by the
loss of some N-terminal parts of the sequence. In addition, to the best of my knowledge,
none of these enzymes have ever been studied, so only some general observations
can be made. The putative protein sequences in this group originate from a variety
of archaeal species (Fig. 13.2) belonging to both Crenarchaeota and Euryarchaeota
(Sulfolobus sp., Ferroplasma, Pyrobaculum, Archaeoglobus, Halobacterium) as
well as bacterial species belonging to the high-GC Gram-positive Bacteria,
Firmicutes (bacilli), the Thermus/Deinococccus group, cyanobacteria and several
a-Proteobacteria. In addition individual sequences from a green non-sulfur bacterium,
a Planctomycete, a Solibacter and an Acidobacterium are also present.
Protein sequence identities in the archaeal group are between 30 and 40%, while
the sequences from bacteria have about 3547% identity. Between these groups the
amino acid identity levels fall to 2533%. Despite this, there is some conservation
of residues surrounding the conserved, Mo-binding cysteine between the two
groups (Fig. 13.3) and the conserved residues found in group 3 are most similar to
those found in the group 1 SOEs.
The vast majority of enzymes in this group lack an export signal (Signal P or
TatP programmes, Bendtsen et al. 2005; Nielsen et al. 1999), and transmembrane
helices appear to be absent from all of the group 3 enzymes, which suggest that
group 3 SOEs are cytoplasmic enzymes (Fig. 13.4).
The genes encoding the group 3 proteins are not found in a conserved genetic
environment. The only exception is a major group of -Proteobacterial sequences
that originate from Rhodopseudomonas, Bradyrhizobium and Xanthobacter species.
The genes encoding these proteins appear to occur together with genes encoding an
OsmC-like protein (COG1765, also COG 2945); however, the gene encoding this
OsmC-like protein may not be another subunit for these group 3 SOEs, but may
simply be conserved because all the sequences found in that cluster are from closely
related bacterial species.
13.6 Conclusions
The combination of genome sequencing, advances in techniques for the study of

proteins and the isolation of many new bacterial species from sulfur-containing
habitats has led to great advances in the study of bacterial and archaeal sulfur
metabolism in recent years, as the phylogenetic analysis of the bacterial SOEs pre-
sented here clearly indicates.
There are still many challenges, and analysis of the genetic data alone may help
with the creation of a systematic overview of an enzyme family such as the SO fam-
ily, but only studies of the in vivo function of the enzymes in question can provide
clues as to their true function and how they are integrated into general metabolism.
It is clear from our analyses though that SOEs occur in many bacterial species, only
some of which are known sulfur chemolithotrophs.
Microbial sulfur oxidation occurs over nearly the entire pH scale, and notably
most of the sequences represented in the analysis presented here of the SO
enzyme family originate from neutrophilic microbes. It is interesting that despite
several publications regarding SOEs from Acidothiobacillus ferrooxidans
(reviewed in Kappler and Dahl 2001), no SOE-related enzyme was detected in
the Acidithiobacillus genome. By the same token, only reports of the presence of
sulfite-oxidizing activities are available for the recently discovered alkali and halo-
alkaliphilic sulfur-oxidizing bacteria (Sorokin et al. 2000), and it will be interesting
to inspect the genome sequences of these organisms once they become available.
There is also the open question as to why some bacteria appear to contain a large
number of genes encoding SOE-related enzymes belonging to several of the three
166 U. Kappler
groups described above. This is the case for Rhizobia, Burkholderia, Campylobacter,
Ralstonia, Streptomyces and others.
In conjunction with our analysis of the phylogeny of the SOEs and the genetic
context in which the different enzymes are found, this apparent redundancy of
genes encoding enzymes from the same family also suggests that the number of
different metabolic functions and possibly also reactions that can be carried out by
the bacterial enzymes of the SO family is much greater than recognized at present.
As most of the SOEs that have been studied to date are soluble proteins, it will be
especially interesting to investigate the properties of the membrane-bound SOEs
found in some high-GC Gram-positive bacteria, and to uncover the role of the
archaeal enzymes from both the Crenarchaeota and the Euryarchaeota. An excit-
ing possibility would be that these enzymes might contain tungsten rather than
molybdenum at their active site, which would make them the first tungsten-containing
enzymes in this metalloprotein family.
Note: A more detailed representation of the different branches of the SOE tree
could not be included in this article owing to space restraints.
Acknowledgement. U.K. thanks the University of Queensland for a grant and a fellowship.
References
Aminuddin M, Nicholas DJD (1974) An AMP-independent sulphite oxidase from Thiobacillus

denitrificans: purification and properties. J Gen Microbiol 82:103113
Appia-Ayme C, Little PJ, Matsumoto Y, Leech AP, Berks BC (2001) Cytochrome complex essential
for photosynthetic oxidation of both thiosulfate and sulfide in Rhodovulum sulfidophilum.
Bardischewsky F, Quentmeier A, Rother D, Hellwig P, Kostka S, Friedrich CG (2005) Sulfur
dehydrogenase of Paracoccus pantotrophus: the heme-2 domain of the molybdoprotein
cytochrome c complex is dispensable for catalytic activity. Biochemistry 44:70247034
Bendtsen JD, Nielsen H, Widdick D, Palmer T, Brunak S (2005) Prediction of twin-arginine
signal peptides. BMC Bioinformatics 6:167176
Brokx SJ, Rothery RA, Zhang GJ, Ng DP, Weiner JH (2005) Characterization of an Escherichia
coli sulfite oxidase homologue reveals the role of a conserved active site cysteine in assembly
and function. Biochemistry 44:1033910348
Carroll KS, Gao H, Chen HY, Stout CD, Leary JA, Bertozzi CR (2005) A conserved mechanism
for sulfonucleotide reduction. PLoS Biol 3:14181435
Chamulitrat W (1999) Activation of the superoxide-generating NADPH oxidase of intestinal lym-
phocytes produces highly reactive free radicals from sulfite. Free Radic Biol Med 27:411421
Cook A, Denger K, Smits T (2006) Dissimilation of C3-sulfonates. Arch Microbiol 185:
8390
deJong GAH, Tang JA, Bos P, de Vries S, Kuenen GJ (2000) Purification and characterization
of a sulfite:cytochrome c oxidoreductase from Thiobacillus acidophilus. J Mol Catal B
8:6167
Denger K, Smits THM, Cook AM (2006) L-Cysteate sulpho-lyase, a widespread pyridoxal
5-phosphate-coupled desulphonative enzyme purified from Silicibacter pomeroyi DSS-3T.
Biochem J 394:657664
DErrico G, Di Salle A, La Cara F, Rossi M, Cannio R (2006) Identification and characterization

of a novel bacterial sulfite oxidase with no heme binding domain from Deinococcus radiodurans.
Domenech P, Reed MB, Barry CE (2005) Contribution of the Mycobacterium tuberculosis MmpL
protein family to virulence and drug resistance. Infect Immun 73:34923501
Doonan CJ, Kappler U, George GN (2006) Structure of the active site of sulfite dehydrogenase
from Starkeya novella. Inorg Chem 45:74887492
Eilers T, Schwarz G, Brinkmann H, Witt C, Richter T, Nieder J, Koch B, Hille R, Haensch R,
Mendel RR (2001) Identification and biochemical characterization of Arabidopsis thaliana
sulfite oxidase a new player in plant sulfur metabolism. J Biol Chem 276:4698946994
Enemark JH, Cosper MM (2002) Molybdenum enzymes and sulfur metabolism. Met Ions Biol
Syst 39:621654
Enemark JH, Astashkin AV, Raitsimring AM (2006) Investigation of the coordination structures
of the molybdenum(V) sites of sulfite oxidizing enzymes by pulsed EPR spectroscopy. Dalton
Trans 35013514
Fischer K, Barbier GG, Hecht HJ, Mendel RR, Campbell WH, Schwarz G (2005) Structural basis
of eukaryotic nitrate reduction: crystal structures of the nitrate reductase active site. Plant Cell
17:11671179
Griffith OW (1987) Mammalian sulfur amino acids metabolism: an overview. Methods Enzymol
143:366376
Hayes MK, Taylor GT, Astor Y, Scranton MI (2006) Vertical distributions of thiosulfate and sulfite
in the Cariaco Basin. Limnol Oceanogr 51:280287
Hemann C, Hood BL, Fulton M, Hansch R, Schwarz G, Mendel RR, Kirk ML,
Hille R (2005) Spectroscopic and kinetic studies of Arabidopsis thaliana sulfite oxidase: nature
of the redox-active orbital and electronic structure contributions to catalysis. J Am Chem Soc
127:1656716577
Hille R (1996) The mononuclear molybdenum enzymes. Chem Rev 96:27572816
Hille R (2005) Molybdenum-containing hydroxylases. Arch Biochem Biophys 433:107116
Kappler U, Bailey S (2005) Molecular basis of intramolecular electron transfer in sulfite-oxidizing
enzymes is revealed by high resolution structure of a heterodimeric complex of the catalytic
molybdopterin subunit and a c-type cytochrome subunit. J Biol Chem 280:2499925007
Kappler U, Dahl C (2001) Enzymology and molecular biology of prokaryotic sulfite oxidation.
FEMS Microbiol Lett 203:19
Kappler U, Bennett B, Rethmeier J, Schwarz G, Deutzmann R, McEwan AG, Dahl C (2000)
Sulfite: cytochrome c oxidoreductase from Thiobacillus novellus purification, characteriza-
tion and molecular biology of a heterodimeric member of the sulfite oxidase family. J Biol
Chem 275:1320213212
Kappler U, Friedrich CG, Trper HG, Dahl C (2001) Evidence for two pathways of thiosulfate
oxidation in Starkeya novella (formerly Thiobacillus novellus). Arch Microbiol 175:102111
Kappler U, Bailey S, Feng CJ, Honeychurch MJ, Hanson GR, Bernhardt PV, Tollin G, Enemark
JH (2006) Kinetic and structural evidence for the importance of Tyr236 for the integrity of the
Mo active site in a bacterial sulfite dehydrogenase. Biochemistry 45:96969705
Kisker C, Schindelin H, Pacheco A, Wehbi WA, Garrett RM, Rajagopalan KV, Enemark JH, Rees
DC (1997a) Molecular basis of sulfite oxidase deficiency from the structure of sulfite oxidase.
Cell 91:973983
Kisker C, Schindelin H, Rees DC (1997b) Molybdenum-cofactor-containing enzymes: Structure
and mechanism. Annu Rev Biochem 66:233267
Lester MR (1995) Sulfite sensitivity significance in human health. J Am Coll Nutr 14:
229232
168 U. Kappler
Loschi L, Brokx SJ, Hills TL, Zhang G, Bertero MG, Lovering AL, Weiner JH, Strynadka NCJ
(2004) Structural and biochemical identification of a novel bacterial oxidoreductase. J Biol
Chem 279:5039150400
Marchler-Bauer A, Anderson J, Cherukuri P, DeWeese-Scott C, Geer L, Gwadz M, He S, Hurwitz D,
Jackson J, Ke Z, Lanczycki C, Liebert C, Liu C, Lu F, Marchler G, Mullokandov M,
Shoemaker B, Simonyan V, Song J, Thiessen P, Yamashita R, Yin J, Zhang D, Bryant S (2005)
CDD: a Conserved Domain Database for protein classification. Nucleic Acids Res
33:D192196
McEvily AJ, Iyengar R, Otwell WS (1992) Inhibition of enzymatic browning in foods and
beverages. Crit Rev Food Sci Nutr 32:253273
McEwan AG, Lewin A, Davy SL, Boetzel R, Leech A, Walker D, Wood T, Moore GR (2002) PrrC
from Rhodobacter sphaeroides, a homologue of eukaryotic Sco proteins, is a copper-binding
protein and may have a thioldisulfide oxidoreductase activity. FEBS Lett 518:1016
Mendel RR (2005) Molybdenum: biological activity and metabolism. Dalton Trans 34043409
Mendel RR, Bittner F (2006) Cell biology of molybdenum. Biochim Biophys Acta 1763:
621635
Mitsuhashi H, Nojima Y, Tanaka T, Ueki K, Maezawa A, Yano S, Naruse T (1998) Sulfite is
released by human neutrophils in response to stimulation with lipopolysaccharide. J Leukoc
Biol 64:595599
Mitsuhashi H, Yamashita S, Ikeuchi H, Kuroiwa T, Kaneko Y, Hiromura K, Ueki K, Nojima Y
(2005) Oxidative stress-dependent conversion of hydrogen sulfide to sulfite by activated
neutrophils. Shock 24:529534
Moller S, Croning MDR, Apweiler R (2001) Evaluation of methods for the prediction of
membrane spanning regions. Bioinformatics 17:646633
Myers JD, Kelly DJ (2005) A sulphite respiration system in the chemoheterotrophic human
pathogen Campylobacter jejuni. Microbiology 151:233242
Nielsen H, Brunak S, VonHeijne G (1999) Machine learning approaches to the prediction of
signal peptides and other protein sorting signals. Protein Eng 12:39
Pukall R, Buntefu D, Frhling A, Rohde M, Kroppenstedt RM, Burghardt J, Lebaron P, Bernard L,
Stackebrandt E (1999) Sulfitobacter mediterraneus sp. nov., a new sulfite-oxidizing member of
the alphaproteobacteria. Int J Syst Evol Microbiol 49:513519
Quentmeier A, Kraft R, Kostka S, Klockenkamper R, Friedrich CG (2000) Characterization of a
new type of sulfite dehydrogenase from Paracoccus pantotrophus GB17. Arch Microbiol
173:117125
Raitsimring AM, Kappler U, Feng CJ, Astashkin AV, Enemark JH (2005) Pulsed EPR studies of
a bacterial sulfite-oxidizing enzyme with pH invariant hyperfine interactions from exchangeable
protons. Inorg Chem 44:72837285
Rajagopalan KV (1980) Sulfite oxidase (sulfite: ferricytochrome c oxidoreductase). In: Coughlan
MP (ed) Molybdenum and molybdenum-containing enzymes. Pergamon, Oxford,
pp 243272
Ratthe C, Pelletier M, Roberge CJ, Girard D (2002) Activation of human neutrophils by the pollutant
sodium sulfite: effect on cytokine production, chemotaxis, and cell surface expression of cell
adhesion molecules. Clin Immunol 105:169175
Roy AB, Trudinger PA (1970) The chemistry of some sulfur compounds. In: Roy AB, Trudinger
PA (eds) The biochemistry of inorganic sulfur compounds. Cambridge University Press,
London, pp 742
Schrader N, Fischer K, Theis K, Mendel RR, Schwarz G, Kisker C (2003) The crystal structure
of plant sulfite oxidase provides insights into sulfite oxidation in plants and animals. Structure
11:12511263
Sorokin DY (1995) Sulfitobacter pontiacus gen. nov., sp. nov. a new heterotrophic bacterium
from the black sea specialized on sulfite oxidation. Microbiology 64:295305
Sorokin DY, Kuenen GJ, Jetten MSM (2000) Denitrification at extremely high pH values by the
alkaliphilic, obligately chemolithoautotrophic, sulfur-oxidizing bacterium Thioalkalivibrio
denitrificans strain ALJD. Arch Microbiol 175:94101

Wagner-Doebler I, Biebl H (2006) Environmental biology of the marine Roseobacter lineage.
Annu Rev Microbiol 60:255280
Wodara C, Bardischewsky F, Friedrich CG (1997) Cloning and characterization of sulfite
dehydrogenase, two c-type cytochromes, and a flavoprotein of Paracoccus denitrificans
GB17: Essential role of sulfite dehydrogenase in lithotrophic sulfur oxidation. J Bacteriol
179:50145023
Yamanaka T, Yoshioka T, Kimura K (1981) Purification of sulphite cytochrome c reductase of
Thiobacillus novellus and reconstitution of its sulphite oxidase system with the purified
constituents. Plant Cell Physiol 22:613622
Zhang X, Vincent AS, Halliwell B, Wong KP (2004) A mechanism of sulfite neurotoxicity direct
inhibition of glutamate dehydrogenase. J Biol Chem 279:4303543045
Chapter 14
Sulfonates and Organotrophic Sulfite
Metabolism
Alasdair M. Cook, Theo H.M. Smits, Karin Denger
Abstract One is used to considering sulfite oxidation as part of a lithotrophic

process (e.g. SorAB or Sox system), much of which involves neutral or ionic
inorganic sulfur species on the outer surface of the cytoplasmic membrane. In
contrast, the processes referred to in this chapter involve organic compounds,
which (1) include a highly stable sulfonate substituent (CSO3), (2) are involved
in the organotrophic growth of the organism and (3) much of whose metabolism
takes place in the cytoplasm. Many phenomena are associated with this life-style.
The sulfonate may be a natural product, e.g. taurine or sulfoquinovose, whose
synthesis can involve sulfite, or a xenobiotic laundry detergent, but it is effec-
tively always a charged species, so an uptake system is essential. Two known
systems are mentioned, ATP binding cassette transporters and tripartite ATP-
independent periplasmic transporters. Annual dissimilation of megatonnes of
organosulfonates essentially always involves intracellular sulfite generated by
diverse enzymic cleavages in bacteria, archaea and possibly eukarya. The fate of this
sulfite in anaerobes is often sulfide. Aerobes occasionally excrete sulfite directly;
more frequently, the sulfite is oxidized to sulfate. Many aerobic bacteria excrete only
sulfate, but many others excrete also some sulfite, which is rapidly oxidized to sulfate,
even under anoxic conditions. The nature and location of these sulfite dehydrogenases
are still unclear, but periplasmic SorAB is apparently used by some bacteria. In
contrast to the cytochrome c coupled SorAB, there is at least one widespread, unchar-
acterized sulfite dehydrogenase, which is assayed with ferricyanide as the electron
acceptor. Dissimilation of a sulfonate releases sulfite (sulfate) in about 500-fold excess of
the sulfur requirement for growth, so exporters, two classes of which have been
detected, are essential to prevent the cell from suffering osmotic stress.
14.1 Introduction
The sulfur cycle has many facets of different magnitudes, and this conference
(ISMSM) examined processes involving the major lithotrophic mass fluxes and
placed emphasis on membrane-associated processes. Here, the focus is moved to
organotrophy with biosynthesis, biotransformation and dissimilation of a group of
170
Springer 2008
14 Sulfonates and Organotrophic Sulfite Metabolism 171
Fig. 14.1 Representative aliphatic organosulfonates from the atmosphere, vertebrates, spiders, bacteria,
archaea, plants and algae. The arrows indicate some degradative routes in the literature (Cook and Denger
2002; Cook et al. 2006). Natural sulfonates are obviously ubiquitous, and the CSO3 bond is not
degraded by, e.g., mammals, which excrete organosulfonates (Huxtable 1992). The widespread utilization
of organosulfonates is by microbes, whereby up till now largely bacteria were meant (Cook and Denger
2002; Cook et al. 1999, 2006): utilization by archaea, suggested by sequence data (Rein et al. 2005), has
been supported by the first experimental data (J. van der Oost and T.H.M Smits, unpublished data), and
utilization by a dinoflagellate is suspected (Mayer et al. 2006)
organic compounds, namely organosulfonates, which contain sulfur in oxidation

state +5 (Vairavamurthy et al. 1994; see also Chap. 20 by Prange). These are
ancient compounds, some of which have only recently been discovered, e.g.
methanesulfonate (Fig. 14.1) with its significance in the sulfur cycle (Baker et al.
1991). Another sulfonate, discovered before organic chemistry became a synthetic
science, is taurine, 2-aminoethanesulfonate (Fig. 14.1): it was initially called
Gallen-Asparagin (Tiedemann and Gmelin 1827), and the name Taurin was
attributed to Gmelin (Demaray 1838). A century later, den Dooren de Jong (1926)
included taurine in some of his studies on microbial nutrition. Many natural
organosulfonates of increasing complexity, and sometimes at enormous concentrations,
are cited in reviews (Huxtable 1992; Jacobson and Smith 1968; Yancey et al. 2002)
172 A.M. Cook et al.
Fig. 14.2 A natural arylsulfonate and three commercially available arylsulfonates. The natural product
(Hickford et al. 2004) is juxtaposed with one known desulfonation (during ring cleavage) (Junker et al.
1994a): a typical desulfonation prior to ring cleavage (Junker et al. 1994b) and desulfonation subse-
quent to ring cleavage (Feigel and Knackmuss 1993; Schleheck et al. 2004) are illustrated
and research papers (Abraham et al. 2004; Suzuki et al. 2002; Vollrath et al. 1990)
(Fig. 14.1), which lead us to conclude that large quantities of natural sulfonates are
being cycled in the food webs in marine and terrestrial environments. Natural
aliphatic sulfonates are widely known, but natural arylsulfonates are also being
found (Budzikiewicz et al. 1998; Hickford et al. 2004; Ovenden and Capon 1999),
one of which is shown in Fig. 14.2.
The degradation of these natural arylsulfonates is presumably the background to
the degradation of anthropogenic arylsulfonates. Mankind now uses megatonnes of
sulfonated surfactants annually, especially linear alkylbenzenesulfonate (Knepper

and Berna 2003; Fig. 14.2). Dyestuffs often contain sulfonated moieties (Fig. 14.2).
It is normal to formulate cationic pharmaceuticals with sulfonates, both aromatic
and aliphatic (ONeil 2001): recently, a major aid in assisting patients to abstain
from alcohol, and that is about 4% of the population, is a sulfonate, which is dosed
at 2 g day1 (Cook et al. 2006). Candidate drugs (sulfonates) for stroke and
Alzheimers disease are also dosed at high levels (Cook et al. 2006).
Kondos group made the first attempts to elucidate degradative pathways for
taurine and isethionate (Kondo and Ishimoto 1972), with the consequence that
sulfite was recognized as a degradative intermediate. About the same time, biode-
gradable sulfonated surfactants were introduced in commerce, and Cain recognized
that the degradation of model arylsulfonates involved sulfite as an intermediate
(Johnston et al. 1975): it now appears that all enzymic desulfonation reactions
studied involve the release of sulfite (Cook and Denger 2002; Cook et al. 1999,
2006; Lie et al. 1998). Biosynthesis of natural sulfonates can also involve sulfite
(see later). The amounts of sulfite in these organotrophic processes do not compete
with those in lithotrophic metabolism, but these several megatonnes of sulfite, the
biodiversity in the metabolic pathways and the physiological problems generated
(and solved) in these pathways are the backgrounds to this review.
14.2 Biosynthesis of Organosulfonates
A major part of our understanding of the biosynthesis of non-carbohydrate, aliphatic

sulfonates comes from White (1984, 1986), who set out to characterize the biosyn-
thesis of coenzyme M in methanogenic archaea, and of a sulfolipid in bacteria (capnine;
Fig. 14.1). The two turned out to be related (Fig. 14.3), and the later discovery that
coenzyme M is involved in the cleavage of epoxides in aerobic bacteria involved in
alkene degradation (Coleman and Spain 2003) broadens the relevance of Whites
central pathway. Whites pathway is involved not only in the generation of sulfolac-
tate for bacterial endospore formation, and l-cysteate for capnine-like lipids (Graham
et al. 2002), but, presumably, also for taurine and taurolipids (Fig. 14.3). The key to
the sulfonation reaction is a Michael addition of sulfite to the double bond of phos-
phoenolpyruvate by phosphosulfolactate synthase (EC 4.4.1.19; ComA) (Fig. 14.3).
The synthesis of sulfoquinovose (Fig. 14.1) is also an addition of sulfite to an
activated compound, in this case UDP-glucose (Sanda et al. 2001). Half the sulfur
in plants is apparently present as sulfoquinovose, and the corresponding sulfolipid
is found in photosynthetic algae, protists and bacteria, so this reaction also con-
sumes considerable amounts of sulfite in the biosphere.
The mammalian synthesis of taurine involves an oxygenation of cysteine to
cysteine sulfinate, a decarboxylation and an unknown oxidation to taurine
(Stipanuk 2004; Fig. 14.4). The important fact here is that the taurine is excreted.
Mammals cannot cleave the CSO3 bond (Huxtable 1992). Dietary cysteate is
apparently excreted as sulfolactate and sulfopyruvate (Weinstein and Griffith 1988).
Fig. 14.3 The biosynthesis of coenzyme M in methanogens (bold arrows) and our interpretation
(normal arrows) of the generalized pathway to supply different microorganisms with sulfolactate
(spore-formers), l-cysteate for sulfolipids (Cytophagales), taurine for sulfolipids (marine bacteria
and some algae) and coenzyme M for the aerobes which also use the cofactor in biodegradation
Fig. 14.4 Synthesis of taurine in mammals and spiders, and excretion of sulfonates. Taurine has
many functions in mammals (Huxtable 1992), but after being functional, the compound is
excreted, largely in urine. l-Cysteate in mammals is dietary, and transamination and excretion are
indicated (Weinstein and Griffith 1988). Large amounts of sulfonates are involved in the function
of spiders webs (Vollrath et al. 1990)
Another major source of extracellular sulfonates is spiders webs (Vollrath et al.

1990; Fig. 14.4). So not only do many organisms produce sulfonates (Fig. 14.3),
which will be released when herbivores graze, but many organisms excrete
sulfonates (Fig. 14.4) directly into the environment, where biotransformation or
dissimilation can take place.
Recent research (Dominy et al. 2006) shows that the oxygenation reaction
(EC 1.13.11.20) found in mammals is also present in some bacteria. Apparently
this can function as a second pathway to supply sulfolactate for spore formation.
14.3 Dissimilation of Organosulfonates
We have seen that the sulfonates are costly to generate. Either a cell invests a
high-energy bond to obtain a sulfonate, or it risks oxidative stress by involving an
oxygenase. The carbonsulfonate bond is a strong bond, about as strong as a
carboncarbon bond, and the first organic chemists were astonished at the resistance
of taurine to strong acid or alkali. The consequence for the biodegradation of orga-
nosufonates is that a very stable bond must be broken. We believe that the natural
organosulfonates are phylogenetically ancient entities (Huxtable 1992; Kelly and
Murrell 1999), and that the biodiversity we see in desulfonation mechanisms
reflects the long exposure of microbes to organosulfonates.
The diversity seen in the desulfonation of arylsulfonates will serve as an
introduction. Note that we are talking about dissimilation: M. Kertesz, T. Tralau
and A. Schmalenberger (personal communication) introduced a different set of
desulfonative enzymes involved in the assimilation of sulfonate sulfur. Some aryl-
sulfonates are desulfonated concomitantly with activation of the ring by multicom-
ponent dioxygenases, as for 4-toluenesulfonate (Fig. 14.2). One case is known in
which desulfonation is concomitant with the simpler dioxygenation involved in
ring cleavage (Fig. 14.2). And in the third example, (di)oxygenations generate the
molecule which can be subtly manipulated and desulfonated by hydrolysis, as in
the case of linear alkylbenzenesulfonate (Fig. 14.2). In each case, sulfite is the
stoichiometric product of the enzyme reaction.
When we consider the aliphatic sulfonates (Fig. 14.5), we again see sulfite as the
stoichiometric product of desulfonation. Methylsulfonate monooxygenase is
another multicomponent oxygenase (Kelly and Murrell 1999), and we presume it
to be archetypal for many similar reactions (Cook et al. 2006). Suitably placed
substituents on sulfonates allow less spectacular desulfonations, as can be seen for
sulfolactate and l-cysteate (Fig. 14.5) (Cook et al. 2006). The reaction we know
best, inasmuch as we have sketches of complete pathways (see later), is sulfoacetal-
dehyde acetyltransferase (Xsc) (Fig. 14.5). As in all the desulfonation reactions in
Fig. 14.2, the enzymes are soluble and in the cytoplasm.
Now that desulfonations have been introduced, it is relevant to draw attention to,
e.g., l-cysteate sulfo-lyase (Fig. 14.5) in a different manner. The substrate carries
three charged moieties, and the three products carry one each. None of these
Fig. 14.5 Some desulfonation reactions. Methanesulfonate monooxygenase is a multicomponent

system (MsmABCD), which generates formaldehyde and sulfite from the substrate. Sulfolactate
sulfo-lyase has two subunits (SuyAB) and tightly bound Fe(II). l-Cysteate sulfo-lyase (CuyA)
represents a third desulfonation mechanism with cofactor pyridoxal 5-phosphate (PLP). The
sulfoacetaldehyde acetyltransferase (Xsc) reaction involves thiamin diphosphate (ThDP) as a
cofactor
compounds will pass through a protein-free bimolecular lipid leaflet. The cell
needs to keep its carbon source in the cell for energy conservation and growth, so
the pyruvate disappears. This potentially leaves the cell with problems, because the
nitrogen supply is in about fourfold excess and the sulfur supply is in about 500-fold
excess. Not only that, this sulfur source, sulfite, is considered to be toxic. Exploding
may be one answer to toxin at high osmotic pressure, but it seems a bit extreme,
and considering the amount of desulfonation in extant microorganisms (Figs. 14.2,
14.5), it is obviously not the response that cells have developed.
14.4 The Detoxification or Fate of Sulfite
We would like to introduce the critical situation gradually, with the desulfonation
reaction that we know from pathways whose genes are found in many genomes
(Brggemann et al. 2004; Cook and Denger 2002, 2006; Cook et al. 2006; Denger
et al. 2006a; Gorzynska et al. 2006; Rein et al. 2005). Our hypothesis for the
dissimilation of taurine in anaerobic Desulfotalea psychrophila is given in Fig. 14.6.
Fig. 14.6 Hypothetical pathway of taurine degradation in Desulfotalea psychrophila LSv54.

Given that the organism grows with taurine (R. Rabus, unpublished data), we deduced the path-
way from the genome sequence (Rabus et al. 2004) and our experience with related pathways in
other organisms (Denger et al. 2006a; Gorzynska et al. 2006). Ack acetate kinase, Ald alanine
dehydrogenase, Tpa taurine:pyruvate aminotransferase, DsrAB dissimilatory sulfite reductase,
Pta phosphotransacetylase
Gene candidates for a tripartite ATP-independent periplasmic transporter, TauKLM,

are present, as are genes to encode a taurine:pyruvate aminotransferase and for the
oxidative release by alanine dehydrogenase of the ammonium ion, which presumably
is exported by the AmtB facilitator. So here we have presumably resolved the
problem of accumulation of the ammonium ion. Desulfonation by Xsc is presuma-
bly followed by energy conservation involving dissimilatory sulfite reductase
(see Chap. 2 by Fritz et al. and Chap. 3 by Pereira), so there is nothing unusual
about a sulfate reducer letting HS (H2S?) diffuse out of the cell. The cell has a secondary
source of ATP, via acetate kinase, which will lead to the disposal of acetate; again,
this is nothing unusual in an anaerobe.
Another of our isolates, Desulfonispora thiosulfatigenes, excretes thiosulfate in
place of sulfide (Denger et al. 1999). Yet another isolate, Desulfovibrio sp. strain
GRZCYSA, generates sulfate and sulfide (Laue et al. 1997). Disposal of these
oxyanions is more difficult to explain, but we have no further data on these processes
in anaerobes, so let us consider the corresponding process in aerobes.
The DOE Joint Genomic Institute has recently reannotated the genome sequence of
Burkholderia xenovorans LB400 and incorporated our suggestions about taurine
metabolism (Brggemann et al. 2004; Rein et al. 2005; Ruff et al. 2003), while
V. Sauv and B. Berks (personal communication) and C. Dahl (personal communication;
see Chap. 9 by Grimm et al.) recommended locating cytochrome c in the periplasm or
the cytoplasmic membrane. This allows the development of the following scheme
(Fig. 14.7). The inducible pathway presumably consists of, in the carbon-relevant
aspect, an ATP binding cassette transporter, a cytochrome c coupled, membrane-bound
taurine dehydrogenase (TauXY), the desulfonation by Xsc, phosphate acetyltransferase
(phosphotransacetylase; Pta) and further metabolism of acetylcoenzyme A. The inorganic
aspects involve ammonia and sulfite. We presume that the AmtB facilitator is present to
Fig. 14.7 Degradative pathway for taurine in Burkholderia xenovorans LB400. The genome
sequence, experimental data on taurine dehydrogenase (TDH), Xsc and Pta (Ruff et al. 2003), with
support for sulfite dehydrogenase (SorAB) from a different strain (Table 14.1), and other data
(Denger et al. 2006a; Gorzynska et al. 2006; Rein et al. 2005) form the basis for this figure
excrete ammonia. We assume that there is at least one sulfite dehydrogenase present,
SorAB (see later), and the organism presumably uses the gene product of orfX, in the
xsc-pta-orfX cluster (Brggemann et al. 2004), to excrete sulfite to the periplasm, where
most of the oxyanion is oxidized immediately. Some transient sulfite is detected outside
the cell (Table 14.1), which presumably indicates a faster excretion of sulfite via OrfX
than oxidation via SorAB.
The overall picture is thus quite complex. There is the desulfonative pathway
itself, including the sulfite exporter, but apparently with an independent cytochrome
c for taurine dehydrogenase, an independent AmtB and an independent sulfite
dehydrogenase. The glyoxylate pathway is also needed. So our intelligent microbes
have a sophisticated strategy to deal with the complexities of metabolizing this
apparently simple molecule, taurine.
There is considerable biodiversity in this small pathway (Figs. 14.6, 14.7). The
diversity continues in the unresolved details of the sulfite dehydrogenases.
14.5 Sulfite Dehydrogenases in Sulfonate Metabolism
Lectures at this conference dealt with three major sulfite dehydrogenases, periplas-
mic SorAB (see Chap. 13 by Kappler), an aspect of the periplasmic Sox system (see
Chap. 12 by Friedrich et al.; V. Sauv and B. Berks, personal communication) and
the intracellular, indirect pathway via adenosyl phosphosulfate (see Chap. 2 by
Table 14.1 The nature of the sulfite dehydrogenases involved in aerobic growth of bacteria utilizing taurine. Paracoccus spp. can usually express the sox
genes, so the parentheses indicate that this property has not been confirmed in these strains
Cytochrome-c-coupled Ferricyanide-coupled Sulfite excreted
Organism soxCD sorAB sulfite dehydrogenase sulfite dehydrogenase during growth
Paracoccus pantotrophus NKNCYSAa (+) No data None detected Inducible No
Paracoccus denitrificans NKNISb (+) No data None detected None detected Yes
Paracoccus versutus N-MTc (+) No data None detected Inducible Yes
Silicibacter pomeroyi DSS-3d One None None detected Inducible No
Rhodobacter sphaeroides 2.4.1e None None None detected None detected No
Burkholderia sp. strain ICDf No data No data Inducible No data Yes
Burkholderia xenovorans LB400g None 2? No data No data Yes
Comamomas sp. strain SFCD1h No data No data Inducible No data Yes
Delftia acidovorans NATi No data No data None detected Inducible No
Alcaligenes faecalis MT-1c No data No data None detected Inducible Yes
14 Sulfonates and Organotrophic Sulfite Metabolism
The columns referring to genes contain information derived or inferred from genome sequences. The three right-hand columns refer to biochemical and
physiological data.
a
Rein et al. (2005).
b
Brggemann et al. (2004).
c
Weinitschke et al. (2006).
d
Gorzynska et al. (2006).
e
Denger et al. (2006a).
f
King et al. (1997).
g
Unpublished data (from JGI and from S. Weinitschke, K. Denger and S.M. Cook).
h
King and Quinn (1997).
i
Mayer et al. (2006).
179
Fritz et al. and Chap. 3 by Pereira) in anoxygenic phototrophs. We have not found
the indirect pathway in aerobes (Denger et al. 2006a), which is unsurprising, given
the sensitivity of the enzymes to oxygen (see Chap. 2 by Fritz et al.). We suspect
that the Sox system is seldom involved, because no organism with the genes on its
genome seems to express a cytochrome c coupled sulfite dehydrogenase (Table 14.1).
We suspect that SorAB is sometimes involved (Fig. 14.7), because Quinns group
found cytochrome c coupled sulfite dehydrogenase in their Burkholderia sp. strain
ICD (King et al. 1997; Ruff et al. 2003), which could correspond with the pres-
ence of candidate sorAB genes in B. xenovorans LB400 (Table 14.1). Quinns
group (King and Quinn 1997) also found candidate SorAB in Comamonas sp.
strain SFCD1.
However, the option to oxidize sulfite in the periplasm seems to be only one
possibility, and some organisms use different options with different substrates.
Silicibacter pomeroyi DSS-3 utilizes taurine, induces a sulfite dehydrogenase and
excretes sulfate via an unknown exporter; no sulfite is observed. However, when
the organism utilizes l-cysteate with induction of sulfite dehydrogenase, it excretes
sulfite almost quantitatively, apparently via CuyZ, a paralogue of TauZ, the pre-
sumed sulfate exporter in many Alphaproteobacteria (Denger et al. 2006a, b;
Gorzynska et al. 2006; Rein et al. 2005). This sulfite dehydrogenase is arguably
cytoplasmic, because a periplasmic enzyme would not allow sulfite to accumulate
to significant amounts extracellularly.
We suspect that this sulfite dehydrogenase in S. pomeroyi DSS-3 represents a
major group of unknown sulfite dehydrogenases, which is found in many of our
isolates (Table 14.1). The enzyme was discovered by Reichenbecher et al. (1999)
in a strain of Delftia acidovorans, and only in D. acidovorans have we been able to
elute active enzyme from a chromatography column (K. Denger, unpublished data).
We hope to be able to characterize this enzyme in the near future.
Some organisms have no detectable sulfite dehydrogenase (Table 14.1), one of
which (Paracoccus denitrificans NKNIS) leaks some sulfite during sulfate forma-
tion and one of which (Rhodobacter sphaeroides 2.4.1) does not. We are, thus,
uncertain whether yet more sulfite dehydrogenases await discovery, or whether the
assay conditions used were unsuitable.
14.6 Conclusions
Organosulfonates are widespread in Nature. Large amounts of sulfite are used to

generate these compounds. Some sulfonates are generated via enzymic or atmos-
pheric oxygenation of sulfhydryl groups.
These organosulfonates are sometimes functional as synthesized, but larger
compounds (e.g. lipids, surfactants and even nucleic acids) are also generated.
All these natural compounds are subject to biodegradation in food webs, and
there are indications that not only bacteria, but also archaea and protists can cleave
the stable carbonsulfonate bond. There is a broad range of enzymes involved in

desulfonation and desulfonative pathways.
The product of desulfonation is always sulfite. This sulfite has many possible
fates, depending on the type of metabolism in the host organism. Similar degrada-
tive pathways are found in strict anaerobes, facultative anaerobes (where they function
under both oxic and anoxic conditions) and strict aerobes. Under aerobic conditions,
the diversity of sulfite dehydrogenases is incompletely understood.
Cells have developed fairly complex sets of responses to the multiplicity of
charged compounds generated during the degradation of these charged organosul-
fonates. We think that we can describe these phenomena, but our understanding at
the molecular level is still very limited.
Acknowledgements. We are grateful to R. Rabus for growing D. psychrophila and to

the DOE Joint Genomic Institute for making sequence data available on their Web
site, especially for B. xenovorans LB400 and R. sphaeroides 2.4.1. Many under-
graduate students in our advanced teaching laboratory also contributed to the data.
The research in our laboratory was funded by the University of Constance, the
DFG, the European Union (SUITE), ECOSOL and CLER, and the LBS Stiftung
Umwelt und Wohnen.
References
Abraham W-R, Strmpl C, Vancanneyt M, Bennasar A, Swings J, Lnsdorf H, Smit J, Moore

ERB (2004) Woodsholea maritima gen. nov., sp. nov., a marine bacterium with a low diversity
of polar lipids. Int J Syst Evol Microbiol 54:12271234
Baker SC, Kelly DP, Murrell JC (1991) Microbial degradation of methanesulphonic acid: a missing
link in the biogeochemical sulphur cycle. Nature 350:627628
Brggemann C, Denger K, Cook AM, Ruff J (2004) Enzymes and genes of taurine and isethionate
dissimilation in Paracoccus denitrificans. Microbiology 150:805816
Budzikiewicz H, Fuchs R, Taraz K, Marek-Kozaczuk M, Skorupska A (1998) Dihydropyoverdin-
7-sulfonic acids unusual bacterial metabolites. Nat Prod Lett 12:125130
Coleman NV, Spain JC (2003) Distribution of the coenzyme M pathway of epoxide metabolism
among ethene- and vinyl chloride-degrading Mycobacterium strains. Appl Environ Microbiol
69:60416046
Cook AM, Denger K (2002) Dissimilation of the C2 sulfonates. Arch Microbiol 179:16
Cook AM, Denger K (2006) Metabolism of taurine in microorganisms: a primer in molecular
diversity? Adv Exp Med Biol 583:313
Cook AM, Laue H, Junker F (1999) Microbial desulfonation. FEMS Microbiol Rev 22:399419
Cook AM, Denger K, Smits THM (2006) Dissimilation of C3-sulfonates. Arch Microbiol
185:8390
Demaray H (1838) Ueber die Natur der Galle. Ann Pharm 27:270291
den Dooren de Jong LE (1926) Bijdrage tot de kennis van het mineralisatieproces. Nijgh & van
Ditmar, Rotterdam
Denger K, Stackebrandt E, Cook AM (1999) Desulfonispora thiosulfatigenes gen. nov., sp. nov.,
a widespread, taurine-fermenting, thiosulfate-producing, anaerobic bacterium. Int J Syst
Denger K, Smits THM, Cook AM (2006a) Genome-enabled analysis of the utilization of taurine as sole
source of carbon or nitrogen by Rhodobacter sphaeroides 2.4.1. Microbiology 152:31973206
Denger K, Smits THM, Cook AM (2006b) L-Cysteate sulfo-lyase, a widespread, pyridoxal
5-phosphate-coupled desulfonative enzyme purified from Silicibacter pomeroyi DSS-3T.
Dominy JE Jr, Simmons CR, Karplus PA, Gehring AM, Stipanuk MH (2006) Identification and
characterization of bacterial cysteine dioxygenases: a new route of cysteine degradation in
eubacteria. J Bacteriol 188:55615569
Feigel BJ, Knackmuss H-J (1993) Syntrophic interactions during degradation of 4-aminobenze-
nesulfonic acid by a two species bacterial culture. Arch Microbiol 159:124130
Gorzynska AK, Denger K, Cook AM, Smits THM (2006) Inducible transcription of genes
involved in taurine uptake and dissimilation by Silicibacter pomeroyi DSS-3T. Arch Microbiol
185:402406
Graham DE, Xu H, White RH (2002) Identification of coenzyme M biosynthetic phosphosulfolactate
synthase: a new family of sulfonate biosynthesizing enzymes. J Biol Chem 277:1342113429
Hickford SJH, Kpper FC, Zhang G, Carrano CJ, Blunt JW, Butler A (2004) Petrobactin sulfonate,
a new siderophore produced by the marine bacterium Marinobacter hydrocarbonoclasticus.
J Nat Prod 2004:18971899
Huxtable RJ (1992) Physiological actions of taurine. Physiol Rev 72:101163
Jacobson JG, Smith LH (1968) Biochemistry and physiology of taurine and taurine derivatives.
Physiol Rev 48:424511
Johnston JB, Murray K, Cain RB (1975) Microbial metabolism of aryl sulphonates. A reassessment
of colorimetric methods for the determination of sulphite and their use in measuring desulpho-
nation of aryl and alkylbenzene sulphonates. Antonie Van Leeuwenhoek 41:493511
Junker F, Field JA, Bangerter F, Ramsteiner K, Kohler H-P, Joannou CL, Mason JR, Leisinger T,
Cook AM (1994a) Oxygenation and spontaneous deamination of 2-aminobenzenesulphonic
acid in Alcaligenes sp. strain O-1 with subsequent meta ring cleavage and spontaneous
desulphonation to 2-hydroxymuconic acid. Biochem J 300:429436
Junker F, Leisinger T, Cook AM (1994b) 3-Sulphocatechol 2,3-dioxygenase and other dioxygenases
(EC 1.13.11.2 and EC 1.14.12.-) in the degradative pathways of 2-aminobenzenesulphonic,
benzenesulphonic and 4-toluenesulphonic acids in Alcaligenes sp. strain O-1. Microbiology
140:17131722
Kelly DP, Murrell JC (1999) Microbial metabolism of methanesulfonic acid. Arch Microbiol
172:341348
King JE, Jaouhari R, Quinn JP (1997) The role of sulfoacetaldehyde sulfo-lyase in the mineralization
of isethionate by an environmental Acinetobacter isolate. Microbiology 143:23392343
King JE, Quinn JP (1997) Metabolism of sulfoacetate by environmental Aureobacterium sp. and
Comamonas acidovorans isolates. Microbiology 143:39073912
Knepper TP, Berna JL (2003) Surfactants: properties, production, and environmental aspects.
In: Knepper TP, Barcel D, de Voogt P (eds) Analysis and fate of surfactants in the aquatic
environment. Elsevier, Amsterdam, pp 150
Kondo H, Ishimoto M (1972) Enzymatic formation of sulfite and acetate from sulfoacetaldehyde,
a degradation product of taurine. J Biochem 72:487489
Laue H, Denger K, Cook AM (1997) Fermentation of cysteate by a sulfate-reducing bacterium.
Lie TL, Leadbetter JR, Leadbetter ER (1998) Metabolism of sulfonic acids and other organosulfur
compounds by sulfate-reducing bacteria. Geomicrobiol J 15:135149
Mayer J, Denger K, Smits THM, Hollemeyer K, Groth U, Cook AM (2006) N-Acetyltaurine
dissimilated via taurine by Delftia acidovorans NAT. Arch Microbiol 186:6167
ONeil MJ (2001) International nonproprietary names (INN) for radicals and groups proposed for
pharmaceutical substances by the World Health Organization. In: The Merck index. Merck,
Whitehorse Station
Ovenden SPB, Capon RJ (1999) Echinosulfonic acids A-C and echinosulfone A: novel bromoindole
sulfonic acids and a sulfone from a southern Australian marine sponge, Echinodictyum. J Nat
Prod 62:12461249

Becker I, Amann J, Gellner K, Teeling H, Leuschner WD, Glckner F-O, Lupas AN, Amann
R, Klenk H-P (2004) The genome of Desulfotalea psychrophila, a sulfate-reducing bacterium
Reichenbecher W, Kelly DP, Murrell JC (1999) Desulfonation of propanesulfonic acid by
Comamonas acidovorans strain P53: evidence for an alkanesulfonate sulfonatase and an atypical
sulfite dehydrogenase. Arch Microbiol 172:387392
Rein U, Gueta R, Denger K, Ruff J, Hollemeyer K, Cook AM (2005) Dissimilation of cysteate via
3-sulfolactate sulfo-lyase and a sulfate exporter in Paracoccus pantotrophus NKNCYSA.
Ruff J, Denger K, Cook AM (2003) Sulphoacetaldehyde acetyltransferase yields acetyl phos-
phate: purification from Alcaligenes defragrans and gene clusters in taurine degradation.
Sanda S, Leustek T, Theisen MJ, Garavito RM, Benning C (2001) Recombinant Arabidopsis
SQD1 converts UDP-glucose and sulfite to the sulfolipid head group precursor UDP-
sulfoquinovose in vitro. J Biol Chem 276:39413946.
Schleheck D, Knepper TP, Fischer K, Cook AM (2004) Mineralization of individual congeners of
linear alkylbenzenesulfonate (LAS) by defined pairs of heterotrophic bacteria. Appl Environ
Stipanuk MH (2004) Sulfur amino acid metabolism: pathways for production and removal of
homocysteine and cysteine. Annu Rev Nutr 24:539577
Suzuki T, Wada T, Saigo K, Watanabe K (2002) Taurine as a constituent of mitochondrial tRNAs:
new insights into the functions of taurine and human mitochondrial diseases. EMBO J
21:65816589
Tiedemann F, Gmelin L (1827) Einige neue Bestandtheile der Galle des Ochsen. Ann Phys Chem
9:326337
Vairavamurthy A, Zhou W, Eglinton T, Manowitz B (1994) Sulfonates: a new class of organic
sulfur compounds in marine sediments. Geochim Cosmochim Acta 58:46814687
Vollrath F, Fairbrother WJ, Williams RJP, Tillinghast EK, Bernstein DT, Gallagher KS, Townley
MA (1990) Compounds in the droplets of the orb spiders viscid spiral. Nature 345:526528
Weinitschke S, Denger K, Smits TMH, Hollemeyer K, Cook AM (2006) The sulfonated osmolyte
N-methyltaurine is dissimilated by Alcaligenes faecalis and by Paracoccus versutus with
release of methylamine. Microbiology 152:11791186
Weinstein CL, Griffith OW (1988) Cysteinesulfonate and -sulfopyruvate metabolism. Partitioning
between decarboxylation, transamination, and reduction pathways. J Biol Chem
263:37353743
White RH (1984) Biosynthesis of the sulfonolipid 2-amino-3-hydroxy-15-methylhexadecane-
1-sulfonic acid in the gliding bacterium Cytophaga johnsonae. J Bacteriol 159:4246
White RH (1986) Intermediates in the biosynthesis of coenzyme M (2-mercaptoethanesulfonic
acid). Biochemistry 25:53045308
Yancey PH, Blake WR, Conley J (2002) Unusual organic osmolytes in deep-sea animals: adapta-
tions to hydrostatic pressure and other perturbants. Comp Biochem Physiol A Mol Integr
Physiol 133:667676
Chapter 15
Oxidation of Sulfur and Inorganic
Sulfur Compounds in Acidianus ambivalens
Arnulf Kletzin
Abstract Mechanisms of archaeal sulfur and inorganic sulfur compound oxidation

were almost exclusively studied in Acidianus species, extremely thermophilic and
acidophilic (pHopt 23), coccoid microorganisms living in acidic volcanic envi-
ronments (solfataras) worldwide. They utilize H2, H2S, S0, polythionates, and
metal sulfides as the most important sources of metabolic energy for CO2 fixation
during aerobic growth. The sulfur oxidation pathways include a soluble sulfur oxy-
genase reductase (SOR), membrane-bound thiosulfate and sulfite oxidoreductases,
a soluble tetrathionate hydrolase, and an oxidative adenosine 5-phosphosulfate
reductase pathway. Here, the current knowledge of the biochemistry of these
enzymes is discussed with a special focus on the implications of the recently
published 3D structure of the SOR.
15.1 Introduction
Oxidation and reduction of elemental sulfur (S0) and inorganic sulfur compounds
(ISCs) for energy conservation is a common property of (hyper-) thermophilic
Archaea. This is not surprising given the abundance of ISCs in volcanic environ-
ments. Most cultivated isolates are anaerobes and thrive by reduction of S0 with
inorganic gases or organic nutrients as electron donors (Schnheit and Schfer
1995; Kletzin 2007). Other Archaea found predominantly in acidic hydrothermal
environments oxidize S0 and ISCs aerobically to sulfuric acid (Huber and
Prangishvili 2005; Kletzin 2006). Most of these isolates belong to the Sulfolobales
order within the Crenarchaeota kingdom: they comprise the strictly aerobic genera
Sulfolobus and Metallosphaera, the strictly anaerobic Stygiolobus, and the faculta-
tively anaerobic Acidianus and Sulfurisphaera (the sixth genus, Sulfurococcus, is
probably lost) (Huber and Prangishvili 2005). Sulfolobales inhabit solfataras,
which are small, steam-heated pools of boiling surface water or mud, named after
the Solfatara caldera near Naples, Italy (Fig. 15.1). Optimal growth conditions are
pH 23 and 6592C in the laboratory, whereas the in situ temperatures are typi-
cally at the ambient boiling point. Members of Sulfolobales also contribute to
bioleaching of base and precious metals and to the formation of acidic drainage
184
Springer 2008
15 Oxidation of Sulfur and Inorganic Sulfur Compounds 185
Fig. 15.1 Solfataric fumarole and hot spring at Caldeira Velha, So Miguel, Aores, Portugal.
The dark color is the result of plant debris dropping into the solfatara from the surrounding area.
The fumarole was the source of Stygiolobus acoricus, an obligatory anaerobic Crenarchaeote of
the Sulfolobales order (Segerer et al. 1991). (Photo, Arnulf Kletzin)
downstream of mines and self-heating slug heaps (Huber and Prangishvili 2005).
In contrast, bacterial sulfur oxidizers are physiologically and phylogenetically
diverse and include anaerobic, phototrophic as well as aerobic, chemolithoau-
totrophic or mixotrophic bacteria.
Several members of the Sulfolobales developed into archaeal model organisms,
especially Sulfolobus acidocaldarius, S. solfataricus, S. tokodaii, and Acidianus
ambivalens. S. solfataricus and S. acidocaldarius grow on various organic sub-
strates. S. acidocaldarius, the first hyperthermophile to be isolated, was originally
described as a facultatively autotrophic sulfur oxidizer (Brock et al. 1972).
However, the strain presently available in culture collections (DSM 639) is not able
to do so anymore (Norris and Johnson 1998). It is assumed that the original
cultures, which were isolated by successive rounds of serial dilution, had not been
strictly pure but consisted of a mixture of microscopically indistinguishable het-
erotrophic and autotrophic strains. Todays type strains were probably grown
from single colonies after plating techniques for hyperthermophiles had been
improved.
Acidianus species and especially A. brierleyi are metabolically more versatile.
They gain energy by autotrophic sulfur or hydrogen oxidation with air or S0 as
electron acceptors (Huber et al. 1992). A. brierleyi has been shown to grow by aero-
bic chemolithotrophic sulfur and hydrogen oxidation, by oxidation of pyritic metal
ores, or by anaerobic sulfur reduction (Huber and Prangishvili 2005). It also grows
heterotrophically on organic substrates with or without sulfur and even by anaero-
186 A. Kletzin
bic molybdate respiration (Brierley and Brierley 1982). The biochemistry of

archaeal sulfur metabolism was most thoroughly studied in A. ambivalens and in
the obligatorly anaerobic heterotrophic euryarchaeote Pyrococcus furiosus (Kletzin
2006). A. ambivalens has the advantage that it is facultativly anaerobic and
can grow both by hydrogen and sulfur oxidation with sulfur and oxygen as electron
acceptors, respectively. A. ambivalens is also part of the aerobic bioleaching com-
munity used for biohydrometallurgical metal recovery from low-grade ores.
A more general overview of the biochemistry of dissimilatory sulfur oxidation
and reduction in Archaea was given recently in Kletzin (2006). This contribution
will summarize advances of the physiology and especially the biochemistry of
sulfur and ISC oxidation in Acidianus species in more detail, focusing on
A. ambivalens and especially on its sulfur oxygenase reductase (SOR).
15.2 Sulfur and Sulfur Oxidation
Sulfur is the 14th most abundant element in the earths crust. The bulk of the sulfur
deposits are found as sulfidic metal ores or as sulfate sediments (Middelburg 2000).
A significant amount of ISCs gets into the circulation owing to volcanic activity. S0 and
ISCs are prevalent in hydrothermal exhalations and can amount up to 10% of the
dry volume.
Sulfur is an element with a complex inorganic chemistry. S0 is almost insoluble in
water (5 g l1 at 25C, solubilities at higher temperatures are unknown) (Boulegue
1978). H2S, polysulfides, metal sulfides (MeS and MeS2), S0, and the sulfur oxy-
anions sulfite, thiosulfate, polythionates, and sulfate are the biologically relevant sul-
fur species (Roy and Trudinger 1970). Sulfur compounds have the tendency to form
homoatomic chains and rings reactly with each other easily. Thus, many ISCs will
react rapidly at elevated temperatures to form the thermodynamically most stable
product under the given conditions (Steudel 2000).
The oxidation of S0 to sulfuric acid proceeds in several steps and involves inter-
mediates like sulfite, thiosulfate, tetrathionate, and even sulfide. Several pathways
are distinguished depending on the organisms, the environment, and the pH of the
medium (reviewed in Takakuwa 1992; Kelly et al. 1997; Friedrich et al. 2005;
Kletzin 2006). The Sox complex is currently the best-understood ISC-oxidizing
enzyme system. It is found in the periplasm of chemolithotrophic aerobic or
phototrophic anaerobic members of the Bacteria growing at more or less neutral
pH. Its composition is modular: at least eight polypeptides collaborate to oxidize
most ISCs in an oxygen-independent way with cytochrome c as an electron acceptor
and without formation of free intermediates (Friedrich et al. 2001, 2005).
Sox complexes or genes thereof are neither found in Archaea nor in acidophilic S0
or ISC-oxidizing Bacteria. These microorganisms, regardless of whether they are
mesophiles or (hyper-) thermophiles, possess an array of different enzymes. A coher-
ent model of ISC oxidation in acidophiles comparable to that of the Sox complex is
lacking. S0 is oxidized by a cytoplasmic SOR, a remarkable enzyme that catalyzes an
oxygen-dependent sulfur disproportionation reaction (Emmel et al. 1986; Kletzin

1989; He et al. 2000; Fig. 15.2). In contrast, mesophilic and acidophilic bioleaching
Bacteria like Acidithiobacillus thiooxidans possess a periplasmic sulfur oxygenase
instead (Rohwerder and Sand 2003). Molecular details are available only for the
Fig. 15.2 Hypothetical model of S0 oxidation in Acidianus ambivalens and of the reaction
mechanism of the sulfur oxygenase reductase (SOR). a Enzymes, enzyme locations and activities,
and possible nonenzymic reactions (not stoichiometric) in Acidianus ambivalens. b Hypothetical
reaction mechanism of the SOR. CM cytoplasmic membrane, SAOR sulfite:acceptor oxidoreduct-
ase, SQR sulfide:quinone oxidoreductase, TQO thiosulfate:quinone oxidoreductase, CQ caldariella
quinone, TTH tetrathionate hydrolase; APS adenosine 5-phosphosulfate, APSR adenosine
5-phosphosulfate reductase, APAT adenosine 5-phosphosulfate:phosphate adenylyltransferase,
AK adenylate kinase, straight arrows enzyme reactions, dotted arrows nonenzymic reactions
188 A. Kletzin
SORs from several hyperthermophiles but not for the Acidithiobacillus sulfur
oxygenases. The oxidation products of the Acidianus SOR are utilized by other oxi-
doreductases, including sulfide:quinone oxidoreductase (SQR), tetrathionate-forming
thiosulfate oxidoreductase (membrane-bound), sulfite oxidoreductase (membrane-
bound), and tetrathionate hydrolases (TTH; soluble) (Fig. 15.2a).
To unravel the pathways and mechanisms of sulfur oxidation in acidophilic
Archaea and to fill in some of these gaps, we purified and characterized several of
the dissimilatory sulfur enzymes from A. ambivalens in the last few years (Fig.
15.2a) and established the role of A. ambivalens as the model organism for sulfur
oxidation in thermoacidophilic Archaea.
15.3 A. ambivalens and A. tengchongensis SORs
The initial enzyme in the archaeal S0 oxidation pathway is unique in several aspects.
The SOR catalyzes an oxygen-dependent sulfur disproportionation reaction to
sulfite, thiosulfate, and hydrogen sulfide in a 1:1 stoichiometry of the oxidized and
reduced products. SOR activity is measured under aerobic conditions using finely
dispersed sulfur in a detergent-containing reaction buffer. The enzyme does not
require external cofactors for activity (Kletzin 1989; He et al. 2000) (Eq. 15.1):
4S0 + O2 + 4H 2 O 2HSO3 + 2H 2 S + 2H + ( sum ) . (15.1)
Thiosulfate formation is probably the result of nonenzymic sulfite condensation

with excess S0 (Eq. 15.2):
S0 + 2HSO3
pH 6

S2 O3 + H + ( thiosulfate formation ) . (15.2)
pH 4
It is not known whether thiosulfate is a primary product of the SOR or

whether it is always formed nonenzymically. The ratio between the observed
thiosulfate and sulfite production is pH-dependent and temperature-dependent;
the thiosulfate fraction will increase with temperature and/or pH (Kletzin 1989,
and unpublished data). This observation suggests a nonenzymic thiosulfate
formation and argues against a reaction mechanism that would include direct
thiosulfate formation by the enzyme.
The SOR reaction (Eq. 15.1) can be formally divided into two partial reactions:
an oxygenase and a disproportionation reaction (Eqs. 15.3, 15.4):
S0 + O2 + H2 O HSO3 + H + ( oxygenase ) (15.3)
and
4S0 + O2 + 4H 2 O 2HSO3 + 2H 2 S + 2H + ( sum ) . (15.4)

Equation 15.4 is identical to the hydroxyl-catalyzed sulfur disproportionation

that occurs at alkaline pH and at elevated temperatures (Roy and Trudinger 1970;
Kletzin 1989). In contrast, the glutathione-dependent sulfur oxygenase of mes-
ophilic Acidithiobacillus species does not have a reductase or disproportionase
activity.
KM and Kcat values of the A. ambivalens SOR are not very favorable (23 mM and
2.2 s1, respectively; Urich et al. 2004); however, they are understandable given the
complex 3D structure (Sect. 15.3.1) and the poor solubility of the substrate.
The SOR activity is inhibited by thiol-binding reagents, pointing to the involvement
of cysteines in catalysis (Kletzin 1989; Urich et al. 2004; Chen et al. 2005). Three
conserved cysteine residues are present in the various SOR sequences (Urich et al.
2004; Fig. 15.3). Site-directed mutagenesis showed that only one of these (Cys31
in A. ambivalens numbering) is indispensable and cannot be replaced by alanine or
serine, while mutagenesis of the other two cysteines resulted in reduced activities
(Chen et al. 2005; Urich et al. 2005). Cys101/Cys104 double mutants retained up
to 30% of wild type activity, thereby confirming that these residues are not
essential (Fig. 15.6b; Urich et al. 2005).
Electron paramagnetic resonance spectroscopy and redox titration showed that the
isolated A. ambivalens SOR contains a mononuclear non-heme iron center in the high-
spin Fe3+ state. The center has an uncommonly low reduction potential (E0 = 268 mV,
protein as isolated). The signal disappeared upon reduction with dithionate or incuba-
tion of the SOR with S0 at elevated temperature (Urich et al. 2004). It was intriguing
to find that the reduction potential was more than 300 mV lower than usually found for
this type of iron center and that it was low enough to explain the S0 reducing activity
of the enzyme [E0 (H2S/S0) = 270 mV; Thauer et al. 1977].
A SOR characterized from the related strain A. tengchongensis was highly similar
(88% identity; Fig. 15.3). A third enzyme with similar properties was also described
from a phylogenetically uncharacterized S. brierleyi isolate (the isolate was most
probably an Acidianus species); however, a reductase activity was not reported (Table
15.1). Other sor genes were identified in the genomes of the crenarchaeote S. tokodaii,
of the Euryarchaeota Ferroplasma acidarmanus and Picrophilus torridus, and of the
hyperthermophilic bacterium Aquifex aeolicus. They shared 3569% identical residues
with the two Acidianus enzymes (Fig. 15.3). Additional sor genes were identified in
community studies of a commercial gold bioleaching plant recently (Chen et al. 2007).
One of the deduced SOR sequences was identical to that of Ferroplasma acidarmanus
(SORFA), another (SORSB) was only approximately 50% identical to the archaeal
sequences and belonged to a novel mesophilic Acidithiobacillus SM-1 strain isolated
from the bioreactor at an enrichment temperature of 45C. The third sequence, SORSA,
was similar but not identical to the SM-1 sequence (75% identity). When the SORSB
gene was expressed in Escherichia coli, the resulting protein showed the typical SOR
reaction at an optimal temperature of 7580C. It was speculated that the bacterium
acquired the sor gene from a hyperthermophile recently (Chen et al. 2007).
The SORs constitute a unique protein family without similarities to other families
(Pfam 07682; Wellcome Trust Sanger Institute 2006; Urich et al. 2004). Interestingly,
190 A. Kletzin
Fig. 15.3 Multiple alignment of SOR sequences available in public databases with details from
the Acidianus ambivalens 3D structure. A1A9 and B1B8 -helices and -sheets, respectively;
#, conserved cysteine residues; +, iron-coordinating residues; gate-keeping phenylalanine
residues from the pore at the enzymes fourfold symmetry axis; , methionine residues at the
pore to the active-site pocket. Accession numbers as follows: Acidianus ambivalens, P29082;
Acidianus tengchongensis, AAK58572; Sulfolobus tokodaii, NP_377053; Ferroplasma acidar-
manus, ZP_00608922; Picrophilus torridus, YP_023579; Acidithiobacillus strain SM-1,
DQ480733 (Chen et al. 2007); Aquifex aeolicus, AAC06723; uncultured bacterium with SOR-SA,
DQ480731, and DQ480732 (Chen et al. 2007). An in-frame stop codon was found immediately
upstream of the Ferroplasma acidarmanus Faci1674 open reading frame (X, position 49). It was
treated as an unknown residue, because of the similarity of the deduced amino acid sequences
of the upstream region to the other SOR sequences (Urich et al. 2004). It is not known
whether this represents a pseudogene or whether Ferroplasma acidarmanus produces active
SOR under suitable conditions. Similar, a reading frame shift at position 96 of the SORSA
sequence was corrected for the purpose of the alignment (Chen et al. 2007). (Extended from
Urich et al. 2004, 2006)
sor genes were missing in the S. solfataricus and S. acidocaldarius genomes, which
were originally described as facultative chemolithoautotrophic, sulfur-dependent
aerobes (Brock et al. 1972; Zillig et al. 1980). The observation is in accordance with
other reports that both strains cannot grow chemolithotrophically on S0 anymore.
15.3.1 SOR 3D Structure
Electron microscope preparations of A. tengchongensis cells treated with

immunogold-labeled anti-SOR antisera showed that the enzyme is most likely
associated with the cytoplasmic membrane (Chen et al. 2005). Hollow globular
Table 15.1 Properties of the sulfur oxygenase reductases (SORs) and sulfur oxygenase
Acidianus Acidithiobacillus
ambivalens Sulfolobus brierleyi Acidianus tengchongensis sp. strain SM-1
Recombinant SOR Recombinant SOR Recombinant SOR
Source Wild-type SOR (Escherichia coli)a Wild-type sulfur oxygenase (Escherichia coli) (Escherichia coli)
Subunit molecular mass 35,187b 36,311b 35,000 35,172b 34,491
c c
Holoenzyme molecular mass 844,488 871,464 550,000 NR NR
pHopt/pH range 77.4/48 NR 6.57.5 /NR 5/3.59 7.5/NR
Topt/Tmax 85C/108C NR 65C/>80C 70C/>90C 7580C/NR
Specific oxygenase activity at 10.6 2.8 0.9 753f 3.76
optimal temperatured (U mg1) 29.7
Specific reductase activity at 2.6 0.66 NR 45.2f NR
15 Oxidation of Sulfur and Inorganic Sulfur Compounds
optimal temperaturee (U mg1) 3.3

Reference (Kletzin 1989) (Urich et al. 2004) (Emmel et al. 1986) (He et al. 2000) (Chen et al. 2007)
NR not reported.
a
Including ten amino acid C-terminal Streptag.
b
From a sequence without N-terminal methionine.
c
From sequence and X-ray crystallography without N-terminal methionine.
d
1 U was defined as 1 mol of sulfite plus thiosulfate formed per minute assuming that thiosulfate is formed nonenzymically.
e
1 U was defined as 1 mol of H2S per minute.
f
Wild-type and recombinant enzymes.
191
192 A. Kletzin
Fig 15.4 Structural model of the SOR holoenzyme and holoenzyme assembly viewed from the
noncrystallographic fourfold symmetry axis. a Secondary structure model of the SOR holoen-
zyme. Orange -helices, green -sheets, gray coils, blue spheres iron atoms. b Molecular surface
representation of the holoenzyme. Gray carbon, blue nitrogen, red oxygen c Cross section of the
holoenzyme showing the interior large cavity and the subunits. The trapezoid denotes the dimer
shown in d and in Fig. 15.5b; the arrows denote the entrances to the active-site pockets. d Model
of holoenzyme assembly via homodimers. The figure was prepared with PyMOL (DeLano 2002).
(After Urich et al. 2006)
particles of 15.5 nm in diameter appeared in electron microscope pictures of the

purified A. ambivalens SOR (Kletzin 1989; Urich et al. 2004). X-ray crystallo-
graphic analysis to 1.7- resolution showed that the SOR subunits assemble to
a spherical homoicosatetramer (i.e., 24 subunits) with 432 point group symme-
try and an external diameter of 150 (Fig. 15.4a). The subunits surround an
empty cavity with a diameter of 71107 (Urich et al. 2006; Fig. 15.4c). The
resulting molecular mass of 844 kDa for the native SOR (871 kDa for the
recombinant enzyme; Table 15.1) was higher than anticipated from biochemical
analyses (550730 kDa; Urich et al. 2004, 2006). Narrow pores at the fourfold
symmetry axes provide an entrance to the cavity (Figs. 15.4c, 15.5a). Two rings
of four phenylalanine side chains each close the pores in the present 3D model
(Fig. 15.5a). The SOR thus provides an enclosed reaction and/or storage compartment
Fig. 15.5 Structural details of the SOR. a Model of the pore at the crystallographic
fourfold axis of the SOR viewed from the side; one of the four subunits that form the pore
was removed for a better view of the interior (Urich et al. 2006). The two rings of four
phenylalanine residues each are highlighted in a ball-and-stick model (F132 and F140).
b Model of a SOR homodimer in surface (upper part) and secondary structure representation
(lower part) (Urich et al. 2006). The iron atom is shown as a sphere; the arrows denote the
entrance pore to the active-site pocket. Two neighboring methionine residues contribute
significantly to pore formation (Met296/Met297, yellow in the upper part and sticks in the
lower part of the model). c Coordination of the iron (dashed gray lines) and potential
hydrogen-bonding network (dashed blue lines) around the iron site; red spheres ordered
water molecules; given are all ON or H 2OO distances below 3.1 . The figure was
prepared with PyMOL (DeLano 2002)
physically separated from the cytoplasm (Fig. 15.4c). Reversible denaturation

experiments had suggested that the holoenzyme assembles via homodimeric
building blocks (Urich et al. 2004). The 3D structure showed that each subunit
contacts five neighboring subunits directly; however, one of these contacts has
a significantly higher intersubunit contact area compared with the rest (16%
vs. 8% or less; Urich et al. 2006; Fig. 15.4d), thus supporting the conclusions
drawn from biochemical work.
The 24 Fe sites are well separated from each other (minimal distance 38 ).
They are not expected to interact during catalysis. They reside inside a pocket
in the interior of each subunit and are accessible through narrow entrance pores
from the internal cavity only (Urich et al. 2006; Figs. 15.4c, 15.6a); thus, sulfur
194 A. Kletzin
and the reaction products both have to pass two bottlenecks restricting access
and exit, respectively. These constrictions could contribute to the high KM and
the low K cat values observed. The narrow pores also suggest that the actual
substrate might be a linear sulfur species (e.g., polysulfides) and not the circular
-S8 ring.
15.3.2 SOR Subunit and Active-Site Structure
Each subunit consists of a -barrel core surrounded by -helices (Urich et al. 2006;
Fig. 15.5b). The spacious active-site pocket (18 18 6 ) is located outside
the barrel and is lined by at least 21 amino acid residues, including the three con-
served cysteines. Residue Cys31 showed additional electron density, which proved
to be a persulfide modification (Css; Fig. 15.6).
The iron, located at the far end of the pocket (Fig. 15.6a), is coordinated in a
structural motif known as 2-His 1-carboxylate facial triad. Two histidine ligands,
a bidentate glutamate, and two water molecules complete the octahedral geometry
(Costas et al. 2004; Urich et al. 2006; Figs. 15.5c, 15.6a, b). Mutation of any of the
three iron ligands to alanine resulted in the loss of activity and iron-binding capa-
bilities, whereas replacement of the glutamate by aspartate resulted in some resid-
ual activity and concomitantly low iron occupancy (approximately 1%; Urich et al.
2005; Fig. 15.6b). It was concluded from the structural and the mutational analysis
that the iron site and Css31 constitute the core of the active site, whereas the roles
of the remaining cysteines are less well defined. The minimal ironcysteine dis-
tance is 7.8 (Cys101), whereas the distance to the Css31 is 8.9 (Fig. 15.6b).
We concluded therefore that sulfur is covalently bound to the persulfide moiety of
Css31 and that the linear enzyme-bound polysulfide chain aligned to the iron site
is the final substrate of the reaction. An interesting effect was observed when
Cys101 was mutated to serine: the activity and iron content of the enzyme dropped
to almost zero, showing that distant mutations can trigger effects on iron incorpora-
tion into the enzyme.
The low reduction potential of the iron site is probably the result of a surrounding
network of hydrogen bonds (Fig. 15.5c). Glu87 seems to be crucial for enzyme
activity since mutagenesis drastically alters not only the specific activity of the
enzyme but also the stoichiometry of the reaction products (K. Seyfarth and
A. Kletzin, unpublished results).
15.3.3 SOR Reaction Mechanism
Some conclusions regarding the reaction mechanism could be derived from properties
of other oxygenases with a mononuclear non-heme iron atom, from the structure,
and from the mutagenesis experiments. The oxygenase reaction (Eq. 15.3) requires
Fig. 15.6 The active site of the SOR and effect of mutations (Urich et al. 2006). a Surface rep-
resentation of the active-site pocket with the iron (magenta), the coordinating histidine and
glutamic acid residues, and the cysteine persulfide (sticks); arrow pocket entrance pore.
b Secondary structure representation of a subunit with iron (magenta) and with important resi-
dues. Dots Fe-coordinating water ligands; dashed lines FeS and SS distances. Mutations as
follows: zero activity; reduced activity; strongly reduced activity; increased activity.
The figure was prepared with PyMOL (DeLano 2002)
196 A. Kletzin
the iron to be in the +2 state for dioxygen binding (Fe3+ is unable to activate O2)
(Costas et al. 2004). Upon reduction, the octahedral coordination sphere of the iron
should change to five ligands in a yet unknown geometry. Dioxygen binding usu-
ally results in a reactive and short-lived Fe4+-peroxo intermediate poised to attack
the substrate (Fig. 15.2b). It is expected that the abundance of free electron pairs in
the sulfur chain of the SOR would immediately refill the electron gap of the Fe4+.
In contrast, the disproportionation reaction (Eq. 15.4) requires the action of a strong
nucleophile (e.g., OH) without the need for the presence of dioxygen. The ordered
water molecule (Wat127; Fig. 15.2c) is a good candidate for the missing
nucleophile.
The core active site of the SOR is thus composed of the iron site and the
modified Css31. Substrate entry has to proceed through the hydrophobic chan-
nels along the fourfold axes of the sphere and through the pore of the active site
(Figs. 15.5a, 15.6a). The presence of a persulfide suggests that S0 is covalently
bound to Css31, a process that is equally possible with sulfur and polysulfides
as substrates (Fig. 15.2b, reactions 1a and 1b). The linear polysulfide chain
aligns to the iron site, thereby replacing the remaining water ligand(s). A mixed
reaction follows starting with a hydrolytic release of hydrogen sulfide from the
chain, thereby forming a sulfino intermediate that is a strong reductant in itself.
A polarized water molecule or hydroxyl ion might provide the nucleophile
required for this attack.
The following sequence of events is less obvious. Oxygen could bind either
to the iron or to the sulfino group and get reduced to the peroxide state. The
peroxide is a strong oxidant that could attack the sulfur chain and release
sulfite. The scheme outlined here predicts that the enzyme is not necessarily a
dioxygenase but rather a monooxygenase. Previous work by Emmel et al.
(1986) showed moderate 18O incorporation from 18O2 into sulfite, supporting the
monooxygenase hypothesis.
15.4 Oxidation of Soluble Sulfur Compounds in Acidianus
15.4.1 Sulfite:Acceptor Oxidoreductase
Sulfite:acceptor oxidoreductase (SAOR) and other sulfite-oxidizing enzymes are

known from many (micro-) organisms (Kappler and Dahl 2001). Two pathways are
important for Acidianus:
1. SAOR activity as part of the Sox complex is found in many bacteria. SAORs
feed electrons typically via c-type cytochromes into the respiratory chain
(Friedrich et al. 2001, 2005). Few of these enzymes are membrane-bound
(Kappler and Dahl 2001).
2. An alternative sulfite oxidation pathway coupled to substrate-level phosphorylation
was identified in Thiobacillus denitrificans and in A. ambivalens. It involves the
indirect sulfite oxidation via an adenosine 5-phosphosulfate (APS) reductase
(Fig. 15.2a) and an adenosine 5-phosphosulfate:phosphate adenylyltransferase

(APAT) (Brser et al. 2000).
The activities of a SAOR (membrane fraction), APS reductase, APAT, and ade-
nylate kinase (soluble fraction) were demonstrated in A. ambivalens (Zimmermann
et al. 1999; Fig. 15.2a) showing that both sulfite oxidation pathways are realized.
ATP sulfurylase activity was not found. The enzymes have not yet been purified
and molecular details are not known.
15.4.2 Thiosulfate:Quinone Oxidoreductase
Periplasmic or soluble tetrathionate-forming thiosulfate oxidoreductases or

dehydrogenases were found in several Bacteria (Visser et al. 1997; Nakamura et
al. 2001). The proteins vary considerably in subunit composition, molecular mass,
and cofactor content. Some contain c-type hemes. Gene or protein sequences are
not known but the proteins do not appear to be similar to thiosulfate-oxidizing
moieties of the Sox complex.
In contrast, a membrane-bound tetrathionate-forming thiosulfate:quinone oxi-
doreductase (TQO) was isolated from aerobically grown A. ambivalens cells
(Mller et al. 2004). It reduced ferricyanide and decyl ubiquinone and used meth-
ylene blue as an electron donor during the reverse reaction. The protein contained
bound caldariella quinone. Optimal activity was observed at 85C and pH 5. The
102-kDa glycosylated holoenzyme consists of 28- and 16-kDa subunits, suggesting
an 22 stoichiometry. Oxygen electrode measurements showed an electron trans-
port from thiosulfate to molecular oxygen via the terminal heme copper quinol:
oxygen oxidoreductase. The TQO subunits were identical to DoxA and DoxD,
originally described as parts of the A. ambivalens terminal oxidase (Mller et al.
2004). Both enzymes were copurified in previous work (Purschke et al. 1997) and
might form a supercomplex in the membrane.
The recently isolated thiosulfate-oxidizing haloarchaeon Natronorubrum
sp. HG 1 (Sorokin et al. 2005) contained a membrane-associated tetrathionate
synthase or a thiosulfate:acceptor oxidoreductase, whose activity depended
specifically on elevated concentrations of Cl. Its function is similar to the
TQO described here but the protein has not been purified yet and the sequence
is not known.
15.4.3 Tetrathionate Hydrolase
The fate of the tetrathionate formed by the TQO is not well understood. We had
proposed that a thiosulfate/tetrathionate cycle might exist because tetrathionate is
unstable in the presence of strong reductants like H2S and sulfite and is reduced to
thiosulfate in vitro at high temperatures (Xu et al. 1998, 2000; Kletzin et al. 2004;
198 A. Kletzin
Fig. 15.2a). Such a cycle would feed electrons indirectly from the S0 disproportionation
into the quinone pool. We found, however, a tetrathionate hydrolase (TTH) activity
in tetrathionate-grown cells of A. ambivalens recently (F. Mller and A. Kletzin,
unpublished data). The enzyme produced sulfate, thiosulfate, and elemental sulfur
from tetrathionate but neither H2S nor sulfite and higher polythionates. A TTH was
also purified from Acidithiobacillus ferrooxidans (Kanao et al. 2006). The
homodimeric and membrane-bound protein produced S0, sulfate, and thiosulfate from
tetrathionate at an acidic pH optimum. N-terminal sequencing allowed the identifica-
tion of the gene and the deduced amino acid sequence from the genome, showing that
the protein belongs to a superfamily of pyrroloquinoline quinone-containing enzymes.
BLAST searches showed that the most similar homologues were not present in other
bacterial genomes but in S. tokodaii (three probably paralogous genes) and A. ambiv-
alens (two paralogs; Kanao et al. 2006; A. Kletzin, unpublished observation). These
genes might eventually encode the protein(s) with TTH activity. There were no
homologues in S. acidocaldarius and S. solfataricus, suggesting that this protein and
its genes are restricted to the true sulfur oxidizers.
15.4.4 Sulfide:Quinone Oxidoreductase
Sulfide:quinone oxidoreductases (SQRs) catalyzing the oxidation of hydrogen

sulfide with quinones as electron acceptors are widely distributed in the microbial
world. They are type II family flavoproteins similar to NADH:quinone oxidore-
ductases (NADH-OR). These enzymes typically consist of a single subunit and are
unable to pump protons during NADH oxidation (Gomes et al. 2001). Multiple
homologues of SQRs/NADH-OR genes are found in archaeal genomes, including
those of members of the Sulfolobales. The only type II NADH-OR known from
Archaea was purified and characterized from A. ambivalens. Sequence compari-
sons showed that the enzyme contains all the cysteine residues conserved in SQRs,
but it remains to be demonstrated that the enzyme has an SQR activity.
15.5 Conclusions
The recent advances in the biochemistry and structure resolution of the SOR allow
for the design of experiments to unravel the reaction mechanism of the enzyme and
eventually also the pathways of its assembly. Extensive mutagenesis experiments
will provide derivatives that can be analyzed with various spectroscopic methods.
Therefore there is hope that we will be able to get a handle on the difficult inorganic
sulfur biochemistry using the SOR as a model system.
Regardless of the recent advances in structure resolution and mutational
analysis, a puzzling question remains unanswered, namely, why a cytoplasmic and
not a periplasmic or membrane-bound enzyme is used for the initial step of sulfur
oxidation (Rohwerder and Sand 2003; Friedrich et al. 2005). One could hypothe-
size that the closed sphere allows the utilization of the high reactivity of the S0 dis-
proportionation at elevated temperature and near-neutral pH, so little activation
energy is required. The highly reactive products of the reaction are another problem
that could be solved elegantly. These could be liberated in a more controlled fashion
posing less danger for damage to other proteins.
The scheme of S0 oxidation pathways and of electron transport in A. ambivalens
presented in Fig. 15.2a is outlined from the presently known enzymes and enzyme
activities; however, it is tentative and many gaps are open. Sulfur is oxidized by the
SOR; the products hydrogen sulfide, thiosulfate, and sulfite are oxidized by mem-
brane-bound oxidoreductases. In addition, there is the APS oxidation pathway.
Unsolved questions include how the sulfur gets into the cell, how sulfur gets inside
the SOR, and how the products get out. Some of these gaps, however, will be hopefully
closed in the near future.
Acknowledgements. I wish to thank Felicitas Pfeifer for her support. This work was supported
by grants from the Deutsche Forschungsgemeinschaft (Kl885/3-1, Kl885/3-2, and Kl885/3-3).
References
Boulegue J (1978) Solublity of elemental sulfur in water at 298 K. Phosphorus Sulfur 5:127128
Brierley CL, Brierley JA (1982) Anaerobic reduction of molybdenum by Sulfolobus species.
Zentralbl Bakteriol Hyg I Abt Orig C 3:289294
Brock TD, Brock KM, Belly RT, Weiss RL (1972) Sulfolobus: A new genus of sulfur-oxidizing
bacteria living at low pH and high temperature. Arch Microbiol 84:5468
Brser T, Selmer T, Dahl C (2000) ADP sulfurylase from Thiobacillus denitrificans is an
adenosine 5-phosphosulfate:phosphate adenylyltransferase and belongs to a new family of
nucleotidyltransferases. J Biol Chem 275:16911698
Chen ZW, Jiang CY, She Q, Liu SJ, Zhou PJ (2005) Key role of cysteine residues in catalysis and
subcellular localization of sulfur oxygenase-reductase of Acidianus tengchongensis. Appl
Chen ZW, Liu YY, Wu JF, She Q, Jiang CY, Liu SJ (2007) Novel bacterial sulfur oxygenase
reductases from bioreactors treating gold-bearing concentrates. Appl Microbiol Biotechnol
74:688698
Costas M, Mehn MP, Jensen MP, Que L Jr (2004) Dioxygen activation at mononuclear nonheme
iron active sites: enzymes, models, and intermediates. Chem Rev 104:939986
DeLano WL (2002) The PyMOL molecular graphics system, version 0.97. DeLano Scientific, San
Carlos http://www.pymol.org
Emmel T, Sand W, Knig WA, Bock E (1986) Evidence for the existence of a sulfur oxygenase
in Sulfolobus brierleyi. J Gen Microbiol 132:34153420
Gomes CM, Baucleines TM, Teixeira H (2001) A new type-II NADH dehydrogenase from the
archaeon Acidianus ambivalens: characterization and in vitro reconstitution of the respiratory
chain. Bioenerg Biomember 33:18
200 A. Kletzin
He Z, Li Y, Zhou P, Liu S (2000) Cloning and heterologous expression of a sulfur oxygenase/

reductase gene from the thermoacidophilic archaeon Acidianus sp. S5 in Escherichia coli.
Huber G, Drobner E, Huber H, Stetter KO (1992) Growth by aerobic oxidation of molecular
hydrogen in archaea a metabolic property so far unknown for this domain. Syst Appl
Microbiol 15:502504
Huber H, Prangishvili D (2005) The order Sulfolobales. In: Dworkin M, Falkow S, Rosenberg
E, Schleifer K-H, Stackebrandt E (eds) The prokaryotes: an evolving electronic resource for
the microbiological community, 3rd edn, release 3.19. Springer, New York
Kanao T, Kamimura K, Sugio T (2006) Biochemical and genetic characterization of tetrathionate
hydrolase from iron-oxidizing bacterium Acidithiobacillus ferrooxidans. In: Querellou
J, Lamy C (eds) Extremophiles 2006. International Society for Extremophiles, Brest
Kletzin A (1989) Coupled enzymatic production of sulfite, thiosulfate, and hydrogen sulfide from
sulfur: purification and properties of a sulfur oxygenase reductase from the facultatively
anaerobic archaebacterium Desulfurolobus ambivalens. J Bacteriol 171:16381643
Kletzin A (2006) Metabolism of inorganic sulfur compounds in Archaea. In: Garrett RA, Klenk
H-P (eds) Archaea. Evolution, physiology, and molecular biology. Blackwell, Oxford,
pp 261274
Kletzin A (2007) General characteristics and important model organisms. In: Cavicchioli R (ed)
Archaea. Molecular and Cellular Biology. ASM-Press, Washington, pp 1492
Kletzin A, Urich T, Mller F, Bandeiras TM, Gomes CM (2004) Dissimilatory oxidation and
reduction of elemental sulfur in thermophilic archaea. J Bioenerg Biomembr 36:7791
Middelburg JJ (2000) The geochemical sulfur cycle. In: Lens PNL, Hulshoff PL (eds)
Environmental technologies to treat sulfur pollution. IWA, London, pp 3346
pathway of sulphur oxidation to dioxygen reduction: characterization of a novel membrane-
bound thiosulphate:quinone oxidoreductase. Mol Microbiol 53:11471160
Nakamura K, Nakamura M, Yoshikawa H, Amano Y (2001) Purification and properties of thio-
sulfate dehydrogenase from Acidithiobacillus thiooxidans JCM7814. Biosci Biotechnol
Biochem 65:102108
Norris PR, Johnson DB (1998) Acidophilic microorganisms. In: Horikoshi K, Grant WD (eds)
Extremophiles: microbial life in extreme environments. Wiley, New York, pp 133154
Purschke WG, Schmidt CL, Petersen A, Schfer G (1997) The terminal quinol oxidase of the
hyperthermophilic archaeon Acidianus ambivalens exhibits a novel subunit structure and gene
organization. J Bacteriol 179:13441353
Rohwerder T, Sand W (2003) The sulfane sulfur of persulfides is the actual substrate of the sulfur-
oxidizing enzymes from Acidithiobacillus and Acidiphilium spp. Microbiology
149:16991710
Roy AB, Trudinger PA (1970) The chemistry of some sulphur compounds. In: The biochemistry
of inorganic compounds of sulphur. Cambridge University Press, Cambridge, pp 729
Schnheit P, Schfer T (1995) Metabolism of hyperthermophiles. World J Microbiol Biotechnol
11:2657
Segerer AH, Trincone A, Gahrtz M, Stetter KO (1991) Stygiolobus azoricus gen. nov., sp. nov.
represents a novel genus of anaerobic, extremely thermoacidophilic archaebacteria of the order
Sulfolobales. Int J Syst Bacteriol 41:495501
Sorokin DY, Tourova TP, Muyzer G (2005) Oxidation of thiosulfate to tetrathionate by an haloar-
chaeon isolated from hypersaline habitat. Extremophiles 9:501504
Steudel R (2000) The chemical sulfur cycle. In: Lens PNL, Hulshoff PL (eds) Environmental
technologies to treat sulfur pollution. IWA, London, pp 132
Takakuwa S (1992) Biochemical aspects of microbial oxidation of sulfur compounds. In: Oae
S (ed) Organic sulfur chemistry: Biochemical aspects. CRC, Boca Raton, pp 143
Urich T, Bandeiras TM, Leal SS, Rachel R, Albrecht T, Zimmermann P, Scholz C, Teixeira
M, Gomes CM, Kletzin, A (2004) The sulphur oxygenase reductase from Acidianus ambivalens
is a multimeric protein containing a low-potential mononuclear non-haem iron centre.
Urich T, Kroke A, Bauer C, Seyfarth K, Reuff M, Kletzin A (2005) Identification of core active
site residues of the sulfur oxygenase reductase from Acidianus ambivalens by site-directed
mutagenesis. FEMS Microbiol Lett 248:171176
Urich T, Gomes CM, Kletzin A, Frazo C (2006) X-ray structure of a self-compartmentalizing
sulfur cycle metalloenzyme. Science 311:9961000
Visser JM, de Jong GAH, Robertson LA, Kuenen JG (1997) Purification and characterization of
a periplasmic thiosulfate dehydrogenase from the obligately autotrophic Thiobacillus sp. W5.
Wellcome Trust Sanger Institute (2006) Pfam. http://www.sanger.ac.uk/pfam
Xu Y, Schoonen MAA, Nordstrom DK, Cunningham KM, Ball JW (1998) Sulfur geochemistry
of hydrothermal waters in Yellowstone National Park: I. The origin of thiosulfate in hot spring
waters. Geochim Cosmochim Acta 62:37293743
Xu Y, Schoonen MAA, Nordstrom DK, Cunningham KM, Ball JW (2000) Sulfur geochemistry
of hydrothermal waters in Yellowstone National Park, Wyoming, USA. II. Formation and
decomposition of thiosulfate and polythionate in Cinder Pool. J Volcanol Geotherm Res
97:407423
Zillig W, Stetter KO, Wunderl S, Schulz W, Priess H, Scholz I (1980) The Sulfolobus-Caldariella
group: Taxonomy on the basis of the structure of DNA-dependent RNA polymerases. Arch
Zimmermann P, Laska S, Kletzin A (1999) Two modes of sulfite oxidation in the extremely
thermophilic and acidophilic archaeon Acidianus ambivalens. Arch Microbiol 172:7682
Chapter 16
A Novel Coenzyme F420 Dependent Sulfite
Reductase and a Small Sulfite Reductase
in Methanogenic Archaea
Eric F. Johnson, Biswarup Mukhopadhyay
Abstract Recently a novel, highly active, coenzyme F420 dependent sulfite

reductase (Fsr) has been discovered in Methanocaldococcus jannaschii. Three
other extremophilic methanogens and an uncultured archaeon from a consor-
tium performing anaerobic oxidation of methane (AOM) carry Fsr homologs.
Methanogens require sulfide and most are sensitive to sulfite. Since Fsr is
induced by sulfite, reduces sulfite to sulfide with H2F420, and seems to be
associated with the membrane, it is a sulfite detoxification and assimilation
enzyme. The N-terminal half of Fsr is a homolog of H2F420 dehydrogenase
(FqoF/FpoF). FqoF/FpoF is the electron input unit of a membrane-bound
electron transport system of late-evolving methylotrophic methanogens and
Archaeoglobus fulgidus, a sulfate reducing archaeon employing the partial
reverse methanogenesis pathway. The C-terminal half (Fsr-C) represents a
dissimilatory sulfite reductase subunit (DsrA). While only four methanogens
carry Fsr, every methanogen carries a small putative sulfite reductase with
sequence features of Fsr-C. These observations lead to following hypotheses.
At one time methanogenesis and sulfate reduction involving a sulfite reductase,
two of the oldest energy-conserving respiratory metabolisms of Earth, existed
in one organism that performed sulfate reduction driven AOM. Fsr gave rise to
FqoF/FpoF and DsrA, or from a small sulfite reductase of methanogens DsrA
and Fsr (a fusion with FqoF/FpoF) evolved.
16.1 Introduction
Hydrogenotrophic methanogenesis and dissimilatory sulfate reduction (Eqs. 16. 1,

16.2; Fig. 16.1A), two of the oldest energy-conserving respiratory systems on
Earth, developed at least 2.7 billion to 3.2 billion and 3.7 billion years ago, respec-
tively (Leigh 2002; Teske et al. 2003). The recent discovery of a novel and highly
active sulfite reductase in Methanocaldococcus jannaschii (Johnson and
Mukhopadhyay 2005) and the presence of putative small sulfite reductases in
methanogens raise the question whether at one time these two ancient metabolisms
202
Springer 2008
16 A Novel Coenzyme F420 Dependent Sulfite Reductase 203
Fig. 16.1 Methanogenesis and sulfate reduction (A), energy metabolism of Archaeoglobus
fulgidus (B) and anaerobic methane oxidation (C). The site of inhibition of methanogenesis
by sulfite is shown. Pyruvate and lactate in B are examples of energy substrates for A. fulg-
idus; the methyl group of lactate or pyruvate enters the reverse methanogenesis pathway as
methyl- tetrahydromethanopterin (Mller-Zinkhan et al. 1989). The dotted line in c indicates
the involvement of two organisms. Methyl (CH3), methylcoenzyme; methylreductase,
methylcoenzyme M reductase.
existed in a single organism. This question is also relevant to anaerobic oxidation

of methane (AOM) (Boetius et al. 2000; Orphan et al. 2002; Eq. 16.3; Fig. 16.1C).
4H 2 + SO 4 2 + H + HS + 4H 2 O, G 0 = 152.2 kJ mol 1 SO 4 2
( Thauer et al. 1977 ) (16.1)
4H 2 + CO2 CH 4 + 2H 2 O, G = 131 kJ mol CH 4
0 1
( Thauer et al. 1977 ) (16.2)

CH 4 + SO 4 + H CO 2 + HS + 2H 2 O, G = 21 kJ mol CH 4
2 + 0 1
( Shima and Thauer 2005) (16.3)

HSO3 + 6e + 6H HS + 3H 2 O, E = 116 mV
+ 0
( Thauer et al. 1977 ) (16.4)

2
Sulfite (SO3 ) is an obligatory intermediate in the eight-electron reduction of
sulfate to sulfide (Fig. 16.1A). This strong nucleophile is toxic to cells of all types
owing to its reactivity toward proteins and sulfhydryl groups (Wedzicha 1992).
Therefore, it must be reduced rapidly to sulfide. Consequently, sulfite reductase,
204 E.F. Johnson and B. Mukhopadhyay
which catalyzes the six-electron reduction of sulfite (SO32) or bisulfite (HSO3)

(Eq. 16.4, Fig. 16.2AD), is a key enzyme in the reduction of sulfate. Two types
of sulfite reductases are known. Assimilatory sulfite reductases (Asr) generate
sulfide for the synthesis of cysteine, which in turn provides sulfur for all sulfur-
containing compounds in a cell (Crane and Getzoff 1996). Asr is used by certain
bacteria, fungi and plants. Dissimilatory sulfite reductases (Dsr) participate in
dissimilatory sulfate reduction, an energy-conserving process in a group of
anaerobic bacteria and archaea. In this metabolism, electrons derived from the
oxidation of complex materials such as carbohydrates and hydrocarbons or sim-
ple compounds such as acetate and H2 are transferred to sulfate (Widdel 1988).
Asr contains siroheme, an iron octacarboxylic tetrahydroporphyrin of the iso-
bacteriochlorin type (LeGall and Fauque 1988). Dsr contains siroamide, a siroheme
with one of the acetate chains amidated (Matthews et al. 1995; Lubbe et al. 2006;
Fig. 16.2). The Asr of Escherichia coli (Sir) is composed of two units: an octameric
flavoprotein component (SirFP) made up of 66-kDa flavin mononucleotide (FMN)
containing- and flavin adenine dinucleotide (FAD) containing subunits and a tetra-
meric heme-protein component (SirHP) composed of siroheme-containing 64-kDa
subunits (Crane and Getzoff 1996; Fig. 16.2A). SirFP derives electrons from
NADPH, a hydride donor or two-electron-restricted electron carrier, and then via
protein-bound FMN and FAD (two-electron/one-electron switch) it passes these
electrons to the [4Fe-4S]siroheme groups of SirHP. SirHP reduces sulfite to
sulfide. In the Asr of Arabidopsis thaliana, a monomeric SirHP-type unit reduces
sulfite with the electrons provided by a NADPH:ferredoxin reductase or photosystem
I via a ferredoxin (Nakayama et al. 2000; Fig. 16.2B). The Dsr enzymes are
tetramers of DsrA and DsrB subunits (Fig. 16.2C), which show low primary
sequence similarities to the E. coli and plant enzymes (Johnson and Mukhopadhyay
2005; Fig. 16.2A). DsrA carries siroheme (Crane and Getzoff 1996; Fig. 16.2C).
DsrA and DsrB share substantial sequence similarities with each other and are
believed to have originated from a gene duplication event (Dhillon et al. 2005).
DsrB lacks a conserved cysteine residue of the consensus siroheme-binding site,
and it probably does not carry the siroheme cofactor (Dahl et al. 1993; Crane and
Getzoff 1996). The electron-donating units for dissimilatory sulfite reductases are
Fig. 16.2 Assimilatory and dissimilatory sulfite reductases (AD) and A. fulgidus H2F420:quinone
oxidoreductase or Fqo complex (E). Most dissimilatory sulfite reductases are 22 proteins, but
222 structures have also been observed, where the function of the -subunit is unknown (Crane
and Getzoff 1996). The quaternary structure for Methanocaldococcus jannaschii coenzyme F420
dependent sulfite reductase (Fsr) is not known. The question mark in d indicates that it is not
known whether Fsr contains bound flavin. e Fqo and Fpo complexes (Deppenmeier 2004). These
are similar to respiratory complex I of Escherichia coli and mitochondria. Sir or SR, sulfite reduct-
ase; SirFP and SirHP, flavoprotein and hemoprotein subunits of sulfite reductase, Fd, ferredoxin;
PS I, photosystem I; Cyt, cytochrome; FAD, flavin adenine dinucleotide; FMN, flavin mononu-
cleotide; MQ, menaquinone; FqoF or FpoF, H2F420 dehydrogenase subunit. (Modified from
Johnson and Mukhopadhyay 2005)
yet to be clearly identified (Crane and Getzoff 1996). For Desulfovibrio desulfuricans
Dsr, a membrane-bound complex (DsrMKJOP) is a strong candidate for this role,
and homologs for the subunits of this complex are present in every sulfate-reducing
organism for which the genome sequence has been determined (Pires et al. 2006).
During anaerobic growth, Salmonella enterica expresses a small ssulfite reductase,
which has been named anaerobic sulfate reductase or Asr (Huang and Barrett
1991). Since this enzyme serves a dissimilatory and not an assimilatory function
(Huang and Barrett 1991), we rename the enzyme SalDsr. The siroheme-containing
subunit of this enzyme (SalDsrC) is a homolog of DsrA (Dhillon et al. 2005;
Johnson and Mukhopadhyay 2005).
16.2 Incompatibility of Methanogenesis and Sulfate Reduction,

Sulfite as the Key Determinant
If an organism has to perform both methanogenesis and dissimilatory sulfate reduc-

tion, it has to deal with the apparent incompatibility of these two metabolisms
(Fig. 16.1A). As mentioned in Sect. 16.1, sulfite, an intermediate in sulfate reduction,
is toxic to all types of cells. Methanogens have an additional target for sulfite. Sulfite
inhibits methanogenesis (Balderston and Payne 1976), the only means of energy pro-
duction for a methanogen (Wolfe 1992). This inhibition is most likely due to the fact
that sulfite reacts with and inactivates methylcoenzyme M reductase (Becker and
Ragsdale 1998; Mahlert et al. 2002; Fig. 16.1A). A similar incompatibility is expected
in sulfate-dependent AOM (Eq. 16.3), which combines reverse methanogenesis
(reverse of Eq. 16.2, involving methylcoenzyme M reductase) with dissimilatory
sulfate reduction (Hinrichs et al. 1999; Boetius et al. 2000; Shima and Thauer 2005;
Eq. 16.1; Fig. 16.1C). Perhaps for this reason AOM is accomplished through a com-
bined action of at least two organisms, one of which is an archaeon catalyzing methane
oxidation and the other is a sulfate-reducing bacterium (Hinrichs et al. 1999; Boetius
et al. 2000; Fig. 16.1C). This antagonism is also reflected in the metabolic properties
and genomic potential of Archaeoglobus fulgidus. This sulfate-reducing archaeon car-
ries most elements of the methanogenesis pathway except methylcoenzyme M reduct-
ase and associated genes (Klenk et al. 1997; Fig. 16.1B). However, it should be noted
that these concepts need to be justified by the Ki value of methylcoenzyme M reductase
for sulfite, which is yet to be determined.
16.3 Inevitable Exposure of a Methanogen to Sulfite

in Hydrothermal Vents and on Early Earth
The submarine hydrothermal vent methanogens receive nutrition from the vent fluid
(Jannasch 1989; McCollom and Shock 1997), which is rich in nutrients for autotrophic
growth. However, the temperature of the vent fluid (300350C) is detrimental to all
known life forms (Jannasch 1989). Cold seawater that permeates through the vent wall
brings the temperature of the vent fluid down to a level where hyperthermohilic
methanogens can grow (McCollom and Shock 1997). This process also brings oxygen
into the vent. Sulfide, which is present at a high level in the vent fluid (57 mM; Jannasch
1989), reacts with this oxygen and helps to establish anaerobic conditions that a strict
anaerobe, such as a methanogen, needs. On the other hand, this reaction between sulfide
and a low level of oxygen has the potential of producing sulfite, an incomplete oxidation
product of sulfide. Therefore, a deep-sea hydrothermal vent methanogen must be able to
tolerate sulfite. It has been suggested that the development of a fully oxic atmosphere on
sulfide-containing early Earth followed a protracted oxygenation period (Shen et al.
2003; Kah et al. 2004; Poulton et al. 2004). This early oxygenation event presented a situ-
ation similar to that described above for the hydrothermal vents and consequently caused
sulfite production and selection of methanogens with a sulfite detoxification ability.
16.4 Use of Sulfite As a Sulfur Source by Methanocaldococcus

jannaschii and Other Methanogens
Methanocaldococcus jannaschii, a deeply-rooted, hyperthermophilic, strictly

hydrogenotrophic, methanogenic archaeon, is an inhabitant of the deep-sea
hydrothermal vents (Jones et al. 1983; Boone et al. 1993). It tolerates sulfite up to a
level of 40 mM, where observable growth and methane formation occur (E.F. Johnson
and B. Mukhopadhyay, unpublished data). With 20 mM sulfite, growth and methanogenesis
is slow, but the final cell density is comparable to that obtained with the optimal sulfite
levels of 0.52 mM. Methanothermococcus thermolithotrophicus, Methanothermobacter
thermautotrophicus, and Methanothermobacter marburgensis, which are thermophiles,
can use sulfite as their sole sulfur source (Daniels et al. 1986); a concentration of 1 mM
for sulfite is optimal and higher levels are inhibitory. A complete inhibition of growth
occurs at 1.25 mM or above for Methanothermococcus the rmolitotrophicus and at 4 mM
or above for Methanothermobacter thermautotrophicus and Methanothermobacter
marburgensis. Methanosarcina acetivorans and Methanococcus maripaludis cannot
grow in the presence of 0.5 mM sulfite (E.F. Johnson and B. Mukhopadhyay, unpublished
data). Methanopyrus kandleri, Methanocaldococcus igneus, Methanococcus vannielli
and Methanoplanus limicola can tolerate sulfite (in the presence of sulfide) but cannot use
this oxyanion as their sole sulfur source (Rothe and Thomm 2000).
16.5 Expression of a Novel Coenzyme F420 Dependent Sulfite

Reductase in Methanocaldococcus jannaschii During
Growth on Sulfite
When grown with sulfite as the sole sulfur source, Methanocaldococcus jannaschii
expresses a 70-kDa polypeptide in a growth-phase-independent manner (Fig. 16.3B).
This polypeptide corresponds to open reading frame (ORF) MJ0870, and it is not
Fig. 16.3 Sodium dodecyl sulfate polyacrylamide gel elecrophoresis of Methanocaldococcus

jannaschii cell lysates. Cells were harvested from cultures grown with sulfide (S2) (A) or sulfite
(SO32) (B) at indicated ages (in hours). A lysate was prepared by boiling cells in a solution
containing 62.5 mM tris(hydroxymethyl)aminomethane hydrochloride buffer pH 6.8, 10% glyc-
erol, 5% 2-mercaptoethanol. The gels were stained with Coomassie blue. The overexpressed band
in the SO32 lane is MJ0870 or Fsr. a, b and g represent -, - and -subunits of methylcoenzyme
M reductase. M molecular mass standards
detectable in cells grown with sulfide (Fig. 16.3A). Although originally annotated
as the -subunit of a coenzyme F420 reducing hydrogenase (FrhB) (Bult et al. 1996),
MJ0870 encodes a coenzyme F420 dependent sulfite reductase (Fsr), a novel enzyme
(Sect. 16.6). Coenzyme F420 is an 8-hydroxy-5-deazariboflavin derivative, which is
found in the methanogens, certain sulfate-reducing archaea and actinomycetes
(DiMarco et al. 1990; Purwantini et al. 1997). Similar to the nicotinamide coen-
zymes, F420 is a hydride carrier and is restricted to two-electron transfer reactions
(DiMarco et al. 1990), and H2F420 (reduced F420) is a more potent reductant than
NAD(P)H (Eqs. 16.5, 16.6).
F420 + 2e + 2H + H 2 F420 , E 0 = 350 mV

( DiMarco et al. 1990 ) (16.5)
NAD ( P ) + 2e + 2H + NAD ( P ) H + H + , E = 320 mV
0
( Thauer et al. 1977 ) (16.6)
HSO3 + 3H 2 F420 HS + 3H 2 O + 3F420 , G 0 = 135 kJ mol 1 HSO3

( Johnson and Mukhopadhyay 2005) (16.7)
The reduction of sulfite to sulfide with H2F420 is exergonic (Eq. 16.7). Extracts of
Methanocaldococcus jannaschii cells grown with sulfite as the sulfur source oxidize
H2F420 with sulfite with a specific activity of 1.31.7 mol min1 mg1 protein, but
cells grown with sulfide lack this activity (Johnson and Mukhopadhyay 2005).
Methanocaldococcus jannaschii has the potential of expressing two F420-reducing
hydrogenases (Bult et al. 1996), which would supply H2F420 for the Fsr reaction.
16.6 Fsr, Combining Structural Components of Two Different

Dissimilatory Metabolic Machineries to Bring About a Sulfite
Reduction Function
The 620-residue MJ0870 polypeptide has two distinct domains (Fig. 16.2D):
1. Fsr-N (residues 1311): A primary structure analysis clearly identifies this part
of Fsr as a homolog of H2F420 dehydrogenase (FqoF/FpoF) (Johnson and
Mukhopadhyay 2005). All specialized sequence features of FqoF/FpoF are
present in Fsr-N. For example, in both sequence and location, C15C18C21C25P26
and C42H47C50C54P55 of MJ0870 correspond to two ferredoxin-type [Fe 4S4]
motifs of FqoF and FpoF. A phylogenetic analysis also identifies Fsr-N as an
FqoF homolog (Fig. 16.4A). FqoF is the electron-funneling unit of a membrane-
based energy transduction system, called the H2F420:quinone oxidoreductase (Fqo)
complex, found in A. fulgidus (Deppenmeier 2004; Fig. 16.2E). A. fulgidus gener-
ates H2F420 from the oxidation of methyl groups from substrates such as lactate
and pyruvate. This oxidation occurs via a partial reverse methanogenesis path-
way (Mller-Zinkhan et al. 1989; Fig. 16.1B). FqoF then oxidizes H2F420 and
introduces the electrons derived from this oxidation into the membrane-resident
Fqo complex (Deppenmeier 2004; Fig. 16.2E). The Fqo complex is similar to
Fig. 16.4 Phylogenetic tree for the N-terminal and C-terminal halves of Methanocaldococcus
jannaschii Fsr and its homologs. A, N-terminal half of Fsr or MJ0870 (residues 1311); B, C-terminal
half of Fsr or MJ0870 (residues 325620). The proteins are identified by NCBIs open reading
frame or accession numbers. Values near the branches are bootstrap confidence levels. Bar number
of substitutions per site. N, N-terminal half; C, C-terminal half; FrhB and FruB, -subunits of
F420-dependent hydrogenases; FdhB, -subunit of F420-dependent formate dehydrogenase; DsrA
and DsrB, subunits of dissimilatory sulfite reductase; SalDsrC, hemoprotein subunit of Salmonella
enterica anaerobic sulfite reductase; RCIX2197 and RCIX2692, putative assimilatory sulfite
reductases of an uncultured methanogenic archaeon from rice rhizosphere.
the bacterial and mitochondrial respiratory NADH dehydrogenase complex or

complex I (Deppenmeier 2004). Methanogens belonging to the genus
Methanosarcina, which are phylogenetically closely related to A. fulgidus
(Woese et al. 1990), carry out partial reverse methanogenesis and possess an
Fqo-type system called H2F420:phenazine oxidoreductase (Fpo). In this complex,
FpoF is the H2F420 dehydrogenase (Deppenmeier 2004). It seems that the Fqo/
Fpo system is specifically associated with the methyl group oxidation and
reverse methanogenesis pathway. Also, the genomes of Methanocaldococcus
jannaschii and Methanothermobacter thermautotrophicus lack readily identifi-
able FqoF/FpoF homologs; these hydrogenotrophs are more deeply rooted com-
pared with A. fulgidus and Methanosarcina (Woese et al. 1990). These
observations led to the assumption that the H2F420 dehydrogenase complex and
FpoF/FqoF are absent in strictly hydrogenotrohic methanogens (Baumer et al.
1998). We now know that an FpoF/FqoF homolog is indeed present in
Methanocaldococcus jannaschii and certain strictly hydrogenotrophic
methanogens as Fsr-N (Fig. 16.4A). Hence, Fsr-N is expected to retrieve
electrons from H2F420 (Fig. 16.2D).
2. Fsr-C (residues 325620): This part of Fsr is a dissimilatory sulfite reductase
homolog (Johnson and Mukhopadhyay 2005). Available data on sulfite
reductases (Crane and Getzoff 1996) suggest that in MJ0870 Arg355 and
Arg423 are involved in binding sulfite and C428C434C468C472 represents the
siroheme[Fe4-S4]-binding element. Phylogenetically, Fsr-C is most similar
to SalDsrC (Fig. 16.4B); therefore, Fsr-C is likely to house siroheme and
carry out sulfite reduction.
In combination, Fsr-N and Fsr-C create an enzyme that would retrieve electrons
from H2F420, perhaps via bound flavin, and transfer these electrons to the siroheme
via Fe4S4 clusters for use in the reduction of sulfite (Fig. 16.2D); hence, function-
ally Fsr is similar to E. coli Sir (Fig. 16.2A). One of the unique properties of Fsr is
that the electron harvesting and the sulfite reduction units are physically linked on
a single subunit (Fig. 16.2D); in other sulfite reductases these are represented by
two polypeptides (Fig. 2AC).
16.7 Purified Fsr Exhibits Properties Predicted

from the Primary Structure
Purified Fsr contains siroheme and converts sulfite to sulfide with H2F420 as the
reductant with an electron transfer rate of 2332 mol min1 mg1 (Johnson and
Mukhopadhyay 2005). The apparent Km values for sulfite and H2F420 are 12 and
21 M, respectively. Therefore, Fsr is a highly active enzyme with high affini-
ties for its substrates. Fsr oxidizes H2F420 with methylviologen (an artificial
one-electron carrier) and reduces sulfite to sulfide with reduced methylviologen,
which suggest the existence of one-electron paths between the retrieval of

reducing equivalents from H2F 420 and reduction of sulfite; such activities are
found in dissimilatory sulfite reductases as well (Lee et al. 1973; Dahl et al.
1993). The Fsr partial reactions transfer electrons at a rate of 90110 mol
min1 mg 1, which is 35 times faster than that of the complete H 2F420-utiliz-
ing reaction (Johnson and Mukhopadhyay 2005). The transfer of electrons
between two functional units might represent a rate-limiting step in the com-
plete reaction. Fsr cannot utilize NADH or NADPH in place of H2F 420 or thio-
sulfate and sulfate in place of sulfite. In the Fsr reaction, about 70% of the
reducing equivalents provided by H 2F 420 are recovered as sulfide. This gap
could be due to either an error in the measurement of sulfide or the production
of partially reduced compounds. Production of partially reduced compounds
such as trithionate and thiosulfate has been observed with certain sulfite
reductases (LeGall and Fauque 1988). For Fsr, the production of thiosulfate
as an intermediate is unlikely because the enzyme is unable to reduce
thiosulfate.
16.8 Fsr, a Sulfite Detoxification Tool

and an Assimilatory Enzyme
In cells receiving sulfite, Fsr and methylcoenzyme M reductase subunits are expressed
at comparable levels (Fig. 16.3b); the latter is a catabolic enzyme and represents up
to 30% of the cellular protein in a methanogen (Rouviere and Wolfe 1987; Thomas
et al. 1987). From the data presented in Sect. 16.7, it can be calculated that the
extracts of Methanocaldococcus jannaschii grown with sulfite reduce this oxyanion
with H2F420 at a rate of 2.73.7 mol min1 mg1 protein. With reduced methylviolo-
gen as the electron donor, this rate would be 8.118.5 mol min1 mg1 protein. With
A. fulgidus, where sulfite reductase is an energy-metabolism enzyme, cell extract
sulfite reductase activity as measured with methylviologen is 0.07 mol min1 mg1
protein (Dahl et al. 1994). This comparison shows that Methanocaldococcus jannaschii
Fsr behaves like a catabolic enzyme. Upon centrifugation at 160,000g, about 26% of
the cell extract Fsr activity is found in the pellet fraction and 65% in the denser sec-
tion of the supernatant (E.F. Johnson and B. Mukhopadhyay, unpublished prelimi-
nary data), suggesting a loose association of Fsr with the membrane. These
observations lead to the hypothesis that in response to an exposure to sulfite,
Methanocaldococcus jannaschii expresses Fsr at a high cellular level and places this
highly active enzyme near the membrane. This arrangement allows the organism to
convert sulfite to sulfide before it enters the cell and thereby to protect its methyl-
coenzyme M reductase from inactivation. Since Methanocaldococcus jannaschii
requires sulfide for growth (Jones et al. 1983), this detoxification process also yields
an essential nutrient and serves an assimilatory purpose. It should be noted that the
appearance of Fsr activity in the 160,000 g pellet fraction of cell extracts could be due
to an association of the enzyme with other soluble components of the cell, which
could create a rather large complex. If that were the case, coenzyme F420 reducing
hydrogenase, which generates H2F420, might belong to such a complex.
16.9 Homologs of Fsr in Other Organisms
Fsr homologs have been found in three methanogens and an uncultured archaeon (ORFs
MTH280, MK0779, Mbur_0619 and GZ27A8_52; Fig. 16.4). Since MTH280 is closely
related to MJ0870 (Fig. 16.4) and its host, Methanothermobacter thermautotrophicus, can
use sulfite as a sulfur source (Daniels et al. 1986), this ORF is likely to represent an Fsr
enzyme. Phylogenetically, MK0779 is a bit distant from both MJ0870 and MTH280 (Fig.
16.4). Also Methanopyrus kandleri cannot use sulfite as a sole sulfur source (Rothe and
Thomm 2000), but has the genomic potential for catalyzing nitrate to ammonia (Slesarev
et al. 2002). Therefore, MK0779 probably represents an F420-dependent nitrite reductase.
When Fsr was discovered, only three Fsr homologs (MJ0870, MTH280 and MK0799)
were known, and these belong to deeply rooted, thermophilic, strictly hydrogenotrophic
hosts (Johnson and Mukhopadhyay 2005). The identification of Mbur_0619 and
GZ27A8_52, which form a new homolog group (Fig. 16.4), extends the host range of this
novel enzyme to certain late-evolving and psychrophilic archaea that perform reverse
methanogenesis and live in a methane-rich environment. Methanococcoides burtonii,
which carries Mbur_0619, is an obligately methylotrophic methanogen belonging to the
Methanosarcinales (Franzmann et al. 1992). It was isolated from a methane-saturated
environment with permanent temperatures of 12C. The source of GZ27A8_52 is an
uncultured anaerobic-methanotrophic archaeon (Hallam et al. 2004). This organism is
peripherally related to the Methanosarcinales and a member of a cold seeps consortium
that performs reverse methanogenesis and sulfate-reduction-driven AOM. It is not known
whether these microorganisms tolerate sulfite and/or use it as a sulfur source.
16.10 Small Sulfite Reductases in Methanogens
A small siroheme sulfite reductase (subunit size, 23 kDa) has been isolated from
Methanosarcina barkeri (Moura et al. 1986). The physiological electron donor and
the in vivo role for this enzyme are not known. From a BLAST search of the respec-
tive genome using Fsr-C as the query we found that every methanogen carries at least
one ORF with the potential of encoding a small (22.437.2 kDa) siroheme sulfite
reductase (Fig. 16.4b). These ORFs are related to Fsr-C (Fig. 16.4b), but are not
linked to an Fsr-N unit. The previously isolated Methanosarcina barkeri sulfite
reductase most likely belongs to this group. Since many methanogens are sensitive
to sulfite (Sects. 16.2, 16.4), it is unlikely that the small sulfite reductases confer an
ability to tolerate or utilize externally supplied sulfite as their sulfur source. The possible
roles for these ORFs are discussed in the following section.
16.11 Conclusion and Hypotheses
It is possible that in addition to its assimilatory and detoxification roles, Fsr

allows a nonmethanogenic mode of energy generation via H2-dependent sulfite
or nitrite reduction in certain methanogens. This system could be a remnant or
a precursor of an ancient sulfate reduction pathway. Fsr could also be the
ancestor of FqoF/FpoF, the electron input unit of the energy-transduction sys-
tems in late-evolving archaea, and the subunits of the dissimilatory sulfite
reductases. If Fsr is simply a detoxification enzyme, what was its role before
the appearance of oxygen on Earth? One possible early role of this enzyme
is in the synthesis of coenzyme M, where sulfite is the proposed precursor
of a sulfonate group (Graham et al. 2002); coenzyme M is essential for
methanogenesis. Methanogens lack sulfate reduction enzymes and require
sulfide for growth. One way to generate the needed sulfite would be to oxidize
sulfide. In the highly reducing, anaerobic environment of early Earth this con-
version was definitely an endergonic process (Eq. 16.8):
HS + 3H 2 O HSO3 + 3H 2 , G 0 = +171.7 kJ mol 1 HSO3

( Thauer et al. 1977 ) (16.8)
It was probably driven by a membrane-associated Fsr that utilized a reverse

electron transport system. Also, a membrane-resident Fsr would have
prepared a methanogen for dealing with the toxicity of sulfite during the early
oxygenation of Earth. However, a limited distribution of Fsr genes and a wide
distribution of DsrAB genes (Stahl et al. 2002) question the hypothesized
ancestral nature of Fsr. On the other hand, it has to be considered that Fsr thus
far has been found in organisms that live in extreme habitats, which have not
been extensively explored. It is also equally possible that the need for sulfite
in coenzyme M biosynthesis was met by the small sulfite reductase that is
found in methanogens, and this enzyme in combination with FqoF gave rise
to Fsr. This hypothesis is reasonable because, unlike Fsr, the small sulfite
reductases are present in every methanogen and therefore must play a vital
role. These proteins are also closely related to Fsr-C (Fig. 16.4b). In all,
sulfite reduction is probably an ancient process in the methanogens. The
existence of Fsr homologs in methane-oxidizing archaea and methyl group
oxidizing methanogens raises the possibility that some of these organisms
may carry both the methanogenesis/reverse methanogenesis and sulfate
reduction machineries. It now seems more likely that sulfate reduction and
methanogenesis at one time existed in one organism.
Acknowledgement. We thank Endang Purwantini for discussions and help in phylogenetic

analysis, Christiane Dahl for a review of the manuscript and helpful suggestions, and Dwi Susanti,
Jason Rodriguez and Carol Volker for comments. This work was supported by NASA
Astrobiology: Exobiology and Evolutionary Biology grant NNG05GP24G to B.M.
References
Balderston WL, Payne WJ (1976) Inhibition of methanogenesis in salt marsh sediments and
whole-cell suspensions of methanogenic bacteria by nitrogen oxides. Appl Environ Microbiol
32:264269
Baumer S, Murakami E, Brodersen J, Gottschalk G, Ragsdale SW, Deppenmeier U (1998)
The F420H2:heterodisulfide oxidoreductase system from Methanosarcina species.
2-Hydroxyphenazine mediates electron transfer from F420H2 dehydrogenase to heterodisulfide
reductase. FEBS Lett 428:295298
Becker DF, Ragsdale SW (1998) Activation of methyl-SCoM reductase to high specific activity
after treatment of whole cells with sodium sulfide. Biochemistry 37:26392647
Boetius A, Ravenschlag K, Schubert CJ, Rickert D, Widdel F, Gieseke A, Amann R, Jorgensen
BB, Witte U, Pfannkuche O (2000) A marine microbial consortium apparently mediating
anaerobic oxidation of methane. Nature 407:623626
Boone DR, Whitman WB, Rouvire P (1993) Microbiology, diversity and taxonomy of
methanogens. In: Ferry JG (ed) Methanogenesis: ecology, physiology, biochemistry and genet-
ics. Chapman and Hall, New York, pp 3580
Bult CJ, White O, Olsen GJ, Zhou L, Fleischmann RD, Sutton GG, Blake JA, FitzGerald LM,
Clayton RA, Gocayne JD, Kerlavage AR, Dougherty BA, Tomb JF, Adams MD, Reich CI,
Overbeek R, Kirkness EF, Weinstock KG, Merrick JM, Glodek A, Scott JL, Geoghagen NSM,
Weidman JF, Fuhrmann JL, Nguyen D, Utterback TR, Kelley JM, Peterson JD, Sadow PW,
Hanna MC, Cotton MD, Roberts KM, Hurst MA, Kaine BP, Borodovsky M, Klenk H-P,
Frasher CM, Smith HO, Woese CR, Venter JC. (1996) Complete genome sequence of the
methanogenic archaeon, Methanococcus jannaschii. Science 273:10581073
Crane BR, Getzoff ED (1996) The relationship between structure and function for the sulfite
reductases. Curr Opin Struct Biol 6:744756
Dahl C, Kredich NM, Deutzmann R, Truper HG (1993) Dissimilatory sulphite reductase from
Archaeoglobus fulgidus: physico-chemical properties of the enzyme and cloning, sequencing
and analysis of the reductase genes. J Gen Microbiol 139(Pt 8):18171828
Dahl C, Speich N, Truper HG (1994) Enzymology and molecular biology of sulfate reduction in
extremely thermophilic archaeon Archaeoglobus fulgidus. Methods Enzymol 243:331349
Daniels L, Belay N, Rajagopal BS (1986) Assimilatory reduction of sulfate and sulfite by metha-
nogenic bacteria. Appl Environ Microbiol 51:703709
Deppenmeier U (2004) The membrane-bound electron transport system of Methanosarcina spe-
cies. J Bioenerg Biomembr 36:5564
Dhillon A, Goswami S, Riley M, Teske A, Sogin M (2005) Domain evolution and functional
diversification of sulfite reductases. Astrobiology 5:1829
DiMarco AA, Bobik TA, Wolfe RS (1990) Unusual coenzymes of methanogenesis. Annu Rev
Biochem 59:355394
Franzmann PD, Springer N, Ludwig W, Conway de Macario E, Rohde M (1992) A methanogenic
archaeon from Ace Lake, Antarctica: Methanococcoides burtonii sp. nov. Syst Appl Microbiol
15:573581
Graham DE, Xu H, White RH (2002) Identification of coenzyme M biosynthetic phosphosulfol-
actate synthase: a new family of sulfonate-biosynthesizing enzymes. J Biol Chem 277:
1342113429
Hallam SJ, Putnam N, Preston CM, Detter JC, Rokhsar D, Richardson PM, DeLong EF (2004)
Reverse methanogenesis: testing the hypothesis with environmental genomics. Science
305:14571462
Hinrichs KU, Hayes JM, Sylva SP, Brewer PG, DeLong EF (1999) Methane-consuming
archaebacteria in marine sediments. Nature 398:802805
Huang CJ, Barrett EL (1991) Sequence analysis and expression of the Salmonella typhimurium
asr operon encoding production of hydrogen sulfide from sulfite. J Bacteriol
173:15441553
Jannasch HW (1989) Chemosynthetically sustained ecosystems in the deep sea. In: Schlegel HG,
Bowien B (eds) Autotrophic bacteria. Springer, New York, pp 147166
Johnson EF, Mukhopadhyay B (2005) A new type of sulfite reductase, a novel coenzyme F420-
dependent enzyme, from the methanarchaeon Methanocaldococcus jannaschii. J Biol Chem
280:3877638786
Jones WJ, Leigh JA, Mayer F, Woese CR, Wolfe RS (1983) Methanococcus jannaschii sp. nov.,
an extreme thermophilic methanogen from a submarine hydrothermal vent. Arch Microbiol
136:254261
Kah LC, Lyons TW, Frank TD (2004) Low marine sulphate and protracted oxygenation of the
Proterozoic biosphere. Nature 431:834838
Klenk HP, Clayton RA, Tomb JF, White O, Nelson KE, Ketchum KA, Dodson RJ, Gwinn M,
McKenney K, Adams MD, Loftus B, Peterson S, Reich CI, McNeil LK, Badger JH, Glodek A,
Zhou L, Overbeek R, Gocayne JD, Weidman JF, McDonald L, Utterback T, Cotton MD,
Spriggs T, Artiach P, Kaine BP, Sykes SM, Sadow PW, DAndrea KP, Bowman C, Fujii C,
Garland SA, Mason TM, Olsen GJ, Fraser CM, Smith HO, Woese CR, Venter JC (1997) The
complete genome sequence of the hyperthermophilic, sulphate-reducing archaeon
Archaeoglobus fulgidus. Nature 390:364370
Lee JP, LeGall J, Peck HD, Jr. (1973) Isolation of assimilatory- and dissimilatory-type sulfite
reductases from Desulfovibrio vulgaris. J Bacteriol 115:529542
LeGall J, Fauque G (1988) Dissimilatory reduction of sulfur compounds. In: Zenhder AJB (ed)
Biology of anaerobic microorganisms. Wiley, New York, pp 587693
Leigh JA (2002) Evolution of energy metabolism. In: Staley JT, Reysenbach AL (eds) Biodiversity
of microbial life: foundation of earth biosphere. Wiley, New York, pp 103120
Lubbe YJ, Youn HS, Timkovich R, Dahl C (2006) Siro(haem)amide in Allochromatium vinosum
and relevance of DsrL and DsrN, a homolog of cobyrinic acid a,c-diamide synthase, for
sulphur oxidation. FEMS Microbiol Lett 261:194202
Mahlert F, Bauer C, Jaun B, Thauer RK, Duin EC (2002) The nickel enzyme methyl-coenzyme
M reductase from methanogenic archaea: In vitro induction of the nickel-based MCR-ox EPR
signals from MCR-red2. J Biol Inorg Chem 7:500513
Matthews JC, Timkovich R, Liu MY, Le Gall J (1995) Siroamide: a prosthetic group isolated from
sulfite reductases in the genus Desulfovibrio. Biochemistry 34:52485251
McCollom TM, Shock EL (1997) Geochemical constraints on chemolithoautotrophic metabolism
by microorganisms in seafloor hydrothermal systems. Geochim Cosmochim Acta 61:
43754391
Mller-Zinkhan D, Brner G, Thauer RK (1989) Function of methanofuran, tetrahydromethanopterin,
and coenzyme F420 in Archaeoglobus fulgidus. Arch Microbiol 152:362368
Moura I, Lino AR, Moura JJ, Xavier AV, Fauque G, Peck HD Jr, LeGall J (1986) Low-spin sulfite
reductases: a new homologous group of non-heme iron-siroheme proteins in anaerobic
bacteria. Biochem Biophys Res Commun 141:10321041
Nakayama M, Akashi T, Hase T (2000) Plant sulfite reductase: molecular structure, catalytic
function and interaction with ferredoxin. J Inorg Biochem 82:2732
Orphan VJ, House CH, Hinrichs KU, McKeegan KD, DeLong EF (2002) Multiple archaeal
groups mediate methane oxidation in anoxic cold seep sediments. Proc Natl Acad Sci USA
99:76637668
Pires RH, Venceslau SS, Morais F, Teixeira M, Xavier AV, Pereira IA (2006) Characterization of
the Desulfovibrio desulfuricans ATCC 27774 DsrMKJOP complex a membrane-bound
Poulton SW, Fralick PW, Canfield DE (2004) The transition to a sulphidic ocean approximately
1.84 billion years ago. Nature 431:173177
Purwantini E, Gillis TP, Daniels L (1997) Presence of F420-dependent glucose-6-phosphate
dehydrogenase in Mycobacterium and Nocardia species, but absence from Streptomyces and
Corynebacterium species and methanogenic Archaea. FEMS Microbiol Lett 146:129134
Rothe O, Thomm M (2000) A simplified method for the cultivation of extreme anaerobic Archaea
based on the use of sodium sulfite as reducing agent. Extremophiles 4:247252
Rouviere PE, Wolfe RS (1987) Use of subunits of the methylreductase protein for taxonomy of
methanogenic bacteria. Arch Microbiol 148:253259
Shen Y, Knoll AH, Walter MR (2003) Evidence for low sulphate and anoxia in a mid-Proterozoic
marine basin. Nature 423:632635
Shima S, Thauer RK (2005) Methyl-coenzyme M reductase and the anaerobic oxidation of meth-
ane in methanotrophic Archaea. Curr Opin Microbiol 8:643648
Slesarev AI, Mezhevaya KV, Makarova KS, Polushin NN, Shcherbinina OV, Shakhova VV,
Belova GI, Aravind L, Natale DA, Rogozin IB, Tatusov RL, Wolf YI, Stetter KO, Malykh AG,
Koonin EV, Kozyavkin SA (2002) The complete genome of hyperthermophile Methanopyrus
kandleri AV19 and monophyly of archaeal methanogens. Proc Natl Acad Sci USA
99:46444649
Stahl DA, Fishbain S, Klein M, Baker BJ, Wagner M (2002) Origins and diversification of sulfate-
respiring microorganisms. Antonie Van Leeuwenhoek 81:189195
Teske A, Dhillon A, Sogin ML (2003) Genomic markers of ancient anaerobic microbial pathways:
sulfate reduction, methanogenesis, and methane oxidation. Biol Bull 204:186191
Thomas I, Dubourguier H-C, Presiner G, Debeire P, Albagnac G (1987) Purification of component C
from Methanosarcia mazei and immunolocalization in Methanosarcinaeae. Arch Micorbiol
148:193201
Wedzicha BL (1992) Chemistry of sulphiting agents in food. Food Addit Contam 9:449459
Widdel F (1988) Microbiology and ecology of sulfate- and sulfur-reducing bacteria. In: Zehnder A
(ed) Biology of anaerobic microorganisms. Wiley, New York, pp 469585
Woese CR, Kandler O, Wheelis ML (1990) Towards a natural system of organisms: proposal for
the domains Archaea, Bacteria, and Eucarya. Proc Natl Acad Sci USA 87:45764579
Wolfe RS (1992) Biochemistry of methanogenesis. Biochem Soc Symp 58:4149
Chapter 17
Archaeal and Bacterial Sulfur
Oxygenase-Reductases: Genetic Diversity
and Physiological Function
Shuang-Jiang Liu
Abstract Many chemolithotrophs oxidize elemental sulfur for energy transfor-

mation under acidothermophilic conditions. Research has revealed that some of
these acidothermophilic sulfur oxidizers adopt sulfur oxygenase-reductase
(SOR) for catalysis of the initial reactions involved in such elemental sulfur
oxidation. Archaeal SORs were firstly purified from Acidianus brierleryi and
Acidianus ambivalens, and were subsequently characterized at molecular levels.
Acidianus tenchongensis represents an example of an acidothermophilic
archaeon from hot springs in China. Oxidation of elemental sulfur by this
archaeon is initiated by SOR, and this SOR gene was cloned, expressed in
Escherichia coli. Furthermore, archaeal SOR gene from Sulfolobus tokodaii
was identified from genome data, and when it was cloned in E. coli, functional
SOR was synthesized. More recently, bacterial SORs have been identified from
a microbial community in a bioleaching reactor by a metagenomic method.
Bacterial SORs have been identified also from Acidithiobacillus species and
Aquifex aeolicus.
17.1 Introduction
Sulfur oxygenase-reductase (SOR) catalyzes the conversion of elemental sulfur

into sulfite and sulfide: 4S0 + O2 + 4H2O 22SO3 + 2H2S. Under the conditions
for enzymatic catalysis, a spontaneous reaction of sulfite and elemental sulfur
occurs and thiosulfate is produced. Thus, the total reaction catalyzed by SOR is
generally expressed as follows:
5S0 + O2 + 4H 2 O H 2 SO3 + H 2 S2 O3 + H 2 S. (17.1)
The first SOR was evidenced and purified from Acidianus brierleyi (formerly
Sulfolobus brierleyi; Emmel et al. 1986). Although the reductase activity was not
reported, the properties of this so-called sulfur oxygenase are quite comparable to
those of the later-described SORs from Acidianus ambivalens (Kletzin 1989) and
Acidianus tengchongensis (He et al. 2000; Sun et al. 2003).
217
Springer 2008
218 S.-J. Liu
Fig. 17.1 Phylogenetic relationships among archaeal and bacterial sulfur oxygenase-reductase
(SOR) genes. Two genes, sorPt and sorFa, have not been functionally identified, and their hosts do
not grow on sulfur as an energy source. The gene sorFa further carries a mutation and has been
recognized as a pseudogene that has lost its function. sorAt (AF267286) from Acidianus tengchon-
gensis; sorAb from Acidianus brierleyi (unfinished genome sequence); sorAa (X56616) from
Acidianus ambivalens; sorSt (BA000023) from Sulfolobus tokodaii; sorPt (AE017261) from
Picrophilus torridus; sorFa from Ferroplasma acidarmanus; sorAqa (AE000657) from Aquifex
aeolicus; sorAct (DQ480734) from Acidithiobacillus strain SM-1; sorSA (DQ480732) from metage-
nomic DNAs of bioleaching bioreactors. Bar one base difference per 1,000 bases
At present, knowledge of SORs is mainly derived from investigations on sulfur

metabolisms in the archaeal species, namely, A. ambivalens and A. tengchongensis.
But SORs from bacterial species have been reported recently from Aquifex aeolicus
and Acidithiobacillus sp. (Chen et al. 2007; N. Pelletier, M. Guiral, G. Leroy, M.-T.
Guidici-Orticoni and C. Aubert, unpublished data). These new discoveries will
certainly provide more information on both the SOR diversity and its function in
sulfur metabolism in the future. The phylogenetic relationship of all the currently
known SORs is shown in Fig. 17.1. The nine SORs are phylogenetically grouped
17 Archaeal and Bacterial Sulfur Oxygenase-Reductases 219
into three subclusters: a large subcluster of six SORs from archaeal strains, a
subcluster of two SORs from a bioleaching reactor and Acidithiobacillus sp., and a
subcluster of a single SOR from A. aeolicus (Fig. 17.1). This chapter focuses on the
SOR diversity, physiology and potential application in bioleaching processes.
The biochemistry of SOR is described in Chap. 15 by Kletzin.
17.2 Diversity of Archaeal SORs
The archaeal SORs represent the majority of the currently known SORs
(Fig. 17.1). So far five SORs have been identified: SORAt from A. tengchongensis
(He et al. 2000), SORAb from A. brierleyi (Emmel et al. 1986; Sun et al. 2003),
SORAa from A. ambivalens (Kletzin 1989), SORSt from Sulfolobus tokodaii
(unpublished data) and SORSm from Sulfolobus metallicus. The genes coding
for SORAb and SORSt were identified according to high identities of amimo acid
residue sequences from genome projects, and recently SOR Ab and SOR St
have been confirmed to be active by cloning and expression in Escherichia
coli. S. tokodaii is unable to grow with elemental sulfur as the sole energy
source; thus, the physiological function of SORSt is still unknown. Evidence
shows that an SOR occurs in S. metallicus (S. Bathe, PE Caldwell and PR
Norris, unpublished data), but this SOR has not been characterized at molecular
levels. Two other SOR-like genes have been detected in the genomes of
Picrophilus torridus and Ferroplasma sp.; however, it is not clear if these
SOR-like genes encode active SOR enzymes.
17.2.1 SORAb from A. brierleyi
A sulfur oxygenase was purified from A. brierleyi (formerly S. brierleyi; Emmel

et al. 1986). This enzyme catalyzed the oxidization of elemental sulfur to sulfite, and
efforts to determine other products were apparently not made by the researchers. The
holoenzyme has a molecular mass of 560,000 kDa, and is composed of a homosubunit
of 35 kDa. Its optimal pH and temperature for activity were determined to be 7.0 and
65C, respectively, and had an apparent Km for sulfur of 0.05 M. SORAb was firstly
termed as a sulfur oxygenase (Emmel et al. 1986), owing to no observation of sulfide
production (indicating the reductase activity of SOR). Evidence that SORAb is similar
to the SORs was obtained from western blotting tests, which showed that A. brierleyi
cells grown with elemental sulfur contained a protein that immunologically reacted
with an antibody specific to the SOR from A. tengchongensis (Sun et al. 2003). Later,
an SOR identical at the protein level to the SOR from A. tengchongensis was discovered
in the genome of A. brierleyi (There is only one nucleotide difference, and this
difference does not result in amino acid change; unpublished data).
220 S.-J. Liu
17.2.2 SORAa from A. ambivalens
SORAa was purified from A. ambivalens (formerly Desulfurolobus ambivalens;

Kletzin 1989) and subsequently the SOR gene was cloned and sequenced (Kletzin
1992). This was the first SOR that was characterized at genetic level and was the first
reported observation where elemental sulfur was simultaneously oxidized and
reduced (an disproportional reaction) during enzymatic catalysis. Similar to SORAb
from A. brierleyi, this holoenzyme of SORAa has molecular mass of 550,000 kDa, and
is composed of a homosubunit of 40 kDa (35.6 kDa calculated from the sequence).
The optimal pH and temperature for activity were 7.4 and 85C, respectively. When
observed under the electron microscope, SORAa had a hollow globular morphology,
with a diameter of 15.6 nm. SORAa has recently been crystallized and its structure was
solved at 1.7- resolution (Urich et al. 2006, see also Chap. 15 by Kletzin).
17.2.3 SORAt from A. tengchongensis
By application of a pair of primers that targeted the conserved motif K-V-C-M-V-Y and
the C-terminus W-R-E-Y-L-N, an 840-bp DNA fragment was amplified from the
thermophilic sulfur-oxidizing A. tengchongensis (He et al. 2000, 2004). This DNA frag-
ment was used to probe the SOR gene from genomic DNAs of A. tengchongensis, and
a 3.7-kb EcoRI fragment was obtained. Sequence analysis of this 3.7-kb DNA fragment
revealed an open reading frame (ORF) that showed 88% identity to SORAa. This ORF
was cloned in E. coli, and recombinant E. coli cells massively synthesized a protein with
SOR activity (He et al. 2000). This work enabled a procedure to be developed for purifica-
tion of large amounts of SOR for further biochemical and structural studies on SOR. The
purified recombinant SORAt has a holoenzyme molecular mass of 550 kDa, and is com-
posed of a homosubunit of 35 kDa. The optimal pH and temperature for activity were
determined to be 5.0 and 70C, respectively, which are lower compared with the values
for SORAa. The lower temperature for SORAt is apparently related to the optimal growth
temperature of A. tengchongensis (70C). By application of site-directed mutagenesis, all
three cysteine residues were identified to be necessary for enzymatic activity, and the
importance of these cysteine residues has been confirmed by crystal structures of SORAt
(unpublished data) and SORAa (Urich et al. 2006).
17.2.4 SORSt from S. tokodaii
Data-mining of the S. tokodaii genome with SORAt as a probe revealed an ORF

homologous to those of SORs. This ORF was cloned by PCR technique and
expressed in E. coli cells with the same method as for SORAt. Enzymatic assays
indicated the recombinant SORSt catalyzed oxidation of elemental sulfur to sulfite
and thiosulfite (unpublished data). SORSt showed 68.1 and 64.9% sequence identities
to SORAa and SORAt/SORAb, respectively, and is the only elemental sulfur oxidizing
enzyme found in Sulfolobus species. S. tokodaii is not able to autotrophically grow
with elemental sulfur as the sole energy source, but coupling of sulfur oxidation and
CO2 fixation was observed in extended cultivation and in the presence of organic
compounds such as alanine.
17.2.5 SORSm from S. metallicus
Very recently, a new SOR gene was identified in S. metallicus by using a subtractive
hybridization approach (S. Bathe, PE Caldwell and PR Norris, unpublished data).
Although details are not available, it is expected that this SORSm will have catalytic
properties similar to those of other SORs.
17.3 Efforts To Identify Bacterial SORs
Although many chemolithotrophic bacteria oxidize elemental sulfur for energy

production, they metabolize elemental sulfur via various enzyme(s) such as the Sox
system. Involvement of enzymes similar to SOR had been indicated in a previous
study (Tano and Imai 1968), but this observation was not confirmed until recently
when functional SORs originating from bacteria were identified.
17.3.1 SORAqa from A. aeolicus
Data-mining of the A. aeolicus genome discovered two putative sulfur metabolism

systems: the thiosulfate-oxidizing multienzyme system and the SOR system.
SORAqa was cloned and expressed in E. coli. Biochemical characterization of
SORAqa revealed this enzyme is similar to the archaeal SORs with respect to holoen-
zyme composition, catalytic properties, etc. Some unique features were reported,
such as a more compact structure at 80C than at 20C and involvement of aro-
matic residues in maintaining a stable structure (N. Pelletier, M. Guiral, G.
Leroy, M.-T. Guidici-Orticoni and C. Aubert, unpublished data).
17.3.2 SORAct from Acidithiobacillus sp. strain SM-1
Novel putative bacterial SOR-like genes that are very phylogenetically different
from the SORAqa gene were identified with metagenomic methods from a microbial
community for preoxidation of gold concentrates. One of the putative genes was
222 S.-J. Liu
cloned from the metagenome of the microbial community, and this gene, namely,
sorSB, was later located in Acidithiobacillus sp. strain SM-1 (Chen et al. 2007).
SORAct (previously SORSB) from Acidithiobacillus sp. strain SM-1 was synthesized
with recombinant E. coli, and SOR activity was detected. Further biochemical
properties and the physiological role of SORAct in elemental sulfur oxidation with
Acidithiobacillus sp. strain SM-1 are currently under investigation.
17.4 SOR Links Elemental Sulfur Oxidation to ATP Synthesis

via Sulfite:Acceptor Oxidoreductase and Thiosulfate:Acceptor
Oxidoreductase
SOR simultaneously oxidizes and reduces elemental sulfur. As can be seen from
Eq. 17.1, neither the oxidation nor the reduction of elemental sulfur by SOR is
directly coupled to ATP generation or to electron transportation across the cytoplasmic
membrane. The linkage between sulfur oxidation and ATP generation was not
understood for many years until a sulfite:acceptor oxidoreductase (SAOR) and a
thiosulfate:acceptor oxidoreductase (TAOR) were discovered in sulfur-oxidizing
So SO42
2e CW
2H+
Electron carriers
? TAOR SAOR
? CM
SOR
ATP
S4O62 S2O62 So SO32 SO42 2H+ + 2e H2O Synthase
+ + 1/2 O2
H2S
CO2 ATP 2H+ ADP + Pi
Fixation
So
Fig. 17.2 Coupling sulfur oxidation and ATP generation in A. tengchongensis and A. ambivalens.
By application of immunogold electron microscopy technique, the SOR moieties were located at
both the cytoplasmic membrane and the periplasmic membrane. Enzymatic activities of SOR,
sulfite:acceptor oxidoreductase (SAOR) and thiosulfate:acceptor oxidoreductase (TAOR) were
simultaneously determined in the membrane fraction of elemental sulfur (S0) grown A. tengchon-
gensis cells (Chen et al. 2005). The TAOR was also purified from the membrane fraction of S0-
grown A. ambivalens cells (Mller et al. 2004). These findings suggest that functional coupling of
the three activities possibly happens at the periplasmic membrane CW cell wall, CM cytoplasmic
membrane, Pi inorganic phosphate
A. tengchongensis (Chen et al. 2005) and A. ambivalens (Mller et al. 2004). It is

now known that the major physiological function of SOR is to supply substrates
(sulfite and thiosulfite) that will be further oxidized by SAOR and TAOR. The
oxidations of sulfite to sulfate by SAOR and of thiosulfate to tetrathionate by
TAOR are coupled with electron transportation across the cytoplasmic membrane
and subsequently with ATP generation (Fig. 17.2). A recently characterized TAOR
from A. ambivalens uses quinone as the electron acceptor (Mller et al. 2004).
17.5 Physiological Regulation of SOR Activity in Archaea
To date, knowledge of regulation of SOR is very limited. Regulation of SOR activity

via posttranslational modification has not been reported. Kletzin (1992) reported
that transcription of the SOR gene under aerobic conditions was much higher that
under anerobic conditions for A. ambivalens. S. metallicus showed upregulation of
SOR gene transcription when grown on sulfur and, to a lesser extent, on pyrite
(S. Bathe, P.E. Caldwell and P.R. Norris, unpublished data).
Acknowledgements. The author acknowledges Z.-W. Chen for his kind assistance in the
preparation of the figures. The research was supported by the National Natural Science
Foundation of China (30621005) and the Ministry of Science and Technology (973 project no.
2004CB719600).
References
Chen ZW, Jiang CY, She Q, Zhou PJ, Liu SJ (2005) Key role of cysteine residues in catalysis and
subcellular localization of sulfur oxygenase reductase of Acidianus tengchongensis. Appl
Chen, ZW, Liu YY, Wu JF, She Q, Jiang CY, Liu SJ (2007) Novel bacterial sulfur oxygenase reductases
from bioreactors treating gold-bearing concentrates. Appl Microbiol Biotechnol 74:688698
Emmel T, Sand W, Koenig WA, Bock E (1986). Evidence for the existence of a sulfur oxygenase
in Sulfolobus brierleyi. J Gen Microbiol 132:3153420
He ZG, Li Y, Zhou P, Liu SJ (2000) Cloning and heterologous expression of a sulfur oxygenase/
reductase from the thermoacidophilic archaeon, Acidianus sp. S5 in Escherichia coli. FEMS
He ZG, Zhong H, Li Y (2004) Acidianus tengchongensis sp. nov., a new species of acidother-
mophilic archaeon isolated from an acidothermal spring. Curr Microbiol. 48:15963
Kletzin A (1989) Coupled enzymatic production of sulfite, thiosulfate, and hydrogen sulfide from
sulfur: purification and properties of a sulfur oxygenase reductase from the facultatively
anaerobic archaebacterium Desulfurolobus ambivalens. J Bacteriol 171:16381643
Kletzin A (1992) Molecular characterization of the sor gene, which encodes the sulfur oxygenase/
reductase of the thermoacidophilic Archaeum Desulfurolobus ambivalens. J Bacteriol
174:58545859
pathway of sulphur oxidation dioxygen reduction: characterization of novel membrane-bound
thiosulphate:quinone oxidoreductase. Mol Microbiol 53:11471160
224 S.-J. Liu
Sun CW, Chen ZW, He ZG, Zhou PJ, Liu SJ (2003) Purification and properties of the sulfur
oxygenase/reductase from the acidothermophilic archaeon, Acidianus strain S5. Extremophiles
7:131134
Tano T, Imai K (1968) Physiological studies on thiobacilli. Part II. The metabolism of colloidal
sulfur by the cell-free enzyme system of Thiobacillus thiooxidans. Agric Biol Chem
32:5154
Urich T, Gomes CM, Kletzin A, Frazo C (2006) X-ray structure of a self-compartmentalizing
Chapter 18
Diversity of Halophilic Sulfur-Oxidizing
Bacteria in Hypersaline Habitats
Dimitry Y. Sorokin
Abstract The culturable diversity of halophilic obligately chemolithoautotrophic

sulfur-oxidizing bacteria (SOB) in various aquatic hypersaline habitats, such as
inland chloridesulfate lakes, sea solar saltern and deep-sea salt brines, was found
to be unexpectedly high. Six different groups of halophilic SOB belonging to the
Gammaproteobacteria were found. Two groups of moderately halophilic strictly
aerobic SOB dominated at 2 M NaCl, including representatives of the genus
Halothiobacillus (at fully aerobic conditions) and the genus Thiomicrospira
(at microoxic conditions). In a few cases, halothiobacilli also dominated at 4 M
NaCl. Under denitrifying conditions at 2 M NaCl, moderately halophilic and facul-
tatively anaerobic SOB capable of complete denitrification of nitrate were found.
They are a member of a new genus, Thiohalomonas, with closest relatives among
marine thiodenitrifyers. At moderate salinity and with thiocyanate as a substrate, a
pure culture of moderately halophilic SOB capable of growth with thiocyanate
and thiosulfate up to 4 M NaCl was obtained, and these SOB are a member of a new
genus Thiohalophilus distantly related to the genus Thiomicrospira. Two groups
of extremely halophilic SOB growing between 2 and 4 M NaCl with an optimum at
3 M NaCl dominated in enrichments at 4 M NaCl. The group of obligately aerobic
extreme halophiles, members of a new genus Thiohalospira, are related to the
Ectothiorhodospiraceae, and facultatively anaerobic nitrate-reducing extreme
halophiles, members of a new genus Thiohalorhabdus, are distantly related to the
genus Acidithiobacillus.
18.1 Introduction
The diversity of halophilic sulfur-oxidizing bacteria (SOB) able to develop

optimally in NaCl brines remains largely unexplored. Apart from a single moderately
halophilic species Halothiobacillus halophilus, discovered 15 years ago in an
Australian hypersaline lake (Wood and Kelly 1991; Kelly et al. 1998; Kelly and
Wood 2000), nothing is known about such chemolithoautotrophic bacteria. Our
225
Springer 2008
226 D.Y. Sorokin
recent research on natronophilic SOB inhabiting saline and highly alkaline soda
lakes provided evidence for the widespread potential of chemolithoautotrophic
SOB to grow at very high concentrations of sodium carbonate/sodium bicarbonate
(Sorokin and Kuenen 2005a, b; Sorokin et al. 2005a). This prompted us to start
similar research on the diversity of SOB in hypersaline chloridesulfate habitats
with neutral pH.
Extremely halophilic heterotrophic haloarchaea growing optimally at 34 M
NaCl were traditionally regarded as dominating prokaryotes in hypersaline habitats,
such as sea salterns and hypersaline lakes (Oren 2002). Recently, however,
evidence started to emerge indicating the importance of bacterial components in
the extremely halophilic prokaryotic communities (Antn et al. 2002; Sorokin et al.
2006a). Among the chemolithotrophic bacteria, SOB have a good chance to adapt
to extreme conditions, such as high salt, owing to a very high energy yield from
complete oxidation of sulfide/thiosulfate to sulfate (Oren 1999). However, so far,
no culturable SOB phenotypes, equal to haloarchaea with regard to their salt
response, are known. Since functional genes of sulfur-oxidation pathways are not
conserved and only recently started to become a subject for molecular analysis
(Friedrich et al. 2001, 2005), the culture-independent approach is not yet available
for diversity analysis of SOB. Therefore, traditional methods of enrichment and
isolation in pure culture remain the main option for biodiversity studies of SOB.
Hypersaline aquatic habitats are divided into marine-dependent (thalassic),
which include sea solar salterns, hypersaline lagoons and deep-sea brines, and
inland (athalassic) lakes formed either by evaporative concentration of incoming
diluted solutions (primary evaporates) or by dissolution of ancient salt depositions
(secondary evaporates). The sea solar salterns have been studied most extensively,
being relatively easy to access and having an advantage for investigators in offering
a whole range of salinity gradients within a short distance (various stages of evapo-
ration). Less is known about microbial communities in hypersaline lakes, located
mostly in remote areas with an evaporative climate. The principal difference
between these two is the much higher magnesium content in the thalassic brines
and, usually, the higher sulfate content in inland lakes. In our search for halophilic
SOB we mainly focused on inland hypersaline lakes, but also used samples from a
sea saltern and from a deep-sea salt brine, formed during dissolution of ancient salt
deposits, for a comparison.
18.2 Description of Habitats Investigated
Hypersaline aquatic habitats in six different regions were examined in this study,
including four sites of hypersaline inland lakes, a sea solar saltern and a deep-sea
salt brine. The main area of study was in the Kulunda Steppe, southwest Siberia,
located along the northeast Kazakhstan border. It harbours numerous salt lakes,
ranging from shallow ponds to very large water bodies (Issachenko 1951) with a
18 Diversity of Halophilic Sulfur-Oxidizing Bacteria in Hypersaline Habitats 227
Table 18.1 Characteristics of the hypersaline habitats investigated

Mineral composition
Total
salts Na+ Na+/ Cl Cl/
Region Type Name pH (g l1) (M) Mg2+ (M) SO42
Kulunda Steppe Inland lakes 26 hypersaline 7.58.5 100380 0.95.5 14b 0.654.4 4.5b
(southwest lakesa
Siberia)
Northeast Mongolia Lake Dzun-Davst 7.58.1 220320 ND ND ND ND
and Lake
Dolon-Davst
South Russia Lake Baskunchak 6.2 360
Crimea peninsula Lake Marfinskoe 8.0 115145
(Ukraine) and Lake
Kayashskoe
Slovenia (Adriatic Sea saltern Seovljec 6.5 320 4.1 5.2 5.2 18.6
coast)
East Medi-terranean Deep-sea Urania Basind 6.8 220 3.10 11.1 3.30 34.7
brine
a
For enrichment purposes sediments were combined into eight groups: group 1, salinity 1015%,
north-central; group 2, salinity 1622%, north-central; group 3, salinity 2228%, south; groups
48, five individual samples from lakes with salinity 3038%.
b
Average data.
c
Data of Gunde-Cimerman (2000).
d
Data of M. Yakimov (personal communication).
total salt content from 10 to 38% (w/v), a pH range from 7.5 to 8.5, and with Na+,
Mg2+, Cl and SO42 as the dominant ions in the brines (Table 18.1). Other lake
provinces, in northeast Mongolia, south Russia (Lake Baskunchak is the biggest
salt lake in Russia and an important source of cooking salt) and in the Crimea
peninsula were studied only briefly (Table 18.1). In addition, a sample from a final
evaporation pond in a Seovlje Adriatic Sea saltern (Gunde-Cimerman et al. 2000)
and a sample of deep-sea brine from the eastern Mediterranean Urania Basin (Sass
et al. 2001; van der Wielen et al. 2005) were included in the analysis.
18.3 Enrichment Strategy
In general, two basic mineral media were used to enrich and isolate moderate and
extreme halophiles, with 2 and 4 M NaCl, respectively. Commonly, thiosulfate
(1020 mM) was used as the energy source and, in some cases, also sulfide,
tetrathionate (5 mM) or thiocyanate (10 mM). NaHCO3 served as a carbon source
228 D.Y. Sorokin
Sediments from hypersaline lakes, sea saltern, deep-sea brines
Enrichments at 2 M NaCl Enrichments at 4 M NaCl
Fully aerobic, Microaerophilic Anaerobic Aerobic, Anaerobic

colonies denitrifying microaerophilic denitrifying
Halothiobacillus Thiomicrospira S2O32/ NO3 NCS/ NO3 Thiohalospira Thiohalorhabdus

halophila
Thiohalomonas Thiohalophilus
Fig. 18.1 General scheme showing culturable diversity of halophilic sulfur-oxidizing bacteria
(SOB) from hypersaline habitats. Halothiobacillus spp. and Thiomicrospira halophila, aerobic
moderate halophiles; Thiohalospira, aerobic extreme halophiles; Thiohalomonas, thiodenitri-
fying moderate halophiles; Thiohalorhabdus, thiodenitrifying extreme halophiles;
Thiohalophilus, facultatively anaerobic and thiocyanate-utilizing moderate halophile. Dashed
lines indicate occasional selections
and additional alkaline buffer (pH 78). To prevent loss of CO2 and evaporation,
aerobic cultivation was performed in closed bottles with 10% liquid volume at
static conditions. Microaerophilic (2% oxygen in the gas phase) and denitrifying
cultures were grown in 100-ml serum bottles with butyl rubber stoppers and with
10 ml (aerobic) to 80 ml(anaerobic) of the medium. With sulfide as a substrate, the
gradient cultivation technique (Nelson and Jannasch 1993) was employed. Solid
medium containing 23 M NaCl was prepared by mixing complete liquid medium
containing 4 M NaCl and 3040 mM thiosulfate with 46% (w/v) agarose at different
ratios at 50C. The plates were incubated in closed jars at 020% O2/5% CO2 (v/v)
in the gas phase.
Various types of enrichments of SOB from hypersaline habitats and their general
results are represented in Fig. 18.1.
18.4 Moderately Halophilic Aerobic SOB
The aerobic enrichments at 2 M NaCl usually developed quite rapidly, oxidizing

20 mM thiosulfate within 1 week. In the lake sediments, direct serial dilutions
indicated the presence of 105107 viable cells in 1 cm3. In fully aerated cultures,
short motile rods forming large sulfur-containing colonies dominated and could be
Fig. 18.2 Typical cell morphology of halophilic SOB from hypersaline habitats. a Moderately
halophilic aerobic Halothiobacillus sp. HL 1, thin section, bar 0.5 m; b moderately halophilic
aerobic Thiomicrospira halophila HL 5, bar 0.5 m; c extremely halophilic aerobic Thiohalospira
HL 4, bar 1 m; moderately halophilic denitrifying Thiohalomonas HLD 1; e, extremely
halophilic denitrifying Thiohalorhabdus HLD 8; f moderately halophilic, facultatively anaero-
bic and thiocyanate-utilizing Thiohalophilus HRhD 2, thin section, bar 0.5 m
easily obtained in pure culture. In static cultures with a high liquid-to-gas ratio and
with sulfide as a substrate in gradient enrichments, a highly motile small vibrio
became the dominant morphotype. It also could produce tiny sulfur colonies on
thiosulfate plates after prolonged incubation. After isolation in pure culture, the
vibrio strains could easily grow at fully aerated culture. Overall, four strains with
rod-shaped cells (Fig. 18.2a) and three strains with vibrio cells (Fig. 18.2b) have
230
Table 18.2 Types of culturable halophilic sulfur-oxidizing bacteria (SOB) in hypersaline habitats: summary
Genetic properties Growth kinetics
Number Salt range Y (mg
of (optimum) Denitri- CNS Gene- protein
Type isolates Habitat Affiliation (M NaCl) fication S intermediate oxidation G+C (mol%) speciesa (h1) mmol1)
1 7 SL, MB Halothiobacillus 0.54.0 (1.0 Sulfur 64.067.7 2 0.200.35 4.04.5
1.5)
2 3 SL Thiomicrospira 0.53.5 (1.5) 56.157.1 1 0.25 3.5

4 10 SL, ST Thiohalomonas 1.03.0 (1.5 + 58.060.0 1 0.030.04 4.04.9
2.0)
6 1 SL Thiohalophilus 1.04.0 (2.5) + + 58.2 1 0.10 5.6
3 20 SL, ST Thiohalospira 2.05.0 (3.0) Tetrathionate 65.867.0 3 0.030.04 2.02.5
5 7 SL, ST Thiohalorhabdus 2.04.5 (3.0) + 65.065.8 1 0.052 4.2
SL inland lakes, MB deep-sea brines, ST sea salterns, m specific growth rate at optimal salinity, Y specific growth yield.
a
On the basis of DNADNA hybridization.
Dimitry Y. Sorokin
been isolated in pure culture from the Siberian and Mongolian lakes (Table 18.2).
The rod-shaped isolates were identified as members of the genus Halothiobacillus
and contained at least two different gene-species (DNADNA hybridization below
species level). The vibrio strains were genetically almost identical to each other and
represent a new species within the genus Thiomicrospira, Thiomicrospira halo-
phila (Sorokin et al. 2006c; Fig. 18.3).
Fig. 18.3 Phylogenetic position of representative strains of halophilic SOB from hypersaline
habitats within the Gammaproteobacteria based on 16S ribosomal RNA gene sequence analysis.
Tree topography and evolutionary distances are given by the neighbour-joining method with Jukes
and Cantor distances. Numbers at the nodes indicate the percentage of bootstrap values for the
clade in 1,000 replications. Only values above 90% are shown
232 D.Y. Sorokin
With deep-sea brines from Urania Basin, only the rod-shaped phenotype was
present in the enrichments at 2 M NaCl. The isolate from the Urania Basin, strain
HL-U1, clustered with Halothiobacillus hydrothermalis according to the 16S
ribosomal RNA (RNA) gene analysis (Fig. 18.3). Two more Halothiobacillus
species, strain HL 20 and strain HL 27, were isolated from the aerobic enrichments
at 4 M NaCl, when a specialized group of extremely halophilic SOB Thiohalospira
(Sect. 18.5) were either at low number or completely absent. Strain HL 20 dominated
an enrichment culture with tetrathionate as a substrate inoculated with a sediment
sample from the Mongolian lakes, while strain HL 27 was one of the dominant
organisms in the enrichment culture from the Crimean lakes with thiosulfate. All
these isolates were moderately halophilic with an optimum around 1 M NaCl.
18.5 Extremely Halophilic Aerobic SOB
Enrichment cultures at 4 M NaCl were much slower than at 2 M NaCl, the first
indication of thiosulfate consumption usually appearing only after 10 days of
incubation. Despite this, positive results were obtained for most of the samples
studied, except for those from the deep-sea brine of Urania Basin. This indicated
the universal presence of SOB populations able to develop at saturating salt
concentrations. Moreover, they were as abundant in the lake sediments as moderate
halophiles (103107 cm3). The dominant phenotype observed at 4 M NaCl in most
cases was a thin motile spirillum (Fig. 18.2c). Since it did not form colonies, the
pure culture isolation was achieved in several rounds of dilution to extinction.
Overall, 20 strains of this phenotype were obtained from salt lakes and a saltern
using medium with 4 M NaCl, and with thiosulfate, sulfide or tetrathionate as
substrates. The group included at least three different gene species, from Siberian
and Mongolian lakes and from a Slovenian saltern. On the basis of the 16S rRNA
gene sequence analysis, it represents a new lineage in the Gammaproteobacteria
with the provisional name Thiohalospira, clustering with the members of the
family Ectothiorhodospiraceae (Fig. 18.3). All strains are extreme halophilies not
known before among the SOB (Table 18.2). Another specific property of this group
was production of large amounts of tetrathionate as an intermediate of thiosulfate
oxidation (up to 80% conversion), which was finally oxidized to sulfate.
18.6 Moderately Halophilic Thiodenitrifyers
Anaerobic enrichments at 2 M NaCl with thiosulfate as the electron donor and

nitrate the as electron acceptor were positive in nine enrichments with various lake
sediments and from a saltern (Table 18.2). Nitrite and N2O were observed as major
nitrogen intermediates and elemental sulfur as an occasional intermediate during
oxidation of thiosulfate to sulfate. Six pure cultures were obtained from the
Kulunda lakes, and a single strain each from the Mongolian lakes, Lake Baskunchak,
the Crimean lakes and the Slovenian saltern. All these isolates were facultatively
anaerobic, denitrifying, moderately halophilic SOB with long, nonmotile,
rod-shaped cells (Fig. 18.2d). Despite the fact that nitrite and N2O were produced
as intermediates during anaerobic growth with nitrate and that washed cells, grown
with nitrate, could reduce both intermediates in the presence of thiosulfate, anaerobic
growth occurred only with nitrate. Growth was also observed under microoxic con-
ditions at O2 concentrations below 5% (v/v) in the gas phase. This SOB group harbours
moderate halophiles with a relatively narrow salt range for growth (Table 18.2).
According to the results of DNADNA hybridization and sequencing of the 16S
rRNA gene, all HLD strains consisted a single gene species and formed a new
lineage within the Gammaproteobacteria, with the closest relatives among a cluster
of marine yet not described thiodenitrifyers (Nercessian et al. 2005; S. Sievert and
G. Muyzer, unpublished data; Fig. 18.3). The provisional name for this genus is
Thiohalomonas.
18.7 Extremely Halophilic Denitrifying SOB
Positive anaerobic enrichments with thiosulfate and nitrate at 4 M NaCl were

obtained from the four lake samples and from the saltern. Despite extremely slow
development (35 mM thiosulfate consumed within 1 month), all positive
enrichments resulted in the isolation of a pure culture of the dominant SOB
morphotype with long, flexible, nonmotile, rod-shaped cells (Fig. 18.2e).
A similar phenotype was also found in two aerobic enrichments at 4 M NaCl.
Strain HL 19 was dominant in an mixotrophic enrichment (acetate/thiosulfate).
During the first stage of this enrichment, heterotrophic haloarchaea utilized acetate
and concomitantly oxidized thiosulfate to tetrathionate (Sorokin et al. 2005b).
When all acetate had been utilized, a mixture of extremely halophilic SOB started
to develop using tetrathionate as the energy source. One of the dominant phenotypes
(strain HL 19) was separated by using dilution series with tetrathionate as a
substrate. It was similar in morphology and its ability to grow anaerobically with
nitrate to HLD strains isolated at 4 M NaCl from denitrifying enrichments. Another
similar strain, HL 28, was isolated from an aerobic enrichment culture at 4 M NaCl
from one of the Crimean lakes, where it was developing in a mixture with the
Thiohalospira (Sect. 18.5).
The group shared several common physiological properties. In contrast to the
moderately halophilic denitrifying HLD strains (see above), they grew well under
microoxic conditions (25% v/v O2) and some of them even in fully aerated
cultures. Tetrathionate was a major intermediate of aerobic thiosulfate oxidation to
sulfate in this group, similar to the aerobic extreme halophiles from the
Thiohalospira group. Under anaerobic conditions, with either thiosulfate or
tetrathionate as substrates, nitrate was only reduced to nitrite, and sulfur accumulated
as an intermediate. Washed cells, grown with nitrate, however, very slowly reduced
234 D.Y. Sorokin
nitrite and, more actively, N2O in the presence of thiosulfate as an electron donor.
These bacteria represent a second group of extremely halophilic SOB found in
hypersaline habitats (Table 18.2).
All these isolates were related at the species level and formed a new deep lineage
within the Gammaproteobacteria (new genus Thiohalorhabdus) distantly related
to the genus Acidithiobacillus (Fig. 18.3).
18.8 Oxidation of Thiocyanate at High Salt
Thiocyanate (NCS) is a difficult substrate for SOB and almost nothing is known
about its utilization at high salt. Despite some growth and thiocyanate consumption
being observed in aerobic enrichments at 2 M NaCl, no pure cultures were obtained
because of the presence of high numbers of heterotrophs. Under anaerobic
conditions with thiocyanate as an electron donor and nitrate as an electron acceptor
at 2 M NaCl, a stable binary culture was selected which eventually resulted in the
isolation of strain HRhD 2 capable of aerobic growth with thiocyanate as the only
substrate (Fig. 18.2f). The final products of thiocyanate metabolism were sulfate
and ammonium. With both thiosulfate and thiocyanate it could grow within a broad
salt range from 1.0 to 4.0 M NaCl (Table 18.2). The bacterium was able to grow
anaerobically with thiosulfate using nitrite (but not nitrate) as the electron acceptor
at low concentrations (below 2 mM) with N2O as an intermediate of denitrification.
COS was detected as an intermediate of thiocyanate metabolism, which indicated
the COS pathway (Kelly and Baker 1990) for the primary thiocyanate degradation
in strain HRhD 2. On the other hand, the presence of high cyanase activity in the
cells, grown with thiocyanate, cannot be rationally explained at this moment.
Phylogenetic analysis of strain HRhD 2 placed it in a new lineage within the
Gammaproteobacteria distantly related to the genus Thiomicrospira, for which a
provisional name Thiohalophilus is suggested (Fig. 18.3).
18.9 Fatty Acids in the Membrane Lipids
Since it is the cell membrane which is essential in the salt out strategy used by
halophilic Proteobacteria in their adaptation to live in brines, the composition of
the membrane lipids is an essential property worth investigating. A comparison of
the fatty acid composition in the type strains of four new genera of halophilic SOB
described in preceding sections provided interesting data (Table 18.3). First of all,
palmitic acid (16:0) was the dominant species in all halophilic SOB. Secondly,
hexadecenic acid (16:1w7) was another dominant species in moderately halophilic
genera Thiohalomonas and Thiohalophilus, but not in extremely halophilic genera.
The latter, represented by the genera Thiohalospira and Thiohalorhabdus, despite
their different phylogenetic position, had quite a similar fatty acid composition
Table 18.3 Comparison of dominant fatty acid composition of the polar lipids in halophilic SOB
Percentage of the total
Moderate halophiles Extreme halophiles
Fatty acid Species Thiohalomonas Thiohalophilus Thiohalospira Thiohalorhabdus
Hexadecenic 16:1w7 27.27 22.00 10.8 7.37
acid
16:1w5 2.97 0.25
16:0 25.41 33.70 31.9 29.36
10-Methyl- 10Me16 1.26 44.5 43.43
hexadecanic
acid
11-Methyl-hep- 11Me17:1 2.53 44.5 43.43
tadecenic
acid
Isoheptadecenic i17:1w5 32.41
acid
Cyclopropane 17cyc 8.38 0.15
heptade-
canic acid
Octadecenic 18:1w9 5.10 0.15 0.31
acid
18:1w7 11.84 1.11 5.8 3.35
Octadecanic 18:0 0.49 0.21 3.3 3.43
acid
Cyclopropane 19cyc 3.26
nonadecanic
acid
with an extremely high content of methylated C16 and C17 species, which can
be considered as a specific feature of these new SOB lineages. Thiohalophilus
also had a very specific molecular marker isoheptadecenic acid (i17:1w5),
which was completely absent in the other genera. Comparison with the other
extremely halophilic (Halovibrio-Halospina) and natronophilic (Thioalkalivibrio)
Gammaproteobacteria indicated that only the presence of significant amount of
16:0 is a common trait among all these extremophiles.
18.10 Conclusions and Future Perspectives
Unexpectedly high culturable diversity of halophilic SOB was detected in hypersa-

line habitats. Two moderately halophilic aerobic groups belong to the known gen-
era in the Gammaproteobacteria, while extremely halophilic aerobes, moderately
and extremely halophilic thiodenitrifyers and moderately halophilic thiocyanate-uti-
lizing SOB all represent new lineages within the Gammaproteobacteria.
236 D.Y. Sorokin
Very similar halophilic SOB species were found in inland (athalassic) salt lakes
and solar salterns (thalassic) in Europe, different climatically and in the chemical
composition of its brines. Perhaps, total extreme NaCl content and very specific
metabolism are more important than the other parameters. This is confirmed by a
very high requirement both for Na+ and Cl in all groups of halophilic SOB found
in hypersaline habitats, both thalassic and athalassic. On the other hand, all of them
could grow at very low Mg content and without any added Ca (data not shown).
The most interesting new SOB discovered in hypersaline habitats are the two
groups of extreme halophiles a previously unknown ecotype of SOB. High viable
cell numbers in the sediments indicate that they may represent one of the dominant
bacterial populations there. Both groups have the so-called tetrathionate pathway of
thiosulfate oxidation to sulfate, which is common in SOB living in extreme habitats,
such as members of Acidithiobacillus, Thermothiobacillus and Halothiobacillus
(Kelly and Wood 2000). Despite harsh conditions, their growth yield seems to
be within the usual range (Kelly et al. 1997), implying that these bacteria may
possess special adjustments in their bioenergetic mechanisms, which would be
most interesting to study.
The array of new halophilic SOB from hypersaline environments, available in
culture, offers interesting prospects for future research on their physiology and
biochemistry. Especially interesting topics might be the mechanisms of salt
tolerance in chemolithoautotrophic SOB, the biochemistry and genetics of their
sulfur-oxidizing and denitrification pathways and the biochemistry of thiocyanate
metabolism in halophiles. A preliminary description of the new groups has recently
been published elsewhere (Sorokin et al. 2006b).
Acknowledgements. This work was supported by an NWO-RFBR grant (047.011.2004.010) by

RFBR grant 07-04-00153 and by the Program on Molecular and Cell Biology RAS. The work
was done in collaboration with G. Muyzer, T.P. Tourova and A.M. Lysenko (genetic and phylo-
genetic analysis). We are grateful to M. Yakimov and L. Gerasimenko for the possibility to work
with their samples.
References
Antn J, Oren A, Benlloch S, Rodrguez-Valera F, Amann R, Rossell-Mora R (2002) Salinibacter

ruber gen. nov., sp. nov., a new species of extremely halophilic Bacteria from saltern
crystallizer ponds. Int J Syst Evol Microbiol 52:485491
Gunde-Cimerman N, Zalar P, de Hoog S, Plemenitas A (2000) Hypersaline waters in salterns
natural ecological niches for halophilic black yeasts. FEMS Microbiol Ecol 32:235240
Issachenko BL (1951) Chloride, sulfate and soda lakes of Kulunda steppe and its biogenic
processes. In: Selected works, vol 2. Academy of Sciences USSR, St Petersburg, pp 143162
(in Russian)
Kelly DP, Baker SC (1990) The organosulfur cycle: aerobic and anaerobic processes leading to
turnover of C1-sulfur compounds. FEMS Microbiol Rev 87:241246
Kelly DP, Wood AP (2000) Reclassification of some species of Thiobacillus to the newly
designated genera Acidithiobacillus gen. nov., Halothiobacillus gen. nov. and Thermithiobacillus
gen. nov. Int J Syst Evol Microbiol 50:511516
Kelly DP, Shergill JK, Lu W-P, Wood AP (1997) Oxidative metabolism of inorganic sulfur
compounds by bacteria. Antonie Van Leeuwenhoek 71:95107
Kelly DP, Stackebrandt E, Burghardt J, Wood AP (1998) Confirmation that Thiobacillus
halophilus and Thiobacillus hydrothermalis are distinct species within the -subclass of the
Proteobacteria. Arch Microbiol 170: 138140
Nelson DC, Jannasch HW (1983) Chemolithoautotrophic growth of a marine Beggiatoa in sulfide-
gradient cultures. Arch Microbiol 136:262269
Nercessian O, Fouquet Y, Pierre C, Prieur D, Jeanthon C (2005) Diversity of Bacteria and Archaea
associated with a carbonate-rich metalliferous sediment sample from the Rainbow vent field
on the Mid-Atlantic Ridge. Environ Microbiol 7:698714
Oren A (1999) Bioenergetic aspects of halophilism. Microbiol Mol Biol Rev 63:34348
Oren A (2002) Halophilic microorganisms and their environments. Kluwer, Dordrecht
Sass A, Sass H, Coolen MJ, Cypionka H, Overmann J (2001) Microbial communities in the
chemocline of a hypersaline deep-sea basin (Urania Basin, Mediterranean Sea). Appl Environ
Sorokin DY, Kuenen J G (2005a) Haloalkaliphilic sulfur-oxidizing bacteria in soda lakes. FEMS
Microbiol Rev 29:685702
Sorokin DY, Kuenen J G (2005b) Alkaliphilic chemolithotrophs from soda lakes. FEMS Microbiol
Ecol 52:287295
Sorokin DY, Banciu H, Robertson LA, Kuenen JG (2005a) Haloalkaliphilic sulfur-oxidizing
bacteria. In: Dworkin, Falkow S, Rosenberg E, Schleifer K-H, Stackebrandt E (eds) The
prokaryotes: an evolving electronic resource for the microbiological community. Release 3.20.
http://141.150.157.117:8080/prokWIP/index.htm
Sorokin DY, Tourova TP, Muyzer G. (2005b) Oxidation of thiosulfate to tetrathionate by a haloar-
chaeon from hypersaline habitat. Extremophiles 9:501504
Sorokin DY, Tourova TP, Galinski EA, Belloch C, Tindall BJ (2006a) Extremely halophilic
denitrifying bacteria from hypersaline inland lakes Halovibrio denitrificans sp. nov. and
Halospina denitrificans gen. nov., sp. nov., and evidence that the genus name Halovibrio
(Fendrich 1989) with the type species H. variabilis should be associated with DSM 3050. Int
J Syst Evol Microbiol 56:379388
Sorokin DY, Tourova TP, Lysenko AM, Muyzer G (2006b) Diversity of culturable halophilic
sulphur-oxidizing bacteria in hypersaline habitat. Microbiology 152:30133023
Sorokin DY, Tourova TP, Kolganova TV, Spiridonova EM, Berg IA,, Muyzer G (2006c).
Thiomicrospira halophila sp. nov., a novel, moderately halophilic, obligately chemolithoau-
totrophic sulfur-oxidizing bacterium from hypersaline lakes. Int J Syst Evol Microbiol
56:23752380
van der Wielen PWJJ, Bolhuis H, Borin S, Daffonchio D, Corselli C, Giuliano L, DAuria G, de
Lange GJ, Huebner A, Varnavas SV, Thomson J, Tamburini C, Marty D, McGenity TJ, Timmis
KN (2005) The enigma of prokaryotic life in deep hypersaline anoxic basins. Science
307:121123
Wood AP, Kelly DP (1991) Isolation and characterisation of Thiobacillus halophilus sp. nov., a
sulphur-oxidizing autotrophic eubacterium from a Western Australian hypersaline lake. Arch
Chapter 19
Sulfur Oxidation at Deep-Sea
Hydrothermal Vents
Stefan M. Sievert, Michael Hgler, Craig D. Taylor, Carl O. Wirsen
Abstract Microbial oxidation of geothermally produced reduced sulfur compounds is

at the nexus of the biogeochemical carbon and sulfur cycles at deep-sea hydrothermal
vents. Available information indicates that microbial symbionts and free-living
gammaproteobacteria of the genera Thiomicrospira, Halothiobacillus, and Beggiatoa
are important sulfur-oxidizers above the seafloor at these systems. In addition, bacteria
belonging to the Epsilonproteobacteria have been identified as a major component of
microbial communities at deep-sea vents. We have previously identified a novel sulfur-
oxidizing epsilonproteobacterium, Candidatus Arcobacter sulfidicus, which produces
sulfur in filamentous form that is morphologically and chemically similar to material
observed before and after submarine volcanic eruptions. In the meantime, many
autotrophic epsilonproteobacteria have been isolated and characterized from deep-sea
vents, providing further evidence that these organisms play an important role in sulfur
and carbon cycling in these environments. These kinds of bacteria may form an impor-
tant component of a subseafloor biosphere, a currently poorly defined, yet potentially
critical component of deep-sea hydrothermal vents. Many autotrophic bacteria and
archaea occurring at deep-sea hydrothermal vents, including epsilonproteobacteria, use
the reductive tricarboxylic acid cycle for autotrophic carbon fixation, questioning the
paradigm of the CalvinBensonBassham cycle being at the base of the food web of
these ecosystems. In the future, integrated geochemical and biological studies are
needed to further advance our understanding of chemoautotrophic sulfur oxidation at
deep-sea vents, which will be greatly facilitated by having the genomes of representative
sulfur-oxidizing bacteria available.
19.1 Introduction
At deep-sea hydrothermal vents, microorganisms mediate the transfer of energy from

the geothermal source to the higher trophic levels. In particular the microbial oxida-
tion of reduced sulfur compounds through chemolithotrophic processes, principally
involving H2S, has been identified to be at the nexus of the biogeochemical carbon
and sulfur cycles of these systems, where H2S is primarily produced via seawater
rock interactions within the high-temperature zone (about 400C) near the sheeted
238
Springer 2008
19 Sulfur Oxidation at Deep-Sea Hydrothermal Vents 239
dikegabbro interface (Jannasch and Mottl 1985; Jannasch 1995; but see Karl 1995
for a very worthwhile critical look at this aspect; Fig. 19.1). On the basis of
thermodynamic modeling, the oxidation of the H2S contained in the hydrothermal
fluids upon contact with oxygenated seawater represents the major energy source
available at deep-sea vents (McCollom and Shock 1997), supporting a physiological
grouping of microorganisms collectively referred to as the colorless sulfur-oxidizing
bacteria. These organisms are characterized by their ability to oxidize H2S or other
partially oxidized sulfur compounds for the mixotrophic or autotrophic incorporation
of CO2 into cellular material by using either oxygen or nitrate as electron acceptors.
At deep-sea hydrothermal vents, sulfur-oxidizing bacteria either exist as free-living
forms in the mixing zone between oxygenated seawater and reduced hydrothermal
fluids, either above or below the seafloor, or in a symbiotic relationship with various
invertebrates (Jannasch and Mottl 1985; Table 19.1). Thermodynamic calculations
suggest that sulfide oxidation is most favorable at relatively low temperatures (below
20C) owing to an increasing availability of oxygen (McCollom and Shock 1997).
This suggests that organisms employing this metabolic strategy, in particular at higher
temperatures, are able to utilize very low oxygen concentrations, to use alternative
electron acceptors, such as nitrate, or to employ energy-efficient metabolic pathways,
e.g., for carbon fixation. While the importance of microbial oxidation of H2S to
support chemoautotrophic production above the seafloor at deep-sea hydrothermal
vents is well established, much less is known about the composition and extent of the
microbial communities in the subseafloor portions of these systems and the global
impact of their activities (Wilcock et al. 2004).
19.2 Types of Sulfur-Oxidizing Bacteria
19.2.1 Symbiotic Sulfur-Oxidizing Bacteria
Although sulfur-oxidizing bacteria living in symbiosis with invertebrates have

never been cultured, several studies have investigated their phylogeny and
physiology (reviewed in Nelson and Fisher 1995; Stewart et al. 2005). In particular,
recent genomic and proteomic investigations have significantly advanced our
understanding of these symbioses (Markert et al. 2007; Cavanaugh 2006; Newton
et al. 2007). To date most sulfur-oxidizing endosymbionts have been shown to
belong to the subdivision of the Proteobacteria (Gammaproteobacteria) (Stewart
et al. 2005), although recently members of the subdivision of the Proteobacteria
(Epsilonproteobacteria) have also been identified as endosymbionts (Suzuki et al.
2005a; Urakawa et al. 2005). Many deep-sea hydrothermal vent invertebrates have
also been found to live in a stable association with an epibiotic community, which,
on the basis of present information, are mostly dominated by Epsilonproteobacteria
(Campbell et al. 2006). Although the metabolism of these epibionts has not been
determined in each case, sulfur oxidation appears to be common.
240
Fig. 19.1 A mid-ocean ridge hydrothermal vent site and potential microbial habitats in the subseafloor. Seawater cycles through the seafloor
where it is geothermally altered. Hot, reducing fluid containing millimolar concentrations of H2S ascend to the seafloor either exiting undiluted
S.M. Sievert et al.
19.2.2 Free-Living Sulfur-Oxidizing Bacteria
19.2.2.1 Gammaproteobacteria
In addition to the symbionts, available information indicates that free-living

bacteria belonging to the genera Thiomicrospira, Halothiobacillus, and Beggiatoa
within the Gammaproteobacteria are important sulfur-oxidizers above the
seafloor at deep-sea hydrothermal vents (Brinkhoff et al. 2005; Durand et al.
1993; Teske and Nelson 2006). Up to now no pure cultures are available for
Beggiatoa spp. found at hydrothermal vents. However, physiological and in situ
studies have contributed to a better understanding of these organisms, which
build extensive, up to several centimeter thick mats at hydrothermal vents in
Guaymas Basin (Teske and Nelson 2006). Thiomicrospira spp. have been fre-
quently isolated or detected from a variety of deep-sea hydrothermal vents
(Brinkhoff et al. 2005), whereas Halothiobacillus hydrothermalis has been iso-
lated from a deep-sea hydrothermal vent site in the Fiji Basin (Durand et al.
1993). Recently, the complete genome of Thiomicrospira crunogena strain XCL-2
was published, revealing many adaptations that have allowed this organism to
live at the dynamic oxicanoxic interface at deep-sea vents (Scott et al. 2006). As
an example, only a low oxygen adapted cbb3-type cytochrome oxidase could be
identified, indicating that the general habitat of this organism is characterized by
microaerobic conditions.
19.2.2.2 Epsilonproteobacteria
Until recently Thiomicrospira spp. and H. hydrothermalis represented the only pure
cultures of mesophilic, obligately chemolithoautotrophic sulfur-oxidizing bacteria
from deep-sea vents. However, in recent years Epsilonproteobacteria have been
increasingly recognized as important members of the microbial communities at deep-sea
vents, ranging from black smoker chimney walls and associations with invertebrates
(epibionts, and endosymbionts) to the shallow subsurface (Campbell et al. 2006).
These bacteria appear to be predominantly chemolithoautotrophic, and frequently iso-
lates have been obtained that are able to generate energy by oxidizing reduced sulfur
compounds (Takai et al. 2003, 2006; Inagaki et al. 2003; Nakagawa et al. 2005).
In fact, organisms related to Sulfurimonas autotrophica might be the most prevalent
free-living sulfur-oxidizers at deep-sea vents (Inagaki et al. 2003). In the meantime, the
Fig. 19.1 (Continued) through black smokers or mixing in varying proportions with seawater in
the subseafloor before being discharged from the seafloor at diffuse-flow vent sites. The latter
creates a range of physicochemical conditions and energy sources that can be exploited by differ-
ent types of microbes living in the subseafloor. The stylized cell depicts a chemolithoautotrophic
sulfur-oxidizing bacterium that can use oxygen or nitrate as an electron acceptor. The growth of
these organisms in the subseafloor is primarily expected to occur at temperatures between 4 and
50C. (Compiled from Jannasch and Mottl 1985, Jannasch 1995, Huber et al. 2003, Tivey 2004,
and Wirsen 2004)
Table 19.1 Free-living and endosymbiotic sulfur-oxidizing bacteria isolated from or identified at deep-sea hydrothermal vent sites and some of their characteristics
Phylo- Growth Sulfur oxi- Carbon
genetic Isolation/observa- temperature pH Electron Electron dation path- fixation
Organisma affiliation tion siteb (C) range m (h1) G+C donors acceptors way pathway References
Thiomicrospira Vestimentiferan 438.5 5.08.5 0.8 44.2 S0, S2, O2 ND ND Jannasch
crunogena tube worm cas- S2O32 (et al.
strain TH-55 ing, 21N EPR 1985)
Thiomicrospira Vestimentiferan ND ND 0.45 43.1 S0, S2, O2 Sox, SQR CBB Scott et al.
crunogena tube worm cas- S2O32 (2006)
strain XCL-2 ing, Galapagos
Rift
Thiomicrospira Polymetal sulfide 441 5.58.5 0.8 44.6 S0, S2, O2 ND ND Wirsen et al.
crunogena rock, TAG site S2O32 (1998)
strain MA-3 MAR
Thiomicrospira Mussel periostra- 1035 5.58.5 0.2 44.4 S0, S2, O2 ND CBB Ruby and
crunogena cum, Galapagos S2O32 Jannasch
strain L-12 Rift (1982)
Thiomicrospira Hydrothermal 1555 5.08.0 0.7 43.8 S0, S2, O2 APSR and CBB Takai et al.
thermophila fumarole, S2O32, TSO (2004,
strain I78 TOTO caldera, OC activity 2005)
Mariana Arc,
Western Pacific
Halothiobacillus Active hydrother- 1145 6.09.0 0.6 67.4 S0, S2, O2 soxB ND Durand et al.
hydrothermalis mal vent, North S2O32, detected (1993)
Fiji Basin S4O62
2
Beggiatoa spp. Guaymas Basin Mesophile ND ND ND S O2, NO32 APS Rubisco Nelson et al.
activity (1989)
Riftia pachyptila East Pacific Mesophile ND ND ND S2 O2, NO32 APS CBB, Nelson and
ES rTCA Fisher
(1995),
Markert
et al.
(2007)
Calyptogena mag- East Pacific Mesophile ND ND ND S2, S2O32 O2 APS CBB Newton et al.
nifica ES (2006)
Bathymodiolus ther- East Pacific Psychrophile ND ND ND S2, S2O32, O2 ATPS Rubisco Nelson and
mophilus ES OC activity activity Fisher
(1995)
Alviniconcha Western Pacific ND ND ND ND ND ND ATPS Rubisco Stein et al.
hessleri ES activity activity (1988),
Suzuki
et al.
(2005b)
Alviniconcha sp. Western Pacific ND ND ND ND ND ND ND Calvinc Suzuki et al.
type 1 ES (2006)
Ifremeria nautilei Western Pacific ND ND ND ND ND ND ND Rubisco Desbruyeres
ES activity et al.
(1994),
Urakawa
et al.
(2005)
Scaly foot gastropod Indian Ocean Ridge ND ND ND ND ND ND ND ND Goffredi et al.
ES (2004)
Strain TB66 Galapagos vent 1542 ND ND ND S2O32, OC O2 ND ND Teske et al.
(2000)
Strain AG33 Water sample, 1542 ND ND ND S2O32, OC O2, NO32 ND ND Teske et al.
Galapagos vent (2000)
Strain NF18 Beggiatoa mat, 1542 ND ND ND S2O32, OC O2, NO32 ND ND Teske et al.
Galapagos vent (2000)
(continued)
Table 19.1 ( continued)
Phylo- Growth Sulfur oxi- Carbon fix-
genetic Isolation/observa- temperature pH Electron Electron dation path- ation path-
Organisma affiliation tion siteb (C) range m (h1) G+C donors acceptors way way References
Arcobacter spp. 9N EPR, 13N ND ND ND ND S2d O2d ND rTCAd Taylor and
EPR Wirsen
(1997),
Taylor et
al. (1999),
Wirsen et
al. (2002),
Moussard
et al.
(2006)
Sulfurimonas Sediments, Mid- 1040 4.59.0 0.5 35.2 S0, S2, O2 SOR activ- rTCA Takai et al.
autotrophica Okinawa Trough S2O32 ity, Sox?e (2005),
strain OK10 hydrothermal Inagaki
field et al.
(2003)
Sulfurimonas par- Paralvinella nest, 435 5.48.6 0.04 37.6 S0, S2O32, O2, NO32 SOR activ- rTCA Takai et al.
alvinellae strain Iheya North H2 ity, Sox?e (2005,
GO25 field, Mid- 2006)
Okinawa Trough
Sulfurovum Gas bubbling 1040 5.09.0 0.46 48.0 s0, S2O32 O2, NO32- Sor activity rTCA Inagaki
lithotrophicum sediment, Iheya Sox? et al., 2004
strain 42BKT North field,
Mid-Okinawa
Trough
Alviniconcha aff. Indian Ocean Ridge ND ND ND ND ND ND ND rTCA Suzuki et al.
hessleri ES (2005a)
Alviniconcha sp. Western Pacific ND ND ND ND ND ND ND rTCAc Suzuki et al.
type 2 ES (2006)
Alviniconcha sp. ES Manus Basin, ND ND ND ND ND ND ND ND Urakawa
Southwestern et al.
Pacific (2005)
Persephonella AF Rebeccas Roost, 5575 4.77.5 0.09 38.5 S0, S2O32, O2, NO32 ND ND Gtz et al.
guaymasenis Guaymas Basin H2 (2002)
strain EX-H2
Persephonella AF Q vent, 9N EPR 5580 4.77.5 0.14 37.4 S0, S2O32, O2, NO32, ND rTCA Gtz et al.
marina strain H2 S0 (2002),
EX-H1 Hgler
et al.
(2007)
ES endosymbiont, Gammaproteobacteria, Alphaproteobacteria, Epsilonproteobacteria, AF Aquificales, EPR East Pacific Rise, APS adenosine 5-phosphosulfate
pathway, ATPS ATP sulfurylase, APSR adenosine 5-phosphosulfate reductase, sox Sox pathway, SOR sulfite:acceptor oxidoreductase, SQR sulfide:quinone oxdidore-
ductase, TSO thiosulfate oxidase, Rubisco ribulose-1,5-bisphosphate carboxylase/oxygenase, rTCA reductive tricarboxylic acid cycle, CBB CalvinBensonBassham
cycle, ND no data, OC organic carbon.
a
Only one representative of vestimentiferans, vesicomyid clams, and mytilids is listed. No epibionts are listed, even though sulfur oxidation is a likely metabolism.
b
For symbionts, the geographical area of occurrence of the host is given.
c
Isotopic evidence.
d
Based on data obtained with Candidatus Arcobacter sulfidicus.
e
Based on the genome of Sulfurimonas denitrificans.
246 S.M. Sievert et al.
genome of the sulfur-oxidizing bacterium Sulfurimonas denitrificans (formerly known

as Tms. denitrificans; Timmer-Ten Hoor 1975) has been sequenced, representing not
only the first autotrophic, but also the first free-living epsilonproteobacterium for which
a genome sequence is available. Although it was isolated from marine sulfidic sedi-
ments, it appears to share quite a few physiological characteristics with S. autotrophica
and S. paralvinellae, two autotrophic sulfur-oxidizing bacteria isolated from deep-sea
hydrothermal vents that are its closest cultured relatives (Inagaki et al. 2003; Takai et
al. 2006). The genome of S. denitrificans has a size of about 2.2 Mb with a coding den-
sity of approximately 94% (Sievert 2006). As has been observed for Tms. crunogena
(Scott et al. 2006), many features of the genome of S. denitrificans indicate that it has
been streamlined for obligate autotrophy (Sievert 2006). Its genome further reveals a
more versatile metabolism than previously thought, possibly explaining the success of
this and closely related species in various environments (Campbell et al. 2006). The
sensitivity of S. denitrificans towards O2 is well known (Timmer-Ten Hoor 1975), and
on the basis of the genome it actually appears that this bacterium is an obligate denitri-
fier that can also grow at low O2 concentrations. Interestingly, S. denitrificans appears
to rely on a periplasmic nitrate reductase rather than a cytoplasmic membrane-bound
one that operates in all known organisms producing N2 from nitrate (Richardson and
Watmough 1999). In the future, it will be interesting to compare this genome with the
genomes of other free-living epsilonproteobacteria that are currently being sequenced
or have been sequenced (Campbell et al. 2006).
19.2.2.3 Aquificaceae
In addition to Gammaproteobacteria and Epsilonproteobacteria, chemolithoau-

totrophic bacteria of the genus Persephonella within the Aquificaceae have been
isolated from deep-sea hydrothermal vents that are capable of oxidizing reduced
sulfur compounds, as well as H2, under microaerobic conditions and with nitrate at
temperatures up to 80C (Gtz et al. 2002). Interestingly, Persephonella marina
strain EX-H1, the genome of which has been sequenced, can also grow under
anaerobic conditions by using S0 as an electron-acceptor. This indicates that
environmental conditions might be highly dynamic, necessitating the need for
metabolic versatility.
19.2.2.4 Carbon Metabolism in Sulfur-Oxidizing Bacteria
Until recently the CalvinBensonBassham (Calvin) cycle was basically the only
autotrophic carbon-fixation pathway known too in colorless sulfur-oxidizing bacteria
(Table 19.1). However, all autotrophic Epsilonproteobacteria and Aquificae studied to
date use the reductive tricarboxylic acid (rTCA) cycle for converting inorganic carbon
into biomass (Hgler et al. 2005, 2007; Takai et al. 2005; Suzuki et al. 2005a). Since
these organisms appear to constitute an important component of microbial communi-
ties at deep-sea hydrothermal vents, carbon fixation through the rTCA cycle might be
more important in these habitats than previously thought (Campbell et al. 2006; Hgler
et al. 2007). Recently, evidence has also been presented that the gammaproteobacterial
endosymbiont of Riftia pachyptila uses the rTCA cycle for autotrophic carbon fixation
(Markert et al. 2007). The fact that the rTCA cycle requires significantly less ATP and
fewer reducing equivalents to synthesize a three-carbon unit compared to the Calvin
cycle could be of relevance in a potentially energy limiting environment, e.g., for
organisms growing under microaerobic conditions. Based on the simultaneous
presence of reduced sulfur and organic compounds the argument has also been made
that the predominant sulfur-oxidizers at deep-sea hydrothermal vents might actually be
mixotrophs rather than obligate autotrophs (Karl 1995); however, to date this
hypothesis has not been rigorously tested. It might be expected that obligate chemo-
lithoautotrophs are the dominant types in the subseafloor portion of deep-sea vents,
whereas facultative autotrophs or mixotrophs might fare better in dense animal patches
where organic matter concentrations are likely to be higher (Karl 1995; Jannasch
1995). Clearly, more work in this area is needed, e.g., measuring the concentration and
composition of dissolved organic matter at various locations within a given vent field
in parallel with quantifying the different metabolic types.
19.3 Sulfur Oxidation Pathways
19.3.1 Types of Pathways
Significant advances have been made in our understanding of sulfur oxidation

pathways in a variety of sulfur-oxidizing bacteria over the last few years, mainly
by the pioneering genetic studies on the phototrophic sulfur-oxidizing bacterium
Allochromatium vinosum and facultatively autotrophic sulfur-oxidizing bacteria, in
particular Paracoccus pantotrophus GB17 (see Chap. 9 by Grimm et al. and Chap.
12 by Friedrich et al. for further details). These studies coupled with the sequencing
of bacterial genomes have revealed that neutrophilic sulfur-oxidizing bacteria basi-
cally use two types of sulfur oxidation pathways. One involves a multienzyme
complex catalyzing the complete oxidation of reduced sulfur compounds to sulfate
(Sox pathway; Kelly et al. 1997; Friedrich et al. 2001), and the other has sulfite and
elemental sulfur as important intermediates (Kappler and Dahl 2001; Pott and Dahl
1998; Shahak et al. 1999). The biochemistry and the genetic basis of these path-
ways are described in more detail elsewhere in this volume, and only a few aspects
of relevance to this chapter will be discussed here.
The pathway involving sulfite and sulfur as intermediates proceeds from the
transformation of sulfide to polysulfide, most likely via sulfide:quinone reductase
(Shahak et al. 1999). Sulfur globules are formed and again remobilized using
dissimilatory siroheme sulfite reductase to produce sulfite (Pott and Dahl 1998).
Sulfite is either oxidized completely to sulfate by sulfite:acceptor oxidoreductase
(SOR), which is encoded by sorAB, or via adenosine 5-phosphosulfate (APS)
using the enzymes APS reductase (APSR) and ATP sulfurylase (ATPS) in a
reaction akin to a reversal of dissimilatory sulfate reduction (APS pathway;
Kappler and Dahl 2001). The Sox multienzyme complex is capable of oxidizing
sulfide, sulfite, sulfur, and thiosulfate to sulfate (Friedrich et al. 2001). It has been
demonstrated that four proteins are required for a fully functional complex in vitro:
SoxB, SoxXA, SoxYZ, and SoxCD (Friedrich et al. 2001). SoxCD has homologies
to SorAB, but in contrast has been shown to act as a sulfur dehydrogenase (Friedrich
et al. 2001). It has recently been shown that organisms that lack soxCD, but do have
soxB, soxXA, and soxYZ, use the Sox system to oxidize thiosulfate to sulfur, which
is either stored inside the cell or excreted (Hensen et al. 2006). It might well be that
organisms missing the Sox system completely might not be able to use thiosulfate
at all. Acidophilic sulfur-oxidizing bacteria and thermoacidophilic sulfur-oxidizing
archaea use different sulfur oxidation pathways that will not be further discussed
here (see Friedrich et al. 2001).
19.3.2 Endosymbionts
Until recently, the sulfur oxidation pathway of sulfur-oxidizing (endo) symbionts has
remained incompletely characterized (Nelson and Fisher 1995). A major advance in
our understanding of sulfur oxidation in endosymbiotic Gammaproteobacteria came
from a recent proteomic study of the endosymbiont of Riftia pachyptila (Markert
et al. 2007). Besides APSR and ATPS, dissimilatory siroheme sulfite reductase could
be identified and a model based on these data has been proposed that is similar to that
of Chlorobaculum tepidum (formerly Chlorobium tepidum) (Eisen et al. 2002),
except that no sox genes have been found. Possibly, this could explain why thiosul-
fate does not appear to be used by the symbiont. A similar pathway might also be
used by the symbiont of Calyptogena magnifica (Newton et al. 2007). Interestingly,
this symbiont contains sox genes (soxZYXAB) like Chlorobaculum tepidum and
Allochromatium vinosum (Eisen et al. 2002; Hensen et al. 2006), in line with its
capability to use thiosulfate and to store sulfur globules inside the cell (Nelson and
Fisher 1995).
19.3.3 Free-Living Sulfur-Oxidizing Bacteria
19.3.3.1 Beggiatoa
Previous work on Beggiatoa has shown that the obligately chemolithoautotrophic

marine strain MS-81-1c uses the APS pathway involving APSR and ATPS or SOR
to oxidize sulfite (Hagen and Nelson 1997). At present, it is not known how sulfide
is oxidized to sulfite, but it is likely that it proceeds similarly as in phototrophic
sulfur bacteria which are also known to store elemental sulfur inside their cells.
It is expected that Beggiatoa found at hydrothermal vents use the same pathway,
but no direct evidence has been yet obtained. However, the genome of a hydrother-
mal vent Beggiatoa is currently being sequenced as part of the marine microbial
genome sequencing project of the Gordon and Betty Moore Foundation (2007) and
may lead to confirmation of this hypothesis.
19.3.3.2 Thiomicrospira crunogena and Epsilonproteobacteria
Until now there were also no indications of which sulfur oxidation pathway might
be used by Tms. crunogena, H. hydrothermalis, or sulfur-oxidizing Epsilonpro-
teobacteria. New information from recently sequenced genomes, however, points to
the importance of the Sox system in free-living sulfur-oxidizing bacteria at deep-sea
hydrothermal vents. This sulfur oxidation pathway appears to be operating in Tms.
crunogena XCL-2 and S. denitrificans (Scott et al. 2006; Sievert 2006), the latter of
which might be representative for other Epsilonproteobacteria. Details of the Sox
system are described in great detail in Chap. 12 by Friedrich et al. and we wish only
to illustrate the main differences from facultative autotrophic sulfur-oxidizers in
which this pathway has been studied (Friedrich et al. 2001). First of all, the
arrangement of sox genes in Tms. crunogena XCL-2 and S. denitrificans is different.
The sox genes do not occur in one cluster, as in the model organism P. pantotrophus
GB17 (Friedrich et al. 2001), but in different parts of the genome (Fig. 19.2). In this
regard, differences also exist between Tms. crunogena and S. denitrificans. S.
denitrificans has basically two clusters, one containing soxZYXAB and another one
containing soxZYCD (Sievert 2006). In Tms. crunogena one cluster contains
soxZYXA; soxCD and soxB are located elsewhere (Scott et al. 2006). At present the
reasons for this difference in arrangement are not known. Interestingly, both S.
denitrificans and Tms. crunogena have soxCD, although their other sox genes are
more closely related to those of C. tepidum, A. vinosum, and Thiobacillus denitrifi-
cans, which do not have soxCD and produce sulfur globules (Hensen et al. 2006).
Tms. crunogena also forms sulfur from thiosulfate (Javor et al. 1991) and appears to
be the first sulfur-oxidizing bacterium to do so even though it has soxCD. Possibly
soxCD is regulated and turned on and off depending on environmental conditions.
Elemental sulfur formation by S. denitrificans has not been reported.
Recently, sulfur oxidation enzymes were also measured in some other autotrophic
epsilonproteobacteria, including Sulfurimonas autotrophica and Sulfurimonas par-
alvinellae (Takai et al. 2005). In this case, SOR activity was detected using an assay
that would not be expected to measure such activity were these organisms to use the
Sox system (C.G. Friedrich, personal communication). This indicates that other
Sulfurimonas autotrophica and Sulfurimonas paralvinellae might either not use the
Sox system or use a modified version of it. In this regard it is interesting to note, that
the soxC sequence identities of S. denitrificans and Tms. crunogena to the soxC
sequences of those organisms that have the complete sox gene set are significantly
lower than when soxC sequences from organisms with a complete sox gene set are
compared among themselves (44% compared with more than 63%) (Scott et al.
2006; Sievert 2006). Thus, taken together with the fact that the genes are not in a
cluster with the other sox genes (Fig. 19.2), the possibility exists that SoxCD in S.
denitrificans and Tms. crunogena is regulated and functions differently, possibly
exhibiting sulfite dehydrogenase activity. It is also worth mentioning that APSR
activity has been found in Tms. thermophila (Takai et al. 2005), although the gene
coding for this enzyme has not been detected in the genome of Tms. crunogena
XCL-2 (Scott et al. 2006). Possibly, different pathways exist in different
Thiomicrospira species. A soxB has been detected in H. hydrothermalis, indicating
that it might also use the Sox pathway for sulfur oxidation (Petri et al. 2001).
19.3.3.3 Oxidation of H2 by Sulfur-Oxidizing Bacteria
The genome of S. denitrificans revealed the presence of a hydrogenase, indicating

that it might also be able to use H2 as an electron donor. This was confirmed in
subsequent growth experiments (S. Sievert, unpublished data), making S. denitrificans
physiologically similar to S. paralvinellae (Takai et al. 2006). It appears that the
ability to use both H2 and reduced sulfur compounds as electron donors might be
more widespread at deep-sea vents than previously thought, representing a very
useful strategy for living in these dynamic environments (Nakagawa et al. 2005,
Takai et al. 2006).
SoxX SoxY SoxZ SoxA SoxB
SoxZ SoxY SoxD SoxC
SoxX SoxY SoxZ SoxA
SoxC SoxD
SoxB
C
SoxX SoxY SoxZ SoxA SoxB SoxC SoxD
Fig 19.2 Gene arrangement of soxXYZABCD in Sulfurimonas denitrificans (a) and Thiomicrospira
crunogena XCL-2 (b), as opposed to that in the facultative autotrophic sulfur-oxidizing
Paracoccus pantotrophus GB17, which contains a complete sox gene cluster (c)
19.4 Snowblower Vents As Signs for Sulfide Oxidation

in the Subseafloor
19.4.1 The Subseafloor Biosphere
The discovery of microbial populations beneath the deep ocean floor at hydrother-
mal vents has far-reaching implications, ranging from speculation on the origins of
life to the biogeochemistry of the oceans (Summit and Baross 2001). Though the
biota residing in the subseafloor ecosystem potentially has a strong influence on a
variety of biogeochemical processes, it is a relatively poorly defined component of
hydrothermal systems (Wilcock et al. 2004). We have yet to determine the nature
and extent of the microbiology of this ecosystem, its contribution to subseafloor
primary production, and its influence on the geosphere. Consequently, little is
known about the importance of sulfur oxidation in the interior and subsurface
portion of hydrothermal systems, which is in stark contrast to our knowledge of the
importance of this process in supporting dense animal communities above the
seafloor. Potentially, hydrothermal fluids flowing through cracks and pores within
the oceanic crust provide rich environments for subseafloor biological communities
(Fig. 19.1). Indeed, sampling of diffuse-flow vents immediately after volcanic erup-
tions or diking events and during subsequent monitoring points to the existence of
a subseafloor biosphere that might contribute considerably to the primary
productivity of these hydrothermal systems (Holland et al. 2004).
19.4.2 Filamentous-Sulfur Formation in the Laboratory
By simulating the conditions that are likely to occur in the upper microaerobic
subseafloor ecosystem at these vent sites, i.e., active mixing of hydrothermal fluid
containing H2S and oxygenated deep-sea-bottom water resulting in a low O2/high
H2S environment (Fig. 19.1), a new type of H2S-oxidizing bacterium was enriched
from coastal sulfidic sediments (Taylor and Wirsen 1997). This highly motile
vibrioid bacterium is unique among prokaryotes in that it excretes sulfur in filamen-
tous form (filaments 0.52.0-m thick by 20500-m long) as a product of its
metabolism (Taylor and Wirsen 1997; Sievert et al. 2007). This bacterium belongs
to the genus Arcobacter within the Epsilonproteobacteria and has been provision-
ally named Candidatus Arcobacter sulfidicus (CAS) (Wirsen et al. 2002). The fixa-
tion of CO2 occurs via the rTCA cycle at rates equivalent to or higher than those of
other sulfur-oxidizing autotrophs utilizing the Calvin cycle (e.g., Tms. crunogena)
(Wirsen et al. 2002; Hgler et al. 2005). Overall, CAS seems to be well adapted to
the conditions supposedly prevailing in the upper subseafloor portion of deep-sea
hydrothermal vents: the organism is microaerophilic, tolerates high sulfide concen-
trations, is able to fix N2, and the entangling filamentous sulfur it produces forms
mats that permit retention in high fluid flow environments.
19.4.3 Snowblowers
The metabolic capability of forming filamentous sulfur is not limited to the coastal
strain, as it has now been documented to occur in other sulfidic environments
(Sievert et al. 2007). Filamentous sulfur produced by the coastal strain of CAS was
first compared with and found to be morphologically and chemically similar to
white flocculent material collected from extensive discharges, so-called blizzard or
snowblower vents, observed during and after a volcanic eruption at the 9N
deep-sea hydrothermal vent site on the East Pacific Rise (EPR) in 1991 (Nelson et
al. 1991; Haymon et al. 1993; Taylor and Wirsen 1997). In this case the filamentous
material accumulated in mats of up to 5-cm thickness (Nelson et al. 1991; Haymon
et al. 1993). The process of microbial filamentous-sulfur formation was documented
to occur in situ on a colonization device deployed at M vent (Titanium Ring for
Alvinella Colonization, TRAC, at 9N EPR) as well as in shipboard experiments
using inocula collected from these sites (Taylor et al. 1999).
19.4.4 Diversity of Filamentous-Sulfur-Forming Bacteria
We could further confirm that the formation of filamentous sulfur at 9N EPR is also
mediated by Arcobacter spp. 16S ribosomal RNA sequences closely related, but not
identical to CAS were retrieved from the TRAC device (M vent TRAC clones) and
from shipboard reactors (shipboard reactor clones) (Fig. 19.3). It is interesting,
however, that the sequences from the shipboard reactors and the coastal laboratory
strain CAS form one cluster that is distinct from the cluster formed by the sequences
obtained from TRAC (Fig. 19.3). This might be related to temperature, as all reactor
enrichments and CAS cultures were incubated between 20 and 25C, whereas the M
vent TRAC enrichments were exposed to temperatures of 4050C (Taylor et al.
1999). The sequences from the in situ incubation device might have originated from
filamentous-sulfur-producing organisms that grow at higher temperatures, and thus
did not ultimately become established in the shipboard reactors. Recently, the
formation of filamentous-sulfur mats by Arcobacter spp. has also been reported
from the 13N deep-sea hydrothermal vent site on the EPR (Moussard et al. 2006).
The sequences obtained in this study form a cluster with the M vent TRAC clones
(Fig. 19.3, L50-sequences) indicating the existence of hydrothermal vent-specific
filamentous-sulfur-forming Arcobacter populations. Furthermore, arcobacter
sequences have also been detected after an eruptive event in the outflow of a diffuse-
flow vent at Axial Volcano, Juan de Fuca Ridge. (Huber et al. 2003), indicating
the presence of these bacteria in the subseafloor at yet another geographic area
(Fig. 19.3, Marker 33 sequences). All of this indicates that filamentous-sulfur
formation may be an important process at hydrothermal vents, extending into the
shallow subsurface biosphere, driven by inorganic nutrients alone (i.e., H2S and
CO2, N2), and thus contributing to overall organic matter production at deep-sea
hydrothermal vent sites (Fig. 19.1).
Clone L50-WB6
M vent TRAC clone a7
Clone L50-WB53
Marker 33-FL74B00
Vestimentiferan symbiont clone (D83061)
M vent TRAC clone a3 Hydro-
Marker 33-FL88B00 thermal
Marker 33-FL70B00 group
Marker 33-PA62B98
Arcobacter
Deep-sea hydrothermal vent chimney clone CHA3-437
Japan Trench deep-sea desiment clone JTB129
Shipboard reactor clone CB2C3
Shipboard reactor clone CB2E4
Shipboard reactor clone CB2B10
Shipboard reactor clone CB2D1
Candidatus Arcobacter sulfidicus
Marker 33-PA 28B00
Marker 33-FL 76B00
Marker 33-FL 58B00
Arcobacter sp. strain Solar Lake
Black Sea sediment clone B4b1
Arcobacter cryaerophilus
Arcobacter skirrowi
Activated sludge clone T31
Arcobacter butzlerii
Arcobacter nitrofigilis
Oilfield sulfur oxidizer strain FWKO-B
Geospirillum barnesii
Sulfurospirillum arcachonense
Campylobacter jejuni
Vent cap clone VC2.1 Bac31
Alvinella pompejana epibiont clone APG44b
Alvinella pompejana epibiont clone APB13b
Rimicaris exoculata epibiont
Pele's vent clone PVB_OTU_3
Sulfurimonas denitrificans
Helicobacter pylori
Wolinella succinogenes
Riftia pachyptila endosymbiont
Thiomicrospira crunogena
Calyptogena magnifica endosymbiont
0.05
Fig. 19.3 16S ribosomal RNA based neighbor-joining distance tree depicting the phylogenetic
relationship of the filamentous sulfur-producing microbe, Candidatus Arcobacter sulfidicus and
sequences obtained from ship-board reactors and an in situ incubation at M vent at 9N East Pacific
Rise (EPR) (M vent TRAC clones), as well as sequences obtained from a filamentous-sulfur mat at
13N EPR (clones L50-WB6 and L50-WB53) and clones obtained from a diffuse-flow hydrothermal
vent habitat (Marker 33) at Axial Volcano, Juan de Fuca Ridge, during a time series after an eruption
(Huber et al. 2003) to select cultured and environmental proteobacterial sequences. The tree was
constructed with sequences containing at least 1,300 bp by using the phylogenetic software ARB
(Technische Universitt Mnchen 2007). The partial Marker 33 sequences where inserted into the
tree by applying parsimony criteria without allowing for changes in the overall tree topology. The
scale bar represents 0.05 estimated changes per nucleotide
19.5 Conclusions and Outlook
Research conducted at deep-sea hydrothermal vents over the last decade has
revealed that sulfur oxidation is mediated by a diverse group of organisms.
Traditionally, Gammaproteobacteria, either free-living or in a symbiotic associa-
tion, were seen as the main sulfur-oxidizers at these systems. Only recently have
we begun to appreciate the importance of Epsilonproteobacteria for autotrophic
carbon production in general and in particular for sulfur oxidation (Campbell et al.
2006). It might well be that organisms related to Sulfurimonas spp. might be the
predominant sulfur-oxidizers at deep-sea vents (Inagaki et al. 2003). The Sox path-
way emerges as a potentially very important sulfur oxidation pathway at deep-sea
hydrothermal vents. Many autotrophic prokaryotes occurring at deep-sea vents
appear to use alternative CO2-fixation pathways, in particular the rTCA cycle,
questioning the paradigm of the Calvin cycle being at the base of the food chain at
deep-sea hydrothermal vents. Future work on this subject promises to be a very
fruitful area of research.
Along these lines, it will be important to (1) quantify epsilonproteobacteria and

other autotrophic organisms, (2) assess whether these organisms are active in their
environment by analyzing RNA or intact polar lipids (Martinez et al. 2006; Sturt et
al. 2004), and (3) implement studies that couple the identity of microorganisms more
directly with their function, e.g., autotrophic carbon fixation, and the measurements
of CO2-fixation rates, to come to a better understanding of what is driving these
unique ecosystems and to arrive at overall productivity estimates. The methodology
to address the latter has been developed (e.g., Gray and Head 2001; Orphan et al.
2001), but so far has largely not been applied to deep-sea hydrothermal vents, mainly
because conducting these studies at vents represents a great challenge, particularly in
light of the physical, chemical, and biological heterogeneity of these systems.
An important task for the future will also be to better define the contribution of the
subseafloor biosphere to the overall productivity of these ecosystems. It was pointed
out soon after the discovery of the deep-sea hydrothermal vents that bacteria grow
most abundantly in the shallow crust where upwelling hot, reducing hydrothermal
fluid mixes with downwelling cold, oxygenated seawater. The predominant produc-
tion of biomass, however, is the result of symbiotic associations between chemolitho-
autotrophic bacteria and certain invertebrates (Jannasch and Mottl 1985). While the
validity of the first part of the statement has been proven in subsequent studies, we
presently do not have enough information to assess whether most of the production
occurs above or below the seafloor. Clearly, more research in this area is warranted.
Finally, the genomes of representatives of the four dominant groups of sulfur-
oxidizing bacteria at deep-sea vents are available, i.e., endosymbionts, as well as free-
living bacteria belonging to the Gammaproteobacteria, the Epsilonproteobacteria,
or Aquificales. Coupled with metagenomic studies (Xu 2006), these genomes
already have and will continue to provide unique opportunities to advance our
understanding of this important process at deep-sea hydrothermal vents.
Acknowledgements. Preparation of this manuscript was supported by National Science

Foundation grant OCE-0452333 and a fellowship from the Hanse-Wissenschaftskolleg (http://
www.h-w-k.de) to S.M.S. Research conducted in our laboratories (S.M.S., C.D.T., C.O.W.) on
filamentous-sulfur formation and alternative carbon-fixation pathways was supported by National
Science Foundation grants OCE-0452333, IBN-0131557, and IBN-9630054, as well as NASA
Astrobiology Institute grant NNA04CC04A. M.H. was supported through a WHOI postdoctoral
scholarship. Thanks are due to Stephen J. Molyneaux for excellent technical support and for help
in preparing this manuscript, as well as to Jack Cook for preparing Fig. 19.1.
References
Brinkhoff, T, Kuever, J, Muyzer, G, Jannasch, HW (2005) The genus Thiomicrospira. In: Brenner DJ,
Krieg NR, Staley JT (eds) Bergeys manual of systematic bacteriology, vol 2, 2nd edn. Springer,
New York, pp 193199
Campbell BJ, Engel AS, Porter ML, Takai K (2006) The versatile epsilonproteobacteria: Key
players in sulphidic habitats. Nat Rev Microbiol 4:458468
Cavanaugh CM (2006) Comparative genomics of chemosynthetic symbionts. In: American
Society for Microbiology, General Meeting, Orlando
Desbruyeres, D, Alayse-Danet A-M, Ohta S, and the Scientific Parties of BIOLAU and STARMER
Cruises (1994) Deep-sea hydrothermal communities in southwestern Pacific back-arc basins (the
North Fiji and Lau Basins): composition, microdistribution, and food web. Mar Geol 116:227242
Durand P, Reysenbach A-L, Prieur D, Pace N (1993) Isolation and characterization of Thiobacillus
hydrothermalis sp. nov., a mesophilic obligately chemolithoautotrophic bacterium isolated
from a deep-sea hydrothermal vent in Fiji Basin. Arch Microbiol 159:3944
tepidum TLS a photosynthetic, anaerobic, green-sulfur bacterium. Proc Natl Acad Sci USA
99:95099514
inorganic sulfur compounds by bacteria: emergence of a common mechanism. Appl Environ
Goffredi SK, Warn A, Orphan VJ, Van Dover CL, Vrijenhoek RC (2004) Novel forms of struc-
tural integration between microbes and a hydrothermal vent gastropod in the Indian Ocean.
Appl Environ Microbiol 70:30823090
Gordon and Betty Moore Foundation (2007) Microbial genome sequencing project. http://www.
moore.org/microgenome/
Gtz D, Banta A, Beveridge TJ, Rushdi AI, Simoneit BRT, Reysenbach A-L (2002) Persephonella
marina gen. nov., sp. nov. and Persephonella guaymasensis sp. nov., two novel, thermophilic,
hydrogen-oxidizing microaerophiles from deep-sea hydrothermal vents. Int J Syst Evol
Gray ND, Head IM (2001) Linking genetic identity and function in communities of uncultured
bacteria. Environ Microbiol 3:481492
Hagen KD, Nelson DC (1997) Use of reduced sulfur compounds by Beggiatoa spp.: enzymology
and physiology of marine and freshwater strains in homogeneous and gradient cultures. Appl
Haymon RM, Fornari DJ, Von Damm KL, Lilley MD, Perfit M, Edmond JM (1993) Volcanic
eruption of the mid-ocean ridge along the East Pacific Rise crust at 945-52N: Direct
submersible observations of seafloor phenomena associated with an eruption event in April,
(1991). Earth Plant Sci Lett 119:85101
Hensen D, Sperling D, Trper HG, Brune DC, Dahl C (2006) Thiosulphate oxidation in the
phototrophic sulphur bacterium Allochromatium vinosum. Mol Microbiol 62:794810
Holland ME, Baross JA, Holden JF (2004) Illuminating the subseaflloor ecosystem using micro-
bial tracers. In: Wilcock WSD, DeLong EF, Kelley DS, Baross JA, Cary SC (eds) The subsea-
floor biosphere at mid-ocean ridges. Geophysics monographs 144. American Geological
Union, Washington, pp 291304
Huber JA, Butterfield DA, Baross JA (2003) Bacterial diversity in a subseafloor habitat following
a deep-sea volcanic eruption. FEMS Microbiol Ecol 43:393204
Hgler M, Wirsen CO, Fuchs G, Taylor CD, Sievert SM (2005) Evidence for autotrophic CO2 fix-
ation via the reductive tricarboxylic acid cycle in members of the epsilon-Proteobacteria.
Hgler M, Huber H, Molyneaux SJ, Vetriani C, Sievert SM (2007) Autotrophic CO2 fixation via
the reductive tricarboxylic acid cycle in different lineages within the phylum Aquificae:
Evidence for two ways of citrate cleavage. Environ Microbiol 9: 8192
Inagaki F, Takai K, Kobayashi H, Nealson KH, Horikoshi K (2003) Sulfurimonas autotrophica
gen. nov., sp. nov., a novel sulfur-oxidizing -proteobacterium isolated from hydrothermal
sediments in the Mid-Okinawa Trough. Int J Syst Evol Microbiol 53:18011805
Inagaki F, Takai K, Nealson KH, and Horikoshi K (2004). Sulfurovum lithotrophicum gen. nov.,
sp. nov., a novel sulfur-oxidizing chemolithoautotroph within the -Proteobacteria isolated
from Okinawa Trough hydrothermal sediments. Int J Syst Evol Microbiol 54: 14771482.
Jannasch HW (1995) Microbial interactions with hydrothermal fluids. In: Humphris SE,
Zierenberg RA, Mullineaux LS, Thomson RE (eds) Seafloor hydrothermal systems.
Geophysics monographs 91. American Geological Union, Washington, pp 273296
Jannasch HW, Mottl MJ (1985) Geomicrobiology of deep-sea hydrothermal vents. Science
229:717725
Jannasch HW, Wirsen CO, Nelson DC, Robertson LA (1985) Thiomicrospira crunogena sp. nov.,
a colorless, sulfur-oxidizing bacterium from a deep-sea hydrothermal vent. Int J Syst Bacteriol
35:422424
Javor BJ, Wilmot DB, Vetter RD (1990) pH-dependent metabolism of thiosulfate and sulfur glob-
ules in the chemolithotrophic marine bacterium Thiomicrospira crunogena. Arch Microbiol
154:231238
Karl DM (1995) Ecology of free-living, hydrothermal vent microbial communities. In: Karl DM
(ed) Microbiology of deep-sea hydrothermal vents. CRC, Boca Raton, pp 35124
Kelly DP, Shergill JK, Lu W-P, Wood AP (1997) Oxidative metabolism of inorganic sulfur com-
pounds by bacteria. Antonie Van Leuwenhoek 71:95107
Markert S, Arndt C, Felbeck H, Feldman RA, Becher D, Sievert SM, Hgler M, Albrecht D,
Robidart J, Bench S, Hecker M, Schweder T (2007) Approaching the uncultivable endosym-
biont of Riftia pachyptila by physiological proteomics. Science 315:247250
Martinez RJ, Mills HJ, Story S, Sobecky PA (2006) Prokaryotic diversity and metabolically active
microbial populations in sediments from an active mud volcano in the Gulf of Mexico. Environ
McCollom T, Shock EL (1997) Geochemical constraints on chemolithoautotrophic metabolism by
microorganisms in seafloor hydrothermal systems. Geochim Cosmochim Acta 61:43754391
Moussard H, Corre E, Cambon-Bonavita M-A, Fouquet Y, Jeanthon C (2006) Novel uncultured
Epsilonproteobacteria dominate a filamentous sulphur mat from the 13N hydrothermal vent
field, East Pacific Rise. FEMS Microbiol Ecol 58:449463
Nakagawa S, Takai K, Inagaki F, Hirayama H, Nunoura T, Horikoshi K, Sako Y (2005)
Distribution, phylogenetic diversity and physiological characteristics of epsilon-Proteobacteria
in a deep-sea hydrothermal field. Environ Microbiol 7:16191632
Nelson DC, Fisher CR (1995) Chemoautotrophic and methanotrophic endosymbiotic bacteria at
deep-sea vents and seeps. In: Karl DM (ed) Microbiology of deep-sea hydrothermal vents.
CRC, Boca Raton, pp 125167
Nelson DC, Wirsen CO, Jannasch HW (1989) Characterization of large, autotrophic Beggiatoa
spp. abundant at hydrothermal vents of the Guaymas Basin. Appl Environ Microbiol
55:29092917
Nelson D, Haymon RM, Lilley M, Lutz R (1991) Rapid growth of unusual hydrothermal bacteria
observed at new vents during ADVENTURE dive program to the EPR crest at 94552N.
EOS Trans Am Geophys Union 72:481
Newton ILG, Woyke T, Auchtung TA, Dilly GF, Dutton RJ, Fisher MC, Fontanez KM, Lau E,
Stewart FJ, Richardson PM, Barry KW, Saunders E, Detter JC, Wu D, Eisen JA, Cavanaugh
CM (2007) The Calyptogena magnifica chemoautotrophic symbiont genome. Science
315:9981000
Orphan VJ, House CH, Hinrichs K-U, McKeegan KD, DeLong EF (2001) Methane-consuming
archaea revealed by directly coupled isotopic and phylogenetic analysis. Science 293:484-487
Petri R, Podgorsek L, Imhoff JF (2001) Phylogeny and distribution of the soxB gene among
thiosulfate-oxidizing bacteria. FEMS Microbiol Lett 197:171178
Pott AS, Dahl C (1998) Sirohaem-sulfite reductase and other proteins encoded in the dsr locus of
Chromatium vinosum are involved in the oxidation of intracellular sulfur. Microbiol 144:18811894
Richardson DJ, Watmough NJ (1999) Inorganic nitrogen metabolism in bacteria. Curr Opin Chem
Biol 3:207219
Ruby EG, Jannasch HW (1982) Physiological characteristics of Thiomicrospira sp. strain L-12
isolated from deep-sea hydrothermal vents. J Bacteriol 149:161165
Shahak Y, Schtz M, Bronstein M, Hauska G, Padan E (1999) Sulfide-dependent anoxygenic
photosynthesis in prokaryotes: sulfide:quinone reductase (SQR), the intial step. In: Peshek
GA, Lffelhardt W, Schmetterer C (eds) The phototrophic prokaryotes. Kluwer/Plenum, New
York, pp 211228
Scott KM, Sievert SM, Abril FN, Ball LA, Barrett CJ, Blake RA, Boller AJ, Chain PSG, Clark JA,
Davis CR, Detter C, Do KF, Dobrinski KP, Faza BI, Fitzpatrick KA, Freyermuth SK, Harmer
TL, Hauser LJ, Hgler M, Kerfeld CA, Klotz MG, Kong MW, Land M, Lapidus A, Larimer
FW, Longo DL, Lucas S, Malfatti SA, Massey SE, Martin DD, McCuddin Z, Meyer F, Moore
JL, Ocampo LH Jr, Paul JH, Paulsen IT, Reep DK, Ren Q, Ross RL, Sato PY, Thomas P,
Tinkham LE, Zeruth GT (2006) The genome of deep-sea vent chemolithoautotroph
Thiomicrospira crunogena XCL-2. PLoS Biol 4:e383. doi:10.1371/journal.pbio.0040383
Sievert SM (2006) The genome of the sulfur-oxidizing bacterium Thiomicrospira denitrificans: a
model for epsilonproteobacterial autotrophs at vents and other redox interfaces. In: American
Society for Microbiology, General Meeting, Orlando
Sievert SM, Wieringa EBA, Wirsen CO, Taylor CD (2007) Growth and mechanism of filamentous-
sulfur formation by Candidatus Arcobacter sulfidicus in opposing oxygen-sulfide gradi-
ents. Environ Microbiol 9:271276
Stein JL, Cary SC, Hessler RR, Ohta S, Vetter RD, Childress JJ, Felbeck H (1988) Chemoautotrophic
symbiosis in a hydrothermal vent gastropod. Biol Bull 174:373378
Stewart FJ, Newton ILG, Cavanaugh CM (2005) Chemosynthetic endosymbioses: adaptations to
oxic-anoxic interfaces. Trends Microbiol 13:439448
Sturt HF, Summons RE, Smith K, Elvert M, Hinrichs KU (2004) Intact polar membrane lipids in
prokaryotes and sediments deciphered by high-performance liquid chromatography/electro-
spray ionization multistage mass spectrometry new biomarkers for biogeochemistry and
microbial ecology. Rapid Commun Mass Spectrom 18:617628
Summit M, Baross JA (2001) A novel microbial habitat in the mid-ocean ridge subseafloor. Proc
Nat Acad Sci USA 98:21582163
Suzuki Y, Sasaki T, Suzuki M, Nogi Y, Miwa T, Takai K, Nealson KH, Horikoshi K (2005a) Novel
chemoautotrophic endosymbiosis between a member of the Epsilonproteobacteria and the
hydrothermal-vent gastropod Alviniconcha aff. hessleri (Gastropoda: Provannidae) from the
Indian Ocean. Appl Environ Microbiol 71:54405450
Suzuki Y, Sasaki T, Suzuki M, Tsuchida S, Nealson KH, Horikoshi K (2005b) Molecular phylo-
genetic and isotopic evidence of two lineages of chemoautotrophic endosymbionts distinct at
the subdivision level harbored in one host-animal type: the genus Alviniconcha (Gastropoda:
Provannidae). FEMS Microbiol Lett 249:105112
Suzuki Y, Kojima S, Sasaki T, Suzuki M, Utsumi T, Watanabe H, Urakawa H, Tsuchida S,
Nunoura T, Hirayama H, Takai K, Nealson KH, Horikoshi K (2006) Host-symbiont relation-
ships in hydrothermal vent gastropods of the genus Alviniconcha from the southwest Pacific.
Appl Environ Microbiol 72:13881393
Takai K, Inagaki F, Nakagawa S, Hirayama H, Nunoura T, Sako Y, Nealson KH, Horikoshi K
(2003) Isolation and phylogenetic diversity of members of previously uncultivated epsilon-
proteobacteria in deep-sea hydrothermal fields. FEMS Microbiol Lett 218:167174
Takai K, Hirayama H, Nakagawa T, Suzuki Y, Nealson KH, Horikoshi K (2004) Thiomicrospira
thermophila sp. nov., a novel microaerobic, thermotolerant, sulfur-oxidizing chemolithomix-
otroph isolated from a deep-sea hydrothermal fumarole in the TOTO caldera, Mariana Arc,
western Pacific. Int J Syst Evol Microbiol 54:23252333
Takai K, Campbell BJ, Cary SC, Suzuki M, Oida H, Nunoura T, Hirayama H, Nakagawa S,
Suzuki Y, Inagaki F, Horikoshi K (2005) Enzymatic and genetic characterization of carbon
and energy metabolisms by deep-sea hydrothermal chemolithoautotrophic isolates of
Epsilonproteobacteria. Appl Environ Microbiol 71:73107320
Takai K, Suzuki M, Nakagawa S, Miyazaki M, Suzuki Y, Inagaki F, Horikoshi K (2006) Sulfurimonas

paralvinellae sp. nov., a novel mesophilic, hydrogen- and sulfur-oxidizing chemolithoautotroph
within the Epsilonproteobacteria isolated from a deep-sea hydrothermal vent polychaete nest,
reclassification of Thiomicrospira denitrificans as Sulfurimonas denitrificans comb. nov. and
emended description of the genus Sulfurimonas. Int J Syst Evol Microbiol 56:17251733
Taylor CD, Wirsen CO (1997) Microbiology and ecology of filamentous sulfur formation. Science
277:14831485
Taylor CD, Wirsen CO, Gaill F (1999) Rapid microbial production of filamentous sulfur mats at
hydrothermal vents. Appl Environ Microbiol 65:22532255
Technische Universitt Mnchen (2007) The ARB project. http://www.arb-home.de
Teske A, Nelson D (2006) The genera Beggiatoa and Thioploca. In: Dworkin M, Falkow S,
Rosenberg E, Schleifer K-H, Stackebrandt E (eds) Proteobacteria: gamma subclass. The prokary-
otes. A handbook on the biology of bacteria, vol 6, 3rd edn. Springer, New York, pp 784812
Teske A, Brinkhoff T, Muyzer G, Moser DP, Rethmeier J, Jannasch HW (2000) Diversity of
thiosulfate-oxidizing bacteria form marine sediments and hydrothermal vents. Appl Environ
Timmer-Ten Hoor A (1975) A new type of thiosulphate oxidizing, nitrate reducing microorgan-
isms: Thiomocrospira denitrificans sp. nov. Neth J Sea Res 9:344350
Tivey MK (2004) The remarkable diversity of seafloor vents. Oceanus 42:6065
Urakawa H, Dubilier N, Fujiwara Y, Cunningham DE, Kojima S, Stahl DA (2005) Hydrothermal
vent gastropods from the same family (Provannidae) harbor epsilon and gammaproteobacterial
endosymbionts. Environ Microbiol 7:750754
Wilcock WSD, DeLong EF, Kelley DS, Baross JA, Cary SC (eds) (2004) The subseafloor biosphere
at mid-ocean ridges. Geophysics monographs 144. American Geological Union, Washington
Wirsen (2004) Is life thriving deep beneath the seafloor? Oceanus 42:7277
Wirsen CO, Brinkhoff T, Kuever J, Muyzer G, Molyneaux S, Jannasch HW (1998) A new
Thiomicrospira strain from the Mid-Atlantic Ridge compared to known hydrothermal vent
isolates. Appl Environ Microbiol 64:40574059
Wirsen CO, Sievert SM, Cavanaugh CM, Molyneaux SJ, Ahmad A, Taylor LT, DeLong EF, Taylor
CD (2002) Characterization of an autotrophic sulfide-oxidizing marine Arcobacter
sp. that produces filamentous sulfur. Appl Environ Microbiol 68:316325
Xu J (2006) Microbial ecology in the age of genomics and metagenomics: concepts, tools, and
recent advances. Mol Ecol 15:17131731
Chapter 20
Speciation Analysis of Microbiologically
Produced Sulfur by X-ray Absorption
Near Edge Structure Spectroscopy
Alexander Prange
Abstract The first part of this chapter presents a basic and brief introduction to
X-ray absorption spectroscopy with special regard to its application in microbiol-
ogy and its great advantages as an in situ method for speciation analysis. The sec-
ond part summarizes X-ray absorption near edge structure spectroscopic
investigations of microbiologically produced sulfur. Two examples are presented in
more detail, the speciation of sulfur in sulfur globules of phototrophic and chemo-
trophic sulfur bacteria and the speciation of elemental sulfur taken up by
Allochromatium vinosum.
20.1 Introduction
Over the last three decades X-ray absorption spectroscopy (XAS) has developed
as an incisive probe of the local structure around selected atomic species in
solids, liquids, and gases. Foremost among its strengths are its applicability to
amorphous materials and its tunability, which means the ability to probe the
environments of different elements in a sample by selecting a suitable incident
X-ray energy. The amount of information available from a single XAS spectrum
is relatively small compared with that available from X-ray diffraction; however,
the information obtained from a well-chosen experiment can be particularly incisive
and in some cases inaccessible by any other technique. The purpose of this chapter
is to provide (for nonphysicists) a basic and brief introduction to XAS as it is been
traditionally practiced. It is written from the point of view of a user of this technique
and focuses on the information which is of interest and should be mentioned when
just thinking about using XAS to investigate (micro-)biological samples as a
nonphysicist (the author is a microbiologist with special interest in biophysical
methods and using XAS). Furthermore, an overview and some examples of the
259
Springer 2008
260 A. Prange
application of XAS for investigating sulfur speciation analysis1 in microbiological

systems are given.
20.2 XAS: X-ray Absorption Near-Edge Structure

and Extended X-ray Absorption Fine Structure
An XAS spectrum is typically divided into two regions: the X-ray absorption
near-edge structure (XANES) and extended X-ray absorption fine structure
(EXAFS). Although both regions have in principle the same physical origin, the
distinction is convenient for the interpretation of the results obtained by this
spectroscopy. The XANES region is strongly sensitive to formal oxidation state
(effective charge), electronegativity of neighboring atoms, and coordination chemistry
(e.g., octahedral, tetrahedral coordination) of an absorbing atom (fine structure
about 10 eV below and up to about 50100 eV above the absorption edge), while
the EXAFS region can be used to determine the radial distances, coordination
number, and neighbors species of an absorbing atom (fine structure from about 100
to 1,000 eV above the absorption edge). More detailed insight into XAS (including
technical details) can be found in the excellent books by Teo (1986), Stoehr (1996),
and Koningsberger and Prins (1988). For more information on the application of
XAS in biology and applied sciences and an elementary introduction to XAS, the
reader is referred to Behrends (1992a, b), Prange and Modrow (2002), and Prange
et al. (2007). For the latest developments and an overview of scientific questions of
current interest, the conference proceedings of the X-ray absorption fine structure
conferences (XAFS 12 2005; XAFS 13 2007) are recommended.
20.2.1 Experimental
Synchrotron light with a continuous spectrum from infrared light with energy of
approximately 0.03 keV to hard X-rays with energy of approximately 500 keV or
less is obtained when a relativistic electron is accelerated vertically to its direction
of movement. In synchrotron radiation (light) sources of different generations
(in Germany ANKA, Karlsruhe, BESSY, Berlin, DELTA, Dortmund, ELSA,
Bonn, HASYLAB, Hamburg; worldwide about 50 storage rings dedicated to basic
and applied research using synchrotron radiation are in operation; lightsources.org
2006), highly relativistic electrons are stored (in high vacuum) to travel along a
1
The term speciation has often been used in the literature in different ways. In this chapter, the
terms speciation and speciation analysis and chemical species, respectively, are used
according to IUPAC definitions. Speciation analysis: analytical activities of identifying and/or
measuring the quantities of one or more individual chemical species in a sample. Chemical
species: specific form of an element defined as to isotopic composition, electronic or oxidation
state, and/or complex or molecular structure (Templeton et al. 2000).
20 Speciation Analysis of Microbiologically Produced Sulfur 261
circular path for many hours at about 99.9999985% of the speed of light (traveling
approximately 300,000 km s1) and emitting synchrotron radiation tangentially.
In simple XAS experiments, e.g., for investigating sulfur speciation at the sulfur
K-edge, polychromatic X-rays are produced by a synchrotron radiation source and
then monochromatized by diffraction from a double-crystal monochromator
(Lemonnier et al. 1978).
A schematic sketch of an XAS experiment (transmission and fluorescence
modes, see later) is displayed in Fig. 20.1; in Fig. 20.2 an authentic XAS experiment
Fig. 20.1 Experimental setup for X-ray absorption spectroscopy measurements A in transmission
mode and B in fluorescence mode. In transmission mode, fluorescence photons are also present,
but they are not drawn for reasons of clarity. Furthermore, in fluorescence mode, the beam is also
transmitted through the sample, but is also not drawn, nor are the free electrons that occur when
the beam hits the sample (electron yield mode)
262 A. Prange
Fig. 20.2 Experimental setup of the beamline BN3 for X-ray absorption near edge structure
(XANES) spectroscopy measurements in transmission mode: A sample holder, B chamber for the
sample, C manual valve, D first ionization chamber (measuring I0), E second ionization chamber
(measuring I1), F first electrometer (displays I0), G second electrometer (displays I1), H wall of
lead cuboids (radiation protection), and I barometer (measuring the pressure inside the chambers).
The monochromator (not shown) is located behind the wall of lead cuboids
(beamline BN3, synchrotron laboratory at ELSA; Physikalisches Institut Universitt

Bonn 2007) is shown with a transmission mode setup. Only the X-ray photons hit-
ting the first crystal that are of the correct wavelength will be redirected to the sec-
ond crystal, whereas the others are absorbed [Si(111) or InSb(111) crystals are
usually used for probing sulfur]. The parallel second crystal is used as a mirror
mainly to restore the beam to its original direction. The monochromatic X-rays then
pass through the sample, which should be prepared in such a way that it absorbs
approximately half of the incident X-rays.
An X-ray absorption spectrum can be measured in three different modes: in
transmission mode, in fluorescence mode, or in electron yield mode. In transmission
mode, the intensity of the X-ray beam is measured before (I0) (incident beam) and
after passing through the sample (I1) (transmitted beam) using (gas) filled ionization
chambers (detectors), measuring the missing photons. In this case, the number of
X-ray photons absorbed by core electrons to create a photoelectron (and a hole in
the shell) is determined. The XAS spectrum then shows the variation of log (I0/I1)
versus the energy (electronvolts). The detection limit is matrix (sample) and element
concentration dependent and can be estimated with 1% of target element within
a surrounding matrix, even for low-Z elements like sulfur. In fluorescence mode,
the fluorescence (If) from the sample matrix, which is as a secondary effect of the
Fig. 20.3 Sulfur K-edge XANES spectra of sulfur globules of Allochromatium vinosum measured
in transmission mode (black line) and in fluorescence mode (gray line) at the DCM beamline,
Center for Advanced Microstructures and Devices, Baton Rouge, LA, USA (Prange 2002). A.
vinosum was grown photoorganoheterotrophically on malate, then sulfide solution was added, and
formation of sulfur globules started. Spectra were recorded 2 h after sulfide addition and are nor-
malized at 2,510 eV
absorption of X-rays, is determined using a fluorescence detector. In this case, the

number of fluorescence photons emitted from the sample, when an electron in the
upper level fills the hole in the core level, is determined. The resulting XAS
spectrum shows the variation of If/I0 versus the energy (Stoehr 1996). The fluorescence
mode is typically used when very small concentrations of a target element are
present and in favorable cases measurements can be performed even below the parts
per million range.
Both modes, transmission and fluorescence, provide as one can expect comparable
results, when the samples are prepared correctly (Sect. 20.2.3). As an example,
Fig. 20.3 illustrates sulfur K-edge XANES spectra of the purple sulfur bacterium
Allochromatium vinosum (with sulfur globules) measured in both modes (for details
on XANES spectroscopy of bacterial sulfur globules, see Sects. 20.3.1, 20.3.2;
Prange et al. 1999, 2002a, b; Franz et al. 2007)
The third possibility to measure an XAS spectrum, the electron yield mode
(detection of electrons created when the synchrotron light hits an element), has to
the best of authors knowledge never been used to probe microbiologically
produced sulfur. Therefore, this mode is not discussed in this chapter (for an
elementary introduction to the electron yield mode, see Prange et al. 2007).
264 A. Prange
20.2.2 Advantages of XANES Spectroscopy
For speciation analysis of an element of interest in a (micro-)biological sample,

XANES spectroscopy is the technique of first choice. The main advantages, from
a biological point of view are as follows. Nearly all elements of interest (Z 6)
can be investigated separately, independent of the other surrounding elements
within a matrix. Only a relatively small amount of sample is necessary, normally
a few hundred milligrams is adequate. The method is (almost) nondestructive (as
an approximation, starting from 1023 particles, the number of particles in 1 mole,
in a sample and from approximately 1010 photons, which can hit the sample
within 1 s, and a measurement duration of 10 min, the fraction of hit atoms can
be estimated to be only 1010). As X-rays have a high penetration strength,
measuring a sample in a vacuum is normally not necessary. Besides solid
materials it is possible to investigate noncrystalline materials like liquids and
gases, and measurements with high spatial resolution can be performed as well as
measurements of specifically resolved (e.g., chemical) reactions. First results, e.g.,
propositions on valencies, can be easily and directly obtained when using the
fingerprint approach, i.e., the comparison of the XANES spectrum of the sample
with those of suitable reference compounds. Considering all these statements, it
becomes clear that in situ measurements are possible, which is the most important
advantage (for more information, see Prange and Modrow 2002).
20.2.3 Sample Preparation
The preparation of samples in general and of biological samples in particular is a

crucial step. Especially homogeneity, thickness, and concentration of the target
element play a key role. A perfect sample is thin and homogeneous, has a low,
but not too low concentration of the target element and is easy to handle, i.e., a
powder. In general, for all types of samples and the mode of measurement, the
preparation has to be optimized to avoid effects which lead to distorted spectra and
thus false results. All three modes of measuring a XANES spectrum imply pitfalls,
which should be avoided: for transmission mode these are pinhole and/or thickness
effects, for fluorescence mode self-absorption effects, and for electron yield mode
accumulation of charge on the sample. Therefore, only a few general remarks are
given related to preparation of samples for XANES measurements in transmission
and fluorescence modes. Measuring samples in transmission mode, one has to
avoid possible pinhole and/or thickness effects (Parrat et al. 1957; Stern and Kim
1981). Especially when measuring powder samples the risk of artifacts is high at
soft X-ray energies (e.g., for sulfur) owing to pinhole effects. Therefore, in addition
to requiring the right thickness (the X-rays must be able to pass through the
sample!), the sample must be uniformly and homogeneously prepared, and free of
pinholes. If these conditions can be met, which is sometimes not so easy,
measurements in transmission mode are simple to perform and yield excellent data.
For samples with very low concentration of the target element (parts per million
level and lower), fluorescence mode is preferred. However, the key problem related
to this mode is self-absorption. Detailed considerations concerning the measure-
ment modes were recently published (George et al. 2002; Prange et al. 2002c) for
the case of measuring sulfur K-edge XANES spectra of bacterial sulfur globules
and elemental sulfur, and the reader is referred to the detailed information presented
on the advantages and disadvantages of the different modes by these authors.
20.2.4 Quantitative Analysis of XANES Spectra
The interpretation of XANES is complicated by the fact that there is not a simple
analytically exact description of XANES; however, there is much chemical information,
notably formal valency and coordination environment, available from XANES
measurements. For example, the chemical shifts of the so-called white line (first
strong maximum in a XANES spectrum) can be used as a ruler to determine the
valency of sulfur in an unknown compound, just by using the energy position of the
white line (Sect. 3 in Prange et al. 2007; Fig. 20.4). XANES analysis is often based
on linear combinations of known spectra from reference compounds (examples of
XANES spectra of different reference compounds are shown in Fig. 20.4), which
can provide ratios of valency states and/or phases, the so-called quantitative analysis.
The fact that the local environment of the absorbing atoms is probed implies that
XANES spectra are additive, i.e., the spectrum of a mixture of substances A and B
can be composed pf the separately measured spectra of A and B, respectively. This
Fig. 20.4 Sulfur K-edge XANES spectra of different reference compounds: a cyclo-octasulfur,
b polymeric sulfur, c methionine sulfone, d cysteic acid, and e zinc sulfate. Spectra are normalized
at 2,510 eV. The typical energy positions of the white lines for different sulfur species are given
266 A. Prange
additivity is the basis for the quantitative analysis of XANES spectra, which
means the decomposition of a sum spectrum into the components of which it is
composed. To achieve this decomposition, a quality function defined by the
difference between experimental data and a linear combination of spectra contained
in a basis set can be minimized (Modrow et al. 2001; Prange et al. 2002a, 2003).
More sophisticated linear algebra techniques such as factor analysis can also be
(and are) applied to XANES spectra. However, ab initio calculations of all spectral
features of a spectrum of a real sample are still difficult to perform and are not always
reliable. This situation is improving, but at this point a fully quantitative treatment of
XANES using ab initio calculations is rarely available (Rehr and Ankudinov 2001).
20.3 Sulfur K-Edge XANES Spectroscopy and Speciation

of Microbiologically Produced Sulfur
Although XANES spectroscopy has become a more or less routine technique in

physics, chemistry, environmental sciences, and geology, only some studies apart from
EXAFS investigations on metalloproteins originating from microorganisms have
been performed in the field of microbiology in general and only very few studies
of sulfur speciation (and of other low-Z elements) in particular.
For more information on sulfur K-edge XANES spectroscopy, a detailed section
is given in the article by Prange et al. (2007) highlighting the possibilities and limits
when investigating different sulfur compounds. In general, when regarding the
interests of a microbiologist working with sulfur bacteria, XANES spectroscopy
can determine in situ the valency and valencies, respectively, of sulfur (and of
course other elements) within a sample (Fig. 20.5). It can distinguish between
different modifications of sulfur with one formal valency, e.g., sulfur chains of
different lengths (up to four sulfur atoms) (Chauvistr et al. 1997) and between
Fig. 20.5 Sulfur K-edge XANES spectrum of sulfide feeding solution, prepared according to
Siefert and Pfennig (1984). The spectrum is normalized at 2,510 eV
polymeric sulfur and sulfur rings of different ring sizes (Prange et al. 1999; Franz
et al. 2007). It also yields information on/determines atoms bound to the sulfur
atom in the second and the third coordination shell as well as the bond itself
(Chauvistr et al. 1997; Prange et al. 2007). Furthermore, it can if one has the
correct or at least very similar reference samples determine the quantitative
speciation (Sect. 20.2.4). Before presenting our current knowledge of speciation of
microbiologically produced sulfur obtained by XANES spectroscopy, the sulfur
K-edge XANES spectrum of a typical feeding sulfide solution for sulfur bacteria
(Siefert and Pfennig 1984) shown in Fig. 20.5 should be considered. This example
illustrates the great potential of XANES spectroscopy to elucidate the sulfur specia-
tion, which might be of great interest for microbiologists working with and feeding
sulfur bacteria in the laboratory.
The spectrum reveals at least five dominant sulfur species (sulfide, S0/disulfide,
sulfoxide, sulfone, sulfonate; cf., Fig. 20.4) present in the solution instead of only
one species, namely, sulfide, which might be expected in such a solution (Prange
2002). However, it has to be kept in mind that this is qualitative information. For
relative percentages of single sulfur species contributing to the spectrum, a quantitative
analysis (Sect. 20.2.4) must be performed.
Table 20.1 summarizes the studies performed so far on sulfur bacteria and
microbiologically produced sulfur by using XANES spectroscopy; two examples
are discussed in the following sections.
20.3.1 Speciation of Sulfur in Sulfur Globules of Phototrophic

and Chemotrophic Sulfur Bacteria
The most detailed studies using sulfur K-edge XANES spectroscopy have been
performed on sulfur in the sulfur globules of the purple sulfur bacterium A. vinosum
(Table 20.1). Furthermore, the sulfur in globules or granules of some other pho-
totrophic and chemotrophic sulfur bacteria has been investigated in detail (Prange
et al. 2002a; Table 20.1). A detailed description of investigations of bacterial sulfur
globules, including a historical outline from the early beginning in the nineteenth
century, is given in Dahl and Prange (2006). XANES spectroscopy revealed at least
three different sulfur speciations in bacterial sulfur globules, reflecting the different
ecological and physiological properties of different metabolic groups of bacteria:
cyclo-octasulfur dominates in the sulfur globules of Beggiatoa alba and
Thiomargarita namibiensis. In the chemotrophic sulfur bacterium Acidithiobacillus
ferrooxidans (grown at pH 2) sulfur occurs predominantly as polythionates. In
sulfur globules of purple and green sulfur bacteria, the stored sulfur mainly consists
of sulfur chains, most probably terminated by an organic group at one or both ends
(mono-organylsulfanes/bisorganylsulfanes) (Prange et al. 2002a).
Here, the investigation of sulfur globules in intact cells of A. vinosum versus
isolated sulfur globules is briefly presented, as this is an excellent example to
268 A. Prange
Table 20.1 Overview of investigations characterizing microbiologically produced sulfur by X-ray

absorption near edge structure spectroscopy
Microbiologically
Microorganism produced sulfur References
Allochromatium vinosum Speciation of sulfur in Prange et al. (1999, 2002ac),
DSMZ 180T intercellulary stored Pickering et al. (2001),
sulfur globules George et al. (2002), Dahl
and Prange (2006)
Elemental sulfur uptake Franz et al. (2007)
Thiocapsa roseopersicina Speciation of sulfur in Prange et al. (1999, 2002a)
DSMZ 219, Marichromatium intercellulary stored
purpuratum DSMZ 1591T sulfur globules
Amoebobacter purpureus Pickering et al. (2001)
Chlorobaculum parvuma DSMZ Speciation of sulfur in Prange et al. (1999, 2002a)
263 extracellulary stored
sulfur globules
Chlorobaculum tepidumb Pickering et al. (2001)
Chloroflexus aurantiacus Speciation of sulfur in Pickering et al. (2001)
extracellulary stored
sulfur globules
Halorhodospira halophila Speciation of sulfur in Prange et al. (1999, 2002a)
DSMZ 244T, Halorhodospira extracellulary stored
abdelmalekii DSMZ 2110 sulfur globules
Beggiatoa alba DSMZ 1416 Speciation of sulfur stored Prange et al. (2002a)
intracellularly
Marine Beggiatoa Pickering et al. (2001)
Thiomargarita namibiensis Speciation of sulfur stored Prange et al. (2002a)
intracellularly
Acidithiobacillus Speciation of sulfur stored Prange et al. (2002a)
ferrooxidans DSMZ 584 intracellularly
Candidatus Acrobacter Sulfur filaments Prange (2002), A. Prange
sulfidicus and S.M. Sievert
(unpublished data)
Acidianus ambivalens Sulfur filaments A. Prange, A. Kletzin, H.
Lichtenberg and J. Hormes
(unpublished data)
Epsilonproteobacteria Sulfur oxidation states in Lopez-Garcia et al. (2003)
bacterial filaments
(micro-XANES)
Epsilonproteobacteria and Sulfur speciation in Engel et al. (2007)
Gammaproteobacteria filaments (natural
samples)
Thermoanaerobacter Sulfur filaments Prange (2002),
sulfurigignens Lee et al. (2007a, b)
Phomopsis viticola DSMZ S0 formed by this fungus A. Prange
1789 (unpublished data)
Oscillatoria limnetica Intracellularly deposited A. Prange and J. Rethmeier
sulfur granules (unpublished data)
XANES X-ray absorption near-edge structure.
a
Formerly Chlorobium tepidum.
b
Formerly Chlorobium vibrioforme (f. thiosulfatophilum).
clearly point out the necessity to use a nondestructive and in situ method like
XANES spectroscopy. Measurements of isolated (isolated under aerobic conditions)
sulfur globules from anaerobically grown A. vinosum showed completely different
spectra from those of intact cells (also prepared under aerobic conditions) (Prange
et al. 2002a). Quantitative analysis of these spectra with suitable reference
compounds (Sect. 20.2.4) showed that sulfur is predominantly present as
cyclo-octasulfur and minor as sulfate. This is in contrast to the chain sulfur structure
for sulfur globules of intact cells of A. vinosum (about 80% sulfur chains; about 20%
CSH/CSSC). During extraction of the globules from the cells, the integrity of
the cells was destroyed and, therefore, the sulfur was directly exposed to oxygen
from the air, probably leading to changes in the chemical speciation.
20.3.2 Speciation of Elemental Sulfur Taken Up by A. vinosum
Elemental sulfur (S0) has the formal valency of zero; however, elemental sulfur
tends to catenate and to form chains with various lengths (S or S) and ring
sizes (Sn) (Steudel 2000; Steudel and Eckert 2003). All sulfur allotropes are hydro-
phobic, not wetted by water, and they hardly dissolve in water. They can be inves-
tigated by XANES spectroscopy and distinguished according to differences in the
spectral features (Prange et al. 1999). The thermodynamically most stable form of
elemental sulfur at ambient temperature and pressure is cyclic, orthorhombic a-
sulfur (a-S8) (cyclo-octasulfur or S8 rings) (Roy and Trudinger 1970). At 20C pure
a-S8 has a green-yellow color, turning to white after cooling to 80C (Steudel
1996). In contrast, the customary commercial typical elemental sulfur (flowers of
sulfur) remains yellow after cooling to 80C (Steudel 1996). It mainly consists
of S8 rings, some polymeric sulfur chains, and traces of S7 rings which are respon-
sible for the yellow color (Steudel and Holz 1988). Elemental sulfur sublimed at
ambient pressure (flowers of sulfur) always contains some polymeric sulfur
(Steudel and Eckert 2003). Polymeric sulfur, which is frequently used in the rubber
industry for vulcanization of natural and synthetic rubbers, consists of chainlike
macromolecules. The bonding energy between SS bonds in polymeric sulfur,
however, is relatively weak (2.4 kJ mol1 weaker than in cyclo-octasulfur) (Steudel
et al. 1985; Steudel 1996; Steudel and Eckert 2003); therefore, chainlike sulfur
(polymeric sulfur), might be the more easily accessible species of elemental sulfur
for microorganisms. This hypothesis gains some support from the recent study of
Urich et al. (2006), who investigated the influence of different sulfur species on
enzyme functions in the sulfur oxygenase-reductase from Aquifex aeolicus.
Theoretical considerations on the basis of the crystal structure of this enzyme led
to the hypothesis that linear sulfur but not cyclic sulfur species can serve as a
substrate for this enzyme (see Chap. 15 by Kletzin).
Following this hypothesis, Franz et al. (2007) investigated by XANES spectroscopy
whether only one sulfur species is used or at least preferred when A. vinosum takes
up elemental sulfur (flowers of sulfur) and forms globules. A. vinosum took up only
270 A. Prange
a part of the elemental sulfur added to the medium, formed sulfur globules, and
oxidized them to sulfate. Some sulfur remained as sulfur platelets in the medium. The
sulfur used and oxidized by A. vinosum was quantified via the sulfate formed and
compared with the contents of polymeric sulfur and cyclo-octasulfur, respectively,
determined by XANES spectroscopy. It was shown that A. vinosum uses only the
polymeric sulfur (sulfur chain) fraction of elemental sulfur and is probably unable to
take up and form sulfur globules from S8 rings. Probably, the speciation of elemental
sulfur plays a key role in bacterial sulfur oxidation in a more general way and Franz
et al. (2007) hypothesize that sulfur chains are also the microbiologically preferred
form when elemental sulfur is taken up by other microorganisms.
Acknowledgements. The results presented were partly gained in collaboration with different
colleagues, who are gratefully acknowledged: J. Hormes and H. Modrow and the SyLi group
(Institute of Physics, University of Bonn); the X-ray spectroscopy group of the Center for Advanced
Microstructures and Devices (Louisiana State University, Baton Rouge); C. Dahl, B. Franz, and
H.G. Trper (Institute for Microbiology & Biotechnology, University of Bonn). J. Hormes is
thanked for many helpful discussions and for critical reading of the manuscript.
References
Behrens P (1992a) X-ray absorption spectroscopy in chemistry. II. X-ray absorption near edge
structure. Trends Anal Chem 11:237244
Behrens P (1992b) X-ray absorption spectroscopy in chemistry. I. Extended X-ray absorption fine
structure. Trends Anal Chem 11:218222
Chauvistr R, Hormes J, Hartmann E, Etzenbach N, Hosch R, Hahn J (1997) Sulfur K-shell
photoabsorption spectroscopy of the sulfanes R-Sn-R, n=24. Chem Phys 223:293302
Dahl C, Prange A (2006) Bacterial sulfur globules: occurrence, structure and metabolism.
In: Shively M (ed) Inclusions in prokaryotes. Microbiology monographs. Springer, Berlin,
pp 2151
Engel AS, Lichtenberg H, Prange A, Hormes J (2007) Speciation of sulfur from naturally-occuring,
filamentous microbial mats from sulfidic cave springs using X-ray absorption near edge spec-
troscopy. FEMS Microbiol Lett 269:5462
Franz B, Lichtenberg H, Hormes J, Modrow H, Dahl C, Prange A (2007) Utilization of solid
elemental sulfur by the phototrophic purple sulfur bacterium Allochromatium vinosum:
a sulfur K-edge X-ray absorption spectroscopy study. Microbiology 153:12681274
George GN, Pickering IJ, Yu EY, Prince RC (2002) X-ray absorption spectroscopy of bacterial
sulfur globules. Microbiology 148:22672268
Koningsberger DC, Prins R (eds) (1988) X-ray absorption: principles, applications, techniques of
EXAFS, SEXAFS and XANES. Wiley, New York
Lee Y-J, Dashti M, Prange A, Rainey FA, Rohde M, Whitman WB, Wiegel J (2007a)
Thermoanaerobacter sulfurigignens sp. nov., a novel anaerobic thermophilic bacterium reducing
1 M thiosulfate to elemental sulfur and tolerating 90 mM sulfite. Int J Syst Evol Microbiol
57:14291434
Lee Y-L, Prange, A, Lichtenberg H, Rohde M, Dashti M, Wiegel J (2007) In situ speciation of
sulfur globules from thiosulfate reduction by Thermoanaerobacter sulfurgignens and
Thermoanaerobacterium thermosulfurigenes. J Bacteriol ( in Press)
Lemonnier M, Collet O, Depautex C, Esteva JM, Raoux D (1978) High vacuum two crystal soft
X-ray monochromator. Nucl Instrum Methods Phy Res Sect A 152:109111
lightsources.org (2006) lightsources.org home. http://www.lightsources.org

Lopez-Garcia P, Duperron S, Philippot P, Foriel J, Susini J, Moreira D (2003) Bacterial diversity
in hydrothermal sediment and epsilonbacterial dominance in experimental microcolonizers at
the Mid-Atlantic Ridge. Environ Microbiol 5:961976
Modrow H, Visel F, Zimmer R, Hormes J (2001) Monitoring thermal oxidation of sulfur crosslinks
in SBR-elastomers by quantitative analysis of sulfur K-edge XANES-spectra. Rubber Chem
Technol 74:281294
Parrat LG, Hemstead CF, Jossem EL (1957) Thickness effect in absorption spectra near absorption
edges. Phys Rev 105:37813787
Pickering IJ, George GN, Yu EY, Brune DC, Tuschak C, Overmann J, Beatty JT, Prince RC (2001)
Analysis of sulfur biochemistry of sulfur bacteria using X-ray absorption spectroscopy.
Biochemistry 40:81388145
Prange A (2002) Molekulargenetische und physikalisch-chemische Untersuchungen an den
Schwefelkugeln photo- und chemotropher Schwefelbakterien. Verlag dissertation.de, Berlin
Prange A, Modrow H (2002) X-ray absorption spectroscopy and its application in biological,
agricultural and environmental research. Rev Environ Sci Biotechnol 1:259276
Prange A, Arzberger I, Engemann C, Modrow H, Schumann O, Trper HG, Steudel R, Dahl C,
Hormes J (1999) In situ analysis of sulfur in the sulfur globules of phototrophic sulfur bacteria
by X-ray absorption near edge spectroscopy. Biochim Biophys Acta 1428:446454
Prange A, Chauvistr R, Modrow H, Hormes J, Trper HG, Dahl C (2002a) Quantitative specia-
tion of sulfur in bacterial sulfur globules: X-ray absorption spectroscopy reveals at least three
different species of sulfur. Microbiology 148:267276
Prange A, Dahl C, Trper HG, Behnke M, Hahn J, Modrow H, Hormes J (2002b) Investigation of
S-H bonds in biologically important compounds by sulfur K-edge X-ray absorption spectroscopy.
Eur Phys J D 20:589596
Prange A, Dahl C, Trper HG, Chauvistr R, Modrow H, Hormes J (2002c) X-ray absorption
spectroscopy of bacterial sulfur globules: a detailed reply. Microbiology 148:22682270
Prange A, Birzele B, Krmer J, Modrow H, Chauvistr R, Hormes J, Khler P (2003)
Characterization of sulfur speciation in low molecular weight subunits of glutenin after reoxi-
dation with potassium iodate and potassium bromate at different pH values using X-ray
absorption near-edge structure (XANES) spectroscopy. J Agric Food Chem 51:74317438
Prange A, Hormes J, Modrow H (2007) X-ray absorption spectroscopy as a tool for the detection
and identification of sulfur compounds in photototrophic organisms. In: Hell R, Dahl C,
Leustek T, Knaff D (eds) Sulfur metabolism in phototrophic organisms. Advances in photo-
synthesis and respiration. Springer, Berlin (in press)
Physikalisches Institut Universitt Bonn (2007) Webseite der Bonner Synchrotronstrahlungsgrup
pe. http://syli04.physik.uni-bonn.de/
Rehr JJ, Ankudinov AL (2001) Progress and challenges in the theory and interpretation of X-ray
spectra. J Synchrotron Radiat 8:6165
Roy AB, Trudinger PA (1970) The biochemistry of inorganic compounds of sulfur. Cambridge
University Press, London
Siefert E, Pfennig N (1984) Convenient method to prepare neutral sulfide solution for cultivation
of phototrophic sulfur bacteria. Arch Microbiol 139:100101
Stern EA, Kim K (1981) Thickness effect on the extended-X-ray-absorption-fine-structure amplitude.
Phys Rev B 23:12281232
Steudel R (1996) Das gelbe Element und seine erstaunliche Vielseitigkeit. Chemie in Unserer Zeit
30:226234
Steudel R (2000) The chemical sulfur cycle In: Lens P, Hulshof Pol L (eds) Environmental
technologies to treat sulfur pollution. IWA, London, pp 131
Steudel R, Holz B (1988) Detection of reactive sulfur molecules (S6, S7, S9, Sm) in commercial
sulfur, in sulfur minerals, and in sulfur metals slowly cooled to 20C. Z Naturforsch B
43:581589
Steudel R, Eckert B (2003) Solid sulfur allotropes. In Steudel R (ed) Elemental sulfur and sulfur-rich
compounds I. Topics in current chemistry 230. Springer, Berlin, pp 179
272 A. Prange
Steudel R, Strauss R, Koch L (1985) Quantitative HPLC-Analyse und Thermodynamik der

Schwefelschmelze. Angew Chem 97:5859
Stoehr J (1996) NEXAFS spectroscopy. Springer series in surface sciences, vol 25. Springer,
Berlin
Templeton DM, Ariese F, Cornelis R, Danielsson L-G, Muntau H, van Leeuwen HP, obinski R
(2000) Guidelines for terms related to chemical speciation and fractionation of elements.
Definitions, structural aspects, and methodological approaches. IUPAC recommendations
2000. Pure Appl Chem 72:14531470
Teo BK (1986) EXAFS: basic principles and data analysis. Inorganic chemistry concepts, vol 9.
Springer, Berlin
Urich T, Gomes CM, Kletzin A, Frazo C (2006) X-ray structure of a self-compartimentalizing
XAFS 12 (2005). Proceedings of the 12th international X-ray absorption fine structure conference
(XAFS 12), Malm, Sweden, June 2327, 2003. Phys Scr T115
XAFS 13 (2007). Conference proceedings of the 13th international conference on X-ray absorption
fine structure, Stanford, CA, USA, July 914, 2006. Am Inst Phys Proc Ser (in press)
Chapter 21
Controls on Isotope Fractionation During
Dissimilatory Sulfate Reduction
Joost Hoek, Donald E. Canfield
Abstract Sulfur isotopes are fractionated during dissimilatory sulfate reduction;

therefore, sulfur isotope studies have been useful in elucidating the role of sulfate
reduction in the sulfur cycle, and in understanding the early evolution of the sulfur
metabolism on Earth. Sulfur isotope fractionation during sulfate reduction occurs
when sulfate is transported into the cell and is reduced to sulfide through a series
of reversible reactions operating with different efficiencies and distinct fractiona-
tion factors. The magnitude of fractionation depends on the relationship between:
(1) the exchange of sulfate across the cell membrane and (2) the exchange of sulfur
between the different internal sulfur pools. Sulfate exchange and sulfate reduction
rates are controlled by different environmental factors that include temperature, and
electron donor and sulfate concentration. Recent considerations of the minor iso-
topes of sulfur, 33S and 36S, provide new insights into the controls on biological
fractionation of sulfur isotopes. This approach has provided a mechanism for dis-
tinguishing the relative contributions of different sulfur metabolisms to the sulfur
isotope record in modern and ancient environments. While recent experimental and
modeling work has improved our understanding of the factors controlling fraction-
ation during sulfate reduction, it has also highlighted several unanswered questions.
For example, precise fractionation factors for individual enzymatic reduction steps
remain unknown. Furthermore, the reversibility of sulfate transport and the revers-
ibility of sulfite reduction in vivo have never been tested. These factors play critical
roles in controlling isotope fractionation during sulfate reduction and affect the
predictive success of any isotope fractionation model.
21.1 Introduction
There are four stable isotopes of sulfur, including the major isotopes 32S and 34S,
with natural abundances of 95.04 and 4.20%, respectively, and the minor isotopes
33
S and 36S, with natural abundances of 0.749 and 0.0156%, respectively. The
relative abundances of sulfur isotopes in nature deviate from these values as a result
of biological and inorganic processes that involve the transformation of sulfur
compounds. These isotope fractionation processes have traditionally been grouped
273
Springer 2008
274 J. Hoek, D.E. Canfield
into several mechanistic categories that include equilibrium and kinetic proc-
esses. The basis for equilibrium fractionation is related to mass-dependent dif-
ferences in bond energies between light and heavy isotopes. For two chemical
species at equilibrium, fractionations result from the minimization of free energy
associated with isotope exchange reactions (Bigeleisen and Mayer 1947; Urey
1947). Kinetic fractionations on the other hand comprise a much larger group of
reactions that are characterized by unidirectional reactions and transport. Kinetic
fractionations result from mass-dependent differences in vibrational energies of
the transition state, those of the reactants, and also the reaction path and its rela-
tion to the potential energy surface that describes the different states (Bigeleisen
and Wolfsberg 1958). They have also been described in terms of the relationship
between velocity and kinetic energy for isotopically substituted species (Mook
2000). Although fractionations resulting from metabolic (biological) processes
are typically considered to result from kinetic fractionations, Johnston et al.
(2005a) make a distinction between fractionations resulting from reactions that
are intrinsic to individual chemical and physical processes, and multistep meta-
bolic and biological processes, which may include multiple equilibrium and
kinetic fractionation effects.
The isotopic composition of sulfur in a sample is always expressed relative to
the major isotope, 32S, with the following d notation:
{(
d 34 S =

3x
S 32
)
S
sample
( 3x
S 32
)
S
standard }
1 1000, (21.1)
where 3xS is 33S, 34S, or 36S, and standard refers to a reference sample (Caon
Diablo Troilite), which has the well-constrained natural isotope abundances, men-
tioned above. Fractionations between two sulfur pools are expressed exactly in
terms of , with units of per mil:
e A B = 1000 (a A B 1) , (21.2)
where (AB) is the fractionation factor between two different sulfur pools A and B.
It has long been observed that organisms metabolizing sulfur compounds,
particularly during dissimilatory sulfate reduction, fractionate sulfur isotopes (Thode
et al. 1951). On the basis of these observations, the isotopic composition of sulfur
compounds in nature has been used to elucidate the role of microbial metabolisms
in the cycling of sulfur, both in modern environments and in ancient environ-
ments preserved in the geologic record. Because the cycling of sulfur is involved
in atmospheric oxygen regulation, sulfur isotopes have also been used to decipher
the history of atmospheric oxygen and, consequently, the oxidation state of
Earths surface environments (Canfield and Teske 1996). This chapter expands on
previous reviews of the biogeochemistry of sulfur isotopes by Canfield (2001a) and
Brchert (2004) by focusing on recent experimental and modeling work, particularly
with the inclusion of the minor sulfur isotopes, which has contributed significant
additional insights into our understanding of the mechanisms of sulfur isotope
21 Controls on Isotope Fractionation During Dissimilatory Sulfate Reduction 275
fractionation during dissimilatory sulfate reduction and sulfur compound dispro-

portionation. A consideration of additional research steps that should be taken, as
revealed by the recent research efforts, will be discussed.
21.2 Sulfur Isotope Fractionation During Dissimilatory

Sulfate Reduction
21.2.1 Pure Cultures
The sequential reduction of sulfate to sulfide during dissimilatory sulfate reduction

leads to a fractionation of sulfur isotopes. The fractionation of 34S during dissimila-
tory sulfate reduction by pure cultures has been extensively studied, especially for
mesophilic Desulfovibrio species (Thode et al. 1951; Harrison and Thode 1958;
Kaplan and Rittenberg 1964; Kemp and Thode 1968; Smejkal et al. 1971; Chambers
et al. 1975; McCready 1975; Bttcher et al. 1999; Bolliger et al. 2001; Detmers et al.
2001; Habicht et al. 2005; Johnston et al. 2005a; Canfield et al. 2006). Results from
these pure-culture studies show wide-ranging fractionations of 34S from 3 to 46,
with an average around 18. While some of this variability results from inherent
differences between organisms (Bolliger et al. 2001; Brchert et al. 2001; Detmers
et al. 2001; Kleikemper et al. 2004), environmental variables such as temperature,
electron donor type and concentration, and sulfate concentration exert significant
control on the magnitude of sulfur isotope fractionation for individual species of
sulfate-reducing microorganisms (Harrison and Thode 1958; Kaplan and Rittenberg
1964; Kemp and Thode 1968; Chambers et al. 1975; Habicht et al. 2005; Canfield
et al. 2006).
Research documenting the influence of different environmental variables on
fractionation has shown that for individual species of sulfate-reducing micro-
organisms, the extent of fractionation depends on factors that influence cell spe-
cific rates of sulfate reduction (expressed in moles per cell per unit time).
Growth temperature, for example, has a significant impact on cell-specific rates
of sulfate reduction. It is well known that cell-specific rates of sulfate reduction
decrease when sulfate reducers are grown at temperatures below their optimal
growth temperature. It has generally been observed that when an organic elec-
tron donor is supplied during fractionation experiments, the magnitude of frac-
tionation increases with decreasing cell-specific sulfate reduction rates that
result from lower growth temperatures (Harrison and Thode 1958; Kaplan and
Rittenberg 1964; Kemp and Thode 1968; Chambers et al. 1975). Different tem-
perature-controlled fractionation patterns, however, have also been observed.
Canfield et al. (2006) conducted fractionation experiments using a temperature
gradient block where they grew Desulfovibrio desulfuricans on lactate, over the
complete range of growth temperatures. They observed a positive trend between
sulfate reduction rates and fractionation. Furthermore, the highest fractionations
were observed at the highest and lowest growth temperatures. Hoek et al. (2006)
performed similar temperature-gradient experiments with a chemolithoau-
totrophic and thermophilic sulfate-reducing bacterium. They obtained similar
fractionation patterns, with the highest fractionations occurring at the lowest and
the highest growth temperatures.
Cell-specific sulfate reduction rates are also controlled by electron donor
concentration. It is generally observed that fractionations decrease with increasing
sulfate reduction rates resulting from increasing concentrations of both organic
substrates and hydrogen (Kaplan and Rittenberg 1964; Hoek et al. 2006).
Interestingly, when H2 is used as an electron donor, fractionations are signifi-
cantly reduced when compared with fractionations produced during sulfate
reduction with organic compounds. The reasons for reduced fractionations with
H2 are unclear, but Kaplan and Rittenberg (1964) suggest that with H2 the
reduction of sulfate to sulfite (through adenosine 5-phosphosulfate, APS) is rate-
limiting, allowing only limited expression of the fractionation during subsequent
enzymatic reductions downstream from this step. It is important to note, however,
that most experiments with H2 as an electron donor have been conducted in batch
culture with H2-saturated headspace. Hoek et al. (2006) measured fractionation
during H2-limited sulfate reduction. They found that fractionations increased
from approximately 3 to 37 when H2 supply was changed from nonlimiting to
limiting growth conditions. Their results highlight the importance of electron
donor concentrations in controlling the magnitude of isotope fractionations. In
addition to lower fractionations with growth on H2, suppressed fractionations
have also been observed under low sulfate concentrations (below about 200 M)
(Harrison and Thode 1958; Habicht et al. 2002).
21.2.2 Natural Populations
The direct determination of isotope fractionation during sulfate reduction for

natural populations of sulfate-reducing bacteria has been explored for microbial
mats and marine sediments (Habicht and Canfield 1996, 1997, 2001; Canfield et al.
2000; Canfield 2001b). In these experiments, natural populations were incubated at
in situ temperatures and amended with a variety of organic substrates and a range
of sulfate concentrations. These studies measured maximum fractionations
(approximately 45) within the same range as is typically observed in pure-culture
studies. The low fractionations frequently observed in pure cultures were not
measured in natural populations metabolizing under in situ conditions, and this is
attributed to the generally lower specific rates of sulfate reduction for natural
populations. Canfield (2001b) correlated the extent of fractionation of sulfur
isotopes by natural populations of sulfate-reducing microorganisms with the
concentration of organic substrate and the specific sulfate reduction rates, with
excess organic substrate producing higher specific rates of sulfate reduction and
reduced fractionations.
21.3 Stepwise Reduction of Sulfate and Sulfur Isotope

Fractionation Models
The fractionation of sulfur isotopes during dissimilatory sulfate reduction results

from a series of sequential biochemical reactions that operate at different efficiencies
and with different fractionation factors. The biosynthetic pathways and associated
fractionation processes of dissimilatory sulfate reduction have been extensively
studied (Harrison and Thode 1958; Peck 1961; Kaplan and Rittenberg 1964; Kemp
and Thode 1968; Rees 1973). On the basis of the observed isotope fractionation
trends summarized above, Rees (1973) developed a kinetic model that describes the
principal steps in the sulfate reduction process as
SO 4 2 ( in ) SO32 e

SO 4 2 ( out )
ATP 2 e 3

APS

1
4
H 2 S. (21.3)

In this reaction network, sulfate is actively taken up by the cell together with
sodium ions or protons to preserve charge balance (step 1). This occurs via
membrane-bound transport proteins, and is reversible (Cypionka 1995), allowing
exchange of sulfate in and out of the cell. A small isotope fractionation of 3 to
0 (eSO4(out)SO4(in)) is thought to be associated with this step. Once sulfate enters
the cell, it is activated with ATP by ATP sulfurylase to form APS (step 2), which
is reduced to sulfite (step 3) by APS reductase. Steps 2 and 3 are both considered
reversible. No fractionation is expected with the activation of sulfate, but a 22
25 isotopic fractionation (eSO4SO3) is assigned to APS reduction to sulfite
(Harrison and Thode 1957, 1958). The final reduction of sulfite to hydrogen
sulfide along step 4 occurs by the dissimilatory sulfite reductase. Although sulfite
reductase enzymes catalyze the oxidation of sulfide to sulfite in oxidative metab-
olisms (Dahl and Trper 1994), the reversibility of sulfite reduction (step 4) in
vivo has never been demonstrated (Canfield 2001a). A 25 isotope fractionation
(eSO H S) has been ascribed to this step (Kemp and Thode 1968; Rees 1973).
3 2
According to the model proposed by Rees (1973), the overall isotope fractiona-
tion expressed during sulfate reduction depends greatly on which steps limit the
sulfate reduction process. If sulfate exchange across the cell membrane is rate-
limiting, then most, if not all, the sulfate entering the cell will be reduced and only
minimal fractionation will be expressed. Conversely, fractionation will be maximized
when isotope exchange between the reversible steps is maximized. In most isotope
fractionation models, this is best achieved when the microbial metabolism is
suppressed. In this case, all the fractionations associated with the individual steps
will be preserved and expressed in sulfide that leaves the cell. Based on these
principles, Canfield et al. (2006) constructed a quantitative model that can be used
to interpret all the observed fractionation patterns summarized in Sect. 21.2. The
model builds on the reaction network for sulfate reduction originally developed by
Rees (1973), and formalized by Farquhar et al. (2003):
4 b ,j 4 ,a 4 3,j ,a
SO 4 2 ( out ) SO 4 ( in )
1,j1 ,a1 4a
2
APS SO32 3 3 H 2 S. (21.4)
2 ,j 2 ,a 2 5a 5 b ,j 5 ,a 5
The numbers designate different steps, represents mass flow, and a is the fractiona-
tion factor associated with each step. Canfield et al. (2006) used the same fractiona-
tion values as those used by Rees (1973). Branching points within the network control
mass balance where material flow has two possible paths. The first branching point
is defined for the transport of sulfate across the cell membrane and a second branch-
ing point is defined for the extent to which sulfite formation is reversible. The mass
flow of sulfur at each branch point is described by a set of flux terms, f3 and f5. For
the first branch point, f3 = j3/(j3+j2) describes the fraction of sulfur leaving the cell as
sulfide. For the second branch point, f5 = j5/(j5 + j3) describes the fraction of sulfite
that is further reduced to sulfide. Using these terms and isotope mass balance,
Canfield et al. (2006) developed a set of equations that describe the influence of f3 and
f5, and the isotopic composition of internal and external sulfate, on the isotopic com-
position of sulfide resulting from sulfate reduction. Depending on the exact relation-
ship between the extent to which (1) sulfate is exchanged across the cell membrane
and (2) sulfur exchanges between the internal sulfur pools, Canfield et al. (2006) were
able to reproduce all observed fractionation patterns. Although f3 and f5 are not unique
for a given fractionation value, generally speaking, there is a much greater range of
possible f5 than f3 for any given fractionation value (Hoek et al. 2006; Fig. 21.1). This
implies that the exchange of sulfate in and out of the cell varies less, and exerts a
greater influence on the extent of fractionation than the exchange of internal sulfur
reservoirs below fractionations of about 20 (eSO4(out)H2S).
Brunner and Bernasconi (2005) recently proposed an alternative to the Rees
network summarized above. Their model differs from that developed by Rees (1973)
0.9
0
0.8
0.7
0.6
10
f3 0.5
0.4
0.3 20
0.2
30
0.1
40
0
0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1
f5
Fig. 21.1 The range of possible f3 and f5 values for given fractionation values. Selected
(SO (out)H S) values of each line are shown. There is only a single possible f3,f5 pair for the extreme
4 2
fractionation values of 3 (f3,f5=1,0) and 47 (f3,f5=0,0). (From Hoek et al. 2006)
in that it incorporates a possible trithionate pathway in the reduction of sulfite, and

that sulfite reduction is considered reversible. Additionally, they propose much larger
fractionation factors (SO3H2S = 58) during the sulfite reduction step. The implica-
tions of this model are that maximum fractionations of more than 70, as may occur
in nature, are possible during dissimilatory sulfate reduction. Owing to the complexity
of this network, however, it is difficult to rigorously test this model using only 34S
isotope analyses. This alternative model can, however, be evaluated by incorporating
the minor isotopes 33S and 36S in addition to 34S (see below).
21.4 Multiple Sulfur Isotopes
The minor isotopes of sulfur, 33S and to a lesser degree 36S, have recently been
included in isotope fractionation studies of dissimilatory sulfate reduction and
sulfur-compound disproportionation (Farquhar et al. 2003; Johnston et al. 2005a, b).
Only limited work has been done with minor sulfur isotopes owing to their low
natural abundance making analyses technically difficult. Results of these studies
provide additional insights into the controls on biological fractionation of sulfur
isotopes.
Mass-dependent fractionations result from mass differences between the differ-
ent sulfur isotopes, with 33S fractionating close to half as much (0.515) as 34S and
36
S fractionating about twice as much (1.91) as 34S, compared with 32S. For 33S this
mass-dependent relationship is reflected as the well-constrained fractionation array
of 33S0.51534S that is observed in the geologic record. It is generally recognized,
however, that this linear relationship is an approximation derived from the power
law of the isotope fractionation factors ():
( )
0.515
a 33 32 = a 34 32 . (21.5)
Because of the dependence of isotope fractionations on the natural logarithm of the

fractionation factors, Mook (2000) defined a factor to describe mass-dependent
fractionations:
q = ln ( 33
a AB ) ln ( 34
)
a AB . (21.6)
Similarly, Miller (2002) recommends expressing observed mass-dependent frac-

tionation in logarithmic form and defines the factor
l=
(
ln 1 + d 33 SA 1000 ) (1 + d 33
SB 1000 )
. (21.7)
ln (1 + d 34
SA 1000 ) (1 + d 34
SB 1000 )
In many cases = , but because measured values of represent net quantities
that can include the fractionation effects of different processes with different frac-
tionation factors (e.g. biological networks), and material transfer, in all cases.
Values for 3334q in mass-dependent fractionations range from 0.500 to 0.516 for
different types of sulfur isotope fractionation processes (e.g., equilibrium, kinetic,

gravitational). Values of 3334 produced by equilibrium isotope exchange of sulfur
between sulfur species have been theoretically constrained to be near 0.515 for 33
34
, while kinetic processes are more variable and fall between 0.500 and 0.515
depending on the nature of the fractionation process producing isotopic fractiona-
tions (Farquhar et al. 2003; Johnston et al. 2005a). Because l values depend on the
structure of sequential chemical reactions in a reaction network and on material
transfer, they can be more variable than values of q for the different steps in a given
network. It is the small differences between q and l that provide the theoretical
framework for modeling mass-dependent fractionations of multiple sulfur isotopes
in biological systems.
In a recent study, Farquhar et al. (2003) explored the isotope fractionations
of the four sulfur isotopes during dissimilatory sulfate reduction. They devel-
oped a theoretical framework that describes the mass flow of sulfur through the
metabolic reaction network (see Eq. 21.4) from Rees (1973), using a set of flux
terms, f3 and f5. f3 = j3/(j3 + j2) and f5 = j5/(j5 + j3), where j is the mass flow of
material along a particular path of the network. f3 indicates the amount of mate-
rial leaving the cell as hydrogen sulfide (pathway 3 in Eq. 21.4) and f5 repre-
sents the internal backflow of sulfur (pathways 5a and 5b in Eq. 21.4). These
flux terms are similar to those used by Canfield et al. (2006) discussed in Sect.
21.3; however, Canfield et al. (2006) defined f5 as the fraction of sulfur that is
further reduced to sulfide. Therefore, when the f5 defined by Canfield et al.
(2006) is 1, the f5 from Farquhar is 0. As discussed in Sect. 21.3, the values of
f3 and f5 depend on different environmental factors, such as temperature and
electron donor and electron acceptor concentrations, as well as on inherent dif-
ferences between organisms, such as the fractionation factors and activity of
the different enzymatic reduction steps. Because these factors affect the magni-
tude and direction of material flow through the network (quantified by f3 and
f5), they directly influence values of l.
Farquhar et al. (2003) introduced a flow net that is contoured by f3 and f5
(Fig. 21.2) to illustrate the dependence of l on the intracellular cycling of sulfur
(f5) and the exchange of sulfate across the cell membrane (f3) during dissimilatory
sulfate reduction. The contour at f5 = 0 reflects the case where transport of sulfate
into the cell and activation of sulfate to APS is limiting and the contour at f3 = 0
reflects the case where the enzymatic reduction of sulfite to hydrogen sulfide is
limiting. Fractionation of 33S and 34S by Archaeoglobus fulgidus produced a range
of 3334l averaging around 0.5117. These values are distinct from 3334l values
calculated for equilibrium exchange and values observed in nature, but can be
accounted for by the reaction network summarized in Sect. 21.3 (Farquhar et al.
2003). Interestingly, the isotope results for A. fulgidus plot along the contour of
f3=0.4 and variable f5 for experiments which were run with different sulfate
concentrations, suggesting that f5 responds to changing concentrations of internal
sulfate. Furthermore, f5 approached 1 as sulfate concentrations increased, and f5
decreased as sulfate concentrations decreased. The implications of these results
are that internal enzymatic reaction rates rather than sulfate exchange across the
Fig. 21.2 The effect of the dissimilatory sulfate reduction network (Rees 1973) on multiple iso-
tope fractionations between hydrogen sulfide and sulfate. Fractionation data from several different
sulfate reduction experiments are plotted along the contours for different values of f3 and f5.
(Modified from Johnston et al. 2005a)
cell membrane control the net fractionation. Similar results were obtained by
Johnston et al. (2005a) with Desulfovibrio jorgensii grown in batch culture, while
the isotopic relationships produced by batch culture experiments with Desulfovibrio
autotrophicum evolved in a different way. Fractionation patterns from D.
autotrophicum follow the contour of f5 = 0.6 and variable f3 (Fig. 21.2), which
suggests that transport of sulfate across the cell membrane varied more than was
observed with A. fulgidus and D. jorgensii. Johnston et al. (2005a) suggest that
this may result from constantly changing sulfate concentrations as the growth in
the batch cultures progressed. Interestingly, one of the data points for D.
jorgensenii falls outside the predicted flow net developed by Farquhar et al.
(2003), which suggests that at least one of the fractionation factors used in the
reaction network (Eq. 21.4) is inaccurate.
An important consequence from this analysis is that l will be less than q for
all values of f other than 0 and 1 if the fractionation factors are less than 1.00.
Conversely, when the fractionation factors are greater than 1, l will be greater
than for all values of f other than 0 and 1. In more complex networks where
some fractionation factors are greater than 1 and others are less than 1, such as in
sulfur-compound disproportionation, l can be greater than or less than q depending
on the values of f and the relative magnitude of the different fractionation factors.
One of the implications of this treatment is that different sulfur metabolisms,
such as dissimilatory sulfate reduction and sulfur-compound disproportionation,
produce resolvable and different 33S/32S fractionations for similar magnitudes of

34 32
S/ S fractionations (Johnston et al. 2005b). This provides an important tool for
distinguishing the relative contribution to the sulfur cycle of biological sulfate
reduction and sulfur-compound disproportionation. This application has subse-
quently been used by Johnston et al. (2005b) to show that sulfur-compound
disproportionation was an active part of the sulfur cycle by 1,300 million years
ago, predating earlier estimates by several hundred million years (Canfield and
Teske 1996).
21.5 Conclusions and Future Research
Although the classic literature from the 1950s and 1960s laid the foundation for
our understanding of sulfur isotope fractionation during dissimilatory sulfate
reduction, recent experimental and modeling work has provided significant
advances in our understanding of the controls on fractionation during sulfate
reduction. While early work focused almost exclusively on pure-culture experi-
ments with a few Desulfovibrio species, recent work has provided a broader
survey of fractionations imposed during sulfate reduction by pure cultures under
a wider range of growth conditions. Furthermore, much work has also focused
on measuring fractionation by natural populations of sulfate reducers. These
experiments have improved our understanding of the environmental factors
controlling fractionation as well as placing a possible limit on the extent of
fractionation during sulfate reduction in nature. Additional critical advances in
our understanding of isotope fractionation during sulfate reduction have come
from incorporating multiple isotope analyses in fractionation experiments.
Although technically challenging, results from these studies have provided a
unique perspective on the factors controlling fractionation. This work has
stimulated serious efforts to develop predictive models for the extent of isotope
fractionation during sulfate reduction as well as for sulfur-compound
disproportionation. These models have already been used to decipher the relative
contributions of dissimilatory sulfate reduction and sulfur-compound
disproportionation to the cycling of sulfur preserved in the geologic record.
Despite the critical advances made in recent years, the new models that have
been proposed highlight several uncertainties that still remain unanswered. For
example, fractionation factors for individual enzymatic reduction steps during
sulfate reduction have never been precisely measured despite the critical reli-
ance of any quantitative fractionation model on accurate fractionation values.
Furthermore, the extent to which sulfate transport into the cell is reversible and
whether or not the dissimilatory sulfite reductase operates in reverse have never
been tested. All these factors play a critical role in controlling the extent of iso-
tope fractionation during sulfate reduction and seriously affect the predictive
success of any isotope fractionation model.
References
Bigeleisen J, Mayer MG (1947) Calculation of equilibrium constants for isotopic exchange

reactions. J Chem Phys 15:261267
Bigeleisen J, Wolfsberg M (1958) Theoretical and experimental aspects of isotope effects in
chemical kinetics. Adv Chem Phys 1:1576
Bolliger C, Schroth MH, Bernasconi SM, Kleikemper J, Zeyer J (2001) Sulfur isotope fractiona-
tion during microbial sulfate reduction by toluene-degrading bacteria. Geochim Cosmochim
Acta 65:32893298
Bttcher ME, Sievert S, Kver J (1999) Fractionation of sulfur isotopes during dissimilatory
reduction of sulfate at 60C. Arch Microbiol 172:125128
Brchert V (2004) Physiological and ecological aspects of sulfur isotope fractionation during
bacterial sulfate reduction. In: Amend JP, Edwards KJ, Lyons TW (eds) Sulfur biogeochemistry.
Special paper 379. Geological Society of America, Boulder, p 205
Brchert V, Knoblauch C, Jorgensen BB (2001) Controls on stable sulfur isotope fractionation
during bacterial sulfate reduction in Arctic sediments. Geochim Cosmochim Acta
65:763776
Brunner B, Bernasconi SM (2005) A revised isotope fractionation model for dissimilatory sulfate
reduction in sulfate reducing bacteria. Geochim Cosmochim Acta 69:47594771
Canfield DE (2001a) Biogeochemistry of sulfur isotopes. In: Valley JW, Cole DR (eds) Stable
isotope geochemistry, vol 43. Mineralogical Society of America, Washington, pp 607636
Canfield DE (2001b) Isotope fractionation by natural populations of sulfate-reducing bacteria.
Geochim Cosmochim Acta 65:11171124
Canfield DE, Teske A (1996) Late Proterozoic rise in atmospheric oxygen concentration inferred
from phylogenetic and sulphur-isotope studies. Nature 382:127132
Canfield DE, Habicht KS, Thamdrup B (2000) The Archean sulfur cycle and the early history of
atmospheric oxygen. Science 288:658661
Canfield DE, Olesen CA, Cox RP (2006) Temperature and its control of isotope fractionation by
a sulfate-reducing bacterium. Geochim Cosmochim Acta 70:548561
Chambers LA, Trudinger PA, Smith JW, Burns MS (1975) Fractionation of sulfur isotopes by
continuous cultures of Desulfovibrio desulfuricans. Can J Microbiol 21:16021607
Cypionka H (1995) Solute transport and cell energetics. In: Barton LL (ed) Sulfate-reducing
bacteria. Plenum, New York, pp 151184
Dahl C, Trper HG (1994) Enzymes of dissimilatory sulfide oxidation in phototrophic sulfur
bacteria. Methods Enzymol 243:400421
Detmers J, Bruchert V, Habicht KS, Kuever J (2001) Diversity of sulfur isotope fractionations by
sulfate-reducing prokaryotes. Appl Environ Microbiol 67:888894
Farquhar J, Johnston DT, Wing BW, Habicht KS, Canfield DE, Airieau S, Thiemens MH (2003)
Multiple sulphur isotopic interpretations of biosynthetic pathways: implications for biological
signatures in the sulphur isotope record. Geobiology 1:2736
Habicht KS, Canfield DE (1996) Sulphur isotope fractionation in modern microbial mats and the
evolution of the sulphur cycle. Nature 382:342343
Habicht KS, Canfield DE (1997) Sulfur isotope fractionation during bacterial sulfate reduction in
organic-rich sediments. Geochim Cosmochim Acta 61:53515361
Habicht KS, Canfield DE (2001) Isotope fractionation by sulfate-reducing natural populations and
the isotopic composition of sulfide in marine sediments. Geology 29:555558
Habicht KS, Gade M, Thamdrup B, Berg P, Canfield DE (2002) Calibration of sulfate levels in the
Archean ocean. Science 298:23722374
Habicht KS, Salling LL, Thamdrup B, Canfield DE (2005) Effect of low sulfate concentrations on
lactate oxidation and isotope fractionation during sulfate reduction by Archaeoglobus fulgidus
strain Z. Appl Environ Microbiol 71:37703777
Harrison AG, Thode HG (1957) The kinetic isotope effect in the chemical reduction of sulphate.
Trans Faraday Soc 53:16481651
Harrison AG, Thode HG (1958) Mechanism of the bacterial reduction of sulphate from isotope
fractionation studies. Trans Faraday Soc 54:8492
Hoek J, Reysenbach AL, Habicht KS, Canfield DE (2006) Effect of hydrogen limitation and
temperature on the fractionation of sulfur isotopes by a deep-sea hydrothermal vent sulfate-
reducing bacterium. Geochim Cosmochim Acta 70:58315841
Johnston DT, Farquhar J, Wing BA, Kaufman A, Canfield DE, Habicht KS (2005a) Multiple
sulfur isotope fractionations in biological systems: a case study with sulfate reducers and sul-
fur disproportionators. Am J Sci 305:645660
Johnston DT, Wing BA, Farquhar J, Kaufman AJ, Strauss H, Lyons TW, Kah LC, Canfield DE
(2005b) Active microbial sulfur disproportionation in the Mesoproterozoic. Science
310:14771479
Kaplan IR, Rittenberg SC (1964) Microbiological fractionation of sulphur isotopes. J Gen
Microbiol 34:195212
Kemp ALW, Thode HG (1968) Mechanism of bacterial reduction of sulphate and of sulphite from
isotope fractionation studies. Geochim Cosmochim Acta 32:7191
Kleikemper J, Schroth MH, Bernasconi SM, Brunner B, Zeyer J (2004) Sulfur isotope fractiona-
tion during growth of sulfate-reducing bacteria on various carbon sources. Geochim
Cosmochim Acta 68:48914904
McCready RGL (1975) Sulfur isotope fractionation by Desulfovibrio and Desulfotomaculum
species. Geochim Cosmochim Acta 39:13951401
Miller MF (2002) Isotopic fractionation and the quantification of O-17 anomalies in the oxygen
three-isotope system: an appraisal and geochemical significance. Geochim Cosmochim Acta
66:18811889
Mook WG (2000) Environmental isotopes in the hydrological cycle, vol I. Introduction theory,
methods, review. UNESCO/IAEA, Geneva, p 280
Peck HD (1961) Evidence for reversibility of reaction catalyzed by adenosine 5-phosphosulfate
reductase. Biochim Biophys Acta 49:621624
Rees CE (1973) A steady state model for sulphur isotope fractionation in bacterial reduction proc-
esses. Geochim Cosmochim Acta 37:11411162
Smejkal V, Cook FD, Krouse HR (1971) Studies of sulfur and carbon isotope fractionation with
microorganisms isolated from springs of western Canada. Geochim Cosmochim Acta
35:787800
Thode HG, Kleerekoper H, McElcheran DE (1951) Sulphur isotope fractionation in the bacterial
reduction of sulphate. Res Lond 4:581582
Urey HC (1947) The thermodynamic properties of isotopic substances. J Chem Soc May
562581
Chapter 22
Bioprocess Engineering of Sulfate Reduction
for Environmental Technology
Piet N.L. Lens, Roel J.W. Meulepas, Ricardo Sampaio, Marcus Vallero,
Giovanni Esposito
Abstract Sulfate reduction can be used in a large number of environmental

technologies. Methanogenic bioreactors treating organic wastewater containing
sulfate can be negatively affected by the sulfide produced; however, it is possible to
combine methanogenesis and sulfate reduction when adequate measures are
applied. For the treatment of inorganic wastewaters containing sulfate, organic
substrates or H2/CO2 are added as electron donors. Alternatively synthesis gas or
methane can be used; however, the sulfate reduction rates with methane are still
extremely low. Heavy metals such as Cu, Zn, Cd, Pb, Ni and Fe can be removed
from waste streams by precipitation with biogenic sulfide. Because of differences
in solubility products the metals can be selectively precipitated. The insoluble metal
sulfides formed can be recovered in order to be reused.
22.1 Introduction
Several of the microbial conversions of the sulfur cycle can be implemented for
pollution control (Table 22.1). This chapter overviews environmental technology
applications that utilize the metabolism of sulfate-reducing bacteria (SRB) as the
key process. Technological utilization of SRB sounds at first somewhat contro-
versial, as sulfate reduction has for many years been considered unwanted, since
the production of H2S causes a multitude of problems, such as toxicity, corro-
sion, odour, increase of the liquid effluent chemical oxygen demand (COD), as
well as reduced quality and amount of biogas (Lens et al. 1998a). The emphasis
of the research in the 19701980s was therefore mainly on the prevention or
minimization of sulfate reduction during methanogenic wastewater treatment
(Colleran et al. 1995). From the 1990s, interest has grown in applying sulfate
reduction for the treatment of specific waste streams, e.g. inorganic sulfate rich
wastewaters such as acid mine drainage, metal-polluted groundwater and flue
gas scrubbing waters.
285
Springer 2008
286 P.N.L. Lens et al.
Table 22.1 Overview of applications in environmental biotechnology that mainly utilize conversions
from the microbial sulfur cycle
Application Sulfur conversion utilized Typical waste stream
Wastewater treatment
Removal of oxidized sulfurous S-oxyanion reduction to Industrial wastewaters, acid
compounds (sulfate, sulfite S2, followed by sulfide- mine drainage and spent
and thiosulfate) removal step sulfuric acid
Sulfide removal Partial S2 oxidation to So Industrial wastewaters
Heavy metal removal SO42 reduction Extensive treatment in wet-
lands or anaerobic ponds
High-rate reactors for process
water, acid mine drainage
and ground water
Nitrogen removal S2, S0 and S2O32 oxidation Domestic wastewater
Removal of xenobiotics SO42 reduction Textile wastewaters
Microaerobic treatment Internal sulfur cycle in a Domestic sewage
biofim
Off-gas treatment
Biofiltration of gases Oxidation of S2 and organo- Biogas, malodorous gases
sulfur compounds from composting and
farming
Treatment of scrubbing waters SO42 and/or SO32reduction, Scrubbing waters of SO2-rich
plus partial S2 oxidation gasses
to S0
Solid-waste treatment
Reduction of waste sludge Internal sulfur cycle in a Sulfur cycle in biofilms
production biofim
Desulfurization of resources Organo-sulfur oxidation Waste rubber, coal, oil, LPG,
spent caustic acid
Bioleaching of metals S2 oxidation Sewage sludge, compost
Gypsum processing SO42 reduction Waste gypsum depots
Treatment of soils and sedi-
ments
Bioleaching of metals S2 oxidation Dredged sediments and spoils
Phytoextraction SO42 uptake by plants Dredged sediments and spoils
Degradation of xenobiotics SO42reduction PCB-contaminated soil
slurries
22.2 Sulfate Reduction in Methanogenic Wastewater

Treatment
Despite the problems associated with sulfate reduction in anaerobic wastewater

treatment, methanogenic treatment of sulfate-rich wastewater is possible if adequate
measures that allow the integration of sulfate reduction with methanogenesis are
applied (Fig. 22.1, Table 22.2). Trends in industries to close water cycles lead to
the accumulation of salts (including sulfates) and heat in the wastewaters, and thus
22 Bioprocess Engineering of Sulfate Reduction for Environmental Technology 287
Fig. 22.1 Process configurations integrating methanogenesis with sulfate reduction and sulfide
removal
Table 22.2 Process technological measures to reduce the reactor sulfide concentration, thus
allowing the integration of methanogenesis and sulfate reduction in anaerobic bioreactors
Dilution of the influent
Non-sulfate-containing process water
Recycle of effluent after a sulfide removal step by sulfide stripping, sulfide precipitation,
biological sulfide oxidation to elemental sulfur (Thiobacillus sp., oxygen; Thiobacillus
denitrificans, nitrate; Chlorobium limicola, sunlight), chemical oxidation to elemental sulfur
(ferric sulfate/silicone supported reactor)
Decrease of the unionized sufide concentration
Elevation of the reactor pH
Elevation of the reactor temperature
Precipitation of sulfide, e.g. with iron salts
Stripping of the reactor liquid using high degree of mixing inside the reactor, recirculation of
biogas after scrubbing, other stripping gas (e.g. N2)
Separation of sufide production and methanogenesis
Two-stage anaerobic digestion with sulfate-reducing bacteria in the acidifying stage
Upflow staged sludge bed with methanogenic bacteria in the bottom and sulfate-reducing
bacteria in the top compartment
Selective inhibition of sulfate-reducing bacteria
Sulfate analogues (e.g. mobybdate)
Transition elements (e.g. copper addition)
Antibiotics
impose the need for the methanogenic treatment of hot and saline wastewaters that
also contain moderate (13 g l1) sulfate levels. Several studies have reported the
feasibility of thermophilic treatment of sulfate-rich wastewaters containing
low-energy substrates, e.g. a 1:1:1 mixture of acetatepropionatebutyrate at 55C,
acetate at 70C, methanol at 65 and 70C, and formate at 75C (Vallero et al. 2007).
For wastewaters rich in unacidified organic matter, additional precautions are
required to cope with the potential reactor acidification or deterioration of the
granular sludge quality due to the excessive growth of acidifiers (Verstraete et al.
1996). One way to overcome these problems is to separate the acidifying and
methanogenic activities in phased or staged reactor designs. If sulfate is present in
the wastewater, sulfate reduction will occur together with acidification in the
acidification phase (Reis et al. 1995) or in the first stages of upflow staged sludge
bed reactors (Lens et al. 1998b). A complete sulfate reduction in the first stage or
phase together with high gas (CO2) production rates during acidification may result
in high H2S-stripping efficiencies, and thus in high sulfur-removal efficiencies in
the acidifying reactor or compartment. Studies on thermophilic (55C) granular
sludge reactors operated under acidifying (pH 6) conditions showed that SRB can
coexist with acidifiers during the treatment of a sucrosepropionatebutyrate
mixture (ratio 2:1:1 on a COD basis) with a COD-to-sulfate ratio of 6.7 (Sipma
et al. 2000) or in synthetic cardboard production wastewater with a COD-to-sulfate
ratio of 10 (Lens et al. 2001, 2002) at organic loading rates up to, respectively, 46
and 35 g COD l1 reactor day1.
22.3 Sulfate-Reducing Bioreactors
22.3.1 High-Rate Sulfate-Reducing Bioreactors
22.3.1.1 Inocula
Initially, the experience that sulfate reduction develops spontaneously during

anaerobic wastewater treatment supported the adoption of bioreactor configura-
tions commonly used in methanogenic wastewater treatment, i.e. upflow anaerobic
sludge bed (UASB) reactors, for high-rate sulfate reduction bioreactors. In UASB
reactors, sulfidogenic granules can be obtained by feeding methanogenic granular
sludge with a sulfate-rich wastewater; however, it can take a very long time before
the sulfate reducers outcompete the methanogens. Using a mathematical model,
Omil et al. (1998) showed that the competition between acetate-utilizing SRB and
methanogenic bacteria (MB) is very time consuming. For a granular sludge with an
inoculum size of 103 and 109 cells of, respectively, acetotrophic SRB and MB, it
was calculated that it will take over 1,000 days before the sizes of both populations
are equal. This time period can be shortened by manipulating the population size of
SRB and MB in the inoculum sludge, i.e. deactivating methanogens or by bioaug-
mentation with pure cultures of SRB.
22.3.1.2 Electron Donor
For the treatment of inorganic wastewaters, the choice of the electron donor is an
important design parameter. One can, for example, supply an organic substrate
(e.g. molasses) as the electron donor, although this increases the risk of residual
pollutants. For high-rate sulfate reduction bioreactors supplied with a H2/CO2
mixture, high conversion rates can be obtained in mesophilic (30C; van Houten
et al. 1994) or thermophilic (55C; van Houten et al. 1997) gas-lift reactors in a
short (10-day) start-up period. In H2/CO2-fed reactor systems, a consortium of SRB
(Desulfovibrio sp.) and homoacetogens (Acetobacterium sp.) develops (van Houten
et al. 1995).
In cases where pure hydrogen gas is not available, one can use synthesis gas (a
mixture of H2, CO2 and CO), either directly (van Houten et al. 1997) or after enrich-
ing its H2 content by means of a water-gas-shift reaction, either chemically or bio-
logically with anaerobic granular sludge (Sipma et al. 2004). Parshina et al. (2005)
isolated Desulfotomaculum carboxydivorans, the first sulfate reducer capable of
hydrogenogenic growth on CO. In the presence of sulfate, the hydrogen formed is
used for sulfate reduction. This organism grows rapidly at 200 kPa CO, pH 7.0 and
55C, with a generation time of 100 min, producing nearly equimolar amounts of H2
and CO2 from CO and H2O. High specific CO conversion rates, exceeding 0.8 mol
CO (g protein)1 h1, make it an interesting candidate for a biological alternative to
the currently employed chemical catalytic water-gas-shift reaction to purify synthe-
sis gas (contains mainly H2, CO and CO2). Furthermore, as D. carboxydivorans is
capable of hydrogenotrophic sulfate reduction at partial CO pressures exceeding
100 kPa, it is also a good candidate for biodesulfurization processes at elevated tem-
peratures, e.g. in biological flue gas desulfurization (Sipma et al. 2006).
H2 is currently produced by reforming methane supplied by natural gas or biogas
(Fig. 22.2). However, the emission of the greenhouse gas CO2 and the costs of the
wastewater treatment would be greatly reduced if methane could be used directly
as an electron donor for biological sulfate reduction (Table 22.3). The first clear
evidence for anaerobic oxidation of methane (AOM) came from in situ geochemical
studies of marine sediments (Krger 2005). These studies revealed that methane
diffusing upwards from deep sites of sediments often disappears long before any
contact with oxygen is possible. In such anoxic zones of sediments, sulfate is the
only electron acceptor that can account for methane oxidation. AOM was demon-
strated by the formation of radiolabelled CO2 upon injection of [14C]methane into
anoxic marine sediments. AOM has been detected at temperatures from 4 to over
30C and at different locations like lakes and seashores. Stable isotope analysis
showed that archaea and sulfate reducers were both involved in AOM.
The main drawback of gas-lift bioreactors is the high pressure drop of the water
column that needs to be overcome when supplying the gaseous substrate (H2).
Cell-suspension bioreactors (Lens et al. 2003) or bubbleless H2 supply by hydropho-
bic membranes (Fedorovich et al. 2000) might be elegant alternative reactor designs.
Cell-suspension bioreactors also allow, via the dilution rate, control of the competition
between SRB and MB on the basis of their growth kinetics (Paulo et al. 2005).
Fig. 22.2 Wastewater (consisting mainly of ZnSO4) treatment process at Zinifex (Budel, The
Netherlands). Full-scale plant (a) and flow sheet of the plant in its present situation and when
methane is used as the electron donor (b)
Alternatively, under thermophilic (5565C) conditions, methanol can be supplied

as this substrate is converted to H2/CO2 at these high temperatures (Vallero et al.
2003). In cases where soluble substrates (such as methanol) are supplied, no H2S
stripping occurs and therefore H2S removal needs to be adopted to prevent its accumu-
lation in toxic concentrations. This can be done by stripping using an external gas
stream (e.g. N2) or via extractive H2S membranes (De Smul and Verstraete 1999).
Table 22.3 Comparison of hydrogen-utilizing sulfate-reducing process with the

process with direct methane utilization
Sulfate reduction with H2 Sulfate reduction with CH4
via CH4 directly
Temperature required 900C Wastewater temperature
Pressure required 16 1
(bar)
CH4 required per 1.88 1
mole of SO42 (mol)
CO2 emission per ton 0.9 0.45
of SO42 (t)
Table 22.4 Solubility products of metal sulfides and hydroxides

Metal ion logKsp (metal sulfide) logKsp (metal hydroxide)
Hg(II) 52.4 52.4
Ag(II) 49.7 7.71
Cu(I) 48.0, 48.5a
Cu(II) 35.1, 36.2b 20.4, 19.7b
Cd(II) 27.7, 25.8a 14.4
Pb(II) 27.0, 27.5a 15.3
Zn(II) 23.8, 24.7b 16.7, 16.9b
Co(II) 21.3 14.8
Ni(II) 20.7, 19.5b 17.2, 13.8b
Fe(II) 17.3 15.2
Data from Peters et al. (1984) except where indicated.
a
Data from Smith and Martell (1976).
b
Data from Brown et al. (1997).
22.3.2 Passive Sulfate-Reducing Systems
Microbial sulfate reduction is also regarded as an effective basic mechanism for

treating acid or neutral waters contaminated with heavy metals and sulfate, which
might simultaneously remove acidity and metals owing to, respectively, the alkalinity
produced during sulfate reduction and the very low solubility of metal sulfides
(Table 22.4). Application of high-rate sulfate reduction systems for the treatment of
mine waters is hampered by too high investment and operating costs. Attempts to
overcome these problems have essentially focused on two strategies. Firstly, estab-
lished industrial technologies can be adapted for the purpose of mine drainage
treatment, e.g. by choosing particular low-cost substrates such as whey, methanol
or even wastewaters (Rose et al. 1998). The second approach uses SRB in passive
processes, e.g. constructed wetlands (Gibert et al. 2004; Markewitz et al. 2004) or
reactive walls (Waybrant et al. 1998; Benner et al. 2002). Passive processes have
been developed on the basis of naturelike habitats such as marshes and wetlands
and use both chemical and biological processes, thereby reducing the need for
sophisticated process technology (Barton and Karathanasis 1999).
22.4 Sulfate Reduction for Metal Recovery/Reuse
22.4.1 Metal Sulfide Precipitation
Heavy metals such as Cu, Zn, Cd, Pb, Ni and Fe precipitate with biogenic sulfide to
form insoluble metal sulfides (Table 22.4), thereby concentrating the metals into an
easy separable and sometimes valuable form. In engineered systems, metal sulfide
precipitation can be optimized with respect to the rate of biogenic sulfide production,
metal precipitate product quality and selective precipitation of metal sulfides. Treatment
processes should focus on recovery of the metals also, as metal resources are depleting.
Reuse of metals can only become economically and technically feasible when metals are
removed selectively and relatively pure metal sludges are produced.
In industry, hydroxide precipitation was by far the most widely used method
in the past for wastewater treatment. However, it is also well known that technologies
based on metal precipitation with sulfide have some fundamental advantages
over hydroxide precipitation (Kim and Amodeu 1983; Peters et al. 1984; Veeken
et al. 2003):
1. Effluent concentrations are orders of magnitude lower: micrograms per litre vs.
milligrams per litre.
2. The interference of chelating agents in the wastewater is less problematic.
3. Selective metal removal gives better opportunities for metal reuse.
4. Metal sulfide sludges have better settling, thickening and dewatering character-
istics than hydroxide sludges.
5. Existing smelters can process sulfide precipitates, thus enabling metal recovery
and eliminating the need for sludge disposal.
22.4.2 Biogenic Sulfide for Metal Sulfide Precipitation
Earlier objections against the use of sulfide, i.e. that it is toxic, malodorous and
corrosive, can today be overcome by adequate safety measures and the use of modern
corrosion-resistant construction materials. Chemical forms of sulfide such as Na2S,
NaHS, CaS and H2S can be used, but these need to be transported to the treatment
site. In general, these sulfide sources are more expensive than lime or limestone.
Moreover, the hazards that accompany transport, handling and storage of the
chemical sulfides lead to additional costs for safety measures. These drawbacks can
be overcome by the on-site production of biogenic sulfide in bioreactors as
described in Sect. 22.3.1.
Several studies have focused on the use of SRB for precipitating metal sulfides
in the same reactor systems where the sulfate reduction activity occurs; however, a
problem associated with this is the metal toxicity to SRB (Chen et al. 2000).
Another problem associated with the use of SRB biomass in the metal-precipitation
reactor is that the precipitated metals are located on the biomass together with the
microbial population, thus increasing the volume of metal-contaminated sludge.
A two-stage process in which the metal-precipitation step is separated from the
SRB bioreactor system is a good alternative that uncouples the process conditions
of the bioreactor and the precipitator (Esposito et al. 2006).
22.4.3 Selective Metal Precipitation
The metal sulfides formed are highly insoluble at neutral pH, while some metal
sulfides (e.g. CuS) are highly insoluble at pH values as low as 2. The great advantage
of sulfide precipitation is the possibility of selective precipitation. Tabak et al.
(2003) have shown the possibility of selective precipitation in acid mine drainage
only by changing pH and temperature. Veeken et al. (2003) showed that stoichiometric
addition of sulfide to a heavy metal (Cu, Zn, Cd and Ni) solution can be achieved
by controlling the sulfide concentration in the precipitator by means of combining
a pH and a sulfide ion selective electrode. This results in very low effluent concen-
trations of both metals and sulfide. Every precipitating metal precipitates at a
unique S2 concentration (pS), which is directly related to the solubility product of
the metal sulfide. The uniqueness of the pS level for each metal was successfully
applied as a control parameter to precipitate metals selectively and to obtain pure
metal sulfide, which have better chances for reuse (Knig et al. 2006). The success
of the precipitation process not only depends on the removal of metal ions from the
soluble phase, but also on the separation of the solid phase (metal sulfide precipitate)
from the liquid phase. Therefore, solidliquid separation processes such as
sedimentation or filtration are of key importance in efficient metal-removal processes
(Esposito et al. 2006).
References
Barton CD, Karathanasis AD (1999) Renovation of a failed constructed wetland treating acid mine
drainage. Environ Geol 39:3950
Benner SG, Blowes DW, Ptacek CJ, Mayer KU (2002) Rates of sulphate reduction and metal
sulfide precipitation in a permeable reactive barrier. Appl Geochem 17:301320
Brown TL, Lemay HE, Bursten BE (1997) Chemistry: the central science, 7th edn. Prentice Hall,
Upper Saddle River
Chen BY, Utgikar VP, Harmon SM, Tabak HH, Bishop DF, Govind R (2000) Studies on biosorption
of zinc(II) and copper (II) on Desulfovibrio desulfuricans. Int Biodeterior Biodegrad
46:1118
Colleran E, Finnegan S, Lens P (1995) Anaerobic treatment of sulphate-containing waste streams.
Antonie Van Leeuwenhoek 67:2946
De Smul A, Verstraete W (1999) The phenomenology and the mathematical modelling of the
silicone-supported chemical oxidation of aqueous sulfide to elemental sulfur with ferric sulfate.
J Chem Technol Biotechnol 74:456466
Esposito G, Veeken A, Weijma J, Lens PNL (2006) Effect of the use of biogenic sulphide on ZnS
precipitation under different process conditions. Sep Purif Technol 51:3139
Fedorovich V, Greben M, Kalyuzhnyi S, Lens P, Hulshoff Pol L, Lettinga G (2000) Use of
membranes for hydrogen supply in a sulfate reducing reactor. Biodegradation 11:295303.
Gibert O, de Pablo J, Cortina JL, Ayora C (2004) Chemical characterisation of natural organic
substrates for biological mitigation of acid mine drainage. Water Res 38:41864196
Knig J, Keesman KJ, Veeken A, Lens PNL (2006) Dynamic modelling and process control of
ZnS precipitation. Sep Sci Technol 41:10251042
Krger M, Treude T, Wolters H, Nauhaus K, Boetius A (2005) Microbial methane turnover in
different marine habitats. Palaeogeogr Palaeoclimatol Palaeoecol 227:617
Kim BM, Amodeo PA (1983) Calcium sulfide process for treatment of metal-containing wastes.
Environ Prog 2:175180
Lens P, Visser A, Janssen A, Hulshoff Pol L, Lettinga G (1998a) Biotechnological treatment of
sulfate rich wastewaters. Crit Rev Environ Sci Technol 28:4188
Lens P, van den Bosch M, Hulshoff Pol L, Lettinga G (1998b) Effect of staging on volatile fatty
acid degradation in a sulfidogenic granular sludge reactor. Water Res 32:11781192
Lens P, Gastesi R, Hulshoff Pol L, Lettinga G (2003) Use of sulphate reducing cell suspension
bioreactors for the treatment of SO2 rich flue gases. Biodegradation 14:229240
Lens PNL, Korthout D, van Lier JB, Hulshoff Pol LW, Lettinga G (2001) Effect of upflow velocity
on thermofilic sulfate reduction under acidifying conditions. Environ Technol 22:183193
Lens PNL, Klijn R, van Lier JB, Hulshoff Pol LW, Lettinga G (2002) Effect of specific gas loading
rate on thermofilic sulfate reduction under acidifying conditions. Water Res 37:10331047
Markewitz K, Cabral AR, Panarotto CT, Lefebvre G (2004) Anaerobic biodegradation of an
organic by-products leachate by interaction with different mine tailings. J Hazard Mater
110:93104
Omil F, Lens P, Visser A, Hulshoff Pol LW, Lettinga G (1998) Long term competition between
sulfate reducing and methanogenic bacteria in UASB reactors treating volatile fatty acids.
Biotechnol Bioeng 57:676685
Parshina SN, Sipma J, Nakashimada Y, Henstra HM, Smidt H, Lysenko AM, Lens PNL, Lettinga G,
Stams AJM (2005) Desulfotomaculum carboxydivorans sp. nov., a novel sulfate reducing bac-
terium capable of growth at 100% CO. Int J Syst Evol Microbiol 55:21592165
Paulo P, Kleerebezem R, Lettinga G, Lens PNL (2005) Cultivation of high-rate sulphate reducing
sludge by pH-based electron donor dosage. J Biotechnol 118:107116
Peters RW, Ku Y, Battacharyya D (1984) Evaluation of recent treatment techniques for removal
of heavy metals from industrial wastewaters. Paper presented at AIChE meeting, Philadelphia,
pp 1922
Reis MAM, Lemos PC, Carrondo MJT (1995) Biological sulfate removal of industrial effluents
using the anaerobic digestion. Med Fac Landbouwwet Univ Gent 60:27012707
Rose PD, Boshoff GA, van Hille RP, Wallace LC, Dunn KM, Duncan JR (1998) An integrated
algal sulphate reducing high rate ponding process for the treatment of acid mine drainage
wastewaters. Biodegradation 9:247257
Sipma J, Lens PNL, Vieira A, Miron Y, van Lier JB, Hulshoff Pol LW, Lettinga G (2000)
Thermofilic sulfate reduction in UASB reactors under acidifying conditions. Process Biochem
35:509522
Sipma J, Meulepas RJW, Parshina SN, Stams AJM, Lettinga G, Lens PNL (2004) Effect of carbon
monoxide, hydrogen and sulfate on thermophilic (55C) hydrogenogenic carbon monoxide
conversion in two anaerobic bioreactor sludges. Appl Microbiol Biotechnol 64:421428
Sipma J, Lettinga G, Stams AJM, Lens PNL (2006) Hydrogenogenic CO conversion in a moderately
thermophilic (55C) sulfate-fed gas lift reactor: competition for CO-derived H2. Biotechnol
Progr 22:13271334
Smith RM, Martell AE (1976) Critical stability constants, vol. 4. Inorganic ligands. Plenum,
New York
Tabak HH, Scharp R, Burckle J, Kawahara FK, Govind R (2003) Advances in biotreatment of acid
mine drainage and biorecovery of metals: 1. Metal precipitation for recovery and recycle.
Biodegradation 14:423436
Vallero MVG, Lens PNL, Hulshoff Pol LW, Lettinga G (2003) Effect of NaCl on thermophilic
(55C) methanol degradation in sulfate reducing reactors. Water Res 37:22692280
Vallero MVG, Camarero E, Lettinga G, Lens PNL (2007) Hyperthermophilic sulfate reduction in
methanol and formate fed UASB reactors. Appl Environ Microbiol (in press)
van Houten RT, Hulshoff Pol LW, Lettinga G (1994) Biological sulphate reduction using gas-lift
reactors fed with hydrogen and carbon dioxide as energy and carbon source. Biotechnol
Bioeng 44:586594
van Houten RT, Oude Elferink SJWH, van Hamel SE, Hulshoff Pol LW, Lettinga G (1995)
Sulphate reduction by aggregates of sulphate-reducing bacteria and homo-acetogenic bacteria
in a lab-scale gas-lift reactor. Bioresour Technol 54:7379
van Houten RT, Yun SY, Lettinga G (1997) Thermophilic sulphate and sulfite reduction in lab-scale
gas-lift reactors using H2 and CO2 as energy and carbon source. Biotechnol Bioeng
55:807814
Veeken AHM, de Vries S, van der Mark A, Rulkens WH (2003) Selective precipitation of heavy
metals as controlled by a sulfide-selective electrode. Sep Sci Technol 38:119
Verstraete W, de Beer D, Pena M, Lettinga G, Lens P (1996) Anaerobic bioprocessing of waste.
World J Microbiol Biotechnol 12:221238
Waybrant KR, Blowes DW, Ptacek CJ (1998) Selection of reactive mixtures for use in permeable
reactive walls for treatment of acid mine drainage. Environ Sci Technol 32:19721979
Chapter 23
Impact of Nitrate on the Sulfur Cycle
in Oil Fields
Gerrit Voordouw
Abstract Production of oil from subsurface reservoirs requires injection of water

or gas to maintain reservoir pressure. Seawater is usually injected on offshore
platforms (as in the North Sea). The combination of abundant electron donors
(selected oil components) and electron acceptors (30 mM sulfate in sea water) can
lead to significant production of sulfide in the subsurface through action of resident
or injected sulfate-reducing bacteria (SRB). Lowering sulfide concentrations in the
produced oilwater mixture is desirable to reduce corrosion risk. Injection of nitrate
has recently emerged as a new technology that can reduce sulfide levels reliably.
Adding low concentrations (50100 ppm) of nitrate continuously to all injected water
can eliminate sulfide from produced water and oil. The mechanism underlying this
technology appears to be largely microbial. Nitrate-reducing, sulfide-oxidizing bac-
teria remove sulfide with production of nitrite and other reactive nitrogen species.
Nitrite is a powerful SRB inhibitor that specifically affects dissimilatory sulfite
reductase, the enzyme that produces the sulfide. Heterotrophic nitrate-reducing bac-
teria can directly oxidize oil components with the injected nitrate, outcompeting SRB.
This results in a desirable subsurface microbial community change that prevents the
formation of sulfide, improving oil quality. Nitrate injection is one of the first reliable,
microbe-based processes that is becoming widely used in oil production to control
the oil field sulfur cycle, making microbiologists partners in discovering how we can
continue to produce the worlds most significant energy resource.
23.1 Introduction
Our society depends heavily on the use of fossil fuels for its energy supply, with
oil, gas and coal contributing 40, 24 and 22% of the worlds growing energy
needs, respectively. Nuclear energy and renewable forms of energy (wind, hydro-
power, biomass) contribute 6 and 8% worldwide. The negative environmental
impact of this heavy reliance on fossil fuels is becoming increasingly apparent
and includes significant increases in the CO2 concentration of the Earths atmos-
phere, with associated global warming. In seeking solutions for these problems
we should aim to reduce per capita energy consumption, to increase the contribution
296
Springer 2008
23 Impact of Nitrate on the Sulfur Cycle in Oil Fields 297
of renewables to our energy supply and to use and extract fossil fuels as efficiently
as possible. Microbiology is one of the disciplines that can help improve the effi-
ciency of fossil-fuel extraction. Although its potential in this regard has been rec-
ognized for over 50 years, microbial processes in fossil-fuel extraction or upgrading
are not yet common. Potential or proven processes include the microbial desulfuri-
zation of coal and oil, targeting pyrite and organic sulfur compounds such as
dibenzothiophene, respectively (Monticello and Finnerty 1985). A more recent
development has been the use of nitrate to manage the sulfur cycle in oil and gas
fields. This application is now being used field-wide in several oil fields, especially
in the North Sea to remove sulfide from oil and associated produced water. The
microbial basis of this process and its application will be reviewed in this chapter.
23.2 The Oil Field Sulfur Cycle
Much of the worlds oil is produced by water injection to maintain reservoir

pressure. As a result an oilwater mixture is produced which is separated into
produced water and produced oil. Depending on water availability, the produced
water is reinjected (produced water reinjection, PWRI) or discharged. PWRI is
common in landlocked reservoirs, but rare in offshore situations where seawater
is plentiful. Oil production by water injection often results in increased sulfide
levels (souring), because sulfate-reducing bacteria (SRB) couple the oxidation of
degradable oil organics present in the water in the reservoir (formation water)
to the reduction of sulfate to sulfide (Fig. 23.1). The problem can be especially
severe when seawater, which has a high sulfate concentration of 30 mM, is
injected. An example is provided by seawater flooding of the Skjold field in the
Danish sector of the North Sea. Total daily production of sulfide increased from
100 kg day1 initially to up to 1,100 kg day1 after 5 years of seawater injection
(Larsen 2002). The sulfide was produced by SRB, presumably in the zone where
sulfate-containing, injected seawater mixes with oil organics-containing formation
water. High concentrations of sulfide are unwanted because of the toxicity, and the
associated risk of corrosion of pipes and aboveground equipment processing the oil,
as well as the potential for reservoir plugging by precipitated sulfides.
The reservoir SRB are either indigenous (Magot 2005) or introduced with the
injection water. They are mesophilic or thermophilic, depending on reservoir depth.
Although some are incomplete oxidizers, converting oil organics to CO2 and acetate
(Fig. 23.1), complete oxidizers, producing CO2, only are also common. The incom-
plete oxidizers include the mesophilic Desulfovibrio spp., which are well known and
easy to isolate, but may represent a minor fraction of the SRB found in mesophilic oil
field environments (Rabus et al. 1996). Thermophilic oil field sulfate-reducing
prokaryotes (SRP) include the completely oxidizing Thermodesulforhabdus and
Archaeoglobus spp. (Beeder et al. 1994, 1995).
Souring can be prevented or reversed by nitrate, which can be added to the
injection water in the appropriate concentration. Nitrate injection stimulates
298 G. Voordouw
S0 N2
(A) lactate 2 NO2-
SO4 - NH3
SRB NR-SOB
acetate+CO2 HS- NO3-
lactate NO3-
(B) hNRB
acetate+CO2 NO2- NH3

N2
Fig. 23.1 Survey of microbial groups impacting the sulfur cycle in oil fields. a Sulfate-reducing
bacteria (SRB) couple incomplete oxidation of oil organics (to acetate and CO2), or complete
oxidation of oil organics to CO2 (not shown) to the reduction of sulfate to sulfide. Nitrate-reducing,
sulfide-oxidizing bacteria (NR-SOB) oxidize sulfide to sulfur or sulfate, with nitrate being reduced
to nitrite and then to either nitrogen (with NO and N2O as intermediates) or to ammonia (without
intermediates). b Heterotrophic nitrate-reducing bacteria (hNRB) couple incomplete (as shown)
or complete oxidation of oil organics to reduction of nitrate to nitrite and then to either nitrogen
or ammonia. Note that some NR-SOB/hNRB do not reduce nitrate beyond nitrite. Also nitrite is
a powerful SRB inhibitor, as explained in the text
nitrate-reducing, sulfide-oxidizing bacteria (NR-SOB) and heterotrophic nitrate-

reducing bacteria (hNRB). The activities of these two groups are also outlined in Fig.
23.1. The hNRB oxidize degradable oil organics, which would otherwise be used by
SRB. Such competitive exclusion was initially postulated as the main mechanism
through which souring was prevented (Hitzman and Dennis 1997). However, early
studies on nitrate injection in Coleville, a medium-temperature field in western
Canada, indicated that Thiomicrospira sp. strain CVO became a major community
component, both in injector and in producing wells, when nitrate was injected (Telang
et al. 1997). Strain CVO is an autotroph, deriving energy for growth from the oxidation
of sulfide to sulfate with sulfur as an intermediate, while reducing nitrate to nitrogen
with nitrite, nitric oxide and nitrous oxide as intermediates. Sulfide concentrations
decreased on average by 70% and this was credited primarily to strain CVO, i.e., the
souring control mechanism at Coleville was as indicated in Fig. 23.1a. A more exten-
sive survey of the presence of SRB, hNRB and NR-SOB in western Canadian oil fields
has indicated the presence of all three microbial groups in many oilfields (Eckford and
Fedorak 2002). The hNRB often appeared to outnumber the NR-SOB, i.e., the micro-
bial ecology found at Coleville appeared to be the exception rather than the rule and
souring control by competitive exclusion is also a likely mechanism.
23.3 Effect of Nitrate Injection on SRB Physiology
By definition, SRB reduce sulfate, although there are some strains that also
reduce nitrate, e.g., Desulfovibrio desulfuricans strain ATCC 27774 (Gonzalez et al.
2006). Peculiarly, such strains may not downregulate genes for enzymes involved
in sulfate reduction when they are reducing nitrate; hence the use of alternative
electron acceptors by SRB serves a very different purpose than, for instance, in
Escherichia coli, where distinct gene-expression patterns are established for cells
grown with oxygen, nitrate or fumarate as the electron acceptor. In these different
gene-expressing states genes encoding oxidoreductases for the electron acceptor
present in the medium are on, whereas genes for electron acceptors not present
in the medium are off. In contrast, in Desulfovibrio spp. genes for sulfate
reduction are on all the time, indicating this to be the primary lifestyle of the
organism. Hence, the function of alternate electron acceptor (nitrate, oxygen)
reduction in Desulfovibrio spp. appears primarily to prevent inhibition of sulfate
reduction. Although the physiology and gene-expression pattern of the majority
of SRB, that do not reduce nitrate, are not affected by the addition of millimolar
concentrations of nitrate, these are strongly affected by nitrite, which is a strong
SRB inhibitor (Haveman et al. 2004; He et al. 2006). Nitrite is bound tightly
by dissimilatory sulfite reductase (DsrAB), the terminal reductase of SRB (Wolfe
et al. 1994), which slowly reduces nitrite to ammonia. These properties make
nitrite a strong competitive inhibitor, preventing reduction of sulfite to sulfide,
the normal physiological function of DsrAB. Addition of millimolar concentra-
tions of nitrite to mid-log-phase cultures of D. vulgaris halts sulfate reduction and
associated growth and downregulates expression of genes for enzymes involved
in sulfate reduction (sulfate adenylyltransferase, pyrophophatase, adenosine
5-phosphosulfate reductase) with the exception of DsrAB (Haveman et al. 2004;
He et al. 2006). Genes for ATP synthase, as well as genes for two membrane-bound
redox protein complexes, QmoABC and DsrMKJOP, are also downregulated. This
indicates that a proton-motive force allowing phosphorylation of ADP to ATP is
lacking under conditions of nitrite inhibition. It also indicates involvement of
QmoABC and DsrMJKOP in sulfate respiration, i.e., electrons for the APS
reductase and DsrAB catalyzed reactions are likely provided by QmoABC and
DsrMJKOP, respectively. Although the detailed bioenergetic mechanism through
which D. vulgaris and other SRB derive energy for growth from sulfate respira-
tion is by no means solved, these studies provided strong evidence for involvement
of membrane-bound complexes in sulfate respiration. Hence, when D. vulgaris
derives energy for growth from coupling the oxidation of lactate to the reduction
of sulfate, reducing equivalents (H+, e) cycle from the cytoplasm to the peri-
plasm to return to the cytoplasm through QmoABC and DsrMJKOP (Haveman et
al. 2004; Mussmann et al. 2005; He et al. 2006). As a consequence, these complexes
appear strongly conserved in all SRP, including in the thermophilic archaeon
Archaeoglobus fulgidus. Hence inhibition of sulfate reduction by nitrite has, in
addition to being of practical significance, given us insight into the mechanism
of sulfate reduction by SRP. In order to prevent inhibition of DsrAB by nitrite,
SRB can have a periplasmic nitrite reductase (NrfHA), which reduces nitrite to
ammonia. The nrfHA genes of D. vulgaris Hildenborough are upregulated upon
addition of nitrite; hence, when nitrite is added, reducing equivalents derived
from lactate oxidation are temporary diverted from sulfate reduction to nitrite
reduction by periplasmic NrfHA, as well as at a slower rate by cytoplasmic
300 G. Voordouw
DsrAB. The lethality of nitrite depends on the time required for an SRB population
to reduce all nitrite to ammonia. This depends on the biomass concentration and
the presence or absence of NrfHA. Mid-log-phase cultures of D. vulgaris
Hildenborough can survive addition of 510 mM nitrite, but an nrfHA mutant can
survive addition of only 0.5 mM nitrite. For single cells on plates, the lowest
possible biomass concentration, the inhibitory nitrite concentration, is only
0.04 mM (Haveman et al. 2004). Interestingly, this low inhibitory concentration
is the same for wild-type and nrfHA-mutant cells, because the binding affinity
(Km) of NrfHA for nitrite is quite high (millimolar); hence, NrfHA does not
contribute to nitrite detoxification at 0.04 mM, which is instead reduced by the
target DsrAB under these conditions. NrfHA thus allows dense SRB populations
to survive millimolar concentrations of nitrite by its rapid reduction to ammonia.
Thermophilic SRP (tSRP; including members of Archaeoglobus) appear to lack
nitrite reductase and are, as a result, much more sensitive to inhibition by nitrite
than are mesophilic SRB.
23.4 Mechanism of Souring Control
Having established that SRP are strongly inhibited by nitrite, the question arises
to what extent this inhibition contributes to souring control. hNRB and NR-SOB,
collectively referred to as NRB, reduce nitrate to nitrite, which is then further
reduced to either nitrogen or ammonia (Fig. 23.1). Many NRB excrete nitrite and
it is not uncommon to find nitrite in produced waters of oil fields subjected to
nitrate injection or in the effluent of upflow bioreactors that aim to model such
fields (Reinsel et al. 1996; Myhr et al. 2002). Studies in which the NR-SOB
Thiomicrospira sp. strain CVO was added to growing cultures of mesophilic SRB
in the presence of nitrate indicated rapid formation of millimolar concentrations
of nitrite under these conditions (Greene et al. 2003). This led to either permanent
or transient inhibition of SRB activity, depending on the presence of nitrite
reductase in the SRB strain. However, because in mesophilic oil field populations
some of the SRB present are likely to have nitrite reductase, it is unlikely that
such a population could ever become permanently inhibited by nitrite. The nitrate
dose required to eliminate sulfide appears dictated by the concentration of degra-
dable oil organics in such systems, as was demonstrated in bioreactor studies by
Hubert et al. (2003).
The situation may be different in thermophilic oil field communities. Nitrite
reductase has so far not been demonstrated in tSRP. As a result tSRP-containing
enrichments from Ekofisk, a North Sea oil field with an in situ temperature of
8090C, were inhibited by very low concentrations of nitrite (0.250.5 mM).
Nitrate does not affect sulfate reduction rates at Ekofisk, because thermophilic
NRB also appear to be absent; hence, nitrate injection may not work, but injection
of nitrite could be effective in this field (Kaster et al. 2007).
23.5 Prospects for Nitrate Injection
Nitrate injection is the first reliable, microbe-based process that is being applied
continuously and field-wide to improve the production of oil (Thorstenson et al.
2002; Larsen et al. 2004). The nitrate dose required to prevent souring needs to be
determined by trial and error. In mesophilic systems the dose is dictated by the
concentration of oxidizable electron donors (sulfide, sulfur and degradable oil
organics), whereas in thermophilic systems inhibition of thermophilic SRB by
nitrite may also contribute, lowering the effective dose required. Adoption of this
successful technology is currently being considered in many fields, including those
subjected to PWRI. Successful adoption of the technology in PWRI situations still
requires considerably more research. Also the effects of continuous, long-term,
field-wide nitrate injection need to be considered. So far the experiences with up to
6 years of continuous injection have been positive and preliminary reports, indicating
that this practice leads to production of additional oil through microbially enhanced
oil recovery, are further fanning interest in this technology. Nitrate injections are
here to stay and may well prove the ideal stepping stone to further expand petroleum
microbiology as a contributing science towards improving the production efficiency
of oil, the most important energy supply in the world today.
Acknowledgements. Research in the authors laboratory has been supported through
Strategic Grants of the Natural Science and Engineering Research Council of
Canada (NSERC) with ConocoPhillips, Baker Petrolite and the Computer
Modelling Group as industrial partners. The research contributions of graduate
students Casey Hubert and Krista Kaster, as well as of postdoctoral fellows Anne
Greene, Alexander Grigoriyan and Mehdi Nemati are gratefully acknowledged.
References
Beeder J, Nilsen RK, Rosnes JT, Torsvik T, Lien T (1994) Archaeoglobus fulgidus isolated from
hot North Sea oil field waters. Appl Environ Microbiol 60:12271231
Beeder J, Torsvik T, Lien T (1995) Thermodesulforhabdus norvegicus gen. nov., sp. nov., a novel
thermophilic sulfate reducing bacterium from oil field water. Arch Microbiol 164:331336
Eckford RE, Fedorak PM (2002) Planktonic nitrate-reducing bacteria and sulfate-reducing
bacteria in some western Canadian oil field waters. J Ind Microbiol Biotechnol 29:8392
Greene EA, Hubert C, Nemati M, Jenneman G, Voordouw G (2003) Nitrite reductase activity of
sulfate-reducing bacteria prevents their inhibition by nitrate-reducing, sulfide-oxidizing
bacteria. Environ Microbiol 5:607617
Gonzalez PJ, Rivas MG, Brondino CD, Bursakov SA, Moura I, Moura JJ (2006) EPR and redox
properties of periplasmic nitrate reductase from Desulfovibrio desulfuricans ATCC 27774.
J Biol Inorg Chem. 11:609616
Haveman SA, Greene EA, Stilwell CP, Voordouw JK, Voordouw G (2004) Physiological and
gene expression analysis of inhibition of Desulfovibrio vulgaris Hildenborough by nitrite.
He Q, Huang KH, He Z, Alm EJ, Fields MW, Hazen TC, Arkin AP, Wall JD, Zhou J (2006)
Energetic consequences of nitrite stress in Desulfovibrio vulgaris Hildenborough, inferred
from global transcriptional analysis. Appl Environ Microbiol 72:43704381
302 G. Voordouw
Hitzman DO, Dennis DM (1997) New technology for prevention of sour oil and gas. In:
Proceedings SPE/DOE exploration and production environmental conference, Dallas, pp
406411
Hubert C, Nemati M, Jenneman GE, Voordouw G (2003) Containment of biogenic sulfide
production in continuous up-flow, packed-bed bioreactors with nitrate or nitrite. Biotechnol
Prog 19:338345
Kaster KM, Grigoryan A, Jenneman G, Voordouw G (2007) Effect of nitrate and nitrite on sulfide
production by two thermophilic, sulfate-reducing enrichments from an oil field in the North
Sea. Appl Microbiol Biotechnol 75:195203
Larsen J (2002) Downhole nitrate applications to control sulfate reducing bacteria activity and
reservoir souring. Corrosion 2002. Paper 02025. NACE International, Houston
Larsen J, Rod MH, Zwolle S (2004) Prevention of reservoir souring in the Halfdan field by nitrate
injection. Corrosion 2004. Paper 04761. NACE International, Houston
Magot M (2005) Indigenous microbial communities in oil fields. In: Ollivier B, Magot M (eds)
Petroleum microbiology. ASM, Washington, pp 2133
Monticello DJ, Finnerty WR. 1985. Microbial desulfurization of fossil fuels. Annu Rev Microbiol
39:371389
Mussmann M, Richter M, Lombardot T, Meyerdierks A, Kuever J, Kube M, Glockner FO, Amann R
(2005) Clustered genes related to sulfate respiration in uncultured prokaryotes support the
Myhr S, Lillebo BLP, Sunde E, Beeder J, Torsvik T (2002) Inhibition of microbial H2S production
in an oil reservoir model column by nitrate injection. Appl Microbiol Biotechnol 58:400408
Rabus R, Fukui M, Wilkes H, Widdel F (1996) Degradative capacities and 16S rRNA-targeted
whole-cell hybridization of sulfate-reducing bacteria in an anaerobic enrichment culture
utilizing alkylbenzenes from crude oil. Appl Environ Microbiol 62:36053613
Reinsel MA, Sears JT, Steward PS, McInerney MJ (1996) Control of microbial souring by nitrate,
nitrite or glutaraldehyde injection in a sandstone column. J Industr Microbiol 17:128136
Telang AJ, Ebert S, Foght JM, Westlake DWS, Jenneman GE, Gevertz D, Voordouw G (1997) The
effect of nitrate injection on the microbial community in an oil field as monitored by reverse
sample genome probing. Appl Environ Microbiol 63:17851793
Thorstenson T, Bdtker G, Sunde E, Beeder J (2002) Biocide replacement by nitrate in sea water
injection in sea water injection systems. Corrosion 2002. Paper 02033. NACE International,
Houston
Wolfe B, Lui SM, Cowan J (1994) Desulfoviridin, a multimeric-dissimilatory sulfite reductase
from Desulfovibrio vulgaris (Hildenborough). Purification, characterization, kinetics and EPR
studies. Eur J Biochem 233:7989
Index
A Arenicola marina, 3739

Acetate kinase, 177 Assimilatory sulfate reduction (Asr), 69,
Acid mine drainage, 184 7173, 204, 206, 209
Acidianus Assimilatory sulfur metabolism, 71
A. ambivalens, 185, 187, 190, 217220, ATP sulfurylase, 14, 60, 95, 112, 113, 197,
222, 223 245, 248
A. brierleyi, 185, 217220 ATP, 27
A. tengchongensis, 188191, 217220, ATP synthesis, 38, 39, 42
222, 223 Axial volcano, 252, 253
Acidithiobacillus, 165, 218219, 221, 222
A. ferrooxidans, 7785, 198 B
A. thiooxidans, 187 Bacteriochlorophyll a, 121
Adenosine-5-phosphosulfate (adenylylsulfate, Bacteriochlorophyll c, 120
APS), 14, 18, 19, 245, 247 Beggiatoa, 8891, 102, 103, 241243,
Adenosine-5-phosphosulfate (adenylylsulfate, 248, 249
APS) reductase, 14, 18, 24, 26, 28, Bioleaching, 184, 186, 187, 189
32, 66, 69, 94, 112, 152, 196, 197, Biomining, 78
248 Bioprocess engineering, 285, 287, 289, 291,
Adenylate kinase, 187, 197 293, 295
Adenylylsulfate:phosphate adenylyltransferase Bioreactor, 287289, 292, 293
(adenosine-5-phosphosulfate:
phosphateadenylyltransferase C
APAT), 187, 197 Caldariella quinone, 187, 197
ADP sulfurylase, 95 Calvin-Benson-Bassham cycle, 245
Alkalilimnicola ehrlichei,71, 109 Canfield oceans, 41, 42
Allochromatium vinosum, 66, 69, 71, 88, 89, Capnine, 173
92, 9496, 106, 109, 259, 263, 268 Chemolithoautotrophy, 238, 241, 246248,
Allochromatium, 247, 248 254
Amoebobacter, 90, 91 Chlorobaculum parvum, 103
Anaerobic, 117, 119, 122, 286289 Chlorobaculum tepidum, 62, 64, 67, 109,
Anaerobic oxidation of methane (AOM), 203, 117124, 248, 249
206, 212, 289 Chlorobaculum thiosulfatiphilum, 128,
aprAB, 69, 73 129, 134
APS reductase, 24, 26, 28, 32, 94, 112, 152, 248 Chlorobii, 61, 73
Aquifex aeolicus, 189, 190, 218 Chlorobium chlorochromatii, 26
Archaea, 184, 186, 188, 198, 204, 208, 212, 213 Chlorobium clathratiforme, 109
Archaeoglobus fulgidus, 25, 26, 29, 30, 93, Chlorobium ferrooxidans, 6467,
110, 203, 206, 209, 210, 211, 280 69, 72, 73
Arcobacter, 244, 245, 251253 Chlorobium limicola f. thiosulfatophilum, 128
303
304 Index
Chlorobium limicola, 94, 109 minor sulfur isotopes, 274, 279

Chlorobium phaeobacteroides, 109 sulfur isotope fractionation during, 273,
Chlorobium tepidum (syn. Chlorobaculum 275, 282
tepidum), 26, 29, 62, 64, 67, Dissimilatory sulfite reductase (Dsr), 2833,
117124 47, 52, 66, 72, 144, 177, 204, 206,
Chloroherpeton thalassium, 64, 66, 67, 70, 209211, 213, 247, 248
72, 73 DNA macroarray, 80, 81, 83
Chlorosomes, 62 dsr gene cluster, 66
Chromatiaceae, 95, 96, 102, 103 dsr genes, 95
Chromatium, 88, 9094 dsr operon, 107, 111
Clostridium thermocellum, 72 Dsr proteins
Coenzyme F420 reducing hydrogenase, phylogeny, 108
208, 212 DsrA, A subunit of Dsr, 67, 68, 73,
Coenzyme F420, 8-hydroxy-5 deazariboflavin 204, 209
derivative, 208 DsrAB, 47, 4956
Coenzyme F420-dependent sulfite reductase DsrB, B subunit of Dsr, 204, 209
(Fsr), 204, 208213 DsrMKJOP, 67
Coenzyme M, 173, 174, 213
Complementation, 122 E
Crystallographic analysis, 192 East Pacific Rise (EPR), 242, 244, 245,
CysH, 69, 72 252, 253
Cysteine persulfide, 195 Ecology, 53, 56
Cysteine sulfinate, 173 Ectothiorhodospira, 90, 93
Cysteine, 189, 190, 194, 195, 198 Electron acceptor, 24, 26, 3032
Cysteine-aminotransferase, 39 Electron donor, 27, 28, 32
Cyteine-disulfide transporter SoxV, 141 Electron paramagnetic resonance, 14
Cytochrome c oxidase, 38 Elemental sulfur, 13, 14, 88, 89, 95, 118, 119,
Cytochrome c3, 14, 30, 31 122, 123, 259, 265, 268270
Cytochrome, 25, 26, 2933 Endosymbiotic theory, 40
Cytoplasm, 24, 25, 2732 Energy conservation, 25, 26
Environmental technology, 285
D Epsilon proteobacteria, 239, 241, 245, 246,
Deep-sea hydrothermal vents, 238, 239, 241, 249, 251, 253, 254
246, 247, 249, 251, 253, 254 EPR spectroscopy, 189
Desulfotalea psychrophila, 25, 26, 30, 110 9N EPR, 244, 245, 252
Desulfitobacterium hafniense, 29, 72, 110 Eukaryotes, 3640, 42
Desulfotomaculum reducens, 26, 27 Evolution, 40, 41, 47, 49, 52, 53, 56
Desulfovibrio Extremophile, 235
D. autotrophicum, 281 Extremophilic, 202
D. desulfuricans, 25, 26, 29, 31, 32, 275
D. jorgensii, 281 F
D. vulgaris, 211, 25, 26, 31, 33, 109 F1 ATPase, 123
Desulfovibrio spp, 70 FAD, 14, 204
Dimethylsulfoxide reductase, 158 FCSD, 141, 144
Disproportionation, 187189, 196, Ferredoxin, 204, 209
198, 199 Ferroplasma acidarmanus, 189, 190
Dissimilatory sulfate reduction, 202, Filamentous sulfur, 251253
204, 206 Flavocytochrome c, 66, 69, 71, 73, 95,
factors controlling isotope fractionation, 106, 111
273, 275, 277, 279281 Flavocytochrome c-sulfide dehydrogenase
isotope fractionation by natural (FCSD), 141, 144
populations, 276 Flavoprotein component of Sir
isotope fractionation models, (SirFP), 204
277279 Flavoprotein SoxF, 140, 141, 144
Index 305
Fluorescence, 121 I
FMN, 204 Inorganic wastewaters, 289
FmoA Protein, 123 Intermediate oxidation state, 41
Formate dehydrogenase, 25, 30, 33 Interprotein disulfide, 140, 145, 146
Formate, 25, 30, 33 Iron oxidation, 63
FqoF, H2F420 dehydrogenase subunit of H2F420: Iron-sulfur cluster, 15, 16, 20, 21
quinone oxidoreductase complex, Isotope fractionation
204, 209 equilibrium, 274, 280
Fsr, 204, 208213 kinetic, 274, 280
Fumarate reductase, 17, 42
J
G Juan de Fuca Ridge, 252, 253
Gammaproteobacteria, 231235, 239, 241,
245248, 253, 254 K
Gene expression, 1 Kulunda Steppe, 226, 227
Geukensia demissa, 37, 38
Giant tubeworm. See Riftia pachyptila L
Glutathioneamide, 111 Lactate oxidation, 6, 7
Green sulfur bacteria, 6164, 7073, 91, 92 Lamprocystis, 90, 91
Lateral gene transfer, 67, 72
H L-cysteate sulfo-lyase, 175, 176
H2F420 dehydrogenase, 204, 209, 210 L-cysteate, 173176, 180
H2F420, reduced F420, 208 Linear alkylbenzenesulfonate, 173, 175
H2F420:quinone oxidoreductase, 204, 209 Lugworm. See Arenicola marina
Halophilic, 225, 226, 228236
Halorhodospira halophila, 62, 109 M
Halothiobacillus, 225, 228232, 236 Magnetococcus sp., 109
9Hc, 31 Magnetospirillum magnetotacticum, 109
Hdr, 26 Magnetotactic bacteria, 102
Heavy metal, 286, 291293 Marine sediments, 37, 38
Heme enzyme SoxXA, 140 Membrane complexes, 25, 30, 33
Heme-protein component of Sir (SirHP), 204 Menaquinol, 27, 28, 32
Hemoglobins, sulfide-transporting, 37 Menaquinone, 25, 2733
Hemoproteins, 124 3-mercaptopyruvate-sulfurtransferase, 39
Heterodisulfide reductase, 26, 66, 70, Metal sulfide, 291293
73, 108 Metallosphaera, 184
Hexahistidine tag, 122 Methanesulfonate, 171, 176
History of sulfur metabolism, 87, 93 Methanocaldococcus igneus, 207
2-His 1-Carboxylate facial triad, 194 Methanocaldococcus jannaschii, 202, 204,
Hmc, 3133 207211
Hme Methanococcoides burtonii, 212
Hydrogen cycling, 25 Methanococcus maripaludis, 207
Hydrogen oxidation, 66 Methanogen, 206, 207, 211213
Hydrogen sulfide Methanogenesis, 26, 202, 203, 206, 207,
H2S, 188, 196, 198, 199 209, 210, 212, 213
Hydrogen, 25, 3032 Methanogenic archaea, 204, 208, 212, 213
Hydrogenase, 14, 25, 26, 29, 30, 32, Methanogenic, 285288
248, 250 Methanoplanus limicola, 207
Hydrogenosomes, 41, 42 Methanopyrus kandleri, 207, 212
Hydrothermal environment, 184 Methanosarcina acetivorans, 207
Hydrothermal vent, 37, 206, 207 Methanosarcina barkeri, 212
Hypersaline, 225231, 234236 Methanosarcina, 26, 207, 210, 212
Hyperthermophile, 185, 188, 189 Methanothermobacter marburgensis,
Hyperthermophilic, 207 26, 207
306 Index
Methanothermobacter thermautotrophicus, Photosynthesis, 89, 91, 92, 94

207, 210, 212 Photosynthetic electron transport, 118
Methanothermococcus thermolithotrophicus, Phototrophic sulfur bacteria, 9294
207 Phototrophic sulfur oxidation, 127, 128
Methylcoenzyme M reductase, 203, 206, 211 Picrophilus torridus, 189, 190
Methylcoenzyme M, 203, 206, 211 Plant nitrate reductases, 161
Methylsulfonate monooxygenase, 175 Polysulfide reductase, 66, 70
Mine(s), 185 Polysulfide-reductase-like complex, 3, 70
Mitochondria, 3840, 42 Polysulfides, 111, 186, 194, 196
Mitochondria, anaerobic, 36 Polythionates, 184, 186, 198
Mitochondria, denitrifying, 42 Precipitation, 287, 292, 293
Mitosomes, 41, 42 Prostecochloris aestuarii, 109
MmpL proteins, 162 Prostecochloris vibrioformis, 109
Molybdenum cofactor, 153 -proteobacteria, 231235, 239, 241, 245248,
Moorella thermoacetica, 29, 110 253, 254
Mutagenesis, 120 -proteobacteria, 239, 241, 245, 246, 249, 251,
253, 254
N Proteomics, 78, 8183, 85
N-acetyltaurine, 182 Proton motive force, 24, 26
NADH, 25 PscD, 123
NADH:quinone oxidoreductase, 198 Psychrophilic, 212
Natronorubrum sp. HG 1, 197 Purple non-sulfur bacteria, 91, 92
Nitrate reductase, 246 Purple sulfur bacteria, 61, 8892, 102, 104,
Nitrate, 297301 112
Nitrate-reducing bacteria, 296, 298 Pyrite, 78, 79, 81, 83, 85
Nitrite, 298301 Pyrococcus furiosus, 186
Nitrogenase, 118
Nonaheme c, 14 Q
Non-heme iron, 189, 194 Qmo complex, 66, 70
Qmo, 26, 28, 32
O qmoABC, 70, 73
Oil field, 297, 298, 300
Ore, 185, 186 R
Oscillatoria limnetica, 67 Redox loop, 25, 28
OsmC-like protein, 165 Redox titration, 189
Outer membrane, 119, 123, 124 Reduction potential, 189, 194
Oxidation, 61 Reductive tricarboxylic acid (TCA) cycle,
Oxidative phosphorylation, 24 118, 245
Oxidoreductase SAOR, 187, 196, 197 Respiratory chain, 37, 38, 42
Oxygen, 37, 38, 4042 Respiratory NADH dehydrogenase
Oxygenase, 184, 186189, 194, 196 complex I, 210
Reverse methanogenesis pathway, 203, 209,
P 210
Paracoccus pantotrophus, 70, 73, 96, Rhodanese, 70, 7981, 104, 105
102105, 119, 139142, 144147 Rhodobacter capsulatus, 67, 92
Paracoccus versutus, 103 Rhodobacter sphaeroides, 92
Paracoccus, 247, 250 Rhodoquinone, 42
Periplasm, 25, 28, 29, 30, 33 Rhodospirillum rubrum, 9092
Periplasmic proteins, 83, 84 Rhodovulum sulfidophilum, 102, 105
Persephonella, 245, 246 Ribbed mussel. See Geukensia demissa
Phage Riftia pachyptila, 37
green sulfur bacteria, 72, 73 RNA virus, 72
Phospho-adenylylphosphosulfate, 151 RNA-directed DNA polymerase, 72
Phosphosulfolactate synthase, 173 RuBisCO-like protein, 67
Index 307
S Sulfide:quinone oxidoreductase, 6668, 87,

Salmonella enterica, 206 96, 106, 111, 187, 188, 198,
Saltern, 226228, 230, 232, 233, 236 242, 245
Schizosaccharomyces pombe, 38, 39 Sulfide, 14, 20, 6169, 7173, 118, 119, 123,
Sir, Asr of Escherichia coli, 204 203, 204, 207, 208, 210, 211, 213,
Siroamide, 107, 112, 204 286, 287, 291293, 297300
Siroheme, 20, 21, 95, 107, 204, 206, oxidation, 92, 9496, 102104, 106,
210, 212 107, 111
Small sulfite reductase, 202, 212, 213 Sulfite, 1317, 19, 20, 24, 26, 29, 30, 32, 33,
Snowblower vents, 251, 252 186188, 191, 196199, 202204,
Solfatara, 184, 185 206213
SorB cytochrome c, 163 redox potential, 152
Souring, 297, 298, 300, 301 Sulfite:acceptor oxidoreductase (SOR), 112,
Sox complex, 87, 103, 105, 186, 196, 113, 187, 196, 245, 247
197, 248 Sulfite assimilation, 202
Sox enzyme system, 139141, 144146 Sulfite dehydrogenase, 96, 102,
sox gene cluster, 70, 73 178181
Sox genes Sulfite detoxification, 207, 211
abundance, 142144 Sulfite oxidase family
complementation, 106 Archaea, 152, 155, 164, 165
inactivation, 106 bacterial enzymes, 153, 155, 164, 166
phototrophic bacteria electron acceptors, 152, 153
green sulfur bacteria, 70, 71, 142 electron transfer, 154, 160
purple sulfur bacteria, 144 phylogenetic analysis, 155, 165
SoxAX, 30 phylogenetic groups, 155, 163
soxCD plant SO, 152, 154, 164
soxJ, 70 properties, 154
soxK, 70 protein structure, 157159, 162
SoxY Roseobacter lineage, 157, 159
component of complex SorAB, 154, 155, 158161, 163
SoxYZ, 127 SoxCD, 155, 160164
conformation, 131, 132 sulfite oxidases, 161
crystal structure, 128 Thiobacillus denitrificans, 157, 159
intersubunit disulfide, 128, 132, YedY-like enzymes, 157
134136 Sulfite reductase, 14, 24, 26, 29, 30, 32, 33,
tetramerization, 129, 134 95, 102, 107, 108, 112, 202204,
thiosulfate binding, 127 206213
homodimer, 145, 146 Sulfiting agents, 151
SoyYZ, 71, 73 Sulfoacetaldehyde acetyltransferase, 175
SQR. See Sulfide:quinone oxidoreductase Sulfolactate sulfo-lyase, 176
Starkeya novella, 96, 105 Sulfolactate, 173176
Stygiolobusi, 184, 185 Sulfolobales, 184, 185, 198
Subseafloor biosphere, 251, 254 Sulfolobus
Succinate, 42 S. acidocaldarius, 185, 190, 198
Sulfate reducing bacteria, 285, 287, S. metallicus, 219, 221, 223
296298 S. solfataricus, 185, 190, 198
Sulfate reduction, 1, 2, 5, 6, 8, 285289, S. tokodaii, 185, 189, 190, 198,
291, 292 217221
Sulfate respiration, 14 Sulfopyruvate, 173
Sulfate thioesterase SoxB, 139141 Sulfoquinovose, 173
Sulfate, 1316, 20, 2426, 2833, 186, 198, Sulfur, 6067, 69, 7173
202204, 206, 208, 211213, sulfur disproportionation, 187189
285289, 291, 292 Sulfur bacteria, 259, 266, 267
Sulfate-reducing microorganisms, 47, 48, 50 Sulfur cycle, 296298
308 Index
Sulfur dehydrogenase, 104, 107 Thiocystis, 90, 91

SoxCD, 139140, 144, 155, 160163 Thiodenitrifyers, 232, 233, 235
SoxD, 162, 163 Thiohalomonas thiohalophilus, 228230,
absence of, 234, 235
Sulfur globules, 63, 66, 67 Thiohalorhabdus, 228230, 234, 235
envelope, 95, 102, 106 Thiohalospira, 228230, 232235
intracellular, 102, 106 Thiomargarita, 90
oxidation, 101, 104, 107 Thiomicrospira denitrificans,
Sulfur isotopes, 273277, 279, 280 143, 144
Sulfur metabolism, 78, 80, 8285 Thiomicrospira, 228231, 234, 241, 242, 249,
Sulfur oxidation 250, 253
chemotrophic, 141, 142 Thioploca, 90
phototrophic, 141, 142 Thioredoxin
Sulfur oxygenase reductase (SOR), 184, 186, SoxS, 139, 141, 144, 146, 147
187, 191 SoxW, 141, 144
diversity, 218, 219 Thiospirillum, 90, 91
physiological function, 219, 223 Thiosulfate, 30, 38, 39, 6166, 6973, 7885,
Sulfur oxygenase, 187189, 191 118, 119, 121, 177, 184, 186188,
Sulfur speciation, 260, 261, 266268 197199, 226229, 232234, 236,
Sulfur-binding protein SoxYZ, 141 245, 248, 249
Sulfur-binding protein, 127, 128 oxidation, 94, 101104, 106, 107,
Sulfuric acid, 184, 186 111, 113
Sulfurimonas, 241, 244246, 249, 250, 253 dehydrogenase, 102
Sulfurisphaera, 184 oxidizing multienzyme complex
Sulfur-oxidizing bacteria (SOB) (TOMES), 159, 162164
free-living, 241249 reductase, 104
SOB, 225, 226, 228236 sulfur transferase, 79, 81
symbiotic, 239, 240 Thiosulfate:quinone oxidoreductase, TQO, 82,
Sulfur-oxidizing microorganisms, 51 187, 197
Symbionts Thiothrix, 90, 91, 102, 103
endosymbionts, 239, 241, 248, 254 Tmc, 3133
epibionts, 239, 241 4-toluenesulfonate, 175
Synchrotron radiation, 260, 261 TpIc3, 3032
Syntrophobacter fumaroxidans, 110 Transcript quantity, 1
Transport systems, 82
T
Tapes philippinarum, 39 U
Taurine dehydrogenase, 177, 178 Ubiquinone, 38, 39
Taurine, 171, 173180
Taurine:pyruvate aminotransferase, 177 V
Tetrathionate 63, 64, 94 Volcanic eruptions, 251
formation, 102
hydrolase, 184, 187, 188, 197, 198
Thermophiles, 207 W
Thermothrix, 103 Wastewater, 285, 286, 288292
Thiobacillus denitrificans, 26, 29, 93, 95, 102,
109, 196 X
Thiocapsa roseopersicina, 94 XANES spectroscopy, 87, 96, 262264, 266,
Thiocapsa, 90, 91, 94, 96 267, 269, 270
Thiocyanate, 227229, 234236 X-ray crystallography, 191, 192

Microbial Sulfur Metabolism

Uploaded by

Document Information

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

Microbial Sulfur Metabolism

Uploaded by

Copyright:

Available Formats

Christiane Dahl Cornelius G.

Microbial Sulfur Metabolism

Library of Congress Control Number: 2007929727

ISBN-13 978-3-540-72679-1 Springer-Verlag Berlin Heidelberg New York

Springer-Verlag is a part of Springer Science+Business Media

Springer-Verlag Berlin Heidelberg 2008

Editor: Dr. Christina Eckey, Heidelberg

Printed on acid-free paper 149/3152-HM 5 4 3 2 1 0

Adam P. Arkin, Howard Hughes Medical Institute, Department of Bioengineering,

Karin Denger, Department of Biology, The University, 78457 Constance, Germany

Carlos A. Jerez, Laboratory of Molecular Microbiology and Biotechnology,

Armin Quentmeier, Lehrstuhl fr Technische Mikrobiologie, Fachbereich Bio-

Marcus Vallero, Subdepartment of Environmental Technology, Agricultural

1 Genetics and Genomics of Sulfate Respiration in Desulfovibrio. . . . 1

2 Living on Sulfate: Three-Dimensional Structure

3 Respiratory Membrane Complexes of Desulfovibrio . . . . . . . . . . . . . 24

3.2.1 The Qmo Complex . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 26

4 Biochemical and Evolutionary Aspects of Eukaryotes

5 Evolution and Ecology of Microbes Dissimilating

6 Genomic and Evolutionary Perspectives on Sulfur

6.1.1Green Sulfur Bacteria . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 61

7 Differential-Expression Proteomics for the Study

8 Sulfur and Light? History and Thiology

8.6 The Age of Enzymology and Isotope Labeling . . . . . . . . . . . . . . . 93

9 Thiosulfate and Sulfur Oxidation in Purple Sulfur Bacteria . . . . . . 101

10 Sulfur Oxidation in Chlorobium tepidum

11 Structural Insights into Component SoxY

12 Redox Control of Chemotrophic Sulfur

13 Bacterial Sulfite-Oxidizing Enzymes Enzymes

13.5.2 Group 2: Classic Sulfite-Oxidizing Enzymes

14 Sulfonates and Organotrophic Sulfite Metabolism . . . . . . . . . . . . . . 170

15 Oxidation of Sulfur and Inorganic Sulfur Compounds

16 A Novel Coenzyme F420 Dependent Sulfite Reductase

17 Archaeal and Bacterial Sulfur Oxygenase-Reductases:

18 Diversity of Halophilic Sulfur-Oxidizing Bacteria

19 Sulfur Oxidation at Deep-Sea Hydrothermal Vents. . . . . . . . . . . . . . 238

20 Speciation Analysis of Microbiologically Produced

21 Controls on Isotope Fractionation During Dissimilatory Sulfate

22 Bioprocess Engineering of Sulfate Reduction

22.4 Sulfate Reduction for Metal Recovery/Reuse . . . . . . . . . . . . . . . 292

23 Impact of Nitrate on the Sulfur Cycle in Oil Fields . . . . . . . . . . . . . . 296

Judy D. Wall, Adam P. Arkin, Nurgul C. Balci, Barbara Rapp-Giles

Sulfate-reducing bacteria (SRB) are Gram-negative deltaproteobacteria, ubiqui-

Log2 RNA/ Genomic DNA

Fig. 1.1 Histogram of transcripts of a hisD, DVU0796, average log2(RNA/genomic DNA) =

1.3 Sulfate Metabolism

The enzymology of sulfate reduction by Desulfovibrio strains is rather mature (Peck

putative sulfate permeases (DVU0053/0279/0746/0747/1999) none was found to be

1.4 Lactate Oxidation

Electron donors preferred by Desulfovibrio tend to be strain-specific, although

For the hydrogen cycle to function, it is necessary to have hydrogenases located

Table 1.5 Expression levels of putative cytoplasmic hydrogenase complexes in

1.6 Transmembrane Electron-Conducting Complexes

BRENDA (1987) Cologne University Bioinformatics Center. http://www.brenda.uni-koeln.de/

Gnter Fritz, Alexander Schiffer, Anke Behrens, Thomas Bchert,

S2 in hydrogen sulfide (Amend and Shock 2001). Interconversions of these species

2.2 Adenosine 5-Phosphosulfate Reductase

APS + 2e AMP + HSO3 ( )

Hydrolytic cleavage of the SOP moiety in APS yields approximately 80 kJ mol1

2.2.1 Molecular Properties of APSR

The molecular parameters of APSR from sulfate-reducing bacteria have been a

dithionite gave EPR signals characteristic for magnetically interacting [4Fe4S]

2.2.2 Three-Dimensional Structure of APSR

2.2.3 Reaction Mechanism of APSR