MTHFR Polymorphisms and Disease

MEDICAL INTELLIGENCE UNIT
Per Magne Ueland and Rima Rozen

UELAND • ROZEN
MTHFR Polymorphisms
and Disease
MIU
MTHFR Polymorphisms and Disease
MEDICAL
INTELLIGENCE
UNIT
MTHFR Polymorphisms
and Disease
Per Magne Ueland, M.D.

LOCUS for Homocysteine and Related Vitamins
University of Bergen
Bergen, Norway
Rima Rozen, Ph.D., FCCMG

Departments of Human Genetics, Pediatrics and Biology
McGill University-Montreal Children's Hospital
Montreal, Canada
LANDES BIOSCIENCE EUREKAH.COM

GEORGETOWN, TEXAS GEORGETOWN, TEXAS
U.S.A. U.S.A.
MTHFR POLYMORPHISMS AND DISEASE
Medical Intelligence Unit
Eurekah.com
Landes Bioscience
Copyright ©2005 Eurekah.com

All rights reserved.
No part of this book may be reproduced or transmitted in any form or by any means, electronic or
mechanical, including photocopy, recording, or any information storage and retrieval system,
without permission in writing from the publisher.
Printed in the U.S.A.
Please address all inquiries to the Publishers:
Eurekah.com / Landes Bioscience, 810 South Church Street, Georgetown, Texas, U.S.A. 78626
Phone: 512/ 863 7762; FAX: 512/ 863 0081
http://www.eurekah.com
http://www.landesbioscience.com
ISBN: 1-58706-217-8
While the authors, editors and publisher believe that drug selection and dosage and the specifications
and usage of equipment and devices, as set forth in this book, are in accord with current recommend-
ations and practice at the time of publication, they make no warranty, expressed or implied, with
respect to material described in this book. In view of the ongoing research, equipment development,
changes in governmental regulations and the rapid accumulation of information relating to the biomedical
sciences, the reader is urged to carefully review and evaluate the information provided herein.
Library of Congress Cataloging-in-Publication Data
MTHFR polymorphisms and disease / [edited by] Per Magne Ueland, Rima Rozen.
p. ; cm. -- (Medical intelligence unit)
Includes bibliographical references and index.
ISBN 1-58706-217-8
1. Methylenetetrahydrofolate reductase--Pathophysiology. 2. Isoenzymes--Pathophysiology. I. Title:
Polymorphisms and disease. II. Ueland, Per Magne. III. Rozen, Rima. IV. Series: Medical intelligence
unit (Unnumbered : 2003)
[DNLM: 1. Polymorphism, Genetic. 2. Genetic Diseases, Inborn. 3. Methylenetetrahydrofolate
Reductase (NADPH2)--deficiency. 4. Methylenetetrahydrofolate Reductase (NADPH2)--physiology.
5. Risk Factors. QU 500 M941 2005]
QP603.M48M68 2005
616'.042--dc22
2005016735
This book is dedicated to our children,
our most important contributions to human biology:
Jostein and Hans Olav (PMU)
Steven and Daniel (RR)
CONTENTS
Preface ................................................................................................ xiii
1. Molecular Biology of Methylenetetrahydrofolate Reductase

(MTHFR) and Overview of Mutations/Polymorphisms ........................ 1
Daniel Leclerc, Sahar Sibani and Rima Rozen
cDNA and Genomic Structure of MTHFR ........................................... 2
Mutations in Severe MTHFR Deficiency .............................................. 7
Polymorphisms in MTHFR ................................................................. 12
Recommendations for Nomenclature of MTHFR Mutations
and Numbering of Bases ................................................................. 16
Animal Model of MTHFR Deficiency ................................................ 17
2. Assays for Methylenetetrahydrofolate Reductase Polymorphisms ........... 21

Arve Ulvik and Per Magne Ueland
Methods for Genotyping MTHFR SNPs ............................................. 21
Preparation of DNA Material .............................................................. 27
Throughput Considerations ................................................................ 27
New Technologies and Future Developments ..................................... 28
3. Biochemical Characterization of Human Methylenetetrahydrofolate

Reductase and Its Common Variants ................................................... 31
Kazuhiro Yamada and Rowena G. Matthews
4. Severe Methylenetetrahydrofolate Reductase Deficiency ...................... 41

Mary Ann Thomas and David S. Rosenblatt
Clinical Presentation ........................................................................... 42
Pathophysiology .................................................................................. 42
Prenatal Diagnosis ............................................................................... 43
Laboratory Findings ............................................................................ 44
Studies on Cultured Cells .................................................................... 44
Treatment ........................................................................................... 48
Genetics .............................................................................................. 49
Genotype-Phenotype Correlations ....................................................... 49
5. Mild MTHFR Deficiency and Folate Status ........................................ 54
Paul F. Jacques and Silvina Furlong Choumenkovitch
Mild MTHFR Deficiency ................................................................... 54
The Role of MTHFR in Homocysteine Metabolism ........................... 55
Effect of Folate Status on the Relation between MTHFR 677C→T
Genotype and Homocysteine .......................................................... 55
Effect of the 677C→T Mutation on the Response
to Homocysteine Lowering Therapy ............................................... 60
Effect of the 677C→T Mutation on Folate Status ............................... 61
MTHFR 1298A→C Mutation and Its Influence on tHcy
and Folate Concentrations ............................................................... 66
6. Riboflavin and Methylenetetrahydrofolate Reductase ........................... 71
Steinar Hustad, Jørn Schneede and Per Magne Ueland
Animal Studies .................................................................................... 72
Human Studies ................................................................................... 73
Implications ........................................................................................ 75
7. The Molecular Dynamics of Abnormal Folate Metabolism

and DNA Methylation: Implications for Disease Susceptibility
and Progression .................................................................................... 78
S. Jill James
MTHFR Activity and Folate Methyl Group Dispersal
for Normal DNA Methylation and DNA Synthesis ........................ 78
AdoMet/AdoHcy Dynamics and DNA Hypomethylation ................... 80
DNA Methyltransferases and the Histone/Chromatin Connection ..... 82
Folate Deficiency and Alterations in DNA Methylation
during Embryonic Development and Aging .................................... 85
DNA Methylation Instability and the Epigenetic Basis for Disease:
Modification by Diet ....................................................................... 87
8. Methylenetetrahydrofolate Reductase 677C→T Polymorphism

and Risk of Arterial Occlusive Disease ............................................... 100
Mariska Klerk and Petra Verhoef
The MTHFR 677C→T Polymorphism and Risk of CHD ................ 101
The MTHFR 677C→T Polymorphism and Risk of Stroke ............... 103
Evidence for a Causal Role of Homocysteine
in Occurrence of AOD .................................................................. 103
Possible Explanations for Study Heterogeneity .................................. 103
Bias ................................................................................................... 108
9. Methylenetetrahydrofolate Reductase and Venous Thrombosis ......... 113

Miranda B.A.J. Keijzer and Martin den Heijer
Meta-Analyses ................................................................................... 116
MTHFR 677C→T and Interaction with Other Risk Factors
for Venous Thrombosis ................................................................. 117
Folate Concentrations, Hyperhomocysteinemia
and Association with Venous Thrombosis ..................................... 119
MTHFR 1298A→C .......................................................................... 119
Compound Heterozygosity ............................................................... 120
10. Neural Tube Defects, Other Congenital Malformations and Single
Nucleotide Polymorphisms in the 5,10 Methylenetetrahydrofolate
Reductase (MTHFR) Gene: A Meta-Analysis ...................................... 125
Stein Emil Vollset and Lorenzo D. Botto
Background and Scope of the Chapter .............................................. 126
Identification of Studies, Data Extraction
and Statistical Approach ................................................................ 126
Main Findings ................................................................................... 127
Discussion ......................................................................................... 138
11. Pregnancy Complications ................................................................... 144
Willianne L.D.M. Nelen and Henk J. Blom
MTHFR 677C→T Polymorphism and Pregnancy-Related
Complications ............................................................................... 145
Treatment ......................................................................................... 155
Pathophysiology ................................................................................ 155
12. Neuropsychiatric Disease
and Methylenetetrahydrofolate Reductase .......................................... 163
Björn Regland
Depression ........................................................................................ 163
Schizophrenia .................................................................................... 164
Dementia .......................................................................................... 166
Parkinson .......................................................................................... 167
13. Methylenetetrahydrofolate Reductase Polymorphisms

and Renal Failure ............................................................................... 170
Manuela Födinger and Gere Sunder-Plassmann
MTHFR 677C→T and 1298A→C and Hyperhomocysteinemia
in Renal Failure Patients ................................................................ 172
MTHFR 677C→T and 1298A→C Polymorphisms
and Folate Status in Renal Failure Patients .................................... 173
Interaction of MTHFR 677C→T with Other Genetic
Polymorphisms and Homocysteine / Folate Status
in Renal Failure ............................................................................. 173
MTHFR 677C→T and Cysteine Levels in Renal Failure Patients ..... 173
MTHFR 677C→T and Cardiovascular Disease Risk
in Renal Disease ............................................................................ 173
Effect of MTHFR 677C→T and 1298A→C Polymorphisms
on Total Homocysteine Lowering Therapy
of Renal Failure Patients ................................................................ 174
MTHFR 677C→T and Kidney Transplant Survival.......................... 174
MTHFR 677C→T and 1298A→C Polymorphisms
and Hyperhomocysteinemia in Children with Renal Failure ......... 175
14. MTHFR Polymorphisms and Colorectal Neoplasia ............................ 179
Jimmy W. Crott and Joel B. Mason
Importance and Metabolism of Folate ............................................... 180
Effects of Folate Depletion ................................................................ 182
MTHFR 677C→T Polymorphism .................................................... 184
15. Methylenetetrahydrofolate Reductase Polymorphisms:

Pharmacogenetic Effects ..................................................................... 197
Bernd Christian Schwahn and Rima Rozen
Pharmacogenetic Effects of MTHFR Variants ................................... 198
Index .................................................................................................. 207
EDITORS
Per Magne Ueland
LOCUS for Homocysteine and Related Vitamins
University of Bergen
Bergen, Norway
Chapters 2, 6
Rima Rozen
Departments of Human Genetics, Pediatrics and Biology
McGill University-Montreal Children's Hospital
Montreal, Canada
Chapters 1, 15
CONTRIBUTORS
Henk J. Blom Martin den Heijer
Department of Paediatrics Department of Endocrinology
University Medical Centre Nijmegen University Medical Center Nijmegen
St. Radboud, The Netherlands Nijmegen, The Netherlands
Chapter 11 Chapter 9
Lorenzo D. Botto Manuela Födinger

National Center on Birth Defects Institute of Medical and Chemical
and Developmental Disabilities Laboratory Diagnostics
Centers for Disease Control University of Vienna
and Prevention Vienna, Austria
Atlanta, Georgia, U.S.A. Chapter 13
Chapter 10
Steinar Hustad
Silvina Furlong Choumenkovitch LOCUS for Homocysteine
Jean Mayer U.S.D.A. Human Nutrition and Related Vitamins
Research Center University of Bergen
Tufts University Bergen, Norway
Boston, Massachusetts, U.S.A. Chapter 6
Chapter 5
Paul F. Jacques
Jimmy W. Crott Jean Mayer U.S.D.A. Human Nutrition
Vitamin and Carcinogenesis Laboratory Research Center
Jean Meyer U.S.D.A. Human Nutrition Tufts University
Research Center on Aging Boston, Massachusetts, U.S.A.
Tufts University Chapter 5
Boston, Massachusetts, U.S.A.
Chapter 14 S. Jill James
Department of Pediatrics
College of Medicine
University of Arkansas
Little Rock, Arkansas, U.S.A.
Chapter 7
Miranda B.A.J. Keijzer Björn Regland
Department of Endocrinology Institute of Clinical Neuroscience
University Medical Center Nijmegen University of Göteborg
Nijmegen, The Netherlands SU/Mölndal
Chapter 9 Mölndal, Sweden
Chapter 12
Mariska Klerk
Wageningen Centre for Food Sciences David S. Rosenblatt
Division of Human Nutrition MUHC-Royal Victoria Hospital
and Epidemiology Montreal, Quebec, Canada
Wageningen, The Netherlands Chapter 4
Chapter 8
Jørn Schneede
Daniel Leclerc LOCUS for Homocysteine
Departments of Human Genetics, and Related Vitamins
Pediatrics and Biology University of Bergen
McGill University-Montreal Children’s Bergen, Norway
Hospital Chapter 6
Montreal, Canada
Chapter 1 Bernd Christian Schwahn
Metabolic Unit
Joel B. Mason Clinic for General Pediatrics
Vitamin and Carcinogenesis Laboratory Heinrich-Heine-University
Jean Meyer U.S.D.A. Human Nutrition Düsseldorf, Germany
Research Center on Aging Chapter 15
Tufts University
Boston, Massachusetts, U.S.A. Sahar Sibani
Chapter 14 Departments of Human Genetics,
Pediatrics and Biology
Rowena G. Matthews McGill University-Montreal Children's
Life Sciences Institute Hospital
and Department of Biological Montreal, Canada
Chemistry Chapter 1
The University of Michigan
Ann Arbor, Michigan, U.S.A. Gere Sunder-Plassmann
Chapter 3 Department of Medicine III
Division of Nephrology and Dialysis
Willianne L.D.M. Nelen University of Vienna
Department of Obstetrics Vienna, Austria
and Gynaecology Chapter 13
University Medical Centre Nijmegen
St. Radboud, The Netherlands Mary Ann Thomas
Chapter 11 MUHC-Royal Victoria Hospital
Montreal, Quebec, Canada
Chapter 4
Arve Ulvik Stein Emil Vollset
Department of Pharmacology LOCUS for Homocysteine
University of Bergen and Related Vitamins
Bergen, Norway University of Bergen
Chapter 2 Bergen, Norway
Chapter 10
Petra Verhoef
Wageningen Centre for Food Sciences Kazuhiro Yamada
Division of Human Nutrition Life Sciences Institute
and Epidemiology and Department of Biological
Wageningen, The Netherlands Chemistry
Chapter 8 The University of Michigan
Ann Arbor, Michigan, U.S.A.
Chapter 3
PREFACE
Methylenetetrahydrofolate reductase (MTHFR) is a critical enzyme in both

folate and homocysteine metabolism. It first achieved medical recognition in
1972 with the report of severe deficiency of MTHFR in a patient with
homocystinuria, an inborn error of metabolism characterized by marked elevation
of homocyst(e)ine in plasma and urine. Although the majority of cases of
homocystinuria are due to a deficiency of the first enzyme in the transsulfuration
pathway for homocysteine metabolism, cystathionine-β-synthase (CBS), disruption
of homocysteine remethylation to methionine can also result in homocystinuria.
With the identification of additional patients with severe MTHFR deficiency,
the heterogeneity of this disorder became manifest. Of particular relevance to the
comments below was the report of a heat-sensitive MTHFR in some homocystinuric
patients, which was assumed to be caused by a deleterious mutation. A comprehen-
sive discussion of severe MTHFR deficiency can be found in Chapter 4 of this book.
Patients with homocystinuria, due to transsulfuration or remethylation de-
fects, frequently suffer from thromboses and display arteriosclerotic occlusive
changes in their vasculature. These types of observations led to the hypothesis that
more moderate elevations in plasma homocysteine could contribute to the risk for
cardiovascular disease. In 1988, a thermolabile form of MTHFR was identified in a
group of American patients with coronary artery disease, following enzymatic assays
in lymphocyte extracts that had been heated at 46˚C for 5 minutes. This heat-
sensitive enzyme appeared to be more common in the patient group compared to
the control group and was associated with a relatively milder deficiency than that
observed in patients with homocystinuria. The identification of some homocystinuric
families with both the severe and mild deficiency further complicated the situation.
Studies of MTHFR were limited at this time, since only a few laboratories, those
with biochemical expertise, could routinely measure MTHFR activity.
In the 1990s, molecular genetic investigations of MTHFR were under-
taken in an attempt to clone the cDNA/gene and to identify the molecular
basis of severe and mild MTHFR deficiency. In 1994, the isolation of the cDNA
was reported, with identification of the first mutations in severe MTHFR deficiency
in homocystinuric patients. This report was quickly followed by identification of a
common variant, a missense mutation at bp 677, which encoded the thermolabile
enzyme and predisposed to mild or moderate hyperhomocysteinemia. Subse-
quent studies demonstrated that the thermolability of MTHFR in some patients
with homocystinuria was due to the presence of the common variant, in addition
to the presence of the two deleterious mutations that contributed to the severe
deficiency and homocystinuria.
MTHFR is a FAD-dependent enzyme which catalyzes the irreversible con-
version of 5,10-methylenetetrahydrofolate to 5-methyltetrahydrofolate.
5-Methyltetrahydrofolate in turn serves as a methyl donor in the remethylation of
homocysteine (Hcy) to methionine. The enzyme therefore resides at an important
metabolic branch point directing the folate pool towards Hcy remethylation and
DNA methylation at the expense of DNA and RNA biosynthesis (Fig. 1). This
explains why individuals with the thermolabile variant are predisposed to elevated
plasma homocysteine under conditions of impaired folate status; low riboflavin may
Figure 1. Methylenetetrahydrofolate reductase and folate distribution. Methylenetetrahydrofolate reduc-
tase (MTHFR) catalyzes the irreversible reduction of 5,10-methylenetetrahydrofolate to
5-methyltetrahydrofolate. It affects the distribution between folate species used for DNA and RNA syn-
theses and the 5-methyltetrahydrofolate form required for homocysteine remethylation to methionine
and subsequent protein synthesis and DNA methylation. AdoMet= S-adenosylmethionine; CH3THF=
5-methyltetrahydrofolate; CH2THF= 5,10-methylenetetrahydrofolate; CHTHF= methenyl-
tetrahydrofolate; CHO= formyltetrahydrofolate; CH3DNA= methylated DNA; DHF= dihydrofolate;
Hcy= homocysteine; Met= methionine; THF= tetrahydrofolate.
also influence this association since riboflavin is the precursor for FAD. MTHFR defi-
ciency influences methylation of DNA and possibly other methyl acceptors and may
affect the levels of nucleotide pools available for DNA synthesis. These metabolic
changes presumably contribute mechanistically to some of the disease associations ob-
served with the MTHFR 677C→T polymorphism.
Mutation, polymorphism, variant and sequence change—these terms are used in-
terchangeably throughout the book to denote the 677C→T nucleotide substitution. A
mutation refers to any nucleotide change; a mutation can be benign or deleterious. A
polymorphism is simply a mutation that is common in the population (>1% prevalence);
similarly, a polymorphism can be benign or deleterious. All the above terminologies are
therefore correct when discussing the substitution at bp 677.
The molecular genetic assay for the polymorphism at bp 677 was straightforward.
The polymorphism was common in virtually all populations. A growing list of pheno-
types appeared to be influenced by homocysteine or folate levels (birth defects, pregnancy
complications, psychiatric disorders, etc.). These three facts laid the foundation for a surge
in the number of reports on MTHFR. Figure 2 depicts the number of publications per
year on MTHFR, showing the rapid rise in articles following the reports of the cDNA
sequence and of the polymorphism in 1994 and 1995, respectively.
Investigations of MTHFR deficiencies have paralleled the transition in medical
genetics from isolation of genes involved in single gene disorders to identification of
variants involved in complex multifactorial diseases. The advent of molecular genetic tech-
nologies in the 1970s and 1980s first resulted in the identification of genes involved in
“classic” genetic disorders such as the hemoglobinopathies, Duchenne muscular dystrophy
Figure 2. Publications on methylenetetrahydrofolate reductase (MTHFR) from 1988 to 2002. Publica-
tions on MTHFR were identified by literature searches of MEDLINE with the keywords “MTHFR” or
“methylenetetrahydrofolate reductase” (filled circles). To identify publications that discussed MTHFR
and mutations/polymorphisms, MEDLINE was searched for (MTHFR or methylenetetrahydrofolate
reductase) AND (polymorphism* OR 677 OR C677T OR variant OR mutation OR genotype OR risk
factor OR thermolabile) (filled squares). Figure was prepared by Dr. Daniel Leclerc, McGill University.
and cystic fibrosis, to name a few; the homocystinurias would also fall into this category.
In the new millennium, following the completion of the sequencing of the human ge-
nome, major efforts are being devoted to identification of single nucleotide polymor-
phisms (SNPs) that influence predisposition or risk for complex traits. MTHFR serves
as an interesting model for both types of disorders.
Predisposition or risk for disease implies that a genetic variant on its own is unlikely
to cause disease; other genetic variants and nongenetic factors are required for manifes-
tation of the phenotype. Investigations on MTHFR have been quite instructive in this regard,
since there is a limited number of polymorphisms in the human genome that clearly influ-
ence risk for complex traits. It has become evident that the single 677C→T variant may
only contribute to a modest increase in risk; other genetic variants and nongenetic factors
(folate status, riboflavin status, medication, or age) can modify the risk contributed by the
variant.
This book covers many of the complex traits that have been reported to be influ-
enced by the well-characterized 677C→T variant; there is less information on the
1298A→C variant, but it is discussed where appropriate. It is quite surprising, and
unique, that a single variant should influence such a wide variety of clinical conditions.
However, given the critical role of folates in DNA synthesis and repair, in homocysteine
regulation, and in the methylation cycle, it is understandable why the interest curve in
MTHFR (Fig. 2) is currently at 200-250 publications per year.
Rima Rozen and Per M. Ueland

CHAPTER 1
Molecular Biology
of Methylenetetrahydrofolate
Reductase (MTHFR) and Overview
of Mutations/Polymorphisms
Daniel Leclerc, Sahar Sibani and Rima Rozen
Abstract
M
ethylenetetrahydrofolate reductase (MTHFR) is a key regulatory enzyme in folate
and homocysteine metabolism. Research performed during the past decade has
clarified our understanding of MTHFR deficiencies that cause hyperhomocysteinemia
with homocystinuria, or mild hyperhomocysteinemia. The cloning of the MTHFR coding
sequence was initially followed by the identification of the first deleterious mutations in MTHFR,
in patients with homocystinuria. Shortly thereafter, the 677C→T variant was identified and
shown to encode a thermolabile enzyme with reduced activity. Currently, a total of 34 rare but
deleterious mutations in MTHFR, as well as a total of 9 common variants (polymorphisms)
have been reported. The 677C→T (A222V) variant has been particularly noteworthy since it
has become recognized as the most common genetic cause of hyperhomocysteinemia. The
disruption of homocysteine metabolism by this polymorphism influences risk for several com-
plex disorders, which are discussed in various chapters throughout this book. In this chapter,
we describe the complex structure of the MTHFR gene, summarize the current state of knowl-
edge on mutations/polymorphisms in MTHFR and discuss some initial findings in a
newly-generated mouse model for MTHFR deficiency.
Introduction
Methylenetetrahydrofolate reductase (MTHFR; EC 1.5.1.20) plays a central role in folate
and homocysteine metabolism by catalyzing the conversion of 5,10-methylenetetrahydrofolate
to 5-methyltetrahydrofolate, the primary circulatory form of folate which is utilized in ho-
mocysteine remethylation to methionine.1 The involvement of MTHFR in disease was first
published by Mudd et al who identified a patient with homocystinuria due to a severe defi-
ciency of the enzyme.2 This type of MTHFR deficiency, despite being the most common
inborn error of folate metabolism, is still relatively rare. In 1988, a thermolabile variant of
MTHFR was identified through enzymatic assays of lymphocyte extracts in patients with car-
diovascular disease.3 This milder deficiency, which appeared to be more common, resulted in a
mild to moderate elevation of plasma total homocysteine, an emerging risk factor for cardio-
vascular disease. The relationships, if any, between the severe and mild deficiencies were un-
clear, particularly since thermolabile MTHFR had also been observed in some families with
MTHFR Polymorphisms and Disease, edited by Per Magne Ueland and Rima Rozen.
©2005 Eurekah.com.
2 MTHFR Polymorphisms and Disease
homocystinuria.4,5 At that time, the investigations of MTHFR deficiency were limited to the
laboratories with expertise in biochemical genetics methodologies. The isolation of the cDNA
in 1994 opened up avenues for molecular genetic approaches to study MTHFR deficiencies.6
The cDNA isolation was quickly followed by identification of a common sequence variant at
bp 677 (C→T) which encoded the thermolabile form of the enzyme.7 This variant has become
recognized as the most common genetic cause of hyperhomocysteinemia and has been exten-
sively investigated as a risk factor for several multifactorial disorders associated with distur-
bances in homocysteine metabolism. The experimental progression, from delineation of the
rare, severe derangement of metabolism to the less deleterious consequences of the milder
677C→T mutation, was facilitated by the elucidation of molecular information on MTHFR.
In this chapter, we review the molecular biology of the MTHFR cDNA and gene. We provide
an overview of the genetic mutations in MTHFR, both rare and common sequence variants,
and briefly describe our Mthfr-deficient mice, which serve as animal models for both the severe
and mild forms of MTHFR deficiency. The reader is referred to other chapters in this book for
additional discussions of severe MTHFR deficiency and of the biochemical and clinical conse-
quences of the substitution at bp677.
cDNA and Genomic Structure of MTHFR

Cloning the MTHFR Coding Sequence
We initially isolated a 90-bp MTHFR cDNA from pig liver RNA, using peptide sequences
of the purified porcine liver enzyme that had been provided by Rowena Matthews and col-
leagues; degenerate oligonucleotides based on these peptide sequences were used for PCR am-
plification of porcine liver RNA.6 Sequence data from the porcine cDNA were then used to
screen a human liver cDNA library by PCR for isolation of a partial human cDNA (1.3 kb),
the predicted amino acid sequence of which showed strong homology to porcine MTHFR and
to bacterial metF genes. Another human cDNA library was then screened by plaque hybridiza-
tion with this 1.3 kb cDNA to obtain the missing C-terminal portion of the MTHFR coding
sequence.7 A short 3' UTR sequence was included in the resulting cDNA clone, followed by a
poly(A) tail (212 bp downstream of the stop codon).
The cDNA sequence deduced from this work was 2.2 kb in length and encompassed 11
exons.8 This cDNA sequence, available under GenBank GenInfo identifier (GI):6174884, was
used as the reference sequence in subsequent reports on the identification of the numerous
mutations in MTHFR (see below). When the deduced sequences of human and mouse MTHFR
proteins are aligned, 90% of amino acids are identical.8 The location of intronic boundaries
and most of the intron sizes are quite similar between the 2 species.
Porcine MTHFR was reported to be a dimer with a total molecular mass of about 150
kDa.9 Based on the structure of the porcine enzyme and initial Western blotting of human
tissues, the major product of the human MTHFR gene appeared to be a 77 kDa protein; a
second human isoform of approximately 70kDa was also observed on these Western blots. Our
expression of the 2.2 kb cDNA in bacterial extracts resulted in a catalytically-active 70kDa
protein, suggesting that additional coding sequences would be required to encode the 77kDa
isoform.7 Complex alternative splicing in the 5' end of MTHFR was reported by our group10
and others.11 More recently, we identified the predicted upstream translational start site of
MTHFR, generated from an alternatively-spliced mRNA, and expressed the cDNA; this cDNA
encodes the larger 77 kDa isoform.12 This deduced N-terminal sequence is conserved in hu-
man, mouse and porcine MTHFR genes.
With the increasing number of publications, sequences and other types of genetic informa-
tion, the Internet is playing an increasingly larger role in the study of fundamental biomedical
problems. Although GenBank was the first repository to contain sequence information on
MTHFR, Table 1 provides a partial listing of the vast array of biological information available
via the Internet about MTHFR. The reader is invited to consult these resources, as well as a few
Molecular Biology of MTHFR and Overview of Mutations/Polymorphisms 3
Table 1. Description of MTHFR gene-related entries in current databases
Tool or Database URL
GDB http://gdbwww.gdb.org/gdb-bin/genera/accno?GDB:370882
(The Genome Database)
a
GenBank http://www.ncbi.nlm.nih.gov/entrez/viewer.fcgi?val=6174884
GeneCards http://bioinfo.weizmann.ac.il/cards-bin carddisp?MTHFR
GeneTests http://www.genetests.org/query?gene=MTHFR
HGMD http://archive.uwcm.ac.uk/uwcm/mg/search/370882.html
(The Human Gene
Mutation Database)
HGNC http://www.gene.ucl.ac.uk/cgi-bin/
(HUGO Gene nomenclature/get_data.pl?hgnc_id=7436
Nomenclature Committee)
LocusLink http://www.ncbi.nlm.nih.gov/LocusLink/LocRpt.cgi?l=4524
Mapviewer – Human (NCBI) http://www.ncbi.nlm.nih.gov/mapview/
maps.cgi?ORG=hum&chr=1&maps=cntg-
r,clone,scan,ugHs,loc&VERBOSE=ON&cmd=
focus&fill=40&size-40&query=MTHFR
Mapviewer – Mouse (NCBI) http://www.ncbi.nlm.nih.gov/mapview/
maps.cgi?org=mouse&chr=4&maps=cntg-
r,bes,scan,ugMm,loc&VERBOSE=ON&query=
Mthfr&cmd=focus&fill=40&size=40"
b
OMIM http://www3.ncbi.nlm.nih.gov/htbin-post/Omim/dispmim?236250
(Online Mendelian
c
Inheritance in Man) http://www3.ncbi.nlm.nih.gov/htbin-post/Omim/dispmim?607093
Swiss-Prot http://ca.expasy.org/cgi-bin/niceprot.pl?P42898
a Used as the reference sequence for numbering MTHFR mutations (see text). b OMIM entry for MTHFR
deficiency. c OMIM entry for MTHFR gene
others that are cited throughout this chapter, to complement the text. However, objective analysis
of the compiled data requires consultation of relevant publications.
MTHFR/Mthfr Transcripts
Northern blot analysis has revealed MTHFR transcripts of approximately 2.8 and 7.2-7.7
kb in all tested tissues, and another of approximately 9.5 kb in brain, muscle, placenta, and
stomach.8,12,13 For Mthfr, transcripts of 7.4, 6.3, 3.0 and 2.8 kb were observed.12 The
different-sized transcripts result from alternate transcription start sites and multiple
polyadenylation signals. The total abundance is low, and the proportion of each transcript
differs among tissues. Overall expression is more intense in testis, intermediate in brain and
kidney, and lower in other examined tissues.
Although difficult to achieve due to their lengths, isolation and analysis of the 5' and 3'
ends of MTHFR/Mthfr cDNAs12 has provided valuable information towards completion of
the gene structure and analyses of regulatory sequences. Knowledge of the 3' UTR length
(approximately 5 kb and 4 kb for the majority of MTHFR and Mthfr mRNA species,
respectively) made it possible to predict the size of 5’UTR sequences. Since the coding se-
quence is about 2 kb and the main MTHFR transcripts on Northern blots do not exceed 9.5 kb
(human) and 7.4 kb (mouse), this suggested that 5’UTR sequences were less than 2.5 kb and
1.4 kb for most of human and murine MTHFR mRNAs, respectively. Two clusters of tran-
scription start sites of MTHFR/Mthfr mRNAs have been mapped by ribonuclease protection
assays and are consistent with observations obtained by RT-PCR methodologies and Northern
blotting.11 The identification of transcription start sites was critical for the prediction and the
subsequent preliminary analysis of two promoters of the gene, that have to lie upstream of the
transcribed sequences (Tran and Rozen, unpublished data).
Proximal transcription start sites in human MTHFR were identified 10 bp and 60 bp up-
stream of the ATG encoding the 77 kDa protein.12 These short 5’UTRs are expected to result
in efficient translation of the 77-kDa enzyme. Some transcripts originating from an upstream
region undergo splicing and do not contain the ATG for the long isoform (it is in a spliced-out
segment); these transcripts are predicted to translate the 70 kDa protein, with a 5’UTR of
approximately 50 nucleotides.12
Mapping
By fluorescence in situ hybridization (FISH), the human MTHFR gene was localized to
1p36.3.6 We also found close physical linkage between MTHFR and CLCN610 which was inde-
pendently confirmed by Gaughan et al.13 Using RFLP analysis of 94 mice from an interspecific
backcross panel, we localized the mouse Mthfr gene to the distal portion of mouse chromosome
4,14 which is the expected position for Mthfr based on known synteny between human and
mouse genomes. This is also consistent with the observed linkage between murine Mthfr and
Clcn6.12 For human MTHFR, we sequenced genomic clones that encompass 11 kb of sequences
upstream of the MTHFR coding sequence.12 This DNA segment overlaps with GenBank entry
AL02115, which shows that MTHFR is flanked by CLCN-6, NPPA, and NPPB at its 5’end. The
location of these genes and the numerous STS markers in the genomic contig NT_004488.3
(which is the contig physically linked to MTHFR) concur with the result obtained by FISH.
Exploring the Introns

Although drafts of the sequence of human and mouse genomes are available and public data-
bases contain a wealth of information about MTHFR-related sequences, the chromosome 1 seg-
ment encompassing the MTHFR gene is still a mosaic of sequences varying in data quality. None-
theless, even in this incomplete state, the available information is useful. For example, several
sequence contigs have made it possible to obtain the entire sequence of introns, although the
relevant BAC clone has not been sequenced and/or assembled in its entirety. Figure 1 displays
preliminary data from two BAC clones that contain a piece of human chromosome 1 DNA
encompassing the MTHFR gene. The data reported as “clone 106H5 sequence” received an am-
biguous annotation and an incorrect chromosomal location, which explains its “dead” status in
NCBI databases. However, it can still be accessed using the NCBI ENTREZ retrieval tool with
GI:15072573. The sequence of a segment of mouse chromosome 4 containing the entire Mthfr
gene has been recently completed; the relevant BAC clone is also shown in Figure 1.
Detailed knowledge of MTHFR intronic sequences is critical for performing mutation screen-
ing across all exons and for the design of DNA diagnostic assays. It is also essential for the
complete understanding of mutations that might involve activation of intronic cryptic splice
sites. Although these types of mutations have not yet been reported for MTHFR, the availabil-
ity of intronic data should facilitate their eventual identification and interpretation. Table 2
provides a list of primers used to amplify individual exons of the human gene, for mutation
identification within exons and adjacent splice sites. The last base of the intronic primers is at
least 20bp away from the exonic sequence, except for exon 9 primers which are 14 and 16 bp
away from the exon. Two exons (exons 4 and 5) were amplified using two fragments.
Figure 1. Comparison of sequence data in current genome databases and previously reported intron sizes
for MTHFR/Mthfr. Sequences of clones 106H5, 196P5 and 139J21 were deposited in GenBank databases
under GI 15072573, 8389575 and 21104044, respectively. The approximative sizes of amplicons using
relevant exonic PCR primers are those reported in reference 8. The BLASTN program (http://
www.ncbi.nlm.nih.gov) was used for comparing GenBank accession nos U09806 and NM_010840 (hu-
man and mouse MTHFR cDNAs) against all indicated genomic databases. “HTGS” and “nr” are the
“High-Throughput Genome Sequences” and “nonredundant” divisions of GenBank. The “Ensembl” ge-
nome browser is accessible through http://www.ensembl.org. The MTHFR-related sequences in clone
106H5 were “deleted” by NCBI staff in October 2001, but they are still available under a “dead” status in
NCBI databases (see text). Sequencing of human intron 3 in BAC clone 196P5 was only partially per-
formed. Sizes are reported as bp.
Mthfr Pseudogene
We have recently shown that the mouse genome contains a pseudogene for Mthfr.15 The
relevant locus (Mthfr-ps, 1259 bp) has been characterized. This nonfunctional pseudogene,
which is similar to the Mthfr paralogous gene, arose by retrotransposition of a mis-spliced
Mthfr transcript. The absence of intron 1, the partial retention of intron 2 and the location of
this gene on chromosome 5 are features that are indicative of a partially-processed pseudogene.
The lack of function was evident from a truncated coding sequence and from failure of
reverse-transcription assays. These findings require consideration when designing PCR-based
assays of the murine gene. Since some DNA is often present in a RNA sample, studies of the
Mthfr genomic or mRNA sequences, including regulatory studies of Mthfr, need to distinguish
between the functional transcript and the partially-processed nonfunctional pseudogene present
in DNA.
6
Table 2. Primers used to amplify individual MTHFR exons
Exon Direction Annealing Length of PCR

Amplified of Primer Locationc Primer Sequence (5'→3')d Temperature Product (bp)
1a Sense Intron 1 AACCTGCCACTCAGGTGTCTTG 66 389

Antisense Intron 2 TGACAGTTTGCTCCCCAGGCAC
2 Sense Intron 2 GGAAGGCAGTGACGGATGGTAT 66 372
Antisense Intron 3 ACCAAGTTCAGGCTACCAAGTGG
3 Sense Intron 3 GTAGAGGGAACAGAAAGGGTCTC 66 233
Antisense Intron 4 TCTCAGCCTCCCTAGCTCCATC
4 (part 1) Sense Intron 4 CTCGCCTTGAACAGGTGGAGG 66 183
Antisense Exon 4 AGCGGAAGAATGTGTCAGCCTC
4 (part 2) Sense Exon 4 TGAAGGAGAAGGTGTCTGCGGGA 68 198
Antisense Intron 5 AGGACGGTGCGGTGAGAGTGG
5 (part 1) Sense Intron 5 CACTGTGGTTGGCATGGATGATG 68 253
Antisense Exon 5 CCACTGGCCAGAAGCTCCTGG
5 (part 2) Sense Exon 5 GATGCTGCCATCCGCAACTATG 68 209
Antisense Intron 6 GGCTGCTCTTGGACCCTCCTC
6 Sense Intron 6 TGCTTCCGGCTCCCTCTAGCC 68 251
Antisense Intron 7 CCTCCCGCTCCCAAGAACAAAG
7 Sense Intron 7 CTGGGcATGTGGTGGCACTGC 66 321
Antisense Intron 8 CGCAGCCTGGCCTGCAGCTGG
8 Sense Intron 8 AGAGGGACTCAGGGTGCCAA 61 297
Antisense Intron 9 GGAACCCACGGGTGCCGGTC
9 Sense Intron 9 GGGTTGGTGACAGGCACC 58 169
Antisense Intron 10 CTGGTACTCTGTGGGAACC
10 Sense Intron 10 CTGTCCAGGCAGTGGGACT 62 268
Antisense Intron 11 CAGtACACGCTTGGGACTTG
11 b Sense Intron 11 GAGCCCTGTTAATCTTGCCTC 64 325
Antisense Exon 11 ACAGGAGTGGCTCCAACGCA
a Exon 1 is partially amplified.12 The amplicon for exon 1 encompasses the translational initiation site of the 70-kDa isoform. b Exon 11 is partially amplified, to
cover the C-terminus area. c Based on ref. 8. d The small underlined letters indicate positions where unintentional mistakes had been introduced into the
oligonucleotides. The “c” had been replaced by a “G” and the “t” by a “C” in these imperfect primers. These mismatches did not affect PCR outcome.
Mutations in Severe MTHFR Deficiency

Cloning of the MTHFR cDNA/gene enabled the identification of disease-causing muta-
tions in patients with severe MTHFR deficiency and homocystinuria. Although only 50 or so
patients have been diagnosed worldwide with this type of deficiency, it is the most common
inborn error of folate metabolism.16 Clinical symptoms displayed by MTHFR-deficient pa-
tients include developmental delay as well as various neurological and vascular problems, such
as seizures, thromboses and vascular lesions. The chapter by MA Thomas and DS Rosenblatt
provides a more comprehensive analysis of the clinical features in severe MTHFR deficiency.
To date, 34 mutations have been identified in MTHFR-deficient patients with
homocystinuria (Table 3, Fig. 2); detection mechanisms have included single strand conforma-
tion polymorphism (SSCP) analysis and direct sequencing of PCR or RT-PCR products.6,17-24
Primers for amplification of individual exons (Table 2) were designed on the basis of published
genomic sequences.8 Identified mutations were predicted to be deleterious based on three cri-
teria: (1) their absence from the general population, (2) the degree of conservation of the
amino acid change, and (3) location of the mutated codon within a conserved region and/or
predicted secondary structure. Whenever possible, mutations were expressed either in an in
vitro bacterial expression system20,25 or in vivo in yeast,26 and assayed for enzyme activity
(Table 4).
Table 3 summarizes all the published mutations found in patients with severe MTHFR
deficiency, along with their ethnic origin, age at onset of symptoms and residual activity of the
enzyme. Twelve of the 34 mutations are present in the homozygous state in patients, while the
remaining 22 are heterozygous. The 34 mutations can be classified into 8 nonsense mutations,
23 missense, 1 deletion and 2 splice variants. Exons 5 and 6 have the most mutations relative to
their size. Since both exons are within the catalytic domain and may be involved in substrate
binding,28 mutations in these exons may be more deleterious.
Nonsense Mutations
The eight nonsense mutations (28A→T, 559C→T, 1084C→T, 1134C→G, 1274G→A,
1420G→T, 1711C→T and 1762A→T) are scattered throughout the MTHFR polypeptide
(Fig. 2). Homozygous nonsense mutations can be very useful as they represent naturally occur-
ring in vivo deletions of various portions of the C-terminus of the protein. Of the four ho-
mozygous mutant patients that have been reported, three mutations (559C→T in the catalytic
domain, 1084C→T in the linker region, and 1762A→T in the regulatory domain) result in
extreme enzyme deficiency (0-1%) in fibroblasts from these patients. It is not known whether
the alleles harboring these mutations are translated or whether the mRNA is degraded by
nonsense-mediated mRNA degradation (NMD);29 nonetheless, it seems that most of the
polypeptide is required for enzyme activity. The fourth homozygous nonsense mutation
(1711C→T in the regulatory region) was reported by Kluijtmans et al.22 The patient’s specific
enzyme activity was approximately 32%. This mutation is located in the C-terminus and does
not appear to exert as dramatic an effect on enzyme activity, although the most distal nonsense
mutation, at bp 1762, had been associated with very low activity.
The effects of the other four heterozygous nonsense mutations are unknown, but some
impact may be predicted based on their location. The 5'-most mutation (28A→T; R6X) is at
the beginning of the coding region and is unlikely to contribute to the patients’ (II1 and II3,
Table 3) residual enzyme activity. Mutation 1134C→G is just downstream of the linker re-
gion, leaving the catalytic domain intact but without potential inhibition from the SAM-binding
site in the regulatory domain. Mutation 1274G→A is within the predicted SAM-binding site,
while mutation 1420G→T is just downstream of it.
Deletion of the C-terminal domain of human MTHFR was carried out by our group, in a
bacterial expression system,20 and by another group that expressed the deleted cDNA in yeast
lacking endogenous MTHFR (met11-/-).26 When protein levels were normalized, the bacterial
expression system yielded a 4-fold increase in enzyme activity relative to the wild type enzyme
8
Table 3. Patients’ genotypes, age at onset of symptoms and enzyme activity in fibroblasts
Allele 1 Allele 2
Amino Amino
Ethnic Age at Activity DNA Acid DNA Acid 677 1298 Thermo-
Patient Origin Onset (% Control) Mutation Change Exon Ref. Mutation Change Exon Ref. Codon Codon lability
354 African 13 yr 19 792+1G→A ∆19 aa 4 17 ND ND ND 17 AA EE No
American
355 African 11 yr 14 792+1G→A ∆19 aa 4 17 ND ND ND 17 AA EE No
American
356 Italian 16 yr 20 985C→T R325C 5 17 985C→T R325C 5 17 VV EE Yes
American
458 Caucasian 11 yr 10 167G→A R52Q 1 17 1015C→T R335C 5 17 VV EE Yes
670 Japanese Died 9 mo 4 458G→T1 G149V 2 18 458G→T G149V 2 18 AA EE Yes
735 African 7 mo 2 692C→T T227M 4 17 692C→T T227M 4 17 AA EE ND
Indian
1084 Caucasian 3 mo 0 1755G→A M581I 10 19 ND ND ND 19 AV AE NA
1396 Caucasian 14 yr 14 167G→A R52Q 1 17 1081C→T R357C 5 17 AV AE Yes
1554 Native 1 mo 0 559C→T R183X 3 6 559C→T R183X 3 6 VV EE ND
American
1569 Caucasian 2 wks 0 1553delAG - 9 20 1420G→T E470X 8 20 AA AE ND
1627 Native 1 mo 1 559C→T R183X 3 6 559C→T R183X 3 6 AA EE ND
American
1767 Afr-Amer/ 1 mo 2 980T→C L323P 5 21 1141C→T R377C 6 18 AV EE No
Caucas.
1772 Caucasian 2 wk 1.6 164G→C R51P 1 21 249-1G→T - 2 18 AA AA No
1779 French 15 yr 6 482G→A R157Q 2 6 1711C→T R567X 10 21 AA AA Yes
Canadian
1794 Pakistani 1 mo 0 1762A→T K584X 10 21 1762A→T K584X 10 21 AA EE ND
1807 Japanese In Ist yr 3 764C→T P251L 4 17 764C→T P251L 4 17 VV EE Yes
Table continued on next page

Table 3. Continued
Allele 1 Allele 2
Amino Amino
Ethnic Age at Activity DNA Acid DNA Acid 677 1298 Thermo-
Patient Origin Onset (% Control) Mutation Change Exon Ref. Mutation Change Exon Ref. Codon Codon lability
1834 French Asymp 7 482G→A R157Q 2 6 1711C→T R567X 10 21 AA AA Yes
Canadian
1863 Caucasian 21 yr 14 482G→A R157Q 2 6 1727C→T P572L 10 21 AV AE Yes
1951 Caucasian 1st yr 5.3 1727C→T P572L 10 21 1025T→C M338T 5 20 VV EE Yes
2006 Caucasian 14 yr 7 1172G→A G387D 6 21 1768G→A E586K 11 21 AV EE Yes
2184 Caucasian 16 yr 29 358G→A A116T 2 21 1134C→G Y374X 6 21 AV EE Yes
2231 Turkish 6 wks 0.2 1027T→G W339G 5 20 1027T→G W339G 5 20 VV EE ND
2255 Saudi Arabia 12 yr 13 1274G→A W421X 7 20 471C→G I153M 2 20 AA EE Yes
2351 Caucasian 1st yr 5.5 1025T→C M338T 5 20 1141C→T R377C 6 20 AV EE No
CM Greek- 2 yr 0 983A→G N324S 5 22 983A→G N324S 5 22 VV EE NA
Macedonian
UB Turkish At birth NA 1027T→G W339G 5 22 1027T→G W339G 5 22 VV NA NA
Molecular Biology of MTHFR and Overview of Mutations/Polymorphisms
K Turkish 1 mo 0 1084C→T R358X 6 22 1084C→T R358X 6 22 VV NA NA

U Turkish 10 mo NA 1711C→T R567X 10 22 1711C→T R567X 10 22 AA NA NA
1 NA 5 yrs NA 1420G→T2 E470X 8 23 1274G→C2 W421S 7 23 NA NA NA
2 NA 4 wks NA 1010T→C2 L333P 5 23 1010T→C2 L333P 5 23 NA NA NA
II1 NA 11.5 yr 7.8 28A→T R6X 1 24 1615C→T R535W 9 24 VV AE NA
II3 NA 3 yr 8 28A→T R6X 1 24 1615C→T R535W 9 24 VV AE NA
Bases and amino acids are numbered in accordance with GI:6174884. NA = not available, ND = not determined. 1 There is another sequence change (459C→T),
which does not change the valine codon. 2 Reported using a different numbering system than the one used in this table.
9
Figure 2. The location of MTHFR mutations within the polypeptide and cDNA. Base pairs are shown below
the protein while the amino acid changes are shown above. Red circles= missense mutations; yellow dia-
monds= nonsense mutations; orange triangle pointing upward= deletion; blue triangles pointing down-
ward= splicing mutations. The polymorphisms at positions 677 and 1298 are shown in green.
(Table 4). Yeast met11-/- cells expressing only the N-terminus of MTHFR grew more slowly
than their parental cells, with an observed residual activity of 24%. Thus, in vitro expression
experiments suggest that the presence of the C-terminus has an inhibitory effect on MTHFR
activity; this finding is consistent with the localization of the SAM binding (inhibitory) do-
main in the C-terminal region. In vivo data in yeast suggest that the C-terminal domain is
critical for cell growth. The C-terminus may stabilize the protein in vivo, since the truncation
leads to reduced MTHFR protein levels.26
Missense Mutations
Most of the missense mutations are found in the catalytic N-terminal domain of MTHFR
(16 out of 23, Fig. 2). Mutations 1141C→T, 1172G→A, and 1274G→C all lie within the
predicted SAM-binding site and may influence SAM binding. Mutations 1615C→T,
1727C→T, 1755G→A and 1768G→A are downstream of the linker, while mutation
1081C→T resides within it. With the exception of mutation 1727C→T, the other three mu-
tations are found in patients diagnosed in their second decade of life, possibly indicating a mild
effect on enzyme activity. Furthermore, in vitro expression of the 1081C→T mutation indi-
cated it had no detectable effect on enzyme activity (Table 4); we therefore suggested that
perhaps it had an effect on enzyme stability.
Seven N-terminal missense mutations exist in the homozygous state in patients: 458G→T,
692C→T, 764C→T, 983A→G, 985C→T, 1010T→C and 1027T→G. All of these muta-
tions, except for 985C→T, are predicted to have a severe deleterious effect on enzyme activity
since the patients were diagnosed within the first decade of life. The detrimental effect of
mutations 458G→T, 692C→T, 983A→G, and 1027T→G was confirmed by various mu-
tagenesis experiments (Table 4). The enzyme carrying a mutation at bp 983 has been shown to
be stabilized by FAD in vitro, similarly to the enzyme carrying the polymorphism at bp 677.
Whether the patient with this mutation would benefit from riboflavin supplementation re-
mains to be determined. Mutation 985C→T is interesting in that the homozygous mutant
patient has 20% enzyme activity, but when the mutation was expressed in vitro, it resulted in
an approximate 5-fold increase in enzyme activity. Again, this may indicate that the mutation
affects enzyme stability rather than activity in vivo.20
Table 4. In vitro assessment of MTHFR activity for different mutations
Mutationa Constructb % Controlc Activity Study
None AE 100 ref. 27d

None VE 45
None AA 68
None VA 41
None AA 100 ref. 25d

None VA 47
167G→A 167AA 74
167G→A 167VA 50
692C→T 692AA 1
980T→C 980AA 0
980T→C 980VA 0
985C→T 985AA 496
985C→T 985VA 177
1015C→T 1015AA 11
1015C→T 1015VA 8
1081C→T 1081AA 99
1081C→T 1081VA 57
1141C→T 1141AA 85
1141C→T 1141VA 70
None AE 100 ref. 20d

None VE 42
983A→G 983AE 36
983A→G 983VE 16.5
1025T→C 1025AE 10
1025T→C 1025VE 3.7
1027T→G 1027AE 20
1027T→G 1027VE 10
C-domain deletion A-Dom 471
C-domain deletion V-Dom 284
None AE 100 ref. 26e

None VE 52
None AA 105
164G→C 164AEf 10
458G→T 458AE 3
980T→C 980AEf 3
1141C→T 1141AEf 145
C-domain deletion N-hMTHFR 24
a Mutations reported in patients suffering from severe MTHFR deficiency. Numbering is based on
GI:6174884. b Names of constructs: The number refers to the nucleotide mutated and the following
two letters refer to the amino acid residues at positions 222 and 429 (corresponding to polymorphisms
at bp677 and 1298, respectively). c Enzyme activities are reported relative to the wild type control in
each study. d These studies used a bacterial in vitro expression system. e This study used a yeast
expression system in which the endogenous MTHFR homologue (met11) had been deleted. We
deduced the numbers from their Fig. 2, using activities observed in galactose medium. f The
construction of these plasmids was not fully described in ref. 26. These plasmids carry the AE
background (Warren D. Kruger, personal communication).
Site-directed mutagenesis was used to examine the effect of some of the heterozygous muta-
tions on enzyme activity. Mutations 164G→C, 980T→C, 1015C→T, 1025T→C all had low
enzyme activity (10%, Table 4). Mutations 167G→A and 1141C→T did not have consider-
able effects on activity. Interestingly, mutation 1025T→C, present in exon 5, caused retention
of intron 5 in vivo. Although there is no clear explanation for this observation, it may allude to
the presence of an exon splicing enhancer in that region.
Other Mutations
Two splice site mutations have been identified. The first, 249-1G→T, is located within the
splice-acceptor dinucleotide (AG) in intron 1.18 The second, 792+1G→A, mutates a 5' splice
donor site, resulting in activation of a cryptic splice site and deletion of 57 nucleotides.17
Only one deletion has been reported in severe MTHFR deficiency - patient 1569 (Table
3).20 This is a two-nucleotide deletion of AG at positions 1553 and 1554, with a predicted
frameshift.
Severe Mutations and Mild Polymorphisms

In vitro and in vivo data suggest that the 677C→T polymorphism may modulate enzyme
activity even in patients with severe MTHFR deficiency. This has been demonstrated in in
vitro studies where the presence of the valine allele of the 677C→T polymorphism decreased
enzyme activity by approximately 50%, both on its own and in cis with severe mutations (Table
4; refs. 20 and 25). A similar phenomenon has been observed in vivo where the 677 polymor-
phism appears to contribute to the thermolability of MTHFR in patients with severe MTHFR
deficiency,18,20,21 although the initial reports4 had assumed that the thermolability was due to
the deleterious mutation (since the polymorphism had not yet been identified). Although the
presence of known deleterious mutations may be a good predictor of enzyme activity, the effect
of mild polymorphisms can add to the complexity of genotype-phenotype analysis.
Polymorphisms in MTHFR
677C→T Variant
Historically, investigations of MTHFR genetic defects had focussed on the characterization
of rare inborn errors leading to severe hyperhomocysteinemia and homocystinuria. With the
identification of a MTHFR polymorphism (i.e., a common mutation) that results in mild
hyperhomocysteinemia, it became clear that some diseases of adulthood, such as cardiovascular
disease, reflect milder versions of the fulminant biochemical lesions present in the newborn or
child with severe MTHFR deficiency.7
Biochemical and Molecular Studies

In an attempt to identify the molecular basis of severe and mild MTHFR deficiency, we had
performed SSCP analyses on several types of patients. Ironically, we first identified the C-to-T
substitution at bp677, that converts an alanine to a valine residue (A222V), in a patient with
severe MTHFR deficiency and homocystinuria.7 This substitution was found to be equally
common in homocystinuric patients and in a control population. Consequently, homocystinuric
patients often have 3 or 4 mutations in the MTHFR gene (two distinct severe mutations and
one or two copies of the A to V change, see ref. 18). To confirm that the substitution at bp 677
altered enzyme function, we performed site-directed mutagenesis of the cDNA and expressed
the mutation in bacterial extracts; the mutagenized cDNA encoded an enzyme with reduced
activity and increased thermolability.7 Furthermore, there was excellent correlation between
reduced specific MTHFR activity in lymphocytes at 37˚C, increased thermolability at 46˚C
and the presence of the A to V mutation in 96 patients with coronary artery disease (CAD).30
The initial report by Frosst et al7 also demonstrated the association of the homozygous
mutant genotype (677TT) with mild hyperhomocysteinemia. However, it soon became evi-
dent that this association was present only in individuals with low folate status.31 In that study
of 365 individuals from the NHLBI Family Heart Study, we found that the homozygous mu-
tant genotype was associated with higher levels of plasma homocysteine.31 However, when the
group was subdivided on the basis of plasma folate, there was a more dramatic increase in
plasma homocysteine in individuals who were homozygous mutant and had plasma folate
values below the median. There was no effect of the genotype on homocysteine levels in indi-
viduals with plasma folate values above the median. These findings suggested that folate supple-
mentation should be effective in treating hyperhomocysteinemia in individuals with the muta-
tion. A comprehensive discussion of this topic (mild MTHFR deficiency, hyperhomocysteinemia
and interaction with folate status) is provided in Chapter 5.
Biochemical studies of wild type and mutant enzymes have provided a rationale for the
protective effect of folate on hyperhomocysteinemia in mutant individuals. Lymphocyte ex-
tracts, heated in the presence and absence of 5-methylTHF, retain higher residual MTHFR
activity after heating with increasing amounts of folate; the protective effect is greater on the
mutant than on the wild type enzyme.28 This phenomenon has been reproduced with the
recombinant wild type and mutant human enzymes expressed in heated bacterial extracts. The
A222V polymorphism in human MTHFR has been mimicked in the E. coli enzyme by intro-
ducing the homologous mutation A177V. The biochemical properties of these bacterial en-
zymes, as well as more recent studies on purified human MTHFR,32 have addressed the mecha-
nisms of this protective effect and are discussed more thoroughly in Chapter 3.
The aforementioned studies also demonstrated that the MTHFR cofactor, FAD, could
protect the mutant enzyme from destabilization, suggesting that riboflavin, the precursor of
FAD, should be considered as a modifier of enzyme activity and consequently of
hyperhomocysteinemia (see Chapter 6). The mutation in bacterial MTHFR (A177V) increases
the propensity for bacterial MTHFR to lose its essential flavin cofactor, and folate may protect
the E. coli enzymes against flavin loss.
Clinical Impact
Since hyperhomocysteinemia had emerged as a risk factor for cardiovascular disease, the
677C→T variant became an excellent candidate for risk modification of this complex trait; the
initial studies33,34 supported this concept while subsequent studies reached different conclu-
sions. A couple of reports35,36 had also demonstrated elevated plasma total homocysteine in
families with neural tube defects. Consequently, soon after its initial identification, the MTHFR
variant was reported to be the first genetic risk factor for neural tube defects.37 The number of
clinical conditions influenced by the 677 variant has grown considerably; the majority of the
studies have used the initial HinfI digestion protocol7 for diagnosis of the variant. This book
devotes individual chapters to the first two disorders (cardiovascular disease and neural tube de-
fects) as well as to many of the more recent clinical conditions reported to be influenced by mild
MTHFR deficiency. Although the mechanisms by which this variant influences disease progres-
sion may be unique to the specific condition, the various possibilities include the following:
1. elevation of plasma homocysteine. Homocysteine or one of its metabolites may have direct
toxic effects on the vasculature38 or on embryo development.39
2. disruption in methionine or S-adenosylmethionine synthesis, with consequent effects on
methylation. A disruption in methylation may also occur due to the conversion of ho-
mocysteine to S-adenosylhomocysteine, an inhibitor of several methyltransferases. Indi-
viduals with the TT genotype have decreased methylation in lymphocytes; this disturbance
may be folate dependent.40 Since altered DNA methylation is associated with changes in
gene expression, mild MTHFR deficiency could influence developmental or oncogenic
processes through this mechanism.
3. altered distribution of folate metabolites—decreased MTHFR activity should result in a
decrease in methyltetrahydrofolate and an increase in other folate forms, such as
methylenetetrahydrofolate and other nonmethyl forms. This has been demonstrated in lym-
phocytes of TT individuals.41 The redistribution of folates could affect thymidine or purine
synthesis, with consequent effects on DNA synthesis or repair.
Figure 3. Population frequency of 677T homozygosity by geographic area. Ethnicity is also indicated for
some populations. Data are from healthy controls, derived from the meta-analysis by Botto and Yang42
(underlined geographic regions) or Ogino and Wilson41 (nonunderlined geographical regions), except for
the percentage in Spain that is derived from ref. 44. Data from 6 countries (indicated by a star) are from
percentages recalculated using results pooled from both Botto and Yang42 and Ogino and Wilson41 reports.
As discussed in subsequent chapters, the 677C→T polymorphism may only be a modest

risk factor for some disorders. However, from a world population standpoint, it may represent
a considerable burden since the variant is so common. The initial report identified homozygos-
ity in 10-15% of Canadian controls.7 Subsequent studies by many groups have revealed re-
gional and ethnic variations of the frequency of the 677C→T homozygous mutant genotype.
Figure 3 shows the prevalence of homozygosity by geographical and racial/ethnic group. The
prevalence of the variant is relatively high in the general population, with homozygosity (TT)
of 6-14% in several White populations. The 677T allele is less common in African Blacks and
in Blacks living outside of Africa (Brazil, United States), with homozygosity frequencies of less
than 2%. The 677T allele appears to be very common among Hispanics. Studies in Hispanic
populations in California and Colombia have shown %TT of 21% and 18%, respectively; this
is likely due to the equally high frequencies reported in their southern Mediterranean ancestral
population.42-44 Although there is considerable variation in population frequencies, the muta-
tion may have only occurred once, on a founder haplotype.45 This finding alludes to a selective
advantage of the 677T allele for maintenance of this high frequency haplotype in many popu-
lations. The variant has been shown to be protective in some neoplasias (see Chapter 14), but
this phenomenon would not explain a significant reproductive advantage. The aforementioned
effects of mild MTHFR deficiency on methylation or on DNA synthesis and repair could have
beneficial effects on early development, but experimental studies in this regard have not been
reported.
Other Polymorphisms in MTHFR

Table 5 lists all reported variants in MTHFR. Only the 677 and 1298 substitutions have
been expressed and confirmed to affect enzyme activity. The functional impact, if any, of the
other polymorphisms remains unknown.
A point mutation in exon 7 (1298A→C) results in a glutamate to alanine substitution
(E429A). Site-directed mutagenesis of the MTHFR cDNA and expression in bacterial extracts
have shown that the activity of the encoded enzyme is decreased (to 68% of the wild type
enzyme), but not as dramatically as that for the 677T allele (for which residual activity is 45%
Table 5. Polymorphic mutations in 5,10-methylenetetrahydrofolate reductase
Mutationa Affected Codonb Exon or Intronc Reference
129C→T P39P Exon 1 46

IVS2+533G→A n/a (intronic) Intron 2 45
677C→T A222V Exon 4 7
1068T→C S352S Exon 6 17
IVS6+31T→C n/a (intronic) Intron 6 18
1298A→C E429A Exon 7 47
1317T→C F435F Exon 7 48
IVS10+262C→G n/a (intronic) Intron 10 45
1793G→A R594Q Exon 11 49
a Nucleotide numbers are based on GenBank GI:6174884. b Amino acid numbering refers to the 70
kDa isoform7 (GenBank GI:6174884). c Exon designations are based on Goyette et al8
of wild type).48 The enzyme mutated at bp 1298 is not thermolabile. Activities in lymphocyte
studies of individuals with this variant closely parallel the results with the recombinant enzyme.
Homozygotes represent approximately 8% of individuals in the tested populations, largely
European (range is from 4% to 12% for most tested populations).50 These homozygotes do
not appear to have higher serum homocysteine levels than controls. However, individuals who
are compound heterozygotes for the 1298C and 677T alleles tend to have a biochemical profile
closer to that seen among 677C→T homozygotes, with increased serum homocysteine levels.48
The 1298A→C mutation was initially examined by PCR and MboII digestion. However, a
silent mutation (1317T→C) also creates a MboII site and results in a digestion pattern ex-
tremely similar to that of the 1298A→C variant. We reported a Fnu4HI diagnostic assay that
makes it possible to detect the mutation at position 1298 without interference by the
genotype at position 1317.48 We occasionally experienced difficulties with the Fnu4HI
diagnostic assay. For this reason, we designed a new ACRS assay (Artificially Created
Restriction Site) based on a MwoI digestion (unpublished), involving the sense primer:
5'-TGGGGGGCGGAGCTGGCCAGTGA-3'. The two underlined letters are intentional mis-
matches creating a MwoI site (GCNNNNNNNGC) that constitutes an internal digestion
control, in addition to introducing a second MwoI site if the 1298C allele is amplified. Following
amplification (35 cycles; 94˚C 1 min, 64˚C 1 min, 72˚C 2 min) with the antisense primer
(5'-AGGCCAGGGGCAGGGGATGAA-3'), the 137 bp amplicon is digested to generate frag-
ments of 127 bp (1298A allele) or 119bp (1298C variant).
Most studies have reported no or few cases with 677T and 1298C alleles in the cis configu-
ration.43,48,51,52 It is likely that these mutations arose independently on different alleles and
recombination has not occurred frequently enough, within the requisite small interval, to place
the two mutations on the same chromosome. Furthermore, a recombinant enzyme containing
both the 677T and 1298C substitutions has the same activity as the recombinant enzyme
containing only the 677T allele,27 suggesting that 677T/1298C homozygotes do not have
decreased survival (compared to 677TT homozygotes).
A third variant, R594Q, results from the change 1793G→A.49 It is less common than the
677C→T or 1298A→C polymorphism, with allele frequencies of 6.9% among Caucasians,
5.8% among Hispanics and 3.1% among African-Americans. Homozygosity was observed
only in Caucasians (a single individual in 159 tested subjects).
The 1068T→C allele (reported as 1059T→C)53 is silent and appeared to be in linkage dis-
equilibrium with the 1298A→C mutation. Trembath et al52 referred to it as 1059T→C instead
of 1068T→C because they used numbering that had been reported in an old version (GI:499223)
of U09806. In this entry, GenBank staff had removed the portion corresponding to the synthetic
linker that was present in the original reference sequence (published as Fig. 1 in ref. 6).
Recommendations for Nomenclature of MTHFR Mutations

and Numbering of Bases
This chapter would not be complete without a discussion of ambiguities related to MTHFR
reference sequences and nomenclature. The common C-to-T change at position 677 is often
referred to as “C677T”. Based on the official nomenclature system for human gene muta-
tions,54 this would correspond to the format for an amino acid codon change i.e., C677T
should stand for “Cysteine at codon 677 substituted by Threonine”. The designation “677C→T”
follows the nomenclature rules for a change in the nucleotide sequence. The reference sequence
was first shown in Figure 1 of ref. 6 for the initial report of MTHFR mutations. Because it
contained linkers used to generate the cDNA library, it was a “synthetic construct” based on
GenBank classification (see GI:6174884 for GenBank accession no U09806). As mentioned
in ref. 6, this sequence was predicted to be missing an upstream ATG and therefore the num-
bering started from the synthetic linker. The same numbering principle was used for describing
the thermolabile variant,7 and this numbering method was conserved through virtually all
subsequent publications. Ideally, the A of the ATG initiator methionine codon should be des-
ignated as nucleotide 1.54 Nucleotide 677 of the reference sequence corresponds to nucleotide
665 of the open reading frame for the 70 kDa isoform (GenBank accession nos XM_030156
and NM_005957, where a second upstream ATG had not yet been identified). Given the large
number of scientific publications in which MTHFR bases were numbered according to the
original report, it would become very confusing to establish a new reference sequence to con-
form to the guidelines suggested by Antonarakis et al54 even though the upstream ATG has
now been identified.12 Consequently, we suggest maintaining GI:6174884-based numbering,
as long as investigators clearly state the reference sequence in their reports. Because several
versions of the same accession number (U09806) are now present in the GenBank database, it
would be appropriate to mention GI:6174884, since older versions of U09806 have a different 5'
end. GI: 6174884 was created specifically to match the numbering used in the initial report.6
The GenBank GI:6174884 is appropriate for designation of bases in the short MTHFR
isoform sequence. There are, however, some specific cases that cannot use this reference se-
quence. If mutations are identified in the segment encompassing the coding sequences specific
to the 77 kDa MTHFR isoform, and are absent in GI:6174884, it would then be appropriate
to use negative numbering based on the initially- identified ATG initiator Met codon (for the
70 kDa isoform), with the nucleotide 5' to this ATG labelled “-1”. This would not create any
ambiguities as long as the proper reference sequence is cited, and it would also be appropriate
and unambiguous for designating bases in the 5’UTR sequences. In these situations, given the
alternative splicing and the alternative usage of two close splice acceptor sites (A2 or A3)12
located between the 2 ATG initiator Met codons, it may become confusing to choose a cDNA
reference sequence. Using a genomic DNA reference sequence would then be important (for
example, such human genomic sequences were submitted to GenBank and are cited in ref. 12).
The specific reference sequence (whether cDNA or genomic) needs to be quoted in all reports
of MTHFR mutations or gene structure.
For amino acid numbering, the protein sequence of the short isoform (associated with
GI:6174884) should be used, since this is already in common usage in the literature. However,
for designation of amino acids that are specific to the 77 kDa isoform, it would be appropriate
to specify which acceptor site was used to generate the mRNA in question (usage of A2 or A3
acceptor sites in ref. 12).
Animal Model of MTHFR Deficiency

To investigate the in vivo pathogenetic mechanisms of MTHFR deficiency, we generated
Mthfr knockout mice.55 Plasma total homocysteine levels in heterozygous and homozygous
knockout mice were 1.6- and 10-fold higher than those in wild type littermates, respectively.
Both heterozygous and homozygous knockouts had either significantly decreased
S-adenosylmethionine levels or significantly increased S-adenosylhomocysteine levels, or both,
with global DNA hypomethylation. Heterozygosity for the knockout allele does not yield an
abnormal phenotype. The homozygotes are smaller, with developmental retardation and cer-
ebellar pathology. Abnormal lipid deposition in the proximal aorta was observed in older het-
erozygotes and homozygotes, alluding to an atherogenic effect of hyperhomocysteinemia in
these mice.
Based on these initial observations, the homozygous knockout mice appear to be a good
animal model for homocystinuria due to severe MTHFR deficiency, based on the complete
enzymatic deficiency and the dramatic elevation in plasma total homocysteine. The heterozy-
gous knockout mice, with approximately 60% residual enzyme activity and a moderate eleva-
tion in plasma total homocysteine, appear to be a good animal model for individuals that are
homozygous for the 677T variant; these individuals have approximately 40% of the activity of
677C wild type homozygotes.30
Microarray analysis of brain RNA from knockout Mthfr mice revealed altered expression of
several genes.56 RT-PCR and, in some cases, Western blots were used as a validation method to
confirm a representative set of observations obtained from the analysis of the high-density
oligonucleotide arrays. Interestingly, Mthfd2 expression was increased in Mthfr-/- mice. This
gene encodes a bifunctional enzyme which can generate the MTHFR substrate
5,10-methyleneTHF; the altered expression is likely to be part of a cellular response triggered
by deficient MTHFR activity. Among other differentially-expressed genes, the decreased ex-
pression of Itpr1 (inositol 1,4,5-triphosphate receptor, type 1) may reflect homocysteine-induced
calcium influx, which is one of the proposed routes for compromised neuronal homeostasis.57
These mice have already proven useful in assessing the impact of betaine in MTHFR defi-
ciency.58 Although many studies had reported the homocysteine lowering effect of betaine in
homocystinuria, betaine administration in moderate hyperhomocysteinemic states had not
been extensively investigated. Betaine supplementation reduced plasma homocysteine in mice
of all 3 Mthfr genotypes (+/+, +/-, and -/-), restored liver betaine and phosphocholine levels in
the deficient mice, and prevented severe steatosis in the homozygous knockout animals. These
observations highlighted the importance of betaine as an alternate methyl donor when
folate-dependent remethylation is compromised and prompted us to examine this phenom-
enon in a human study, where we found a significant negative correlation between plasma
homocysteine and plasma betaine in patients with cardiovascular disease.58
Conclusion
Since the isolation of the cDNA in 1994, the work on the mammalian MTHFR gene has
resulted in significant advances in our understanding of its genomic organization, genetic varia-
tions and involvement in human disorders. Several important issues, however, remain to be
addressed. Little is known about the regulation of this gene despite the fact that the enzyme
links folate and homocysteine metabolism, and is involved in such critical cellular processes as
DNA synthesis and DNA methylation. Investigations are required to understand the regula-
tory regions and their modulation, as well as the factors that affect alternative splicing and
synthesis of the two protein isoforms. These are not easy tasks given the unusually complex
MTHFR gene structure.
Although numerous clinical association studies have been performed on MTHFR variants,
conclusions have been contradictory in some cases, due to the multifactorial nature of the
disorders and our inability to identify the multiple genetic and environmental factors that can
interact with MTHFR polymorphisms to impact disease risk. The biologic and tissue-specific
impact of MTHFR deficiency has also not been adequately addressed since these types of
investigations cannot be readily performed in human subjects; the availability of an animal
model may be useful in this regard.
References
1. Rosenblatt DS. Inherited disorders of folate transport and metabolism. In: Scriver CR, Beaudet
AL, Sly S et al, eds. The Metabolic and Molecular Bases of Inherited Disease. 7th ed. New-York:
McGraw-Hill, 1995:3111-3128.
2. Mudd SH, Uhlendorf BW, Freeman JM et al. Homocystinuria associated with decreased
methylenetetrahydrofolate reductase activity. Biochem Biophys Res Commun 1972; 46:905-912.
3. Kang SS, Zhou J, Wong PW et al. Intermediate homocysteinemia: A thermolabile variant of
methylenetetrahydrofolate reductase. Am J Hum Genet 1988; 43:414-421.
4. Rosenblatt DS, Erbe RW. Methylenetetrahydrofolate reductase in cultured human cells. II. Ge-
netic and biochemical studies of methylenetetrahydrofolate reductase deficiency. Pediat Res 1977;
11:1141-1143.
5. Kang SS, Wong PWK, Bock HGO et al. Intermediate hyperhomocysteinemia resulting from com-
pound heterozygosity of methylenetetrahydrofolate reductase mutations. Am J Hum Genet 1991;
48:546-551.
6. Goyette P, Sumner JS, Milos R et al. Human methylenetetrahydrofolate reductase: Isolation of
cDNA, mapping and mutation identification. Nature Genet 1994; 7:195-200.
7. Frosst P, Blom HJ, Milos R et al. A candidate genetic risk factor for vascular disease: A common
mutation in methylenetetrahydrofolate reductase. Nat Genet 1995; 10:111-113.
8. Goyette P, Pai A, Milos R et al. Gene structure of human and mouse methylenetetrahydrofolate
reductase (MTHFR). Mamm Genome 1998; 9:652-656.
9. Daubner SC, Matthews RG. Purification and properties of methylenetetrahydrofolate reductase
from pig liver. J Biol Chem 1982; 10:140-145.
10. Chan M, Tran P, Goyette P et al. Analysis of the 5' region of the methylenetetrahydrofolate
reductase (MTHFR) gene reveals multiple exons with alternative splicing, and an overlapping gene.
FASEB J 1999; 13:A1375.
11. Homberger A, Linnebank M, Winter C et al. Genomic structure and transcript variants of the
human methylenetetrahydrofolate reductase gene. Eur J Human Genet 2000; 8:725-729.
12. Tran P, Leclerc D, Chan M et al. Multiple transcription start sites and alternative splicing in the
methylenetetrahydrofolate reductase gene result in two enzyme isoforms. Mamm Genome 2002;
13:483-492.
13. Gaughan DJ, Barbaux S, Kluijtmans LAJ et al. The human and mouse methylenetetrahydrofolate
reductase (MTHFR) genes: Genomic organization, mRNA structure and linkage to the CLCN6
gene. Gene 2000; 257:279-289.
14. Frosst P, Zhang Z-X, Pai A et al. The methylenetetrahydrofolate reductase (Mthfr) gene maps to
distal mouse Chromosome 4. Mamm Genome 1996; 7:864-869.
15. Leclerc D, Darwich-Codore H, Rozen R. Characterization of a pseudogene for murine
methylenetetrahydrofolate reductase. Mol Cell Biochem 2003; 252:391-395.
16. Rosenblatt DS, Fenton WA. Inherited disorders of folate and cobalamin transport and metabolism.
In: Scriver CR, Beaudet AL, Sly WS et al, eds. The Metabolic & Molecular Bases of Inherited
Disease. 8th ed. New York: McGraw-Hill, 2001:3897-3933.
17. Goyette P, Frosst P, Rosenblatt DS et al. Seven novel mutations in the methylenetetrahydrofolate
reductase gene and genotype/phenotype correlations in severe MTHFR deficiency. Am J Hum
Genet 1995; 56:1052-1059.
18. Goyette P, Christensen B, Rosenblatt DS et al. Severe and mild mutations in cis for the
methylenetetrahydrofolate reductase (MTHFR) gene, and description of 5 novel mutations in
MTHFR. Am J Hum Genet 1996; 59:1268-1275.
19. Selzer RR, Rosenblatt DS, Laxova R et al. Adverse effect of nitrous oxide in a child with
5,10-methylenetetrahydrofolate reductase deficiency. N Engl J Med 2003; 349:45-50.
20. Sibani S, Leclerc D, Weisberg IS et al. Characterization of mutations in severe methylenetetrahydrofolate
reductase deficiency reveals an FAD-responsive mutation. Hum Mutat 2003; 21:509-520.
21. Sibani S, Christensen B, O’Ferrall E et al. Characterization of six novel mutations in the
methylenetetrahydrofolate reductase (MTHFR) gene in patients with homocystinuria. Hum Mutat
2000; 15:280-287.
22. Kluijtmans LA, Wendel U, Stevens EM et al. Identification of four novel mutations in severe
methylenetetrahydrofolate reductase deficiency. Eur J Hum Genet 1998; 6:257-65.
23. Homberger A, Linnebank M, Sewell A et al. Severe methylenetetrahydrofolate reductase deficiency:
Two novel genotypes with different clinical course. J Inherit Metab Dis 2001; 24(Suppl 1):50.
24. Tonetti C, Amiel J, Munnich A et al. Impact of new mutations in the methylenetetrahydrofolate
reductase gene assessed on biochemical phenotypes: A familial study. J Inherit Metab Dis 2001;
24:833-42.
25. Goyette P, Rozen R. The thermolabile variant 677C→T can further reduce activity when expressed
in CIS with severe mutations for human methylenetetrahydrofolate reductase. Hum Mutat 2000;
16:132-38.
26. Shan X, Wang L, Hoffmaster R et al. Functional characterization of human methylenetetrahydrofolate
reductase in Saccharomyces cerevisiae. J Biol Chem 1999; 274:32613-32618.
27. Weisberg IS, Jacques PF, Selhub J et al. The 1298A-->C polymorphism in methylenetetrahydrofolate
reductase (MTHFR): In vitro expression and association with homocysteine. Atherosclerosis 2001;
156:409-15.
28. Guenther BD, Sheppard CA, Tran P et al. The structure and properties of methylenetetrahydrofolate
reductase from Escherichia coli suggest how folate ameliorates human hyperhomocysteinemia. Nat
Struct Biol 1999; 6:359-365.
29. Frischmeyer PA, Dietz HC. Nonsense-mediated mRNA decay in health and disease. Hum Mol
Genet 1999; 8:1893-1900.
30. Christensen B, Frosst P, Lussier-Cacan S et al. Correlation of a common mutation in the
methylenetetrahydrofolate reductase gene with plasma homocysteine in patients with premature
coronary artery disease. Arterioclerosis Thrombosis and Vascular Biology 1997; 17:569-573.
31. Jacques PF, Bostom AG, Wiliams RR et al. Relation between folate status, a common mutation in
methylenetetrahydrofolate reductase, and plasma homocysteine concentrations. Circulation 1996;
93:7-9.
32. Yamada K, Chen Z, Rozen R et al. Effects of common polymorphisms on the properties of recom-
binant human methylenetetrahydrofolate reductase. Proc Natl Acad Sci USA 2001; 98:14853-14858.
33. Kluijtmans LA, van den Heuvel LP, Boers GH et al. Molecular genetic analysis in mild
hyperhomocysteinemia: A common mutation in the methylenetetrahydrofolate reductase gene is a
genetic risk factor for cardiovascular disease. Am J Hum Genet 1996; 58:35-41.
34. Gallagher PM, Meleady R, Shields DC et al. Homocysteine and risk of premature coronary heart
disease. Evidence for a common gene mutation. Circulation 1996; 94:2154-2158.
35. Mills JL, McPartlin JM, Kirke PN et al. Homocysteine metabolism in pregnancies complicated by
neural-tube defects. Lancet 1995; 345:149-151.
36. Steegers-Theunissen RP, Boers GH, Blom HJ et al. Neural tube defects and elevated homocysteine
levels in amniotic fluid. Am J Obstet Gynecol 1995; 172:1436-1441.
37. van der Put NMJ, Steegers-Theunissen RPM, Frosst P et al. Mutated methylenetetrahydrofolate
reductase as a risk factor for spina bifida. Lancet 1995; 364:1070-1072.
38. Bellamy MF, McDowell IF. Putative mechanisms for vascular damage by homocysteine. J Inherit
Metab Dis 1997; 20:307-15.
39. Rosenquist TH, Ratashak SA, Selhub J. Homocysteine induces congenital defects of the heart and
neural tube: Effect of folic acid. Proc Natl Acad Sci USA 1996; 93:15227-1532.
40. Friso S, Choi S-W, Girelli D et al. A common mutation in the 5,10-methylenetetrahydrofolate
reductase gene affects genomic DNA methylation through an interaction with folate status. Proc
Natl Acad Sci 2002; 99:5606-5611.
41. Bagley PJ, Selhub J. A common mutation in the methylenetetrahydrofolate reductase gene is asso-
ciated with an accumulation of formylated tetrahydrofolates in red blood cells. Proc Natl Acad Sci
1998; 95:13217-13220.
42. Botto LD, Yang Q. 5,10-methylenetetrahydrofolate reductase gene variants and congenital anoma-
lies: A HuGE review. Am J Epidemiol 2000; 151:862-877.
43. Ogino S, Wilson RB. Genotype and haplotype distributions of MTHFR677C>T and 1298A>C
single nucleotide polymorphisms: A meta-analysis. J Hum Genet 2003; 48:1-7.
44. Guillen M, Corella D, Portoles O et al. Prevalence of the methylenetetrahydrofolate reductase
677C>T mutation in the Mediterranean Spanish population. Association with cardiovascular risk
factors. Eur J Epidemiol 2001; 17:255-261.
45. Rosenberg N, Murata M, Ikeda Y et al. The frequent 5,10-methylenetetrahydrofolate reductase
C677T polymorphism is associated with a common haplotype in Whites, Japanese, and Africans.
Am J Hum Genet 2002; 70:758-762.
46. Linnebank M, Homberger A, Nowak-Goettl U et al. Linkage disequilibrium of the common mu-
tation 677C→T and 1298A→C of the human methylenetetrahydrofolate reductase gene as proven
by the novel polymorphisms 129C→T, 1068C→T. Eur J Ped 2000; 159:472-473.
47. Viel A, Dall’Agnese L, Simone F et al. Loss of heterozygosity at the 5,10-methylenetetrahydrofolate
reductase locus in human ovarian carcinomas. Br J Cancer 1997; 75:1105-1110.
48. Weisberg I, Tran P, Christensen B et al. A second genetic polymorphism in methylenetrahydrofolate

reductase (MTHFR) associated with decreased enzyme activity. Mol Genet Metabol 1998;
64:169-172.
49. Rady PL, Szucs S, Grady J et al. Genetic polymorphisms of methylenetetrahydrofolate reductase
(MTHFR) and methionine synthase reductase (MTRR) in ethnic populations in Texas: A report
of a novel MTHFR polymorphic site, G1793A. Am J Med Genet 2002; 107:162-168.
50. Robien K, Ulrich CM. 5,10-Methylenetetrahydrofolate reductase polymorphisms and leukemia risk:
A HuGE minireview. Am J Epidemiol 2003; 157:571-582.
51. van der Put NMJ, Eskes TKAB, Blom HJ. Is the common 677C→T mutation in the
methylenetetrahydrofolate reductase gene a risk factor for neural tube defects? A meta-analysis. Q J
Med 1997; 90:111-115.
52. Zetterberg H, Rymo L, Coppola A et al. Reply to “MTHFR C677T and A1298C polymorphisms
and mutated sequences occuring in cis”. Eur J Human Genet 2002; 10:579-582.
53. Trembath D, Sherbondy AL, Vandyke DC et al. Analysis of select folate pathway genes, PAX3 and
human T in a midwest neural tube defect population. Teratorogy 1999; 59:331-341.
54. Antonarakis S, Ashburner M, Auerbach AD et al. For the nomenclature working group. Recom-
mendations for a nomenclature system for human gene mutations. Hum Mutat 1998; 11:1-3.
55. Chen Z, Karaplis AC, Ackerman SL et al. Mice deficient in methylenetetrahydrofolate reductase exhibit
hyperhomocysteinemia and decreased methylation capacity, with neuropathology and aortic lipid deposi-
tion. Human Mol Genet 2001; 10:433-443.
56. Chen Z, Ge B, Hudson TJ et al. Microarray analysis of brain RNA in mice with methylenetetrahydrofolate
reductase deficiency and hyperhomocysteinemia. Gene Expression Patterns 2002; 1:89-93.
57. Ho PI, Ortiz D, Rogers E et al. Multiple aspects of homocysteine neurotoxicity: Glutamate excitotoxicity,
kinase hyperactivation and DNA damage. J Neuroscience Res 2002; 70:694-702.
58. Schwahn BC, Chen Z, Laryea MD et al. Homocysteine-betaine interactions in a murine model of
5,10-methylenetetrahydrofolate reductase deficiency. FASEB J 2003; 17:512-514.
CHAPTER 2
Assays for Methylenetetrahydrofolate

Reductase Polymorphisms
Arve Ulvik and Per Magne Ueland
T
o date, two functional polymorphisms, 677C→T in exon 41 and 1298A→C in exon 7,2
in the gene encoding the enzyme methylenetetrahydrofolate reductase (MTHFR) have
been found and characterized. Both lead to an amino acid change, and different bio-
chemical properties between the normal and variant enzyme have been demonstrated.3 In ad-
dition, a missense mutation (1793G→A in exon 114), three silent polymorphisms (129C→T
in exon 1,5 1068T→C in exon 6,6 and 1317T→C in exon 77), and three intronic polymor-
phisms (IVS2+533G→A,8 IVS6+31T→C9 and IVS10+262C→G8) have been reported, but
their functional implications are unknown.
Allele frequencies of the polymorphisms 677C→T and 1298A→C vary considerably ac-
cording to ethnicity (4-58%, and 9-37%, respectively,10-12 a more detailed description of the
frequency of the 677 variant is provided in Chapter 2). The 1793G→A allele frequency varied
from 15.5 to 32.2% in four distinct ethnic populations,4 and the 1317T→C change was com-
mon (39%) in an African American cohort,7 but essentially absent in German Caucasians.13
Data on frequencies of the remaining polymorphisms are limited.
The1298C-allele is in linkage disequilibrium with the 677C allele. Among the 9 possible
genotype combinations, the 677TT-1298AC, 677CT-1298CC, and 677TT-1298CC are rarely
observed, although a few cases have been reported.7,14
Figure 1 shows a graphical overview of the SNPs and their relative positions along the gene.
Methods for Genotyping MTHFR SNPs

Currently, there are some 30 different reported methods for the detection of polymor-
phisms of the MTHFR gene, but methods for the detection of 677C→T are by far the most
frequently described. These methods can, broadly, be divided into four categories: PCR/re-
striction fragment length polymorphism based assays (PCR/RFLP), allele-specific PCR assays,
heteroduplex assays, and real-time PCR with fluorogenic probes. In addition, the technique of
minisequencing,15 and the use of mass spectrometry16 has been reported.
Restriction Enzyme Based Assays, the 677C→T Variant

Figure 2 depicts some features of the assay reported by Frosst et al in the paper that first
described the 677C→T polymorphism.1 The enzyme HinfI recognizes the sequence GANTC
(N being any base). This implies that the variant sequence (GAGTC) is cleaved, whereas the
wild-type sequence GAGCC remains undigested. The forward primer was placed close to the
cleavage site and the reverse primer at some distance in the 3´ direction, producing a fragment
length of 198bp, and 175 and 23bp after cleavage. Figure 3, panel A, shows a schematic repre-
sentation of the fragments in a sieving gel. Alongside the three possible genotyping outcomes
(CC, CT and TT), profiles obtained after partial digestion of a homozygous TT genotype or
©2005 Eurekah.com.
Figure 1. Polymorphisms of the MTHFR gene. Exons are numbered, and the positions of SNPs are indicated
along with the associated amino acid change (if any) and restriction enzyme(s) used for detection. *Artifi-
cially created restriction site.
Figure 2. Some features of the method by Frosst et al.1 Distances along the horizontal axis are drawn to scale,
and the PCR-primers are drawn as half-arrows.
Figure 3. Schematic drawing of gel electrophoresis, showing characteristics of different RE-based assays.
Panel A) Assay by Frosst et al.1 Panel B) Assay by Ulvik et al.18 Panel C) Assay by Bravo-Osorio et al.19 The
lanes labeled TT* and xx* denote a TT genotype that is partially digested, and any genotype with no
digestion, respectively. The distances between fragments are proportional to the differences in the logarithm
of the fragment sizes.
Assays for MTHFR Polymorphisms 23
Figure 4. Artificially-created restriction sites. The strategy used by Weisberg et al7 for detecting the 1298A→C
variant is shown. The mutagenic primer changes one base in the template (marked by an asterisk), and the
resulting PCR amplicon contains the Fnu4HI recognition sequence when the 1298C-allele is present.
after no digestion (any genotype) are shown. The figure demonstrates that if HinfI digestion
fails, it is possible to make an erroneous genotype assignment. This problem, which is an
inherent feature of this assay design, was addressed in several subsequent publications. Van
Amerongen et al suggested coamplification with another fragment containing a HinfI recogni-
tion sequence,17 whereas Ulvik et al included a preamplified standard amplicon containing the
recognition sequence before addition of the restriction enzyme.18 The latter assay was opti-
mized for capillary electrophoresis with multiple injections, but the strategy is applicable to
traditional gel formats as well (Fig. 3, panel B). Yet another design was presented by Bravo-Osorio
et al. They engineered an additional restriction site into the amplified fragment by using a
reverse primer that included an extra 25 bases containing the HinfI recognition sequence (Fig.
3, panel C).19 Benson et al demonstrated a multiplexing strategy where fluorescent primers of
different colors were incorporated into amplicons of the 677C→T and other polymorphisms
in separate PCR reactions.18 Restriction enzyme digestion was performed as appropriate, and
products were then pooled and separated by color as well as by size on an automated capillary
electrophoresis system. A digestion control for each SNP was added, similar to Ulvik et al.20
Restriction Enzyme Based Assays, the 1298A→C Variant and Other SNPs
An assay for the 1298A→C variant was first described by van der Put et al using the enzyme
MboII that cuts the 1298A allele.2 Weisberg et al, as they discovered the 1317T→C polymor-
phism, noted that the presence of 1317C generates a MboII recognition sequence that (using the
assay by van der Put et al) produces a restriction pattern almost identical to 1298A. To overcome
this problem, they used artificially- generated restriction sites (explained in Fig. 4) to produce an
alternative assay for 1298A→C as well as an assay for 1317T→C using the enzymes Fnu4HI
and TaqI, respectively.7 Yi et al, modified the assay for 1298A→C by van der Put et al so that
fewer fragments were generated after cleavage. Moreover, they coamplified fragments for the
677 and 1298 polymorphisms, but performed separate cleavages using the HinfI and MboII
enzymes, respectively, followed by pooled gel electrophoresis.21 However, they failed to take into
account the interference by 1317T→C. Meisel et al used two allele-specific long range PCRs
targeting the 677C and 677T alleles separately, followed by the analysis of 1298A→C devised
by Weisberg et al to establish allelic association between the two polymorphic sites.13 A new
assay for 1298A→C has recently been developed by Leclerc et al using the enzyme MwoI (de-
scribed in Chapter 1). An assay has been reported for 1793G→A using the enzyme BsrbI which
cuts the G variant.4 The 129C→T variant creates a AvaI site,5 and the 1068T→C variant
creates a HhaI recognition site,6 but has also been analysed using the enzyme CfoI.5 Figure 1 lists
the reported restriction enzymes used for the analysis of each polymorphism.
Allele-Specific PCR Assays, General Description

This type of assay, as well as the PCR/RFLP assay, was first described in the 1980s. Differ-
ent variations are known under such acronyms as ARMS (amplification refractory mutation
system), ASA (allele-specific amplification), and MS-PCR (mutagenically- separated PCR).
The idea behind these assays is that a PCR primer that ends at the polymorphic position will
only bind completely to one of the variants, which will then be amplified at normal efficiency,
whereas the other variant will not be amplified, due to the mismatch generated at the last base.
In order to design effective assays, some knowledge of the refractoriness of different mismatches
to amplification is advisable. A few studies have been reported (summarized in ref. 22). Gen-
eral findings are that G : T mismatches confer low specificity, and the sequence-specific primer
should not end with an A. In some cases (sequence contexts), substantial amounts of PCR
product from the wrong allele may be generated if reaction conditions (annealing temperature
and number of PCR cycles) are not carefully optimized. Strategies such as shortening the primer,
adding a competitive primer,23 and, most often, adding additional mismatches close to the
3énd24 have been used to enhance selectivity.
Assays have been designed where the wild-type and mutant alleles are detected in separate
reactions, or the two alleles are detected in the same tube (MS-PCR). Both strategies can be
multiplexed to include several SNPs.
Allele-Specific PCR Assays for MTHFR Genotyping

Hill et al described a MS-PCR assay detecting both 677 alleles in the same tube.25 The same
strategy was used by Ulvik et al who designed a multiplex assay for the simultaneous detection
of the 677C→T and factor V 1691G→A variants.26 This was later extended to include
1298A→C (unpublished). Endler et al extended the assay to also encompass the prothrombin
20210G→A polymorphism.27 Two other studies report multiplexing of three SNPs (including
677C→T), carried out in two reaction tubes.28,29 The advantage of allele-specific methods
over the restriction enzyme assays is that allele-specific reaction products, directly detectable by
electrophoresis, are generated during PCR. Furthermore, these methods are readily amenable
for multiplexing. Multiplex MS-PCR, however, rapidly reaches a limit of complexity. A three-way
multiplex MS-PCR assay involves the simultaneous amplification, in an allele-specific and
balanced manner, of up to 6 products, requiring careful control of the relative amount of each
primer. The two-tube strategy is more straightforward, but does not involve the competitive
priming of allele-specific primers as in MS-PCR. In some cases, this could increase the risk of
spurious priming and false results. In addition, depending on the assay design, there may be a
need for an extra control amplicon for confirmation of adequate reaction conditions.29 As long
as the reaction conditions/primers are properly optimized, allele-specific PCR is a rapid and
reliable assay.
Heteroduplex Assays
These assays are robust and well suited for multiplexing. However, they usually require the
formation of so-called heteroduplex generators (HDGs). This is a fairly complex process that
involves the generation of an artificial DNA construct, using site-directed mutagenesis of the
sequence encompassing the polymorphism, followed by confirmation and testing of the con-
struct. Figure 5 illustrates the effect of a heteroduplex generator. This construct is usually
coamplified with the DNA of interest and the assay is somewhat sensitive to the ratio of the
amount of HDG-construct to DNA. Enhanced resolution by separation on polyacrylamide
gels combined with long electrophoresis times is usually necessary. At least three reports using
this method are found in the literature: Clark et al30 for the 677C→T mutation, Bowen et al31
for the 677C→T polymorphism multiplexed with two other SNPs, and Barbaux et al32 for
677C→T with 1298A→C as well as two other polymorphisms. Once the heteroduplex gen-
erators are prepared, and their effects verified, these assays, similarly to the allele-specific PCR
assays, require no post-PCR processing (other than electrophoresis).
Figure 5. Principle of a heteroduplex assay. A heteroduplex generator (HDG) is constructed by site-directed

mutagenesis to contain a short deletion 1-3 bases from the variable site. This construct is coamplified with
target DNA, and after the last cycle of PCR the amplicons are melted and allowed to reanneal. Apart from
correctly annealed homoduplexes (target amplicons and HDG amplicons), heteroduplexes are formed, with
a short bubble due to the deletion in the HDG. The size of the bubble depends on the surrounding sequences
including the variant base. Thus, the heteroduplexes containing the variation are seen by their differential
electrophoretic migration.
Real-Time PCR with Fluorogenic Probes—Homogenous Assays

The development of fluorescent labeling technology, and systems for detection of fluores-
cence during amplification, have facilitated the development of these techniques, which are
fairly recent additions to the methodological repertoire (see refs. 33,34 for reviews). Some of
the reported methods require sophisticated, yet increasingly available, equipment for real-time
PCR detection, but for some methods, a post-PCR reading in an ordinary fluorimeter is ad-
equate. An important reason for the rapid gain in popularity of these methods is that they are
homogenous, meaning that all ingredients for genotyping are added to one tube, and results
are obtained without further manipulations. This also removes the most important source of
contamination in PCR: the reintroduction of PCR products to the PCR setup. At least seven
reports, using these techniques to detect MTHFR polymorphisms, have been published, divis-
ible into two main categories as described below.
5Éxonuclease and Molecular Beacon Assays

The molecular beacon and 5´-exonuclease (also known as TaqMan) assays have a number of
similarities. Both use probes that are doubly labeled oligonucleotides with a reporter fluorophore
at the 5énd and a quencher at the 3énd. The beacon probes contain a short additional se-
quence at both ends, with internal homology, so that a hairpin loop is created. The probe will
then be in an equilibrium between intramolecular hybridization, and hybridization to a target.
In the former state, fluorescence is quenched whereas opening the hairpin structure allows the
reporter to fluoresce.35 Signal generation from the 5éxonuclease probe stems from cleavage of
the probe by the 5éxonuclease activity of the DNA polymerase during PCR, thereby releasing
the fluorophore from the quencher.36
Both types of assays depend on the different binding strengths of the probes to the normal
and variant sequences on target DNA. Giesendorf et al used molecular beacons in separate
reactions for the determination of the 677C→T alleles,37 whereas Happich et al used the
TaqMan format and two differently labeled probes in the same tube for the simultaneous
determination of both 677C→T alleles.38 Ulvik et al demonstrated that the homogenous for-
mat (TaqMan) is compatible with direct analysis on blood without DNA purification. In-
cluded were the 677C→T and 1298A→C polymorphisms.39
The competition between target and internal hybridization of the molecular beacons is said
to enhance the selectivity of this format. A study comparing the 5éxonuclease probes and
molecular beacons, however, found only a marginal difference in the ability to discriminate
between variants40 A recent development of the TaqMan probes is the addition of a minor
groove binder which allows shorter and thereby more selective probes.41 Ulvik et al however,
showed that short (16-20bp) probes can function without the aid of minor groove binders.39
Hybridization Probe Assays

In hybridization probe assays, singly labeled hybridization probes are used. The labeled
probe hybridizes in close proximity to a second fluorophore, either attached to one of the PCR
primers, or an additional hybridized oligonucleotide. Signal is detected as fluorescence reso-
nance energy transfer (FRET) from the probe fluorophore to the second (acceptor) fluorophore.
Genotyping is done by performing a melting curve analysis after PCR. A completely matched
probe melts at a higher temperature than one with a mismatch against the target. The data are
analyzed and plotted as the derivative of signal with respect to temperature (-dF/dT) against
temperature. The probe may be homologous to the normal, or variant allele, and one probe is
sufficient for genotyping.42
Two variants of the assay format outlined above for the analysis of 677C→T have been
published.42,43 Von Ahsen et al demonstrated a hybridization probe assay multiplexed by using
different fluorophores for the two SNPs 677C→T and factor V 1691G→A.44 Crockett et al,
on the other hand, showed that quenching of the probe fluorophore mediated by proximal
guanosines in the target sequence was sufficient as a hybridization-dependent signal for the
generation of melting temperature curves. In their report, the 677C→T was among the in-
cluded SNPs.45
Comparison of the Different Homogenous Formats

The singly labeled probes used in the hybridization probe assays are easier and less costly to
produce than the doubly labeled probes associated with the TaqMan and molecular beacon
(and some other) formats. On the other hand, the melting temperature analysis carried out
with hybridization probes requires specialized equipment and software, whereas the assays us-
ing doubly labeled probes are all compatible with a one-time reading of fluorescence after PCR
(although all three quoted reports make use of real-time PCR equipment). Using the principles
outlined in the paper by Crockett et al,45 hybridization probe assays may be designed, using
one singly labeled probe per SNP, as opposed to two doubly labeled probes per SNP for the
TaqMan/molecular beacon format. Also, the potential for multiplexing several SNPs in the
same tube are greater when melting curves and differently labeled fluorophores are combined,44,46
(reviewed in ref. 33). Finally, with the hybridization probe assay, additional base substitutions
within the boundary of the probe may be more easily discovered, and not compromise the
interpretation of results.33 A minor drawback with the hybridization probe assay has been that
the equipment involves capillary tubes, which do not conform to the industry standard 96-well
format. Recently, other melting curve-based assays have been published,47,48 and a wider vari-
ety of equipment has been introduced.
Assessment of Methods
Frequently, when a new polymorphism is detected, a PCR/RFLP assay is designed. These
assays are easy to perform and do not require expensive or specialized equipment. If a suit-
able restriction enzyme is not available, the format is still applicable by the generation of
artificial restriction sites. A digestion control should be included, and, among the different
variants described above, perhaps the most simple and elegant solution was presented by
Bravo-Osorio et al.19
Allele-specific PCR is an alternative characterized by less handling time, as the step involv-
ing the restriction enzyme digestion is obviated.
Generally, some of the reported PCR/RFLP and allele-specific PCR assays may benefit
from moving one of the primers closer to the variant position, thereby making the relative
difference of fragment sizes larger. This would enhance resolution and shorten migration time
in electrophoresis.
A minor but finite problem associated with many of the described assays is additional sub-
stitutions near the position of the SNP of interest. This may affect the recognition sequence of
the restriction enzyme, the binding of allele-specific primers, and the binding of fluorogenic
probes.49 The heteroduplex method and hybridization probe assay are probably the most resil-
ient to this influence. If anomalies in the migration pattern of heteroduplexes, or melting peaks
of hybridisation probes are detected, reanalysis by sequencing should be performed. An addi-
tional strength of the heteroduplex assay is its multiplexing capability, although multiplexing
can also be obtained with allele-specific PCR and with the real-time PCR strategy described by
von Ahsen et al and others.33,44 The homogenous assay formats are characterized by speed, ease
of operation, and contamination control.
Preparation of DNA Material

An important part of genotyping is the preparation of DNA for subsequent processing
(usually involving PCR). If traditional purification methods, such as phenol-chloroform ex-
traction, are used, this part of the overall workload may require more time and effort than the
actual genotyping. Recently, a number of DNA purification techniques have been developed,
which involve no hazardous chemicals and with the potential for automation. It is often over-
looked that PCR-based genotyping may not require highly purified DNA. Also, the amount of
DNA needed for successful genotyping is small. A few nanograms are sufficient in most cases.
This is less than the amount contained in 1 µl of whole blood. There have been a number of
reports where blood and other biological fluids have been used directly for PCR, either with-
out, or with minimal treatment.50,51 This includes complicated multiplexed assays,26,29 as well
as one example referred to above using a homogenous assay format.39 Also, in studies using
archival material such as paraffin-embedded tissue slices, simple boiling protocols have been
described.52,53 When using blood as template, it is necessary to use a DNA polymerase that is
tolerant to inhibiting substances. If such information for a particular enzyme is lacking, it can
easily be obtained by appropriate tests.
In laboratories where automated DNA purification has been established, there may be lo-
gistic reasons for sample purification. A convenient storage format for future analyses, e.g.,
microtiter plates with dissolved DNA and identifier tags, may then be established. In our
laboratory, we have had good experience with aliquoting purified DNA into PCR tube strips
or PCR plates and letting it air-dry. Plates can then be stored at ambient temperature or 4˚C
for months, shipped to another laboratory, or processed immediately by adding a PCR master
mix. This works with purified DNA (2-20 ng) as well as with unpurified blood (≤ 1µL).
Throughput Considerations
When more than a few hundred SNPs per week need to be analyzed, throughput, or time
spent per sample, becomes important. The way this is addressed partly depends on whether
600 SNPs refers to one SNP in 600 samples, 3 SNPs in 200 samples, or 20 SNPs in 30 samples.
In the first case, the workload falls heavily on the DNA preparation step, and genotyping
without template purification could be considered as a means of increasing throughput and
decreasing cost. The second case seems to be ideal for some of the multiplexed methods de-
scribed above. Multiplexing also ensures that the correct ensemble of SNPs is assigned for any
given sample.
None of the assays described thus far seems ideal for the last case. Most of the PCR/RFLP,
allele-specific PCR, and heteroduplex methods quoted above involve a fair amount of manual
handling (e.g., preparation and loading of gels, photographing etc.) All these assays, however,
are compatible with automated capillary electrophoresis, or equivalent, for fragment analysis.
The homogenous assays require minimal sample handling, mainly the setup of reagents, which
can be carried out by a robotic workstation, and genotype annotation is usually automated.
New Technologies and Future Developments

Thus far, fairly established methods of genotyping have been described. The great interest
in determination of single nucleotide polymorphisms has motivated the development of new
assay formats offering unprecedented levels of automation and throughput. Among recently
developed technologies are the Invader assay,54 fluorescence polarization detection,55
Pyrosequencing, which is a form of chemical sequencing without subsequent gel-separation,56
DNA microarrays, PNA based probing, and mass spectrometry detection.16 Many of the new
methodologies are aimed at large-scale, genome-wide mapping of SNPs, which are beyond the
scope of this chapter. However, the reader should be aware of the rapidly expanding possibili-
ties, including assays on-demand that are currently offered by several companies.
References
mutation in methylenetetrahydrofolate reductase. Nat Genet 1995; 10(1):111-113.
2. van der Put NM, Gabreels F, Stevens EM et al. A second common mutation in the
methylenetetrahydrofolate reductase gene: An additional risk factor for neural-tube defects? Am J
Hum Genet 1998; 62(5):1044-1051.
3. Yamada K, Chen Z, Rozen R et al. Effects of common polymorphisms on the properties of recombi-
nant human methylenetetrahydrofolate reductase. Proc Natl Acad Sci USA 2001; 98(26):14853-14858.
(MTHFR) and methionine synthase reductase (MTRR) in ethnic populations in Texas; a report of
a novel MTHFR polymorphic site, G1793A. Am J Med Genet 2002; 107(2):162-168.
5. Linnebank M, Homberger A, Nowak-Gottl U et al. Linkage disequilibrium of the common muta-
tions 677C > T and 1298A > C of the human methylenetetrahydrofolate reductase gene as proven
by the novel polymorphisms 129C > T, 1068C > T. Eur J Pediatr 2000; 159(6):472-473.
reductase gene and genotype/phenotype correlations in severe methylenetetrahydrofolate reductase
deficiency. Am J Hum Genet 1995; 56(5):1052-1059.
7. Weisberg I, Tran P, Christensen B et al. A second genetic polymorphism in methylenetetrahydrofolate
reductase (MTHFR) associated with decreased enzyme activity. Mol Genet Metab 1998; 64(3):169-172.
8. Rosenberg N, Murata M, Ikeda Y et al. The frequent 5,10-methylenetetrahydrofolate reductase
C677T polymorphism is associated with a common haplotype in whites, Japanese, and Africans.
Am J Hum Genet 2002; 70(3):758-762.
methylenetetrahydrofolate reductase (MTHFR) gene, and description of five novel mutations in
MTHFR. Am J Hum Genet 1996; 59(6):1268-1275.
10. Scholtz CL, Odendaal HJ, Thiart R et al. Analysis of two mutations in the MTHFR gene associ-
ated with mild hyperhomocysteinaemia—heterogeneous distribution in the South African popula-
tion. S Afr Med J 2002; 92(6):464-467.
11. Mutchinick OM, Lopez MA, Luna L et al. High prevalence of the thermolabile
methylenetetrahydrofolate reductase variant in Mexico: A country with a very high prevalence of
neural tube defects. Mol Genet Metab 1999; 68(4):461-467.
12. Pollak RD, Friedlander Y, Pollak A et al. Ethnic differences in the frequency of the C677T muta-
tion in the methylenetetrahydrofolate reductase (MTHFR) gene in healthy Israeli populations. Genet
Test 2000; 4(3):309-311.
13. Meisel C, Cascorbi I, Gerloff T et al. Identification of six methylenetetrahydrofolate reductase
(MTHFR) genotypes resulting from common polymorphisms: Impact on plasma homocysteine lev-
els and development of coronary artery disease. Atherosclerosis 2001; 154(3):651-658.
14. Hanson NQ, Aras O, Yang F et al. C677T and A1298C polymorphisms of the
methylenetetrahydrofolate reductase gene: Incidence and effect of combined genotypes on plasma
fasting and post-methionine load homocysteine in vascular disease. Clin Chem 2001; 47(4):661-666.
15. Zetterberg H, Regland B, Palmer M et al. Increased frequency of combined methylenetetrahydrofolate
reductase C677T and A1298C mutated alleles in spontaneously aborted embryos. Eur J Hum Genet
2002; 10(2):113-118.
16. Ross P, Hall L, Smirnov I et al. High level multiplex genotyping by MALDI-TOF mass spectrom-
etry. Nat Biotechnol 1998; 16(13):1347-1351.
17. Van Amerongen G, Mathonnet F, Boucly C et al. An improved method for the detection of the
thermolabile variant of methylenetetrahydrofolate reductase. Clin Chem 1998; 44(5):1045-1047.
18. Ulvik A, Refsum H, Kluijtmans LA et al. C677T mutation of methylenetetrahydrofolate reductase
gene determined in blood or plasma by multiple-injection capillary electrophoresis and laser-induced
fluorescence detection. Clin Chem 1997; 43(2):267-272.
19. Bravo-Osorio M, Bydlowski SP. Detection of methylenetetrahydrofolate reductase (MTHFR) C677T
and prothrombin G20210A mutations: Second restriction site for digestion control of PCR prod-
ucts. Clin Chim Acta 2000; 301(1-2):219-223.
20. Benson JM, Ellingsen D, Renshaw MA et al. Multiplex analysis of mutations in four genes using
fluorescence scanning technology. Thromb Res 1999; 96(1):57-64.
21. Yi P, Pogribny I, Jill James S. Multiplex PCR for simultaneous detection of 677 C-->T and 1298
A-->C polymorphisms in methylenetetrahydrofolate reductase gene for population studies of cancer
risk. Cancer Lett 2002; 181(2):209.
22. Ayyadevara S, Thaden JJ, Shmookler Reis RJ. Discrimination of primer 3'-nucleotide mismatch by
taq DNA polymerase during polymerase chain reaction. Anal Biochem 2000; 284(1):11-18.
23. Zhu KY, Clark JM. Addition of a competitive primer can dramatically improve the specificity of
PCR amplification of specific alleles. Biotechniques 1996; 21(4):586, 590.
24. Rust S, Funke H, Assmann G. Mutagenically separated PCR (MS-PCR): A highly specific one step
procedure for easy mutation detection. Nucleic Acids Res 1993; 21(16):3623-3629.
25. Hill AE, FitzPatrick DR. MS-PCR assay to detect 677C-->T mutation in the
5,10-methylenetetrahydrofolate reductase gene. J Inherit Metab Dis 1998; 21(6):694-695.
26. Ulvik A, Ren J, Refsum H et al. Simultaneous determination of methylenetetrahydrofolate reduc-
tase C677T and factor V G1691A genotypes by mutagenically separated PCR and multiple-injection
capillary electrophoresis. Clin Chem 1998; 44(2):264-269.
27. Endler G, Kyrle PA, Eichinger S et al. Multiplexed mutagenically separated PCR: Simultaneous
single-tube detection of the factor V R506Q (G1691A), the prothrombin G20210A, and the
methylenetetrahydrofolate reductase A223V (C677T) variants. Clin Chem 2001; 47(2):333-335.
28. Hessner MJ, Luhm RA, Pearson SL et al. Prevalence of prothrombin G20210A, factor V G1691A
(Leiden), and methylenetetrahydrofolate reductase (MTHFR) C677T in seven different popula-
tions determined by multiplex allele-specific PCR. Thromb Haemost 1999; 81(5):733-738.
29. Hezard N, Cornillet-Lefebvre P, Gillot L et al. Multiplex ASA PCR for a simultaneous determina-
tion of factor V Leiden gene, G-->A 20210 prothrombin gene and C-->T 677 MTHFR gene
mutations. Thromb Haemost 1998; 79(5):1054-1055.
30. Clark ZE, Bowen DJ, Whatley SD et al. Genotyping method for methylenetetrahydrofolate re-
ductase (C677T thermolabile variant) using heteroduplex technology. Clin Chem 1998;
44(11):2360-2362.
31. Bowen DJ, Bowley S, John M et al. Factor V Leiden (G1691A), the prothrombin 3'-untranslated
region variant (G20210A) and thermolabile methylenetetrahydrofolate reductase (C677T): A single
genetic test genotypes all three loci—determination of frequencies in the S. Wales population of
the UK. Thromb Haemost 1998; 79(5):949-954.
32. Barbaux S, Kluijtmans LA, Whitehead AS. Accurate and rapid “multiplex heteroduplexing” method
for genotyping key enzymes involved in folate/homocysteine metabolism. Clin Chem 2000;
46(7):907-912.
33. Wittwer CT, Herrmann MG, Gundry CN et al. Real-time multiplex PCR assays. Methods 2001;
25(4):430-442.
34. Foy CA Parkes HC. Emerging homogeneous dna-based technologies in the clinical laboratory.
Clin Chem 2001; 47(6):990-1000.
35. Tyagi S, Kramer FR. Molecular beacons: Probes that fluoresce upon hybridization. Nat Biotechnol
1996; 14(3):303-308.
36. Livak KJ, Flood SJ, Marmaro J et al. Oligonucleotides with fluorescent dyes at opposite ends
provide a quenched probe system useful for detecting PCR product and nucleic acid hybridization.
PCR Methods Appl 1995; 4(6):357-362.
37. Giesendorf BA, Vet JA, Tyagi S et al. Molecular beacons: A new approach for semiautomated
mutation analysis. Clin Chem 1998; 44(3):482-486.
38. Happich D, Madlener K, Schwaab R et al. Application of the TaqMan-PCR for genotyping of the
prothrombin G20210A mutation and of the thermolabile methylenetetrahydrofolate reductase mu-
tation. Thromb Haemost 2000; 84(1):144-145.
39. Ulvik A, Ueland PM. Single nucleotide polymorphism (SNP) genotyping in unprocessed whole
blood and serum by real-time PCR: Application to SNPs affecting homocysteine and folate me-
tabolism. Clin Chem 2001; 47(11):2050-2053.
40. Tapp I, Malmberg L, Rennel E et al. Homogeneous scoring of single-nucleotide polymorphisms:
Comparison of the 5'-nuclease TaqMan assay and Molecular Beacon probes. Biotechniques 2000;
28(4):732-738.
41. Kutyavin IV, Afonina IA, Mills A et al. 3'-minor groove binder-DNA probes increase sequence
specificity at PCR extension temperatures. Nucleic Acids Res 2000; 28(2):655-661.
42. Bernard PS, Lay MJ, Wittwer CT. Integrated amplification and detection of the C677T point
mutation in the methylenetetrahydrofolate reductase gene by fluorescence resonance energy transfer
and probe melting curves. Anal Biochem 1998; 255(1):101-107.
43. Aslanidis C, Nauck M, Schmitz G. High-speed prothrombin G-->A 20210 and
methylenetetrahydrofolate reductase C-->T 677 mutation detection using real-time fluorescence
PCR and melting curves. Biotechniques 1999; 27(2):234-236, 238.
44. von Ahsen N, Oellerich M, Schutz E. A method for homogeneous color-compensated genotyping
of factor V (G1691A) and methylenetetrahydrofolate reductase (C677T) mutations using real-time
multiplex fluorescence PCR. Clin Biochem 2000; 33(7):535-539.
45. Crockett AO, Wittwer CT. Fluorescein-labeled oligonucleotides for real-time pcr: Using the inher-
ent quenching of deoxyguanosine nucleotides. Anal Biochem 2001; 290(1):89-97.
46. Herrmann MG, Dobrowolski SF, Wittwer CT. Rapid beta-globin genotyping by multiplexing probe
melting temperature and color. Clin Chem 2000; 46(3):425-428.
47. Akey JM, Sosnoski D, Parra E et al. Melting curve analysis of SNPs (McSNP): A gel-free and
inexpensive approach for SNP genotyping. Biotechniques 2001; 30(2):358-362, 364, 366-357.
48. Prince JA, Feuk L, Howell WM et al. Robust and accurate single nucleotide polymorphism
genotyping by dynamic allele-specific hybridization (DASH): Design criteria and assay validation.
Genome Res 2001; 11(1):152-162.
49. Lyondagger E, Millsondagger A, Phan T et al. Detection and identification of base alterations
within the region of factor V leiden by fluorescent melting curves. Mol Diagn 1998; 3(4):203-209.
50. Burckhardt J. Amplification of DNA from whole blood. PCR Methods Appl 1994; 3(4):239-243.
51. Nishimura N, Nakayama T, Tonoike H et al. Direct polymerase chain reaction from whole blood
without DNA isolation. Ann Clin Biochem 2000; 37(Pt 5):674-680.
52. Chehab FF, Xiao X, Kan YW et al. Detection of cytomegalovirus infection in paraffin-embedded
tissue specimens with the polymerase chain reaction. Mod Pathol 1989; 2(2):75-78.
53. Frank TS, Svoboda-Newman SM, Hsi ED. Comparison of methods for extracting DNA from
formalin-fixed paraffin sections for nonisotopic PCR. Diagn Mol Pathol 1996; 5(3):220-224.
54. Hsu TM, Law SM, Duan S et al. Genotyping single-nucleotide polymorphisms by the invader
assay with dual-color fluorescence polarization detection. Clin Chem 2001; 47(8):1373-1377.
55. Kwok PY. SNP genotyping with fluorescence polarization detection. Hum Mutat 2002;
19(4):315-323.
56. Ahmadian A, Gharizadeh B, Gustafsson AC et al. Single-nucleotide polymorphism analysis by
pyrosequencing. Anal Biochem 2000; 280(1):103-110.
CHAPTER 3
Biochemical Characterization of Human

Methylenetetrahydrofolate Reductase
and Its Common Variants
Kazuhiro Yamada and Rowena G. Matthews
Abstract
M
ethylenetetrahydrofolate reductase (MTHFR) catalyzes the NADPH-linked
reduction of methylenetetrahydrofolate to methyltetrahydrofolate. The human
enzyme is an ~70 kDa polypeptide with two regions, an N-terminal catalytic region
with significant homology to the smaller bacterial enzymes, and a C-terminal regulatory region
that binds the allosteric inhibitor adenosylmethionine. Two common polymorphisms in hu-
man MTHFR, 677C→T and 1298A→C, have been identified. The biochemical effects of the
677C→T mutation, which occurs in a conserved residue of the catalytic region, have been
characterized both in human MTHFR and in the bacterial homologue, while the 1298A→C
mutant, which occurs in the regulatory region, has been characterized in human MTHFR. The
677C→T mutation, which converts alanine 222 to valine, leads to enhanced dissociation of
FAD and decreased stability of MTHFR. Loss of flavin can be minimized in the presence of
methyltetrahydrofolate or adenosylmethionine. The 1298A→C mutation has no observable
phenotype in the purified human enzyme.
During the initial elucidation of the pathway by which homocysteine (1, Fig. 1) is con-
verted to methionine, Donaldson and Keresztesy described an enzyme involved in the oxida-
tion of prefolicA (later shown to be methyltetrahydrofolate [CH3-H4folate, (2, Fig. 1)] to
tetrahydrofolate and formaldehyde in the presence of menadione.1,2 This enzyme was purified
about 20-fold from pig liver and its activity was shown to be stimulated on addition of FAD.
They established that their enzyme preparation could also catalyze the reduction of
methylenetetrahydrofolate [CH2-H4folate, 3, Fig. 1] to methyltetrahydrofolate and the reduc-
tion of menadione by NAD(P)H. Each of these activities could be explained if reduction of
CH2-H4folate by reduced enzyme were reversible, so that the enzyme can catalyze the reduc-
tion of CH2-H4folate using NAD(P)H as the source of reducing equivalents (equations 1 and
2), or the oxidation of CH3-H4folate using menadione as the oxidant (the reverse of equation
2 and equation 3).
NAD(P)H + H+ + E-FAD → NAD(P)+ + E-FADH2 (1)
E-FADH2 + CH2-H4folate → E-FAD + CH3-H4folate (2)
E-FADH2 + menadione → E-FAD + menadiol (3)
©2005 Eurekah.com.
Figure 1. Structures of L-homocysteine (1), (6-S)CH3-H4folate (2), (6-R)CH2-H4folate (3, where the R
group is the p-aminobenzoylglutamate substituent), and AdoMet (4).
In 1971, Kutzbach and Stokstad reported the properties of a partially purified preparation
of methylenetetrahydrofolate reductase (MTHFR) from porcine liver.3 The enzymatic actvity
was enriched about 400-fold during the course of purification. They determined that NADPH
was a better substrate than NADH for the porcine enzyme. Removal of the FAD cofactor
resulted in complete activity loss. They also made the critical observation that both the
NADPH-menadione and the NADPH-CH2-H4folate oxidoreductase activities were inhibited
by S-adenosylmethionine (AdoMet, (4, Fig. 1), and that S-adenosylhomocysteine (AdoHcy)
could block inhibition by AdoMet. Upon addition of AdoMet, the onset of inhibition was
gradual, requiring several minutes to reach maximal levels of inhibition, and restoration of
activity upon addition of AdoHcy was also slow. Because the enzyme could be desensitized
towards AdoMet, they inferred that the inhibition by AdoMet was allosteric. That is to say,
AdoMet is chemically distinct from any of the substrates. It binds at a site distinct from those
required for substrate binding and communicates the inhibition by subtle influences on the
structure of the protein. AdoMet is derived from methionine, and is the major donor of methyl
groups in biological methylation reactions. Its regulation of methylenetetrahydrofolate reduc-
tase activity controls the flux of one-carbon units into the pathway committed to the conver-
sion of homocysteine to methionine.
Kutzbach and Stokstad recognized that their observations could explain “the effect of me-
thionine on the disturbances of folate metabolism caused by vitamin B12 deficiency.” In B12
deficiency, cobalamin-dependent methionine synthase is less active. They noted that “because
the equilibrium of the methylenetetrahydrofolate reductase reaction favors…methyltetrahydrofolate
synthesis, this folate compound should accumulate and be unavailable for functions in nucleic
acid and protein metabolism. This ‘methyl trap’4,5 could possibly be prevented through the
inhibition of the reductase reaction by AdoMet formed from dietary methionine.”
MTHFR was first purified to homogeneity from porcine liver by Daubner and Matthews.6
The enzyme was isolated with its FAD tightly but noncovalently bound, and reduction of the
FAD cofactor by NADPH and its reoxidation by CH2-H4folate could be directly demon-
strated. The porcine enzyme was shown to be a dimer of identical 77 kDa subunits, each
containing bound FAD.6,7 However, scanning transmission electron microscopy revealed that
Biochemical Characterization of Human Methylenetetrahydrofolate Reductase 33
each subunit contained two spatially distinct domains.7 Digestion of the native enzyme with
trypsin initially separated the 77 kDa subunit into 40 and 37 kDa fragments, thought to
represent the domains visualized by microscopy. The N-terminal 40 kDa fragment was stable
to further proteolysis, while the C-terminal 37 kDa fragment was further degraded. Since
allosteric inhibition by AdoMet was lost during proteolysis, while catalytic activity was re-
tained, these studies led to the proposal that the N-terminal fragment contains the determi-
nants for binding FAD and for catalysis, while the C-terminal fragment binds AdoMet and
mediates allosteric inhibition. The availability of peptide sequence for about 40% of the por-
cine enzyme, and the assignment of some of this peptide sequence to the N- and C-terminal
regions of the protein, allowed Rozen and Matthews to collaborate on the cloning of a human
cDNA specifying MTHFR.8,9 MTHFR isozymes are expressed in a tissue-specific fashion and
an ~77 kDa isozyme is expressed in most tissues, while a smaller ~70 kDa isozyme is expressed
primarily in human liver9 (and see Chapter 1 in this volume). The open reading frame for the
short form specifies a peptide of 74,546 Da (Fig. 2). Discrepancies between predicted molecu-
lar weights and inferred molecular weights based on electrophoretic analysis are not unusual,
although in our hands the short form of recombinant human MTHFR migrates with an esti-
mated molecular weight that is slightly larger than the predicted molecular weight. As shown
in Figure 2, the N-terminal sequence of the short form of the human enzyme is longer than
those of prokaryotic enzymes or those from yeast and plants. The function of this N-terminal
region remains unknown.
The N-terminal region of the deduced amino acid sequence shows extensive homology
with smaller prokaryotic proteins that catalyze reduction of CH2-H4folate, leading to the as-
signment of this region as the catalytic domain. The C-terminal region contains a peptide that
is labeled by ultraviolet irradiation in the presence of an analogue of AdoMet10 and presumably
corresponds to the regulatory domain.
A number of patients with severe MTHFR deficiency have been described. These patients
exhibit hyperhomocysteinemia, homocystinuria and hypomethioninemia. They may present
with developmental delay, with motor dysfunction, with vascular complications, and/or with
neurological abnormalities.11 The availability of the human cDNA sequence allowed the map-
ping of mutations that were associated with severe disease in patients, and led to the identifica-
tion of a common polymorphism, the 677C→T mutation in the cDNA.9 The mutation leads
to the substitution of alanine 222 in the deduced amino acid sequence by valine. As indicated
in Figure 2, Ala222 is located in the region of the protein responsible for catalysis and it is
largely conserved in MTHFR enzymes, which suggests that there may be evolutionary con-
straints favoring its conservation. One of the few sequences in which this residue is not Ala is
Met13 from Saccharomyces cereviseae, in which the homologous residue is a glycine.12
David Kang and his colleagues had described a mild form of MTHFR deficiency, leading to
~50% residual activity, that was present in ~17% of patients with coronary artery disease and at
a lower incidence (5%) in controls.13 This mild deficiency was associated with a thermolabile
form of the enzyme. Thermolability is defined as a difference in the residual activity measured
in crude lymphocyte cell extracts after heating to 46˚C, as compared to the residual activity
measured after the same length of incubation at 37˚C.9 Frosst et al demonstrated that the
677C→T polymorphism conferred thermolability on MTHFR, and that the thermolability of
enzyme from CT heterozygotes was intermediate between that in homozygous CC and TT
genotypes.9 They also reported a method to distinguish variant and wild-type genotypes by
restriction enzyme analysis. Their analysis indicated 12% TT genotype and 51% CT genotype
in a small sample of North Americans of European descent. Thus it was assumed that the
thermolabile polymorphism reported by Kang was in fact the 677C→T polymorphism.
A critical insight into the effect of the polymorphism came from a more extensive study of
the relation between folate status, the polymorphism, and plasma homocysteine concentra-
tions.14 An investigation of 365 individuals from the NHLBI Family Heart Study showed that
individuals with the TT genotype and plasma folate concentrations below the sample median
Figure 2. Alignment of methylenetetrahydrofolate reductase sequences. The sequences shown are for MTHFR
from Homo sapiens (U09806),8,9 with substitution of the mutant 1298C base by an A,18 MTHFR1 from
Arabadopsis thaliana (Af181966),27 Met13 from Saccharomyces cerevisiae (Z72647),12,28 and MetF from
Escherichia coli (P00394).29 Conserved residues are shown in bold and the positions of the two polymor-
phisms discussed in this paper are indicated. The underlined peptide was identified by irradiation of porcine
MTHFR with [methyl-3H]AdoMet, followed by digestion with LysC and sequencing.30 The position of the
lysine upstream of the peptide was inferred. No radioactivity was released in the first 21 cycles, so the
derivatized amino acid presumably is downstream of the underlined portion of the peptide.
(15.4 nmol/L) had 24% higher fasting homocysteine levels than individuals with the CC geno-
type. In individuals with plasma folate concentrations above the sample median, no difference
in the homocysteine levels was seen between individuals with the CC and TT genotype. This
important study revealed an interaction between the MTHFR thermolabile genotype and folate
status, such that only TT individuals with low folate status were likely to have elevated levels of
plasma homocysteine.
At this point, our laboratory became interested in studying the biochemical phenotype
conferred by the 677C→T variant by comparing mutant and wild-type enzymes in vitro.
Unfortunately, the level of expression of the human MTHFR in E. coli was low, and insuffi-
cient to allow purification and characterization of either wild-type or mutant enzyme. For
this reason, we initially chose to construct the 677C→T mutation in MTHFR from Escheri-
chia coli, the product of the metF gene. As shown in Figure 2, the deduced amino acid
sequence of metF is homologous to the N-terminal half of human MTHFR, and in particu-
lar Ala222 in the human sequence aligns with Ala177 in the bacterial sequence. Wild-type
bacterial MTHFR could readily be expressed with a C-terminal histidine tag and purified to
homogeneity by affinity chromatography on nickel-agarose.15 The purified enzyme differs
from its mammalian homologue in being a tetramer of identical 33 kDa subunits, each of
which contains noncovalently bound FAD, rather than a dimer of 77 kDa subunits. There is
no evidence of allosteric regulation of the bacterial enzyme activity, consistent with the ab-
sence of a regulatory domain. The bacterial enzyme is specific for NADH rather than NADPH.
However, aside from these differences, the properties of the bacterial enzyme are very similar
to those of its mammalian counterpart.
A C-terminally histidine-tagged Ala177Val variant of E. coli MTHFR was successfully ex-
pressed and purified to homogeneity and its properties were compared with those of the wild-type
enzyme.16 The catalytic properties of the Ala177 and Val177 enzymes were indistinguishable,
but the Val177 variant was thermolabile, both as defined by assay and by measurements of
denaturation by differential scanning calorimetry. Significantly, both mutant and wild-type
enzymes showed lower melting temperatures when dilute than when concentrated, suggesting
a change in the oligomeric state of the enzyme. Gel filtration established that the concentrated
enzyme was indeed a tetramer, as expected, but on storage in dilute solution, the enzyme
dissociated into dimers. Dilution of the enzyme was also associated with release of the
enzyme-bound FAD and loss of activity, and the Val177 variant lost its flavin more than 10-times
faster than the wild-type enzyme. Analyses of the rate of flavin dissociation on dilution vs. the
concentration of the enzyme after dilution established that flavin dissociation was associated
with conversion of the active tetramer to inactive dimers, and suggested that a rapid and revers-
ible equilibrium between tetramers and dimers was followed by rate-limiting flavin release as
indicated in Figure 3. Although both mutant and wild-type enzymes underwent the same
Figure 3. Schematic diagram of the mechanism for flavin release from diluted E. coli MTHFR. The
holoenzyme tetramer undergoes a rapid and reversible dissociation to form holoenzyme dimers prior to
rate-limiting release of the flavin cofactor. To view available color versions of figures, please go to http://
www.eurekah.com/eurekahlogin.php?chapid=1752&bookid=120&catid=80.
Figure 4. Structure of the E. coli MTHFR monomer. The monomer is an α8β8 barrel with the β-strands
shown as arrows and the connecting α-helices shown as coils. The FAD cofactor is bound at the top of the
barrel and is shown in “ball and stick” mode. The dotted surface at the bottom of the barrel represents the
position of Ala177, which is located in a loop connecting helix α5 with strand β6. Reproduced, with
permission, from Guenther et al.16
reactions on dilution, the mutant enzyme dissociated more quickly. Folates, such as the prod-
uct methyltetrahydrofolate, slowed the rate of dissociation and loss of flavin. Both FAD and
CH3-H4folate protected the mutant and wild-type enzyme in crude cell extracts from thermal
denaturation on incubation at 46˚C for 5 minutes.
These observations suggested a biochemical explanation for the observed inverse correla-
tion between elevated plasma homocysteine in humans with the TT genotype and low folate
status. A high folate status would result in increased intracellular concentration of folate deriva-
tives in the cell, and stabilization of MTHFR against dissociation and loss of flavin and activity.
While both mutant and wild-type enzymes would be protected by folates, the effect would be
more dramatic in the mutant enzyme, with its increased propensity for dissociation and loss of
flavin. The reduced activity seen in lymphocyte extracts from TT individuals might reflect
substantial amounts of inactive monomeric enzyme.
The availability of large amounts of the wild-type E. coli enzyme allowed us to collaborate
with Brian Guenther and Martha Ludwig to determine the structure of the bacterial enzyme by
x-ray crystallography.16 The structure is shown schematically in Figure 4. Each subunit is roughly
barrel-shaped, with eight parallel beta strands forming the staves of the barrel and connected by
eight alpha helices that form the outside of the barrel. The FAD is bound on the inside of the
barrel near its opening at the top. Ala177 is located in a tight loop between the fifth alpha helix
and the sixth beta strand, at the bottom of the barrel. This position is quite far from the site of
flavin binding, consistent with the lack of effect of the mutation on catalytic activity. However,
modeling revealed that the sidechain of Val could not be accomodated in the loop without
displacement of helix alpha 5. The sidechains of amino acids in this helix make a number of
contacts with the flavin cofactor, and displacement of the helix might be expected to result in
weakened flavin binding. Figure 5 shows the arrangement of the subunits of the tetramer.
Figure 5. Positioning of the monomers in the E. coli MTHFR tetramer. The four subunits of the tetramer
are arranged in a planar rosette. The FAD cofactor is again indicated in “ball and stick” mode. The adenine
nucleotide of the cofactor directly interacts with residues in helix α5. Helix α5 also makes the principle
contacts between the left dimer (C,B) and the right dimer (A′,A). The Ala177Val mutation may weaken the
interaction between the FAD and the monomer by shifting the position of helix α5, and may also affect the
strength of the interaction between the left and right dimer.
Helix alpha 5 makes the principle cross-dimer contacts between subunits A' and C and be-
tween subunits B and A. It can be seen that this helix makes a critical contact between the right
and left dimers in the tetramer. We postulate that the displacement of helix alpha 5 also results
in an increased propensity for the tetramer to dissociate into dimers. Thus the structure of the
wild-type E. coli enzyme provides a satisfying rationale for the differences between wild-type
and mutant bacterial enzymes.
The skeptic may argue that the E. coli MTHFR may not be a valid model for the human
enzyme because it is a tetramer of identical 33 kDa subunits, each of which consists of a single
catalytic domain, rather than a dimer of 77 kDa subunits, each of which consists of a catalytic
and a regulatory domain. Thus, the contacts between helix alpha 5 in one subunit and the
adjacent subunit may differ in the two structures. Accordingly, we were interested in expressing
the human enzyme at levels that would permit biochemical characterization of both the wild-type
enzyme and the Ala222 variant. Expression of the N-terminally histidine-tagged enzymes was
accomplished using a baculovirus expression system and Sf9 insect cells in liquid culture. Ex-
pression was achieved at a level of about 3.5 mg of MTHFR per 1.5-liter culture.17 Homoge-
neous MTHFR could be isolated following two chromatographic steps with an overall yield of
45%.
The purified human MTHFR exhibited properties that were very similar to the previously
characterized porcine enzyme. It was isolated with one equivalent of noncovalently bound
Figure 6. Schematic diagram of the mechanism for flavin release from diluted human MTHFR. The
holoenzyme dimer undergoes a rapid and reversible dissociation to form holoenzyme monomers prior to
rate-limiting release of the flavin cofactor.
FAD per subunit, and the specific activity of the homogenous enzyme, 12.4 µmoles per mg for
the CH3-H4folate-menadione oxidoreductase assay, was similar to the value of 19.4 µmoles
per mg obtained for the porcine enzyme. Comparison of the Ala222 MTHFR with the Val222
MTHFR revealed that the catalytic activities of the two variants were indistinguishable, as were
the effects of AdoMet and AdoHcy on enzymatic activity. However, the Val222 variant was
thermolabile, as assessed by residual activity of the purified enzyme after heating, and on dilu-
tion the FAD cofactor was lost about 3-fold more rapidly than from the Ala222 MTHFR.
Again, plots of the rate of flavin release vs. the enzyme concentration after dilution indicated
that both wild-type and mutant enzymes dissociate on dilution, in this case going from dimers
to monomers, and again, a rapid and reversible dissociation is thought to precede rate-limiting
release of FAD (Fig. 6). Addition of CH3-H4folate to the enzyme slows the rate of flavin
dissociation and activity loss on dilution, and this protection is seen with both mutant and
wild-type enzymes. Similar protective effects of added FAD and CH3-H4folate are seen in
measurements of residual activity after heat treatment of purified mutant and wild-type en-
zymes. Thus our experiments with the human enzyme appear to validate the model developed
on the basis of experiments with MTHFR from E. coli.
In the same paper,17 the properties of another polymorphism in human MTHFR, the
1298A→C mutation, were also characterized. The 1298A→C mutation results in the replace-
ment of Glu429 in the regulatory domain by alanine. This mutation is also associated with
somewhat decreased enzyme activity in assays of human lymphocytes,18,19 but the biochemical
phenotype of the purified variant is indistinguishable from that of the wild-type enzyme. The
regulatory domain may not simply serve as a regulator of activity, but may also stabilize the
enzyme, as evidenced by the fact that plant MTHFRs retain the regulatory domain, but show
no change in activity on addition of AdoMet.20
One observation about the effect of the 677C→T mutation that could not have been in-
ferred from studies with the E. coli enzyme is that AdoMet protects both wild-type and variant
enzymes against loss of flavin on dilution, and the protective effect is blocked by AdoHcy. Thus
we believe that flavin release is hindered when the enzyme is in the inactive AdoMet-bound
state, and facilitated when the enzyme is in the active state induced by AdoHcy. These two
states do not appear to differ in their oligomeric status; the concentrated enzyme is dimeric in
both active and inactive forms. These observations may seem hard to reconcile with the obser-
vation that the product, CH3-H4folate, also reduces the rate of flavin dissociation on loss of
activity. As yet unpublished studies on the structure of E. coli MTHFR by Robert Pejchal and
Martha Ludwig may provide insights into this dilemma. Methyltetrahydrofolate binds adja-
cent to the flavin in the barrel and might be expected to physically block dissociation of the
flavin from the active state. Thus the protection against flavin dissociation from enzyme in the
active state exerted by CH3-H4folate may have an entirely different physical explanation than
the protection afforded by conversion of the enzyme to the inactive state.
These studies of the human MTHFR provide a biochemical rationale for clinical observa-
tions of the effect of the 677C→T polymorphism. As discussed, hyperhomocysteinemia in TT
individuals is only seen when the folate status is also low, and presumably reflects the increased
propensity of the flavin to dissociate from MTHFR in the absence of folates that is observed in
vitro. Our observations would suggest that TT individuals with low riboflavin status might
also be at risk for hyperhomocysteinemia.21 Indeed, in the rat, MTHFR activity is particularly
sensitive to riboflavin deficiency.22,23 Three studies have now been published linking riboflavin
deficiency to elevated plasma homocysteine in individuals with the TT genotype.24-26
In addition to the influences of riboflavin and folate status on homocysteine levels in hu-
mans with the TT genotype, studies with the human enzyme indicate that the AdoMet status
is also very important. Here we have the conundrum that AdoMet both inhibits MTHFR
activity and stabilizes MTHFR against dissociation and loss of flavin and activity. The ratio of
AdoMet to AdoHcy is high when the diet has an adequate methionine content, suggesting that
under these circumstances MTHFR will be retained in an inactive but stable form. When the
diet is deficient in methionine, and the ratio of AdoMet to AdoHcy is lowered, MTHFR will
be both activated and more labile. What is not yet clear is whether the changes in activity occur
over the same range of ratios as the changes in stability. We should also be cognizant that
mutations that lead to elevated levels of AdoHcy and reduced AdoMet/AdoHcy ratios will
affect MTHFR stability as well as activity.
References
1. Donaldson KO, Keresztesy JC. Naturally occurring forms of folic acid. I. “Prefolic A”: Preparation
of concentrate and enzymatic conversion to citrovorum factor. J Biol Chem 1959; 234:3235-3240.
2. Donaldson KO, Keresztesy JC. Naturally occurring forms of folic acid. II. Enzymatic conversion of
methylenetetrahydrofolic acid to prefolic A-methyltetrahydrofolate. J Biol Chem 1962;
237:1298-1304.
3. Kutzbach C, Stokstad ELR. Mammalian methylenetetrahydrofolate reductase: Partial purification,
properties, and inhibition by S-adenosylmethionine. Biochim Biophys Acta 1971; 250:459-477.
4. Noronha JM, Silverman M. On folic acid, vitamin B12, methionine and formiminoglutamic acid
metabolism. In: Heinrich HC, ed. Second European Symposium on Vitamin B12 and Intrinsic
Factor. Stuttgart Enke 1962:728-736.
5. Herbert V, Zalusky R. Interrelations of vitamin B12 and folic acid metabolism: Folic acid clearance
studies. J Clin Invest 1962; 41:1263-1276.
6. Daubner SC, Matthews RG. Purification and properties of methylenetetrahydrofolate reductase
from pig liver. J Biol Chem 1982; 257:140-145.
7. Matthews RG, Vanoni MA, Hainfeld JA et al. Methylenetetrahydrofolate reductase: Evidence for
spatially distinct subunit domains obtained by scanning transmission electron microscopy and lim-
ited proteolysis. J Biol Chem 1984; 259:11647-11650.
cDNA, mapping and mutation identification. Nat Genet 1994; 7:195-200.
10. Sumner J, Jencks DA, Khani S et al. Photoaffinity labeling of methylenetetrahydrofolate reductase
with 8-azidoadenosylmethionine. J Biol Chem 1986; 261:7697-7700.
AL, Sly WS et al, eds. The Metabolic Bases of Inherited Disease. Vol II. 7th ed. New York:
McGraw-Hill; 1995:3111-3128.
12. Raymond RK, Kastanos EK, Appling DR. Saccharaomyces cerevisiae expresses two genes encoding
isozymes of methylenetetrahydrofolate reductase. Arch Biochem Biophys 1999; 372:300-308.
13. Kang S-S, Wong PWK, Susmano A et al. Thermolabile methylenetetrahydrofolate reductase: An
inherited risk factor for coronary artery disease. Am J Hum Genet 1991; 48:536-545.
14. Jacques PF, Bostom AG, Williams RR et al. Relation between folate status, a common mutation in
methylenetetrahydrofolate reductase, and plasma homoycsteine concentrations. Circulation 1996;
93:7-9.
15. Sheppard CA, Trimmer EE, Matthews RG. Purification and properties of NADH-dependent
5,10-methylenetetrahydrofolate reuctase (MetF) from Escherichia coli. J Bacteriol 1999; 181:718-725.
reductase from Escherichia coli suggest how folate ameliorates human hyperhomocysteinemia. Na-
ture Struct Biol 1999; 6:359-365.
binant human methylenetetrahydrofolate reductase. Proc Natl Acad Sci USA 2001; 98:14853-14858.
reductase (MTHFR) asociated with decreased enzyme activity. Mol Genet Metab 1998; 64:169-172.
19. van der Put NM, Gabreels F, Stevens EM, et al. A second common mutation in the
Hum Genet 1998; 62:1044-1051.
20. Roje S, Chan SY, Kaplan F et al. Metabolic engineering in yeast demonstrates that S-adenosylmethionine
controls flux through the methylenetetrahydrofolate reductase reaction in vivo. J Biol Chem 2002;
277:4056-4061.
21. Rozen R. Methylenetetrahydrofolate reductase: A link between folate and riboflavin? Am J Clin
Nutr 2002; 76:301-302.
22. Narisawa K, Tamura T, Tanno K et al. Tetrahydrofolate-dependent enzyme activities of the rat
liver in riboflavin deficiency. Tohoku J exp Med 1968; 94:417-430.
23. Bates CJ, Fuller NJ. The effect of riboflavin deficiency on methylenetetrahydrofolate reductase
(NADPH) (EC 1.5.1.20) and folate metabolism in the rat. Brit J Nutr 1986; 55:455-464.
24. Hustad S, Ueland PM, Vollset SE et al. Riboflavin as a determinant of plasma total homocystine:
Effect modification by the methylenetetrahydrofolate reductase C677T polymorphism. Clin Chem
2000; 46:1065-1071.
25. McNulty H, McKinley MC, Wilson B et al. Impaired functioning of thermolabile methylenetetrahydrofolate
reductase is dependent on riboflavin status: Implications for riboflavin requirements. Am J Clin
Nutr 2002; 76:436-441.
26. Jacques PF, Kalmbach R, Bagley PJ et al. The relationship between riboflavin and plasma total
homocysteine in the Framingham Offspring cohort is influenced by folate status and the C677T
transition in the methylenetetrahydrofolate reductase gene. J Nutr 2002; 132:283-288.
27. Rohe S, Wang H, McNeil SD et al. Isolation, characterization, and functional expression of cDNAs
encoding NADH-dependent methylenetetrahydrofolate reductase from higher plants. J Biol Chem
1999; 274:36089-36096.
28. Cerdan E, Rodriguez-Torres AM, Rodriguez-Belmonte E et al. GenBank accession number Z72647
for ORFYGL125w from Saccharomyces cerivisiae, unpublished data. 1997.
29. Saint-Girons I, Duchange N, Zakin MM et al. The nucleotide sequence of metF, the E. coli
structural gene for 5-10 methylene tetrahydrofolate reductase, and of its control region. Nucleic
Acids Res 1983; 11:6723-6732.
30. Sumner J. Structural and mechanistic investigations of methylenetetrahydrofolate reductase and
their functional implications. Ann Arbor: Biological Chemistry. University of Michigan, 1992.
CHAPTER 4
Severe Methylenetetrahydrofolate Reductase

Deficiency
Mary Ann Thomas and David S. Rosenblatt
Abstract
S
evere methylenetetrahydrofolate reductase (MTHFR) deficiency is an inborn error of
folate metabolism that is associated with elevated levels of homocysteine and decreased
levels of methionine and S-adenosylmethionine. The clinical spectrum of severe MTHFR
deficiency ranges from the neonatal onset of significant neurological problems to milder adult
onset cases. There have also been several asymptomatic adult cases reported. The majority of
patients present in the first few years of life with developmental delay and other neurological
problems, such as seizures. Although treatment is difficult, the addition of betaine has im-
proved neurological development in some patients and halted the deterioration in others. This
chapter is a summary of the clinical presentation, pathophysiology, laboratory investigations,
prenatal diagnosis, treatment and current knowledge of genotype-phenotype correlations in
severe MTHFR deficiency.
Introduction
The enzyme 5,10-methylenetetrahydrofolate reductase (methylene-H4Folate reductase,
MTHFR) plays a key regulatory role in folate metabolism. MTHFR reduces
5,10-methylenetetrahydrofolate to 5-methyltetrahydrofolate. 5-methyltetrahydrofolate is in-
volved in the remethylation of homocysteine to methionine, which is then converted to
S-adenosylmethionine, the predominant methyl donor in man. In cases of severe MTHFR
deficiency, the decreased levels of methionine and S-adenosylmethionine adversely affect my-
elination and are thought to be an important cause of the neurological problems. There is also
increased homocysteine, which accounts for thrombosis being a feature in some of the patients.
Alterations in the gene for MTHFR have been associated with two broad categories of
medical conditions. In the first category, common polymorphisms in MTHFR, such as 677C→T
and 1298A→C, are associated with an increased predisposition to several medical problems.
Although less prevalent in Africans and Asians, homozygosity for the polymorphism 677C→T
is present in 5-18% of many European and North American populations.1 The residual en-
zyme activity in homozygotes for this polymorphism is 35-50% that of controls,2 which is
sufficient to increase plasma homocysteine if folate intake is inadequate. Homozygosity for
677C→T has been postulated to increase the risk of developing cardiovascular disease and of
women having children with neural tube defects.3 This is explained in more detail in other
chapters of this book. This chapter will concentrate on the second type of medical condition
that is caused by mutations that decrease the specific activity of MTHFR to less than 20% of
controls.4 Individuals with these mutations often have a severe clinical phenotype, including
developmental delay, motor and gait abnormalities, seizures and psychiatric features.
©2005 Eurekah.com.
Clinical Presentation
A review of at least 85 published cases with severe MTHFR deficiency1,5-58 indicates that the
clinical phenotype varies greatly from individual to individual. This disease can be broadly clas-
sified into neonatal onset, late infancy/early childhood onset and late childhood/adult forms.40
The more severe cases typically present at an earlier age with neurological deterioration.
In the neonatal form, the pregnancy and delivery are usually uneventful. Patients present
with decreased muscle tone, drowsiness, poor feeding, apnea, seizures and even coma. This is
occasionally preceded by an infection.44 Brain imaging typically shows brain atrophy and white
matter disease (demyelination). Electroencephalograms (EEG) generally demonstrate abnor-
mal background activity or seizure activity if the patient has clinical seizures. There is no spe-
cific seizure type in these patients. In cases tested, visual evoked responses are abnormal and
reflect demyelination. Prior to the inclusion of betaine in the treatment regimen, affected indi-
viduals often died within the first year of life, mostly from respiratory failure secondary to
central (CNS) causes or aspiration pneumonia. Some cases treated with betaine caught up with
growth and psychomotor development, although long term neurological outcome of treat-
ment is not known.45,52
Presentation in infancy/early childhood ranges from the age of 3 months to 10 years. Typi-
cally, the developmental delay is not as striking in the first few months of life. Patients present
to medical attention when developmental milestones, such as learning to sit or walk, are not
met. Patients may show developmental regression after an infection.44 Some affected children
present later with seizures and mental retardation of unknown cause. A number of children in
this group have microcephaly,23,28,30,33,46 although this is not a consistent finding.10 Neuro-
logical symptoms differ among cases and can even be conflicting at times. These include hypo-
tonia,30,46,48 hypertonia,10,33,38 spastic paraparesis,10 weakness,47 upper motor neuron signs
(brisk deep tendon reflexes and an upgoing Babinsky reflex), pyramidal tract involvement,37
extrapyramidal movements,33 ataxic gait,10,33 lack of coordinated eye movements30 and pe-
ripheral neuropathy.38 The majority of patients have seizures and all have developmental delay,
usually severe. One ten year-old boy exhibited a history of developmental delay and physical
signs of Angelman syndrome.57 Brain imaging typically shows abnormalities. Before MRI was
available, CT scans were performed and often demonstrated dilated ventricles10,33,48 and corti-
cal atrophy.23,48 When MRI became available, white matter abnormalities, such as demyelina-
tion, were more evident.46 As in earlier onset cases, EEG abnormalities reflect the patient’s clinical
seizure. Studies of visual evoked responses or auditory evoked responses demonstrate abnormali-
ties consistent with abnormal myelination. Treatment regimens that include betaine often im-
prove levels of homocysteine and clinical symptoms. Many cases that did not receive betaine as
part of their treatment died several years after diagnosis, often from respiratory failure.
The later childhood or adulthood form of the disease can present with some of the same
features as the early childhood cases. These include similar neurological features, mental retar-
dation and seizures. Others have prominent peripheral neuropathy,38,47 ataxia, arterial throm-
bosis10,54 and/or psychiatric problems.6,58 There are adults who are asymptomatic and are diag-
nosed because of a more severely affected sibling. In one family, a younger brother developed limb
weakness, incoordination, paresthesias, and memory lapses at age 15 years and was
wheelchair-bound by his early 20s, whereas his older brother was asymptomatic at age 37 years.38
Pathophysiology
Several patients had an autopsy that confirmed the brain abnormalities seen on imaging,
notably a small brain, cerebral atrophy, enlarged ventricles and demyelination. Macrophage infil-
tration and gliosis have also been noted.33,36,37 It has been suggested that decreased methionine
and S-adenosylmethionine cause demyelination.37 Monkeys exposed to nitrous oxide, which blocks
the activity of methionine synthase, were found to have decreased methionine formation and
demethylation. These monkeys became ataxic and, at autopsy, the spinal cord and peripheral
nerves demonstrated changes of subacute combined degeneration. Methionine administration to
Severe Methylenetetrahydrofolate Reductase Deficiency 43
monkeys exposed to nitrous oxide decreased the severity of the associated demethylation. This
group of monkeys did not develop ataxia during the time they were followed and had little or no
changes of subacute combined degeneration.59
There is at least one known child who had severe worsening of his disease with exposure to
nitrous oxide. Prior to the diagnosis of severe MTHFR deficiency, this 3-month-old boy re-
ceived nitrous oxide, administered during the surgical removal of a fibrosarcoma. Seventeen
days after the surgery, he was readmitted to hospital with seizures, apneic episodes, severe
hypotonia and absent reflexes. He died at around 4.5 months of age from a respiratory arrest.
The autopsy demonstrated asymmetric cerebral atrophy and severe demyelination with
astrogliosis and oligodendroglial cell depletion in the mid-brain, medulla and cerebellum.60
Several other patients with MTHFR deficiency presented with subacute combined degen-
eration of the cord, similar to that observed in patients with untreated cobalamin deficiency.21,33
Individuals given betaine have improved CSF levels of S-adenosylmethionine, although treat-
ment with betaine does not always increase CSF methionine levels. This suggests that de-
creased S-adenosylmethionine levels, more than decreased methionine levels, are responsible
for the demyelination.
Several individuals had thromboses of arteries and cerebral veins, which appear to have
been the cause of their death.10 However, cerebral thromboses do not appear to be the cause of
the neurological symptoms in the majority of patients. It has been suggested that the combina-
tion of mutations in MTHFR and factor V Leiden can contribute to the vascular pathology in
some patients.61
There are other proposed explanations for the neurological symptoms in these patients.
One is impaired purine and pyrimidine synthesis in the brain. This has been proposed because
in some cases, there have been neurological symptoms despite normal CSF methionine levels.
Several authors have suggested that the only natural folate that can cross the blood brain barrier
is methyltetrahydrofolate, the product of the MTHFR reaction.16,62 Deficiency of MTHFR
may result in functionally low folate levels in the brain. With the current level of understanding
of this disease, it is unclear what the relative contributions of low folate, low methionine and
low neurotransmitter levels are in the CNS pathology of severe MTHFR deficiency.44
Prenatal Diagnosis
There are at least 5 published reports of prenatal testing for severe MTHFR deficiency using
enzymatic studies.26,32,48,51,63 One case demonstrated low enzymatic activity in amniocytes and
the pregnancy was continued for religious reasons. The first urine contained homocystine and
cord blood showed low MTHFR activity, confirming an affected newborn.26 Another case had
testing in both chorionic villi (11 weeks) and amniocytes (18 weeks). Both samples showed
decreased enzymatic activity in the heterozygous range. This was confirmed 9 months post-natally.
In our laboratory, using the specific activity of MTHFR in confluent amniocytes, we have ex-
cluded the diagnosis of severe MTHFR deficiency in 9 cases and diagnosed one affected fetus.
However, a recent publication demonstrated the potential difficulty in interpreting biochemical
results since enzymatic activity was in the heterozygous range in the prenatal studies, but the
child’s enzymatic activity was very low after birth.51 Another family exemplified the complex
relationship between residual enzyme activity and clinical findings. The mother had severely low
enzyme function, in the range of her severely affected child, but had no clinical symptoms.48
Given the complicated association between residual enzyme activity and clinical severity, mo-
lecular testing is the ultimate prenatal test. This would consist of testing for known causal muta-
tions or performing linkage analysis, rather than assessing enzymatic activity.
Since MTHFR polymorphisms, such as 677C→T and 1298A→C, are relatively common,
prenatal diagnosis may be possible in some families using linkage analysis, even when the mu-
tations are not known. Because identification of mutations can be time-consuming, linkage
analysis may actually be the preferred molecular test. A sample from the affected proband is
necessary to track the mutant allele in the family, and parents need to be heterozygous for at
least one variant in the MTHFR gene. However, since a number of other SNPs (single nucle-
otide polymorphisms), in addition to 677C→T and 1298A→C, are known in the MTHFR
gene (Table 5, Chapter 1), linkage analysis should be possible in most families. Prenatal diag-
nosis for severe MTHFR deficiency has been performed by linkage analysis for three families in
our institution (R. Rozen, personal communication); in the one family that requested amnio-
centesis, instead of chorionic villus sampling, enzymatic analysis in our laboratory confirmed
the prenatal result obtained by DNA testing (unpublished data).
Laboratory Findings
The major biochemical findings include moderate homocystinuria and hyperhomocysteinemia
with low or low-normal levels of plasma methionine. Whereas patients reported in the early
papers had free plasma homocystine measured, current publications usually include total plasma
homocysteine values. In Erbe’s clinical review in 1986,64 homocystinuria was present in all
patients, with a reported range of 15 to 667 µmol/24 h and a mean value of 130 µmol/24 h.
Homocystine, not normally detected in urine or free in plasma, was found in the plasma: mean
value 57 µM (range, 12 to 233 µM). Recent data on total plasma or serum homocysteine
(tHcy) reveal levels (before treatment) of 42-220 µM (controls 4-20 µM).38,46,48,50,51,56,65,66
Plasma methionine levels were low in all patients, ranging from 0 to 18 µM, with a mean of 12
µM; normal is 23-35 µM,64 although values vary among laboratories.
Although homocystinuria was consistently seen in all patients, and indeed is the clinical
sign by which the diagnosis of MTHFR deficiency is made, the excretion of homocystine in
urine is much less than that found in homocystinuria due to cystathionine synthase deficiency.
Indeed, it may not be detected on spot testing, which should not be used in isolation to diag-
nose severe MTHFR deficiency.67 Methionine levels in MTHFR deficiency are usually low.
This again distinguishes these patients from those with cystathionine synthase deficiency, who
generally have high levels of methionine. Although serum folate levels were not always low,
many of the patients with MTHFR deficiency had serum folate levels that were low on at least
one determination. In contrast, serum cobalamin levels were almost always normal. Although
the levels of neurotransmitters in the cerebrospinal fluid have been measured in only a minor-
ity of patients, they have usually been low.47,64
Another group of inborn errors of metabolism that can have homocystinuria are the cobal-
amin (vitamin B12) abnormalities. These patients are functionally deficient in methionine bio-
synthesis because of abnormalities in methylcobalamin formation (complementation groups
cblC, cblD, cblE (methionine synthase reductase deficiency), cblF, and cblG (methionine syn-
thase deficiency)), and differ from patients with MTHFR deficiency by having megaloblastic
anemia. In addition, in contrast to patients with the cblC, cblD, and cblF disorders, patients
with MTHFR deficiency have no methylmalonic aciduria. Tests to assess for megaloblastic
anemia and methylmalonic aciduria should be performed to distinguish cobalamin abnormali-
ties from MTHFR deficiency.
Studies on Cultured Cells

A deficiency of MTHFR has been confirmed in studies of liver, leukocytes, cultured fibro-
blasts and lymphoblasts. The MTHFR reaction is irreversible in vivo, but the enzyme activity
can be measured in the reverse (nonphysiological) direction in vitro. Traditionally, this is the
enzyme assay used to measure MTHFR activity and uses radioactive methyltetrahydrofolate as
a substrate and menadione as the electron acceptor. This is the conventional assay for practical
reasons, including lack of commercial availability of radiolabelled 5,10-methylene-
tetrahydrofolate. Enzyme activity is extremely sensitive to the stage of the culture cycle of
fibroblasts, with the specific activity in control fibroblast cells being highest in confluent cul-
tures.68 This variability is sufficiently great to allow for the misclassification of controls and
heterozygotes if the stage of the culture cycle is not taken into account. In general, there is a
rough correlation between residual enzyme activity and the clinical severity (Table 1).
Table 1. Residual MTHFR activity and clinical presentation
MTHFR Activity
(% Control) Mutation Clinical Presentation Patient Reference
Diagnosis between 0-3 months

0 1762A→T/1762A→T Pakistani male presented in the first month of life with neurological symptoms and 1794 83
failure to thrive. Responded to betaine and folate.
0 1553delAG/1420G→T Caucasian male presented at 2 weeks of life with vomiting. Over the next few weeks, 1569 84
developed stridor, hypotonia and head lag.
0 1084C→T/1084C→T Turkish male, presented at 1 month of age with psychomotor delay and severe K 67
hypotonia. He was not treated. At 10 months, there was severe psychomotor delay,
severe hypotonia and no social interaction. Betaine was started. He died at 7 years
from hyperpyrexia and had severe mental retardation.
0 559C→T/559C→T 1 month-old Native American (Hopi) male with hypotonia. Developed seizures and 1554 80
Severe Methylenetetrahydrofolate Reductase Deficiency
corneal clouding.
0 1755G→A/ Presumed Caucasian male diagnosed at 3 months with an infantile fibrosarcoma. He was 1084 60
heterozygote administered nitrous oxide during surgery. Returned to hospital with marked
hypotonia and apneas, and died at 4 months.
0.2 1027T→G/1027T→G Turkish male referred at 5 weeks with hypotonia, developmental delay, apnea 2231 84
and poor suck.
1 559C→T/559C→T 1 month-old Native American (Choctaw) presented with apnea, failure to thrive, 1627 80
unresponsiveness, seizures and anemia.
1.6 164G→C/249-1G→T 2 week-old Caucasian male presented with failure to thrive and irregular breathing 1772 82, 83
2 980T→C/1141C→T African American/Caucasian female with lethargy and failure to thrive at 1 month 1767 82
of age. At 3 months, had seizures, apnea and hypotonia
N/A 1027T→G/1027T→G Turkish male, treated with betaine, starting at 6 days of life (poor compliance), UB 67
because of a positive family history. At 4 years, had severe mental retardation
and cerebral demyelination.
N/A 1010T→C/1010T→C 4 week-old with severe muscular hypotonia, died at 4 months. Severe cerebral 2 43
demyelination.
45
46
Table 1. Continued
MTHFR Activity
Diagnosis between 3 months-10 years

0 983A→G/ 983A→G Greek female, presented at 2 years with psychomotor retardation, microcephaly, CM 67
hypotonia, restlessness and inability to sit unsupported. At 17 years, had severe
mental retardation.
2 692C→T/692C→T African Indian female presented at 7 months with microcephaly, progressive 735 81
deterioration of mental development, apnea and coma.
3 764C→T/764C→T Japanese female with delayed walking and speech at 2 years, seizures at 6 years and 1807 81
gait disturbance with peripheral neuropathy at 16.
4 458G→T/458G→T Japanese female with developmental delay and seizures who died at 9 months of age. 670 82
5.3 1727C→T/1025T→C Caucasian male presented in the first year of life with developmental delay and 1951 84
seizures. At age 4 years, had gait problems and hyperactivity.
7.8 28A→T/1615C→T Female who presented at 11.5 years with moderate delay (sister of II3). At 2 years, II1 1
she had speech delay, attention deficit and hyperactivity and, at 3 years, was
overweight (+4SD). She had no seizures.
8 28A→T/1615C→T Male diagnosed at 3 years (brother of II1) when his sister was diagnosed. At the time, II3 1
he had a short concentration span and speech delay.
N/A 1420G→T/1274G→C Presented at 5 years with psychomotor retardation, epilepsy and hyperkinetic 1 43
movements. Improved on betaine and folate.
N/A 1711C→T/1711C→T Turkish female, presented at 10 months with psychomotor delay, severe U 67
microcephaly. Received betaine, and, at 4 years, she had severe mental retardation.
Table 1. Continued
MTHFR Activity
Diagnosis after 10 years

5.5 1025T→C/1141C→T Caucasian male referred at 13 years with developmental delay, noted during the first 2351 84
year of life, with a history of seizures, excessive growth, and immature behavior.
7 482G→A/1711C→T 37 year-old asymptomatic French Canadian male (brother of 1779). 1834 80, 83
8 482G→A/1711C→T French Canadian male (brother of 1834) presented at age 15 with weakness, 1779 80, 83
incoordination, paresthesiae and memory lapses. He was wheelchair bound by
his twenties.
8.2 1172G→A/1768G→A 14 year-old Caucasian female presenting with 4-year history of dementia and 3-year 2006 83
history of dysthymia.
Severe Methylenetetrahydrofolate Reductase Deficiency
10 167G→A/1015C→T Caucasian male diagnosed at 12 years with ataxia and marginal school performance. 458 81
13 1274G→A/471C→G Saudi Arabian female had school difficulties at age 12, a stroke at 15 years and 2255 84
spastic paraplegia and seizures at 16.
14 792+1G→A/? African American female (sister of 354) presented at 15 years with anorexia, tremor, 355 81
hallucinations and progressive withdrawal.
14 167G→A/1081C→T Caucasian female presented at 14 years with ataxia, foot drop, and inability to walk. 1396 81
During childhood, she was clumsy and had global developmental delay. She
developed deep vein thrombosis and bilateral pulmonary emboli.
14.2 482G→A/1727C→T 21 year-old Caucasian male, presenting with gait abnormalities of 2-3 years, 1863 80, 83
and found to have spastic paraparesis.
19 792+1G→A/? African American female (sister of 355) diagnosed at 13 years with mild 354 81
mental retardation.
20 985C→T/985C→T Italian male presented at 16 years with muscle weakness, abnormal gait, 356 81
and flinging movements of the upper extremities.
29.1 358G→A/1134C→G Caucasian presented at 16 years with slow neurological deterioration, 2184 83
including changes in mental ability and difficulty walking,
47
There are several reported problems with the reverse direction assay. The assay uses organic
solvent extraction with incomplete recovery and less than optimal specificity. The blank values
can be variable and even high because of impurities in the substrate and dependence on protein
concentration. Recently, a sensitive assay in the physiological direction has been reported.69
MTHFR activity is measured by assessing the conversion of 5,10-methylenetetrahydrofolate,
with NADPH, to 5-methyltetrahydrofolate. This is accomplished with HPLC and fluores-
cence detection. The mean activity with the physiological assay was 2.5-3 fold higher than the
reverse assay and can be used to detect residual activities as low as 2.6%.
Other measures to assess the function of MTHFR include: (1) The synthesis of methionine
from homocysteine using labeled formate.22 Methionine synthesis, in the presence of normal
methionine synthase activity, is a function of MTHFR activity. The goal is to measure the
appearance of label in methionine; (2) Assessing the proportion of folate present in cultured
cells as methyltetrahydrofolate, which correlates with clinical severity. Studies in cultured fi-
broblasts8,15 and liver21,30 have determined the levels and distribution of folate derivatives. In
both control and mutant fibroblasts, most of the folates present were polyglutamates, and the
proportion of polyglutamates relative to folate monoglutamates was similar. In cultured fibro-
blasts, a decrease in the proportion of cellular methyltetrahydrofolate (as a fraction of total
folate) is correlated with worse clinical symptoms and decreased residual activity. This indicates
that the distribution of the different folates may be an important control of intracellular folate
metabolism;4,15 (3) Cultured fibroblasts from patients with severe MTHFR deficiency do not
grow in tissue culture medium lacking methionine, an essential amino acid for these cells. This
is in contrast to control fibroblasts which can grow when homocysteine, along with folate and
cobalamin, is substituted in the culture medium for methionine;9,70 and (4) A differential
microbiologic assay, which makes use of the fact that Lactobacillus casei can utilize
methyltetrahydrofolate for growth but Pediococcus acidilactici (previously known as Pediococcus
cerevisiae) cannot. This is a useful screening test for methylenetetrahydrofolate reductase defi-
ciency since analysis only requires small numbers of cultured fibroblasts.8
Treatment
Interestingly, therapy with methionine alone or with methyltetrahydrofolate has not been
particularly effective in most cases, even though S-adenosylmethionine deficiency in the cen-
tral nervous system appears to play a major role in the pathogenesis of this disease.44 Individu-
als with MTHFR deficiency have been treated with a variety of agents including folates, me-
thionine, pyridoxine, cobalamin, carnitine, betaine, and riboflavin, either alone or in
combination. The rationale for therapy has included: (1) folates, such as folic acid or folinic
acid, in an attempt to maximize any residual enzyme activity; (2) methyltetrahydrofolate to
replace the missing product; (3) methionine to correct the cellular methionine deficiency; (4)
pyridoxine to lower homocysteine levels, because of its role as a cofactor for cystathionine
synthase (enhancing the transsulfuration pathway);40 (5) cobalamin, because of its role as a
cofactor for methionine synthase and at least one case who developed subacute combined
degeneration of the cord when treated with methyltetrahydrofolate alone;33,71 (6) carnitine,
since deficiency can occur because its synthesis requires S-adenosylmethionine; (7) betaine,20
because it is a substrate for betaine:homocysteine methyltransferase,72 a liver-specific enzyme
which converts homocysteine to methionine; and (8) riboflavin, because of the flavin require-
ment of MTHFR.
Treatment is considered successful if there is reduction of the plasma homocysteine levels,
elevation of plasma methionine levels to normal and improvement in the clinical picture.64
In most cases, several of the agents mentioned above have been used in combination, and it
is somewhat difficult to assess the efficacy of a single one. Prior to the addition of betaine to
the treatment regimen, most cases were very resistant to treatment.39,40,64,71,73 There are
exceptions to this, including a 7 1/2 month-old who showed rapid improvement on me-
thionine, pyridoxine, folinic acid, and cobalamin.23 Another patient responded to high doses
of folic acid (400 mg/day) with the disappearance of homocystine in the urine and increased
methionine in the plasma.74 The majority of cases, however, did not improve.10,30,45,47,48,75
One patient’s clinical deterioration was attributed to pyridoxine.28 When betaine is added to
the treatment, there is often decreased homocystine levels, elevated methionine levels and a
variable degree of clinical improvement.44-47,51,52,54-56
Thus, betaine25,34,45,50 appears to be the most promising agent for therapy of MTHFR
deficiency, although some of the other therapies have been partially successful. There is not a
great deal of data on the optimum dose of betaine in these patients because of limited experi-
ence. Ronge and Kjellman suggested a dose of 6 g/day (2 x 3 g). Ten g/day of betaine was tested
and was not found to further improve the clinical or biochemical features of this disease.50
Ogier de Baulny and colleagues suggested a dose of 2-3 g/day in young infants and 6-9 g/day in
children and adults.40 Sakura and colleagues studied the relationship of serum total homocys-
teine and betaine levels during treatment of a patient with oral betaine in doses of between 20
and 120 mg/kg/day.66 They found that serum levels of total homocysteine decreased propor-
tionately until betaine levels reached 400 µM. They suggested that this was the therapeutic
threshold for serum betaine.
Many authors41,50,64,75 have stressed the importance of early diagnosis and therapy because
of the poor prognosis in this disorder once there is evidence of neurologic involvement. Even
with early diagnosis, it is not clear that any of the therapeutic regimens are universally success-
ful, and it is possible that genetic heterogeneity in the disease itself is responsible for some of
the variability in clinical response to therapy.
Genetics
Autosomal recessive inheritance of MTHFR deficiency has been assumed based on clinical
information. Consanguinity has been reported.13,64 The disease has occurred in siblings in
several families, both males and females have been affected and there is decreased activity of the
enzyme in the fibroblasts 9 and lymphocytes 11 of obligate heterozygotes. The clinical suspicion
was confirmed following cloning of the gene, which is on chromosome 1p36.3 and has eleven
exons.76 Most mutations are missense, although nonsense and splice site mutations have been
reported in patients with MTHFR deficiency. Each mutation has been reported in only one or
two families.1,43,65,77-81 Thirty-four different mutations causing severe disease are known, in
addition to polymorphisms which may contribute to disease in the general population. Chap-
ter 1 contains a list of all known mutations as well as information on functional impact of some
of these sequence changes.
Genotype-Phenotype Correlations
Genetic heterogeneity in the severe form of this disorder was suggested by the fact that
fibroblast extracts from two of the original families showed differential heat inactivation at 55
degrees.9 Although several of the later-onset patients had a thermolabile reductase under these
conditions, thermolability was also found in patients with early-onset disease.82 In some pa-
tients, this has been shown to be due to the presence of severe MTHFR mutations in combina-
tion with the common 677C→T polymorphism, which is responsible for the majority of en-
zyme thermolability in the general population.79,83 Recent evidence has shown that having a
severe mutation in cis with the 677C→T polymorphism produced lower enzyme activity than
the severe mutation alone.2 The presence of the 677C→T polymorphism, in combination
with a severe mutation, resulted in an additional decrease of 50%.81
Although there is a correlation between residual enzyme activity and clinical severity, it is
still difficult to make genotype-phenotype correlations. There are many different mutations in
the 32 cases with identified mutations (Table 1). In addition, many of these patients are com-
pound heterozygotes. This makes it difficult to associate a particular mutation with a specific
amount of residual enzyme activity. There can also be clinical variability among family mem-
bers harboring the same mutations. In the patient described with an adverse reaction to nitrous
oxide exposure, only a single mutation was found in combination with the two common MTHFR
polymorphisms.60 In the 32 cases with identified mutations, the range of residual enzyme ac-
tivity correlates directly with the age of onset of symptoms. The 9 cases with onset between 0-3
months had a range of enzyme activity that was 0-2%, average 0.5%. The range in the 7 cases
with onset between 3 months and 10 years was 0-8%, average 3%. In the group over 10 years,
the range was 6-23%, average 13%. One case was asymptomatic and had 7% residual enzyme
activity. Four cases (2 in the first and 2 in the second groups) had unknown residual enzyme
activity. Therefore, in general, individuals with severely decreased enzyme activity present at a
younger age with a more severe phenotype.
Conclusion
MTHFR deficiency is an inborn error of folate metabolism that is associated with decreased
methionine and S-adenosylmethionine, and elevated homocysteine levels. There are numerous
mutations in MTHFR that cause a severe reduction in the enzyme activity. In general, the more
severely reduced the enzyme activity is, the more severe is the phenotype. Clinical presentation
can occur anytime from the neonatal period to adulthood. There are often neurological abnor-
malities associated with abnormal brain pathology, occasional thromboses and, rarely, psychi-
atric symptoms. There are also cases that are asymptomatic. This is a very difficult disease to
treat. The addition of betaine in the treatment regimen has halted the neurological deteriora-
tion in many patients and has even improved the development in others.
References
1. Tonetti C, Amiel J, Munnich A et al. Impact of new mutations in the methylenetetrahydrofolate
reductase gene assessed on biochemical phenotypes: A familial study. J Inherit Metab Dis 2001;
24:833-842.
2. Goyette P, Rozen R. The thermolabile variant 677C-->T can further reduce activity when ex-
pressed in cis with severe mutations for human methylenetetrahydrofolate reductase. Hum Mutat
2000; 16:132-138.
3. Botto LD, Yang Q. 5,10-methylenetetrahydrofolate reductase gene variants and congenital anoma-
4. Rosenblatt D, Fenton WA. Inherited disorders of folate and cobalamin transport and metabolism.
In: Scriver CR, Beaudet AL, Sly WS et al, eds. The Metabolic & Molecular Bases of Inherited
Disease. 8th ed. New York: McGraw-Hill, 2001:3897-3933.
5. Freeman JM, Finkelstein JD, Mudd SH et al. Homocystinuria presenting as reversible “schizophre-
nia”: A new defect in methionine metabolism with reduced 5,10-methylenetetrahydrofolate reduc-
tase activity. Pediatr Res 1972; 6:423.
6. Freeman JM, Finkelstein JD, Mudd SH. Folate-responsive homocystinuria and “schizophrenia”: A
defect in methylation due to deficient 5,10-methylenetetrahydrofolate reductase activity. N Engl J
Med 1975; 292:491-496.
7. Kanwar YS, Manaligod JR, Wong PWK. Morphologic studies in a patient with homocystinuria
due to 5,10-methylenetetrahydrofolate reductase deficiency. Pediatr Res 1976; 10:598-609.
8. Cooper BA, Rosenblatt DS. Folate coenzyme forms in fibroblasts from patients deficient in
5,10-methylenetetrahydrofolate reductase. Biochem Soc Trans 1976; 4:921-922.
9. Rosenblatt DS, Erbe RW. Methylenetetrahydrofolate reductase in cultured human cells. II. Studies
of methylenetetrahydrofolate reductase deficiency. Pediatr Res 1977; 11:1141-1143.
10. Wong PWK, Justice P, Hruby M et al. Folic acid nonresponsive homocystinuria due to
methylenetetrahydrofolate reductase deficiency. Pediatrics 1977; 59:749-756.
11. Wong PWK, Justice P, Berlow S. Detection of homozygotes and heterozygotes with methylene-
tetrahydrofolate reductase deficiency. J Lab Clin Med 1977; 90:283-288.
12. Baumgartner ER, Schweizer K, Wick H. Different congenital forms of defective remethylation in
homocystinuria. Clinical, biochemical, and morphological studies. Pediatr Res 1977; 11:1015
13. Narisawa K, Wada Y, Saito T et al. Infantile type of homocystinuria with N5,10-methylenetetrahydrofolate
reductase defect. Tohoku J Exp Med 1977; 121:185-194.
14. Rosenblatt DS, Cooper BA. Methylenetetrahydrofolate reductase deficiency: Clinical and biochemical
correlations. In: Botez MI, Reynolds EH, eds. Folic acid in Neurology, Psychiatry, and Internal
Medicine. New York: Raven Press, 1979:385-390.
15. Rosenblatt DS, Cooper BA, Lue-Shing S et al. Folate distribution in cultured human cells. Studies
on 5,10-CH2- H4PteGlu reductase deficiency. J Clin Invest 1979; 63:1019-1025.
16. Narisawa K. Brain damage in the infantile type of 5,10-methylenetetrahydrofolate reductase defi-
ciency. In: Botez MI, Reynolds EH, eds. Folic acid in Neurology, Psychiatry, and Internal Medi-
cine. New York: Raven Press, 1979:391-400.
17. Singer HS, Butler I, Rothenberg S et al. Interrelationships among serum folate, CSF folate, neu-
rotransmitters, and neuropsychiatric symptoms. Neurology 1980; 30:419
18. Baumgartner R, Wick, Ohnacker H et al. Vascular lesions in two patients with congenital
homocystinuria due to different defects of remethylation. J Inher Met Dis 1980; 3:101-103.
19. Cederbaum SD, Shaw KNF, Cox DR et al. Homocystinuria due to methylenetetrahydrofolate re-
ductase (MTHFR) deficiency: Response to a high protein diet. Pediatr Res 1981; 15:560
20. Allen RJ, Wong PWK, Rothenberg SP et al. Progressive neonatal leukoencephalomyopathy due to
absent methylenetetrahydrofolate reductase, responsive to treatment. Ann Neurol 1980; 8:211
21. Narisawa K. Folate metabolism infantile type of 5,10-methylenetetrahydrofolate reductase deficiency.
Acta Paediatr Jap 1981; 23:82.
22. Boss G, Erbe RW. Decreased rates of methionine synthesis by methylenetetrahydrofolate
reductase-deficient fibroblasts and lymphoblasts. J Clin Invest 1981; 67:1659-1664.
23. Harpey JP, Rosenblatt DS, Cooper BA et al. Homocystinuria caused by 5,10-methylenetetrahydrofolate
reductase deficiency: A case in an infant responding to methionine, folinic acid, pyridoxine, and
vitamin B12 therapy. J Pediatr 1981; 98:275-278.
24. Harpey JP, Lemoel G, Zittoun J. Follow-up in a child with 5,10-methylenetetrahydrofolate reduc-
tase deficiency. J Pediatr 1983; 103:1007
25. Wendel U, Bremer HJ. Betaine in the treatment of homocystinuria due to 5,10-methylenetetrahydrofolate
reductase deficiency. Eur J Pediatr 1984; 142:147-150.
26. Christensen E, Brandt NJ. Prenatal diagnosis of 5,10-methylenetetrahydrofolate reductase deficiency.
N Engl J Med 1985; 313:50-51.
27. Nishimura M, Yoshino K, Tomita Y et al. Central and peripheral nervous system pathology of
homocystinuria due to 5,10-methylenetetrahydrofolate reductase deficiency. Pediatr Neurol 1985;
1:375-378.
28. Haan E, Rogers J, Lewis G et al. 5,10-methylenetetrahydrofolate reductase deficiency: Clinical and
biochemical features of a further case. J Inher Met Dis 1985; 8:53-57.
29. Hyland K, Smith I, Howells DW et al. The determination of pterins, biogenic amino metabolites,
and aromatic amino acids in cerebrospinal fluid using isocratic reverse phase liquid chromatogra-
phy within series dual cell coulometric electrochemical and fluorescence determinations - Use in
the study of inborn errors of dihydropteridine reductase and 5,10-methylenetetrahydrofolate reduc-
tase. In: Wachter H, Curtius H, Pfleiderer W, eds. Biochemical and Clinical Aspects of Pteridines.
Berlin: Walter de Gruyter, 1985:4:85-99.
30. Baumgartner ER, Stokstad ELR, Wick H et al. Comparison of folic acid coenzyme distribution
patterns in patients with methylenetetrahydrofolate reductase and methionine synthetase deficien-
cies. Pediatr Res 1985; 19:1288-1292.
31. Berlow S. Critical review of cobalamin-folate interrelations (letter). Blood 1986; 67:1526
32. Shin YS, Pilz G, Enders W. Methylenetetrahydrofolate reductase and methylenetetrahydrofolate
methyltransferase in human fetal tissues and chorionic villi. J Inher Med Dis 1986; 9:275-276.
33. Clayton PT, Smith I, Harding B et al. Subacute combined degeneration of the cord, dementia and
Parkinsonian due to an inborn error of folate metabolism. J Neurol Neurosurg Psychiatry 1986;
49:920-927.
34. Brandt NJ, Christensen E, Skovby F et al. Treatment of methylenetetrahydrofolate reductase defi-
ciency from the neonatal period. The Society for the Study of Inborn Errors of Metabolism. In:
Anonymous The Netherlands (Abstract): Amersfoort, 1986:23.
35. Fowler B. Homocystinuria, remethylation defects. Methionine synthesis and cofactor response in
cultured fibroblasts. Society for the Study of Inborn Errors of Metabolism, 24th Annual Sympo-
sium, Amersfoort, the Netherlands, Sept 1986; 9-12.
36. Beckman DR, Hoganson G, Berlow S et al. Pathological findings in 5,10-methylenetetrahydrofolate
reductase deficiency. Birth Defects: Original Article Series 1987; 23:47-64.
37. Visy JM, Le Coz P, Chadefaux B et al. Homocystinuria due to 5,10-methylenetetrahydrofolate
reductase deficiency revealed by stroke in adult siblings. Neurology 1991; 41:1313-1315.
38. Haworth JC, Dilling LA, Surtees RAH et al. Symptomatic and asymptomatic methylenetetrahydrofolate
reductase deficiency in two adult brothers. Am J Med Genet 1993; 45:572-576.
39. Fowler B. Genetic defects of folate and cobalamin metabolism. Eur J Pediatr 1998; 157:S60-S66.
40. Ogier de Baulny H, Gerard M, Saudubray JM et al. Remethylation defects: Guidelines for clinical
diagnosis and treatment. Eur J Pediatr 1998; 157:S77-S83.
41. Abeling NGGM, van Gennip AH, Blom H et al. Rapid diagnosis: Basis for a favourable outcome
in a patient with MTHFR deficiency. J Inher Metab Dis 1998; 21:21 Abstract.
42. Sewell AC, Neirich U, Fowler B. Early infantile methylenetetrahydrofolate reductase deficiency: A
rare cause of progressive brain atrophy. J Inher Metab Dis 1998; 21:22 Abstract.
43. Homberger A, Linnebank M, Sewell A et al. Severe methylenetetrahydrofolate reductase deficiency:
Two novel genotypes with different clinical course. J Inher Metab Dis 2001; 24(suppl 1):50 (ab-
stract)
44. Hyland K, Smith I, Bottiglieri T et al. Demyelination and decreased S-adenosylmethionine in
5,10-methylenetetrahydrofolate reductase deficiency. Neurology 1988; 38:459-462.
45. Holme E, Kjellman B, Ronge E. Betaine for treatment of homocystinuria caused by
methylenetetrahydrofolate reductase deficiency. Arch Dis Child 1989; 64:1061-1064.
46. Engelbrecht V, Rassek M, Huismann J et al. MR and proton MR spectroscopy of the brain in
hyperhomocysteinemia caused by methylenetetrahydrofolate reductase deficiency. AJNR 1997;
18:536-539.
47. Kishi T, Kawamura I, Harada Y et al. Effect of betaine on S-adenosylmethionine levels in the
cerebrospinal fluid in a patient with methylenetetrahydrofolate reductase deficiency and peripheral
neuropathy. J Inher Metab Dis 1994; 17:560-565.
48. Marquet J, Chadefaux B, Bonnefont JP et al. Methylenetetrahydrofolate reductase deficiency pre-
natal diagnosis and family studies. Prenat Diagn 1994; 14(1):29-33.
49. Walk D, Kang S, Horwitz A. Intermittent encephalopathy, reversible nerve conduction slowing,
and MRI evidence of cerebral white matter disease in methylenetetrahydrofolate reductase defi-
ciency. Neurology 1994; 44 (2):344-347.
50. Ronge E, Kjellman B. Long term treatment with betaine in methylenetetrahydrofolate reductase
deficiency. Arch Dis Child 1996; 74:239-241.
51. Tonetti C, Burtscher A, Bories D et al. Methylenetetrahydrofolate reductase deficiency in four
siblings: A clinical, biochemical, and molecular study of the family. Amer J Hum Genet 2000;
9:363-367.
52. Al Essa M, Sakati NA, Dabbagh O et al. Inborn error of vitamin B12 metabolism: A treatable
cause of childhood dementia/paralysis. J Child Neurology 1999; 13:239-243.
53. Al Tawari AA, Ramadan DG, Neubauer D et al. An early onset form of methylenetetrahydroflate
reductase deficiency a report of a family from Kuwait. Brain Development 2002; 24:304-309.
54. Tonetti C, Ruivard M, Rieu V et al. Severe methylenetetrahydrofolate reductase deficiency re-
vealed by a pulmonary embolism in a young adult. Br J Heamatology 2002; 119:397-399.
55. Fattal-Valevski A, Bassan H, Korman SH et al. Methylenetetrahydrofolate reductase deficiency:
Importance of early diagnosis. J Child Neurol 2000; 15:539-543.
56. Abeling NGGM, van Gennip AH, Blom H et al. Rapid diagnosis and methionine administration:
Basis for a favourable outcome in a patient with methylene tetrahydrofolate reductase deficiency. J
Inher Metab Dis 1999; 22:240-242.
57. Arn PH, Williams CA, Zori RT et al. Methylenetetrahydrofolate reductase deficiency in a patient
with phenotypic findings of Angelman syndrome. Amer J Med Genet 1998; 77:198-200.
58. Pasquier F, Lebert F, Zittoun J et al. Methylenetetrahydrofolate reductase deficiency revealed by a
neuropathy in a psychotic adult [letter]. J Neurol Neurosurg Psychiatry 1994; 57(6):765-766.
59. Scott JM, Dinn JJ, Wilson P et al. Pathogenesis of subacute combined degeneration: A result of
methyl group deficiency. Lancet 1981; 2:334-337.
60. Selzer RR, Rosenblatt DS, Laxova R et al. Nitrous oxide and 5,10-methylenetetrahydrofolate re-
ductase deficiency. N Engl J Med 2003; 349:49-50.
61. Goyette P, Rosenblatt DS, Rozen R. Homocystinuria (methylenetetrahydrofolate reductase defi-
ciency) and mutations of Factor V gene. J Inher Metab Dis 1998; 21:690-691.
62. Levitt M, Nixon PF, Pincus JH et al. Transport of folates in cerebrospinal fluid:A study using
doubly labelled 5-methyltetrahydrofolate and 5-formyltetrahydrofolate. J Clin Invest 1971; 50:1301
63. Wendel U, Claussen U, Dickmann E. Prenatal diagnosis for methylenetetrahydrofolate reductase
deficiency. J Pediatr 1983; 102:938-940.
64. Erbe RW. Inborn errors of folate metabolism. In: Blakley RL, Whitehead VM, eds. Folates and
Pterins Nutritional, Pharmacological and Physiological Aspects. New York: Wiley, 1986; 3:413-466.
65. Kluijtmans LAJ, Wendel U, Stevens EMB et al. Identification of four novel mutations in severe
methylenetetrahydrofolate reductase deficiency. Eur J Hum Genet 1998; 6:257-265.
66. Sakura N, Ono H, Nomura H et al. Betaine dose and treatment intervals in therapy for
homocystinuria due to 5,10-methylenetetrahydrofolate reductase deficiency. J Inher Metab Dis 1998;
21:84-85.
67. Fowler B, Jakobs C. Post- and prenatal diagnostic methods for the homocystinurias. Eur J Pediatr
1998; 157:S88-S93.
68. Rosenblatt DS, Erbe RW. Methylenetetrahydrofolate reductase in cultured human cells. Growth
and metabolic studies. Pediatr Res 1977; 11:1137-1141.
69. Suormala T, Gamse G, Fowler B. 5,10 methylenetetrahydrofolate reductase (MTHFR) assay in the
forward direction: Residual activity in MTHFR deficiency. Clin Chem 2002; 48:835-843.
70. Mudd SH, Uhlendorf BW, Freeman JM et al. Homocystinuria associated with decreased
methylenetetrahydrofolate reductase activity. Biochem Biophys Res Commun 1972; 46:905-912.
71. Cooper BA. Anomalies congenitales du metabolisme des folates. In: Zittoun J, Cooper BA, eds.
Folates et cobalamines. Paris: Doin, 1987:193-208.
72. Gaull GE, Von Berg W, Raiha NCR et al. Development of methyltransferase activities of human
fetal tissues. Pediatr Res 1973; 7:527-533.
73. Zittoun J. Congenital errors of folate metabolism. In: Wickringamasinghe S, ed. Megaloblastic
anaemias. London: Bailière Tindall, 1995:603-616.
74. Takenaka T, Shimomura T, Nakayasu H et al. Effect of folic acid for treatment of homocystinuria
due to 5,10-methylenetetrahydrofolate reductase deficiency [Japanese]. Rinsho-Shinkeigaku-Clinical
Neurology 1993; 33(11):1140-1145.
75. Erbe RW. Genetic aspects of folate metabolism. Adv Hum Genet 1979; 9:293-354.
76. Goyette P, Pai A, Milos R et al. Gene structure of human and mouse methylenetetrahydrofolate
reductase (MTHFR). Mamm Genome 1998; 9:652-656.
cDNA, mapping and mutation identification. Nat Genet 1994; 7(2):195-200.
reductase gene and genotype/phenotype correlations in severe methylenetetrahydrofolate reductase
deficiency. Amer J Hum Genet 1995; 56:1052-1059.
methylenetetrahydrofolate (MTHFR) gene, and description of 5 novel mutations in MTHFR. Amer
J Hum Genet 1996; 59:1268-1275.
80. Sibani S, Christensen B, O‘Ferrall E et al. Characterization of six novel mutations in the
methylenetetrahydrofolate reductase (MTHFR) gene in patients with homocystinuria. Hum Mutat
2000; 15(3):280-7.
81. Sibani S, Leclerc D, Weisberg IS et al. Characterization of mutations in severe methylenetetrahydrofolate
reductase deficiency reveals an FAD-responsive mutation. Hum Mutat 2003; 21(5):509-20.
82. Rosenblatt DS, Lue-Shing H, Arzoumanian A et al. Methylenetetrahydrofolate reductase (MR)
deficiency: Thermolability of residual MR activity, methionine synthase activity, and methylcobalamin
levels in cultured fibroblasts. Biochem Med Met Biol 1992; 47(3):221-225.
CHAPTER 5
Mild MTHFR Deficiency and Folate Status

Paul F. Jacques and Silvina Furlong Choumenkovitch
Abstract
M
ild methylenetetrahydrofolate reductase (MTHFR) deficiency can result from a
missense mutation, a cytosine-to-thymidine transition at base pair 677 of the MTHFR
gene (677C→T). This mutation results in an MTHFR enzyme with reduced spe-
cific activity and higher in vitro thermolability. The complex relationship between the MTHFR
677C→T genotype and folate status is reviewed. This mutation reduces plasma and serum
folate levels, but the relation between the 677C→T polymorphism and red cell folate concen-
trations is confounded by the influence of this mutation on the distribution of different folate
forms. Folate status also affects the expression of the mutation as demonstrated by an enhanced
elevation of homocysteine concentrations among individuals who are homozygous for this
mutation when folate levels are inadequate. The evidence relating a second polymorphism, an
adenine to cytosine transition at nucleotide 1298 (1298A→C), is also reviewed. Unlike the
677C→T mutation, 1298A→C mutation alone does not appear to be associated with higher
plasma homocysteine or lower plasma folate concentrations. However, combined heterozygos-
ity for the 677C→T and the 1298A→C variants results in lower enzyme activity, higher
homocysteine levels, and possibly reduced folate levels when compared with the levels observed
in heterozygous for either variant alone
Mild MTHFR Deficiency

Methylenetetrahydrofolate reductase (MTHFR) is the enzyme that catalyzes the irrevers-
ible reduction of 5, 10- methylenetetrahydrofolate to 5-methyltetrahydrofolate, the major cir-
culatory form of folate and the methyl donor for the remethylation of homocysteine into
methionine.1 Severe defects in the MTHFR gene that result in loss of enzyme function are rare
and result in severe hyperhomocysteinemia and hyperhomocystinuria (elevated concentration
of homocysteine in blood and urine, respectively) and low folate levels in blood, red blood cells
and cerebrospinal fluid.2-4 These patients suffer from serious clinical symptoms that can occur
in childhood or even in early adulthood, including developmental delays, seizures, psychomo-
tor retardation, neurologic abnormalities, psychiatric disturbances and occlusive vascular dis-
ease.3,5-7
A milder form of MTHFR deficiency was initially recognized in 1988. Kang et al8 identi-
fied two unrelated patients who had a variant of the MTHFR enzyme that had a lower catalytic
activity and was more susceptible to heat inactivation (“thermolabile”) than the normal en-
zyme. Shortly after the isolation of the cDNA and mapping of the human MTHFR gene,9
Rozen and colleagues identified the mutation responsible for the thermolabile variant.10 This
missense mutation, a cytosine-to-thymidine transition at base pair 677 of the MTHFR gene
(677C→T), results in a substitution of Ala 222 by Val in the amino acid sequence. This poly-
morphism results in an MTHFR enzyme with reduced specific activity and higher in vitro
thermolability.10 An individual can be homozygous (TT genotype), heterozygous (CT
©2005 Eurekah.com.
Mild MTHFR Deficiency and Folate Status 55
genotype) or wild type (CC genotype) for this mutation. This is an autosomal recessive muta-
tion, and the prevalence of homozygosity for this common variant was shown to be different in
different ethnic groups: approximately 10% in Caucasians of Northern European descent,
20% in Italian populations, 25% in Hispanic Americans, and uncommon in people of African
descent.2,11,12
A second polymorphism in the MTHFR gene is also associated with decreased MTHFR
activity in vitro, but not in thermolability13,14 This MTHFR variant is an adenine to cytosine
transition at nucleotide 1298 (1298A→C), which results in a Glu to Ala substitution in the
MTHFR protein.
The Role of MTHFR in Homocysteine Metabolism

Homocysteine is a sulfur containing amino acid whose metabolism is at the intersection of
two metabolic pathways: remethylation and transsulfuration. In remethylation, homocysteine
acquires a methyl group from N-5 methyltetrahydrofolate (methylTHF) to form methionine.1
This reaction is catalyzed by a methyltransferase (methionine synthase), which requires vita-
min B12 as a cofactor. Homocysteine can also accept a methyl group from betaine in a reaction
catalyzed by betaine-homocysteine methyltransferase. However, this betaine-dependent reac-
tion is found only in the liver and kidney. In the transsulfuration pathway, homocysteine con-
denses with serine to form cystathionine in a reaction catalyzed by the enzyme cystathionine
β-synthase (CBS) that utilizes pyridoxal-5'-phosphate (the active circulating form of vitamin
B6) as a cofactor. Cystathionine is then hydrolyzed into cysteine and α-ketobutyrate in a reac-
tion catalyzed by another pyridoxal-5'-phosphate containing enzyme, γ-cystathionase. Thus,
the vitamins folate, B12 and B6 are necessary for the normal metabolism of homocysteine.1
Either a deficiency in one or more of the vitamins that are precursors or cofactors in homocys-
teine metabolism or a defect in one of the enzymes of homocysteine metabolism can disrupt
the remethylation and/or the transsulfuration pathway and potentially lead to excess homocys-
teine. Several studies carried out in humans and in animals have shown that deficiencies of
folate or B12, or even borderline normal levels of these vitamins, result in elevated circulating
total homocysteine (tHcy) concentrations.1,15-18 Because of the relationship between tHcy con-
centrations and status of these vitamins, tHcy is a functional indicator of folate and vitamin
B12 status.
MTHFR is one of the key enzymes involved in homocysteine metabolism. It catalyzes the
irreversible conversion of 5, 10- methylenetetrahydrofolate (methyleneTHF) to 5-methylTHF,
the methyl donor to remethylate homocysteine into methionine.1 If the decreased enzyme
activity of the 677C→T mutant enzyme observed in vitro occurred in vivo, one would expect
a decreased production of 5-methylTHF, which would result in impaired homocysteine
remethylation and higher circulating tHcy concentrations. Studies carried out in several differ-
ent populations have shown that this is the case. Homozygotes for the thermolabile enzyme
have higher circulating tHcy levels than CT or CC individuals,2,19,20 about 25% higher than
individuals with the CC genotype.19
Effect of Folate Status on the Relation between MTHFR 677C→T

Genotype and Homocysteine
The discovery by Kang and colleagues8,21-23 of the relation between thermolabile MTHFR
and mild hyperhomocysteinemia stimulated great interest in the mutant enzyme because mild
hyperhomocysteinemia was becoming recognized as a novel risk factor for vascular disease.
This work also provided two very useful pieces of evidence for further understanding of the
phenotypic relation between thermolabile MTHFR and elevated homocysteine concentrations.
First, it was apparent from their work that not all individuals with the thermolabile enzyme
experienced elevations in plasma tHcy.21,23 This suggested that the presence of the mutant
enzyme was not sufficient to cause mild hyperhomocysteinemia. Moreover, folic acid treat-
ment normalized elevated tHcy in patients with this enzyme defect,8 suggesting that folate
status played an important role in determining the phenotypic expression of the thermolabile
enzyme. The task of examining factors that might affect this phenotypic response was greatly
simplified when Rozen and colleagues identified the mutation responsible for the thermolabil-
ity.9,10
For reasons noted above, the most reasonable candidate to examine as a possible modifier of
the 677C→T mutation was folate status. In collaboration with Dr. Rozen, our laboratory was
the first to demonstrate that folate did influence the relation between MTHFR 677C→T geno-
type and tHcy.24 In a population of 365 participants of the NHLBI Family Heart Study, we
observed that individuals homozygous for the 677T mutation in the MTHFR gene had higher
tHcy in plasma compared to CC or CT only when their folate status was below the sample
median (15.4 nmol/L). When their folate status was above the sample median, there was no
difference in tHcy levels among the three genotypes. This observation, which demonstrated
that folate status affects the phenotypic expression of the mutant enzyme, was quickly con-
firmed by Ma and colleagues.25 They demonstrated, in both incident cases of myocardial inf-
arction and normal controls from the Physicians’ Health Study, that the TT genotype was only
associated with elevated homocysteine concentrations in the lowest quartile category of plasma
folate concentrations. Table 1 summarizes these and several other studies, which confirmed
this gene-nutrient interaction with both plasma and RBC folate in a variety of different healthy
and patient populations.24-43 The consistency of this association across the 20 studies listed in
Table 1 is striking.
Figure 1 displays the relationship between genotype, folate status and plasma tHcy concen-
trations before implementation of folic acid fortification in the United States using previously
unpublished data from the Framingham Offspring Study. This figure clearly shows that the
elevated tHcy concentrations are present in those with the TT genotype only at plasma folate
levels below approximately 7 nmol/L.
While Table 1 summarizes the studies that presented the relation between genotype and
homocysteine within folate categories, other investigators examined this interaction by pre-
senting the relationship between folate and homocysteine within genotype categories. These
studies affirm the presence of the interaction between genotype and folate status. For example,
McQuillan et al44 reported that the relation between plasma tHcy and folate status and intake
was stronger in TT individuals compared with CC among 1111 participants of the Australian
National Heart Foundation Perth Risk Factor Prevalence Survey. Accordingly, the difference in
tHcy between the highest and lowest quartile categories of serum folate was 5.1 µmol/L in TT
compared to 2.1 µmol/L in CC.
Other investigators compared the regression models relating tHcy and folate for individuals
with different genotypes. In a group of 161 vascular disease patients and 349 healthy siblings,
de Jong and coworkers45 reported that the negative slope relating higher serum folate to tHcy
levels was 50% greater in TT relative to CC individuals (P<0.001). In a combined population
of 170 patients with early onset thrombosis and 182 healthy controls, D’Angelo et al46 re-
ported that the inverse regression slope relating folate and tHcy was nearly twice as steep in TT
than in CC or CT individuals (P<0.001). In a population of patients with premature cardio-
vascular disease and healthy controls, Verhoeff and colleagues47 considered the effect of geno-
type on the relation between tHcy and a combined measure of folate and vitamin B12 status.
Those with the TT genotype had an inverse regression slope that was again twice as steep as the
slope in the other two genotypes (P<0.001), suggesting a stronger relation of tHcy and these
vitamins in TT individuals. Nakamura et al48 observed that the negative regression slopes relat-
ing folate and tHcy concentrations were also approximately two-times greater in dialysis pa-
tients with the TT and CT genotypes than in patients with the CC genotype. In a study
conducted in a population of families with spina bifida-affected children and a healthy control
group, van der Put and coworkers49 observed a significant negative Spearman’s correlation
coefficient of -0.4 (P<0.05) between plasma folate and tHcy in TT individuals, but these fac-
tors were uncorrelated in CT (-0.12, P<0.2) and CC (0.04, P<0.6) individuals.
Table 1. Summary of studies that examined the influence of folate status on the relation between the MTHFR 677C→T polymorphism and
total plasma homocysteine concentrations
Total Homocysteine Concentrations (µmol/L)

Low Folate High Folate
Folate Cutoff
Reference Population Description n (nmol/L) CC CT TT CC CT TT
Jacques et al, 199624 Normal and high 365 15.4 9.81 9.51 12.1 7.8 7.5 7.9
CVD risk subjects
Ma et al, 199625 Male physicians 583 <5.0, >15.3 12.31 12.91 16.0 9.4 9.3 9.7
Harmon et al, 199626 Healthy men 611 11.0 7.41 7.91 11.2 6.3 6.5 6.8
Bostom et al, 199627 Dialysis patients 75 66 21.82 29.2 22.3 19.0
Schwartz et al, 199728 Healthy women 338 <8.4, ≥15.6 12.73 18.2 9.2 7.4
Verhoef et al, 199729 CAD cases and controls 318 7904 12.05 13.2 21.4 11.45 12.0 12.1
Christensen et al, 199730 Male CAD cases 94 8.6 11.31 12.1 14.6 9.6 10.2 11.0
Male controls 86 7.5 10.51 10.11 15.8 10.1 11.0 8.2
Dunn et al, 199831 CAD cases aged < 50 yrs 145 11.1 16.31 16.31 23.8 15.4 14.5 14.3
CAD cases aged > 65 yrs 148 12.9 19.01 19.41 24.8 19.9 18.8 16.6
Morita et al, 199832 Stroke patients 141 9.1 11.31 12.21 15.1 8.9 10.4 9.9
Motti et al, 199833 Healthy volunteers 155 10.6 10.96 10.5 16.4 11.0 11.0 11.0
Girelli et al, 199834 CAD cases and controls 415 11.5 14.51,7 16.71 23.1 13.0 14.1 12.7
57

58
Table 1. Continued
Total Homocysteine Concentrations (µmol/L)

Low Folate High Folate
Folate Cutoff
Reference Population Description n (nmol/L) CC CT TT CC CT TT
Thuillier et al, 199835 Vascular disease 182 14.2 11.91 12.31 14.6 12.0 12.6 10.1
patients and controls
Tokgozoglu et al, 199936 CAD patients and controls 242 12.9 15.26 19.5 28.4 14.0 14.5 12.0
Gemmati et al, 199937 Thromboembolic cases 220 8.4 11.41 13.11 21.7 6.5 7.9 8.6
Controls 220 9.3 6.71 8.71 19.9 6.6 7.8 8.9
Yoo & Park, 200038 CAD cases 187 17.2 10.01 9.81 17.7 7.8 9.4 10.8
Controls 122 17.2 8.71 9.91 17.1 9.4 9.4 9.3
Potena et al, 200139 Heart transplant recipients 57 13.8 166 21 34 16 18 20
Saw et al, 200140 Population-based sample 477 13.7 10.73 13.8 9.1 9.4
Geisel et al, 200141 Volunteers 280 <12.9, >18.6 14.55 13.3 16.2 7.55 8.6 7.2
Moriyama et al, 200242 Healthy volunteers 965 Tertiles 8.56 8.7 12.0 7.36 7.9 10.1
McNulty et al, 200243 Healthy volunteers 283 Median4 8.96 9.2 12.0 7.2 7.7 9.6
CVD = cardiovascular disease, CAD = coronary artery disease. 1 Significantly different from TT (P<0.05). 2 CC significantly different from pooled TT+CT (P<0.05).
3 Pooled CC+CT significantly different from TT (P<0.05). 4 RBC folate concentrations. 5 Results of statistical tests for the comparison of means were not provided.
6 Individual differences not provided, but statistical test for overall differences between genotype categories was statistically significant (P<0.05). 7 Significantly
different from CT (P<0.05).

Figure 1. Relation between fasting total plasma homocysteine and folate concentrations by MTHFR 677C→T
genotype. Data from the 5th examination of the Framingham Offspring Study cohort. Geometric mean
plasma homocysteine concentrations and 95% confidence intervals are presented for quintile categories
(fifths) of plasma folate and are plotted at the median plasma folate value for each genotype within folate
categories.
Finally, a study from Guttormsen and colleagues50 examined the folate-genotype interac-
tion on tHcy levels in 67 individuals with plasma tHcy ≥ 40 µmol/L and 329 controls who
were part of the Hordaland Homocysteine Study. They reported that the negative relation
between plasma tHcy and plasma folate was intensified with the number of T alleles. After
stratifying this group into those with plasma folate above and below 3.7 nmol/L, the risk of
having tHcy ≥ 40 µmol among those with the T allele was 36 times greater than for CC
individuals at low folate levels and only 3.1 times greater for those with higher folate.
The influence of folate status on the phenotypic expression of the 677C→T mutation may
be due to the ability of folate to maintain the MTHFR enzyme in its active form. Folate may
stabilize the holoenzyme by enhancing the affinity of the enzyme for a critical cofactor, flavin
adenine dinucleotide (FAD), which transfers reducing equivalents from NAD(P)H to
methylenetetrahydrofolate.20,51,52 The 677C→T mutation affects a residue in the catalytic
domain of the enzyme.51,52 Studies using the Escherichia coli enzyme as a model for the cata-
lytic domain of human MTHFR showed a much higher rate of dissociation of the flavin from
the mutant MTHFR enzyme compared to the rate of dissociation in the wild type enzyme.52
This model also suggested that the wild type enzyme was susceptible to flavin loss.
MethyleneTHF prevented flavin loss and activity loss in both the wild type and mutant en-
zyme. However, the mutant MTHFR required much higher concentrations of folate to de-
crease the initial rate of flavin loss to values comparable to those in the wild type enzyme. These
observations in the bacterial enzyme model were confirmed when human enzyme was ex-
pressed in a baculovirus system in insect Sf9 cells.53 The mutant enzyme had a higher rate of
flavin dissociation and loss of activity compared with the wild type enzyme, and folate retarded
this loss of flavin in both the wild type and mutant enzyme.
The results from these model systems are consistent with observations from free-living hu-
mans. We previously reported that low riboflavin status among participants in the Framingham
Offspring Study was associated with high tHcy, but only under conditions of low folate status
in individuals homozygous for the 677T mutation.54 Hustad and colleagues55 also observed
that low riboflavin was associated with elevated tHcy in those with the mutant T allele, but not
among those with the CC genotype. However, these investigators did not see an effect of folate
status on the relationship, perhaps as a consequence of lower folate levels in their population.
Effect of the 677C→T Mutation on the Response to Homocysteine

Lowering Therapy
The administration of folic acid supplements clearly lowers plasma homocysteine levels.56
Several studies have also examined the impact of MTHFR 677C→T genotype on the
tHcy-lowering effects of folic acid supplements.
Fohr and coworkers57 reported a larger response to the tHcy lowering effect of a 400 µg/day
folic acid supplement in TT and CT individuals than in those with the CC genotype (20% in
TT, 15% in CT and 8% in CC) after 4 weeks of treatment. The tHcy levels after treatment
were 6.8 µmol/L in the TT, 7.1 µmol/L in the CT and 7.7 µmol/L in the CC individuals.
Malinow and colleagues58 showed that 1 or 2 mg/day folic acid supplement treatment given to
242 men and women for 3 weeks decreased plasma tHcy by 10% in TT, 3% in CT and 3% in
CC subjects who took multivitamin supplements during the trial and by 21% in TT, 13% in
CT and 7% in CC subjects who did not use multivitamin supplements. tHcy values were
comparable across genotype groups for individuals who took multivitamins at baseline (8.1
µmol/L TT, 8.4 µmol/L CT and 8.3 µmol/L CC) and after treatment (7.2 µmol/L in the TT,
8.0 µmol/L in the CT and 7.9 µmol/L in the CC). The final tHcy concentrations in the
individuals who did not use multivitamins were 9.1 µmol/L for TT, 8.5 µmol/L for CT and
9.0 µmol/L for CC genotypes. In a study of women who suffered unexplained recurrent mis-
carriages, Nelen et al59 investigated the effect of supplementation with 0.5 mg/day folic acid on
tHcy. After 2 months, homozygous mutant women had a significantly larger decrease in plasma
tHcy (41%) compared to the other two genotypes, but the final homocysteine concentrations
were again similar in the three genotype categories (8.8 µmol/L TT, 9.5 µmol/L CT and 9.7
µmol/L CC).
In addition to the use of supplements to manipulate folate intake, other studies used folate
restricted and folate enriched diets with or without supplements to examine the influence of
the 677 C→T genotype on tHcy concentrations. McDowell and coworkers60,61 administered
3 interventions of 4 months of duration each in 126 healthy individuals (42 of each genotype):
a folic acid exclusion diet (usual diet, excluding foods fortified with folic acid providing ap-
proximately 200 µg total folate/day), a folate-rich diet (providing approximately 400 µg/day
from foods naturally rich in folate and fortified with folic acid), and a folic acid supplement
(exclusion diet plus a 400 µg/day folic acid supplement). TT individuals had higher tHcy
compared to CT or CC individuals at baseline, during the folic acid exclusion diet and after
receiving the folate-rich diet. After the supplementation intervention, plasma tHcy levels de-
clined 24%, 12% and 8% in the TT, CT and CC individuals, respectively. tHcy concentra-
tions after four months of supplementation were 9.5 µmol/L in the TT individuals and 8.1
µmol/L in both the CT and CC individuals. The differences between the genotypes were not
statistically significant. In a controlled metabolic study carried out by Kauwell et al62 in 11 CC,
15 CT and 7 TT women aged 60-85 years, volunteers went through a folate depletion phase of
7 weeks and a repletion period in which they were randomly assigned to receive either 200 or
415 µg folate/day. During the depletion phase, tHcy concentrations increased by 47% in the
TT, 20% in the CT and 13% in the CC individuals. When both repletion groups were com-
bined, tHcy levels of TT individuals decreased by 19% during repletion compared to a 1%
increase in the CT and an 8% decrease in the CC individuals. There was no difference in
plasma tHcy between genotypes after folate repletion (10.1 µmol/L TT, 10.9 µmol/L CT and
10.4 µmol/L CC).
Two folate intervention studies failed to note a genotype effect on tHcy response. Silaste et
al63 evaluated the effect of the MTHFR mutation on the response of plasma tHcy and serum
folate to increased levels of dietary folate in 37 healthy female volunteers (5 TT). With the
increase in folate in the diet (from 200 to 600 µg/day without fortified foods), plasma tHcy
decreased in the three genotypes; however, there was no difference in this response between
genotypes. The lack of any significant effect of genotype may be due to the small sample size
and the low baseline tHcy concentrations (9.1 µmol/L for TT and 8.0 µmol/L for CC and
CT). After the high folate diet, tHcy levels were 7.0 µmol/L for TT, 7.0 µmol/L for CT and
6.9 µmol/L for CC. Woodside et al64 also failed to note a greater response of TT individuals to
treatment with a high-dose B vitamin supplement. They randomly assigned 132 healthy men
ages 30-49 years with tHcy concentrations ≥ 8.3 µmol/L to a supplement containing 1 mg
folic acid, 7.2 mg pyridoxine, and 0.02 mg cyanocobalamin or placebo. In a subset of 50 men
on whom they determined MTHFR genotype, the response of tHcy concentrations to treat-
ment was similar. However, the authors of this study indicated that the trial was not designed
to examine the effect of genotype on response to supplementation and was consequently un-
derpowered.
These intervention studies demonstrate that TT individuals have higher tHcy concentra-
tions than CC individuals before folic acid therapy, but all genotypes tend to achieve similar
post-therapy tHcy levels. This results in a greater response for those with the TT genotype.
However, this may not indicate a metabolic difference in response to supplementation. Rather,
it may just reflect the initial higher tHcy levels of the TT individuals. Overall, the studies
summarized above provide little evidence that TT and CC individuals with similar baseline
tHcy or folate levels respond differently to folic acid treatment. Nor do they suggest that TT
individuals require higher folate intakes than CC individuals to normalize their tHcy levels.
However, these studies were not specifically designed to address such questions. Nonetheless,
the available evidence from these intervention studies does not suggest that the greater respon-
siveness of the TT individuals is a consequence of the genotype per se, but rather a conse-
quence of the higher pretreatment tHcy concentrations.
Effect of the 677C→T Mutation on Folate Status

Table 2, which summarizes the evidence of a possible effect of MTHFR genotype on folate
status, demonstrates the extraordinary amount of attention that this relationship has received.
Studies that have examined the influence of genotype on folate status have assessed folate status
using red blood cell (RBC) and plasma or serum concentrations. There are striking inconsis-
tencies between the studies that used RBC folate concentrations to assess folate status whereas
the evidence based on plasma and serum concentrations is very consistent. The lack of agree-
ment across studies using RBC folate may be a result of two factors: a differential effect of
MTHFR genotype on distinct forms of folate and the use of different assays that are not equally
specific to the different folate forms. To help clarify the relationship between genotype and
folate status, the studies based on RBC and plasma or serum folate are summarized separately.
677C→T Mutation and RBC Folate

In the first report relating RBC folate to MTHFR 677C→T genotype, van der Put and
colleagues49,65 observed that individuals with the TT genotype had higher RBC folate levels.
The next published study on this relation reported lower RBC folate levels among TT indi-
viduals.66 The subsequent studies have been as likely to report no association with genotype29,57,67
or either higher14,59,68,69 or lower43,70,71 RBC folate levels among TT compared with CC indi-
viduals.
As previously suggested by others,20,72 the explanation of these inconsistencies may be found
in the methods used to measure folate. Molloy et al72 addressed this issue directly by compar-
ing RBC folate concentrations measured with both a radioassay and a microbiological assay.
They reported that TT individuals had higher folate levels when measured by radioassay than
when measured by the microbiological method, while CC and CT genotypes had higher levels
for the microbiological assay and lower levels for the radioassay. The difference is likely a result
of the specificity of the two assays for different forms of folate combined with the effect of the
62
Table 2. Summary of studies that examined the relation between the MTHFR 677C→T polymorphism and folate status
MTHFR 677 C→T Genotype

Reference Population Description n CC CT TT
Erythrocyte Folate (nmol/L)
van der Put et al, 199549,65 NTD cases and their parents and controls 392 5411 5171 643
Molloy et al, 199766 Healthy, pregnant women 242 7861 7271 571
Healthy, non-pregnant women 318 7861 711 643
Verhoef et al, 199729 CAD cases and controls 318 No difference between genotypes; data not presented in paper.
Zittoun et al, 199870 Healthy adults 52 5351 522 353
van der Put et al, 199814 NTD cases and their parents and controls 186 5301 5261 630
Nelen et al, 199859 Women with recurrent miscarriages 49
Baseline 485 520 620
Folic acid supplementation: 2 months 775 920 975
Folic acid supplementation: 4 months No difference between genotypes; data not presented in paper.
Christensen et al, 199967 NTD cases 56 560 611 526
Mothers of NTD cases 59 575 489 520
Controls 95 675 749 836
Mothers of controls 90 708 700 612
Wiltshire et al, 200173 Diabetic children (type I) and controls 137 Folate higher in TT than the CC; data not presented in paper.
Fohr et al, 200257 Healthy women 160 397 405 353
McNulty et al, 200243 Healthy men and women 286 8021 7321 610
Kaye et al, 200268 Diabetic patients 746 7761 7631 1167
1
Friso et al, 200271 CAD cases and controls 292 440 290
Castro et al, 200369 Healthy subjects 117 6171 5441 638
Serum or Plasma Folate (nmol/L)
van der Put et al, 199549,65 NTD cases and their parents 392 12.81 12.81 9.5
Harmon et al, 199626 Healthy men 625 12.31, 2 10.81 9.3

Table 2. Continued

Ma et al, 199625 Male MI cases 293 7.9 7.5 5.8

Male controls 290 9.21 8.5 5.9
Molloy et al, 199766 Healthy, pregnant women 242 11.11 10.41 7.1
Healthy, non-pregnant women 318 12.9 12.6 13.2
Schwartz et al, 199728 Healthy women 338 16.41 16.41 11.1

Christensen et al, 199730 Male CAD cases 94 11.8 9.7 10.7
Male controls 86 9.7 8.4 7.7
Morita et al, 199832 Stroke patients 141 9.7 8.6 7.9
Nelen et al, 199859 Women with recurrent miscarriages 49
Baseline 131 121 7.3
Folic acid supplementation: 2 months 511 451 35
Folic acid supplementation: 4 months No difference between genotypes; data not presented in paper.
van der Put et al, 199814 NTD cases and their parents and controls 186 14.41 13.01 11.2
Girelli et al, 199834 CAD cases 278 11.7 11.3 10.3
Controls 137 No difference between genotypes; data not presented in paper.
Zittoun et al, 199870 Healthy adults 52 23.31 20.2 10.9
Verhoeff et al, 199847 Premature arterial disease cases 257 7.51 7.3 6.3
Mager et al, 199974 Early onset CAD cases 65 19.63 19.5 15.0
Late onset CAD cases 94 22.9 20.3 21.7
McQuillan et al, 199944 Population-based sample 1111 17.71 14.3
Yoo & Park, 200038 CAD cases 187 22.01 19.01 14.7
Controls 122 18.8 18.4 17.4
63
64
Table 2. Continued

Silastre et al, 200163 Healthy women: 37

Baseline 11.4 10.6 10.4
Low folate diet 11.2 10.7 11.0
High folate diet 19.9 19.3 17.1
Geisel J et al, 200141 Older adults 182 13.11 13.61 10.4
Ashfield-Watt et al, 200260 Healthy subjects
and Pullin et al, 200161 Baseline 126 20.63 17.2 15.2
Folic acid exclusion diet 110 19.03 17.7 14.8
Folate-rich diet 108 27.5 25.0 24.7
Folic acid supplements 108 36.8 30.0 27.9
Nakamura et al, 200248 Dialysis patients 450 16.51 15.21 13.8
Zychma et al, 200275 Dialysis patients 157 13.31 11.5 9.4
Fohr et al, 200257 Healthy women 160 16.41 16.11 12.1
McNulty et al, 200243 Healthy men and women 286 18.6 16.6 14.6
Kaye et al, 200268 Diabetic patients 746 21.5 20.41 22.9
Friso et al, 200271 CAD cases and controls 292 12.71 10.4
Moriyama et al, 200242 Healthy men 324 11.93 10.9 10.1
Healthy women 641 13.03 12.0 11.1
Castro et al, 200369 Healthy subjects 117 15.61 15.01 13.0
NTD= neural tube defect, CAD = coronary artery disease. 1 Significantly different from TT (P<0.05). 2 Significantly difference from CT (P<0.05). 3 Individual
differences not provided, but statistical test for overall differences between genotype categories was statistically significant (P<0.05).
MTHFR mutation on distribution of folate forms (see below 677C→T Mutation and Folate
Status – Mode of Action). The studies summarized in Table 2 support the observations of
Molloy and coworkers: studies using a microbiological assay generally report lower RBC folate
levels among those who are homozygous for this mutation43,66,70,71 while those using a radioassay
or other nonbiological methods generally report higher RBC folate levels in those with the TT
genotype.14,49,59,65,68,69,73
677C→T Mutation and Plasma or Serum Folate

Unlike the data based on RBC folate levels, the relationship between MTHFR 677 C→T
genotype and folate measured in plasma or serum is very consistent (Table 2). The majority of
the published studies report lower circulating folate concentrations among individuals with the
TT genotype compared with the CC genotype.14,25,26,28,41,42,44,47-49,57,59,65,66,69-71,74,75 Although
a small number of studies failed to note any significant association between genotype and
plasma or serum folate,30,32,34,38,43,63,68 no studies reported significantly higher levels among
the TT than CC individuals. (One study conducted in diabetic patients reported higher folate
levels in TT than CT patients, but the levels were comparable in the CC and TT individuals.68)
Table 2 also summarizes a number of studies that examined the effect of folate interventions
on folate status across genotypes. Nelen et al59 investigated the effect of supplementation with
folic acid in a dose of 0.5 mg/day for 4 months on folate concentrations in a population of
women who suffered unexplained recurrent miscarriages. Before the supplementation, TT in-
dividuals had lower serum folate when compared to the other two genotypes. After 2 months
of supplementation, TT individuals still had lower serum folate when compared to the other
two genotypes; however, there was no difference in serum folate levels among genotypes after 4
mo of supplementation. McDowell and coworkers60,61 administered 3 interventions of 4 months
of duration each in 126 healthy individuals (42 of each genotype): a folic acid exclusion diet
(usual diet, excluding foods fortified with folic acid providing approximately 200 µg total
folate/day), a folate-rich diet (providing approximately 400 µg/day from foods naturally rich
in folate and fortified with folic acid), and a folic acid supplement (exclusion diet plus a 400
µg/day folic acid supplement). Individuals with the TT genotype had the lowest levels of plasma
folate at baseline and on the folic acid exclusion diet, but this difference between genotypes was
no longer apparent for the folate-rich diet or the folic acid supplement interventions. Silaste et
al63 evaluated the effect of the MTHFR mutation on the response of serum folate to increased
levels of dietary folate in 37 healthy female volunteers (5 with the TT genotype). With the
increase in folate in the diet (from 200 to 600 µg/day without fortified foods), serum folate
increased in the three genotypes; however, there was no difference in this effect between geno-
types at either baseline or after the interventions, but this is possibly a consequence of the
study’s very small sample size.
677C→T Mutation and Folate Status – Mode of Action

Sufficient production of 5-methylTHF requires both adequate amounts of reduced folate
and proper function of the MTHFR enzyme. Severe mutations in the MTHFR enzyme result
in altered distribution of the different forms of folate in cultured fibroblasts with a lower pro-
portion of 5-methylTHF and a higher proportion of formylated folates when compared to
controls.76 This altered distribution of the different forms of folate is seen because impairment
of MTHFR activity results in less 5,10 methyleneTHF being reduced to 5-methylTHF allow-
ing greater 5,10 methyleneTHF availability for oxidation to formylated folate forms.
The 677C→T mutation results in an enzyme with lower specific activity in vitro, but it was
not known if mild MTHFR deficiency would decrease the production of 5-methylTHF in
vivo. Bagley and Selhub77 were the first to demonstrate that the mutant enzyme results in a
difference in the distribution of the different forms of folate in vivo. Using a chromatographic
method to measure the different forms of folate, these investigators reported that erythrocytes
of individuals homozygous for the 677C→T mutation contained variable amounts of formylated
THF polyglutamates (from 0 to 59%) in addition to methylated folate, while erythrocytes of

wild type individuals contained exclusively methylated folate derivatives with different
polyglutamate chain lengths. However, total RBC folate was similar in the wild type and mu-
tant genotypes. These observations by Bagley and coworkers are consistent with a variant of the
MTHFR enzyme that is less efficient in synthesizing 5-methylTHF than the wild type en-
zyme.
More recently, Friso and colleagues71 using the same methodology confirmed the previous
observation in 292 age- and sex-matched subjects from Northern Italy. The erythrocytes of
individuals who were homozygous for the mutation contained about 30% formylated THF
polyglutamates compared to wild type individuals whose erythrocyte folate was exclusively
5-methylTHF. Other studies reported similar findings. Quere et al78 reported a dose effect of
the T allele on content of methylfolates in erythrocytes in a sample of patients with deep
venous thrombosis. Zittoun et al70 reported that TT individuals had decreased levels of total
erythrocyte folate compared to CC, decreased levels of methylated folates in erythrocytes com-
pared to CC and CT, and a decreased ratio of methylfolate to total folate in TT erythrocytes
compared to CT and CC.
Since 5-methylTHF is the main form of folate in plasma or serum, a decreased activity in
MTHFR should result in a decreased level of plasma or serum folate. Moreover, the assay
method should not affect the results for plasma or serum folate in the same manner as it might
for RBC folate given the lack of other forms of folate.
MTHFR 1298A→C Mutation and Its Influence on tHcy and Folate

Concentrations
As with the 677C→T mutation, the 1298A→C mutation results in an enzyme that has
reduced activity.13,14,53 It is a common variant, with homozygosity for the 1298A→C muta-
tion occurring in approximately 10% of the Caucasians of Northern European ancestry. Un-
like the 677C→T mutation, the 1298A→C mutation alone does not appear to be associated
with higher plasma tHcy or lower plasma folate concentrations. However, combined heterozy-
gosity for the 677C→T and the 1298A→C variants results in lower enzyme activity, higher
tHcy levels, and possibly reduced folate levels when compared with the levels observed in
heterozygotes for either variant alone.14,69,79-81
Summary
There is a complex relationship between MTHFR 677C→T genotype and folate status.
Not only is there evidence that this mutation reduces circulating folate levels, but folate status
also affects the expression of the mutation as demonstrated by an enhanced elevation of ho-
mocysteine concentrations among individuals with the TT genotype relative to the other geno-
types when folate levels are inadequate. The latter has the obvious potential to increase the risk
of conditions associated with hyperhomocysteinemia, such as cardiovascular disease. However,
the potential consequences of the mutation among persons with adequate folate status are not
as clear. In the presence of apparently adequate folate levels, there appears to be sufficient
5-methylTHF produced for methionine, and consequently S-adenosylmethionine, metabo-
lism as evidenced by normal homocysteine concentrations. With adequate folate for the
remethylation reactions, the mutation would appear to make the surplus folate available for
other folate-dependent reactions. What we have learned about the 677C→T mutation might
also guide the investigation of the impact of the 1298A→C mutation on folate status. How-
ever, much remains to be learned about the influence of mild MTHFR deficiency on the
metabolic fate of folate and folate requirements.
References
1. Selhub J. Homocysteine metabolism. Annu Rev Nutr 1999; 19:217-246.
2. Rozen R. Genetic modulation of homocysteinemia. Semin Thromb Hemost 2000; 26(3):255-261.
AL, Shy WS et al, eds. New York: McGraw-Hill, 1995:3111-3128.
4. Rosenblatt DS. Methylenetetrahydrofolate reductase. Clin Invest Med 2001; 24(1):56-59.
5. Rosenblatt DS, Whitehead VM. Cobalamin and folate deficiency: Acquired and hereditary disor-
ders in children. Semin Hematol 1999; 36(1):19-34.
6. Ogier de Baulny H, Gerard M, Saudubray JM et al. Remethylation defects: Guidelines for clinical
diagnosis and treatment. Eur J Pediatr 1998; 157 Suppl 2:S77-83.
7. Visy JM, Le Coz P, Chadefaux B et al. Homocystinuria due to 5,10-methylenetetrahydrofolate
reductase deficiency revealed by stroke in adult siblings. Neurology 1991; 41(8):1313-1315.
methylenetetrahydrofolate reductase. Am J Hum Genet 1988; 43(4):414-421.
cDNA, mapping and mutation identification. Nat Genet 1994; 7(2):195-200.
mutation in methylenetetrahydrofolate reductase. Nat Genet 1995; 10(1):111-113.
11. Schneider JA, Rees DC, Liu YT et al. Worldwide distribution of a common methylenetetrahydrofolate
reductase mutation. Am J Hum Genet 1998; 62(5):1258-1260.
12. Botto LD, Yang Q. 5,10-Methylenetetrahydrofolate reductase gene variants and congenital anoma-
lies: A HuGE review. Am J Epidemiol 2000; 151(9):862-877.
reductase (MTHFR) associated with decreased enzyme activity. Mol Genet Metab 1998;
64(3):169-172.
Hum Genet 1998; 62(5):1044-1051.
15. Stabler SP, Marcell PD, Podell ER et al. Elevation of total homocysteine in the serum of patients
with cobalamin or folate deficiency detected by capillary gas chromatography-mass spectrometry. J
Clin Invest 1988; 81(2):466-474.
16. Miller JW, Nadeau MR, Smith J et al. Folate-deficiency-induced homocysteinaemia in rats: Dis-
ruption of S- adenosylmethionine’s coordinate regulation of homocysteine metabolism. Biochem J
1994; 298(Pt 2):415-419.
17. Ubbink JB. The role of vitamins in the pathogenesis and treatment of hyperhomocyst(e)inaemia. J
Inherit Metab Dis 1997; 20(2):316-325.
18. Ward M. Homocysteine, folate, and cardiovascular disease. Int J Vitam Nutr Res May 2001;
71(3):173-178.
19. Brattstrom L, Wilcken DE, Ohrvik J et al. Common methylenetetrahydrofolate reductase gene
mutation leads to hyperhomocysteinemia but not to vascular disease: The result of a meta- analy-
sis. Circulation 1998; 98(23):2520-2526.
20. Ueland PM, Hustad S, Schneede J et al. Biological and clinical implications of the MTHFR C677T
polymorphism. Trends Pharmacol Sci 2001; 22(4):195-201.
21. Kang SS, Wong PW, Zhou JM et al. Thermolabile methylenetetrahydrofolate reductase in patients
with coronary artery disease. Metabolism 1988; 37(7):611-613.
22. Kang SS, Wong PW, Bock HG et al. Intermediate hyperhomocysteinemia resulting from com-
pound heterozygosity of methylenetetrahydrofolate reductase mutations. Am J Hum Genet 1991;
48(3):546-551.
23. Kang SS, Wong PW, Susmano A et al. Thermolabile methylenetetrahydrofolate reductase: An in-
herited risk factor for coronary artery disease. Am J Hum Genet 1991; 48(3):536-545.
93(1):7-9.
25. Ma J, Stampfer MJ, Hennekens CH et al. Methylenetetrahydrofolate reductase polymorphism, plasma
folate, homocysteine, and risk of myocardial infarction in US physicians. Circulation 1996;
94(10):2410-2416.
26. Harmon DL, Woodside JV, Yarnell JW et al. The common ‘thermolabile’ variant of methylene
tetrahydrofolate reductase is a major determinant of mild hyperhomocysteinaemia. QJM 1996;
89(8):571-577.
27. Bostom AG, Shemin D, Lapane KL et al. Folate status is the major determinant of fasting total
plasma homocysteine levels in maintenance dialysis patients. Atherosclerosis 1996; 123(1-2):193-202.
28. Schwartz SM, Siscovick DS, Malinow MR et al. Myocardial infarction in young women in relation
to plasma total homocysteine, folate, and a common variant in the methylenetetrahydrofolate re-
ductase gene. Circulation 1997; 96(2):412-417.
29. Verhoef P, Kok FJ, Kluijtmans LA et al. The 677C-->T mutation in the methylenetetrahydrofolate
reductase gene: Associations with plasma total homocysteine levels and risk of coronary atheroscle-
rotic disease. Atherosclerosis 1997; 132(1):105-113.
30. Christensen B, Frosst P, Lussier-Cacan S et al. Correlation of a common mutation in the
coronary artery disease. Arterioscler Thromb Vasc Biol 1997; 17(3):569-573.
31. Dunn J, Title LM, Bata I et al. Relation of a common mutation in methylenetetrahydrofolate
reductase to plasma homocysteine and early onset coronary artery disease. Clin Biochem 1998;
31(2):95-100.
32. Morita H, Kurihara H, Tsubaki S et al. Methylenetetrahydrofolate reductase gene polymorphism
and ischemic stroke in Japanese. Arterioscler Thromb Vasc Biol 1998; 18(9):1465-1469.
33. Motti C, Gnasso A, Bernardini S et al. Common mutation in methylenetetrahydrofolate reductase.
Correlation with homocysteine and other risk factors for vascular disease. Atherosclerosis 1998;
139(2):377-383.
34. Girelli D, Friso S, Trabetti E et al. Methylenetetrahydrofolate reductase C677T mutation, plasma
homocysteine, and folate in subjects from northern Italy with or without angiographically docu-
mented severe coronary atherosclerotic disease: Evidence for an important genetic-environmental
interaction. Blood 1998; 91(11):4158-4163.
35. Thuillier L, Chadefaux-Vekemans B, Bonnefont JP et al. Does the polymorphism 677C-T of the
5,10-methylenetetrahydrofolate reductase gene contribute to homocysteine-related vascular disease?
J Inherit Metab Dis 1998; 21(8):812-822.
36. Tokgozoglu SL, Alikasifoglu M, Unsal et al. Methylene tetrahydrofolate reductase genotype and
the risk and extent of coronary artery disease in a population with low plasma folate. Heart 1999;
81(5):518-522.
37. Gemmati D, Previati M, Serino ML et al. Low folate levels and thermolabile methylenetetrahydrofolate
reductase as primary determinant of mild hyperhomocystinemia in normal and thromboembolic
subjects. Arterioscler Thromb Vasc Biol 1999; 19(7):1761-1767.
38. Yoo JH, Park SC. Low plasma folate in combination with the 677 C-->T methylenetetrahydrofolate
reductase polymorphism is associated with increased risk of coronary artery disease in Koreans.
Thromb Res 2000; 97(2):77-84.
39. Potena L, Grigioni F, Viggiani M et al. Interplay between methylenetetrahydrofolate reductase
gene polymorphism 677C-->T and serum folate levels in determining hyperhomocysteinemia in
heart transplant recipients. J Heart Lung Transplant 2001; 20(12):1245-1251.
40. Saw SM, Yuan JM, Ong CN et al. Genetic, dietary, and other lifestyle determinants of plasma
homocysteine concentrations in middle-aged and older Chinese men and women in Singapore. Am
J Clin Nutr 2001; 73(2):232-239.
41. Geisel J, Zimbelmann I, Schorr H et al. Genetic defects as important factors for moderate
hyperhomocysteinemia. Clin Chem Lab Med 2001; 39(8):698-704.
42. Moriyama Y, Okamura T, Kajinami K et al. Effects of serum B vitamins on elevated plasma ho-
mocysteine levels associated with the mutation of methylenetetrahydrofolate reductase gene in Japa-
nese. Atherosclerosis 2002; 164(2):321-328.
43. McNulty H, McKinley MC, Wilson B et al. Impaired functioning of thermolabile methylene-
tetrahydrofolate reductase is dependent on riboflavin status: implications for riboflavin require-
ments. Am J Clin Nutr 2002; 76(2):436-441.
44. McQuillan BM, Beilby JP, Nidorf M et al. Hyperhomocysteinemia but not the C677T mutation
of methylenetetrahydrofolate reductase is an independent risk determinant of carotid wall thicken-
ing. The Perth Carotid Ultrasound Disease Assessment Study (CUDAS). Circulation 1999;
99(18):2383-2388.
45. de Jong SC, Stehouwer CD, van den Berg M et al. Determinants of fasting and post-methionine
homocysteine levels in families predisposed to hyperhomocysteinemia and premature vascular dis-
ease. Arterioscler Thromb Vasc Biol 1999; 19(5):1316-1324.
46. D’Angelo A, Coppola A, Madonna P et al. The role of vitamin B12 in fasting hyperhomocysteinemia
and its interaction with the homozygous C677T mutation of the methylenetetrahydrofolate reduc-
tase (MTHFR) gene. A case-control study of patients with early-onset thrombotic events. Thromb
Haemost 2000; 83(4):563-570.
47. Verhoeff BJ, Trip MD, Prins MH et al. The effect of a common methylenetetrahydrofolate reduc-
tase mutation on levels of homocysteine, folate, vitamin B12 and on the risk of premature athero-
sclerosis. Atherosclerosis 1998; 141(1):161-166.
48. Nakamura T, Saionji K, Hiejima Y et al. Methylenetetrahydrofolate reductase genotype, vitamin
B12, and folate influence plasma homocysteine in hemodialysis patients. Am J Kidney Dis 2002;
39(5):1032-1039.
49. van der Put NM, van den Heuvel LP, Steegers-Theunissen RP et al. Decreased methylene
tetrahydrofolate reductase activity due to the 677C-->T mutation in families with spina bifida
offspring. J Mol Med 1996; 74(11):691-694.
50. Guttormsen AB, Ueland PM, Nesthus I et al. Determinants and vitamin responsiveness of inter-
mediate hyperhomocysteinemia (> or = 40 micromol/liter). The Hordaland Homocysteine Study. J
Clin Invest 1996; 98(9):2174-2183.
51. Matthews RG. Methylenetetrahydrofolate reductase: A common human polymorphism and its bio-
chemical implications. Chem Rec 2002; 2(1):4-12.
Struct Biol 1999; 6(4):359-365.
53. Yamada K, Chen Z, Rozen R et al. Effects of common polymorphisms on the properties of re-
combinant human methylenetetrahydrofolate reductase. Proc Natl Acad Sci USA 2001;
98(26):14853-14858.
transition in the methylenetetrahydrofolate reductase gene. J Nutr 2002; 132(2):283-288.
55. Hustad S, Ueland PM, Vollset SE et al. Riboflavin as a determinant of plasma total homocysteine:
2000; 46(8 Pt 1):1065-1071.
56. Lowering blood homocysteine with folic acid based supplements: Meta-analysis of randomised tri-
als. Homocysteine Lowering Trialists’ Collaboration. BMJ 1998; 316(7135):894-898.
57. Fohr IP, Prinz-Langenohl R, Bronstrup A et al. 5,10-Methylenetetrahydrofolate reductase genotype
determines the plasma homocysteine-lowering effect of supplementation with 5- methyltetrahydrofolate
or folic acid in healthy young women. Am J Clin Nutr 2002; 75(2):275-282.
58. Malinow MR, Nieto FJ, Kruger WD et al. The effects of folic acid supplementation on plasma
total homocysteine are modulated by multivitamin use and methylenetetrahydrofolate reductase
genotypes. Arterioscler Thromb Vasc Biol 1997; 17(6):1157-1162.
59. Nelen WL, Blom HJ, Thomas CM et al. Methylenetetrahydrofolate reductase polymorphism af-
fects the change in homocysteine and folate concentrations resulting from low dose folic acid supple-
mentation in women with unexplained recurrent miscarriages. J Nutr 1998; 128(8):1336-1341.
60. Ashfield-Watt PA, Pullin CH, Whiting JM et al. Methylenetetrahydrofolate reductase 677C-->T
genotype modulates homocysteine responses to a folate-rich diet or a low-dose folic acid supple-
ment: A randomized controlled trial. Am J Clin Nutr 2002; 76(1):180-186.
61. Pullin CH, Ashfield-Watt PA, Burr ML et al. Optimization of dietary folate or low-dose folic acid
supplements lower homocysteine but do not enhance endothelial function in healthy adults, irre-
spective of the methylenetetrahydrofolate reductase (C677T) genotype. J Am Coll Cardiols 2001;
38(7):1799-1805.
62. Kauwell GP, Wilsky CE, Cerda JJ et al. Methylenetetrahydrofolate reductase mutation (677C-->T)
negatively influences plasma homocysteine response to marginal folate intake in elderly women.
Metabolism 2000; 49(11):1440-1443.
63. Silaste ML, Rantala M, Sampi M et al. Polymorphisms of key enzymes in homocysteine metabo-
lism affect diet responsiveness of plasma homocysteine in healthy women. J Nutr 2001;
131(10):2643-2647.
64. Woodside JV, Yarnell JW, McMaster D et al. Effect of B-group vitamins and antioxidant vitamins
on hyperhomocysteinemia: A double-blind, randomized, factorial-design, controlled trial. Am J Clin
Nutr 1998; 67(5):858-866.
65. van der Put NM, Steegers-Theunissen RP, Frosst P et al. Mutated methylenetetrahydrofolate re-
ductase as a risk factor for spina bifida. Lancet 1995; 346(8982):1070-1071.
66. Molloy AM, Daly S, Mills JL et al. Thermolabile variant of 5,10-methylenetetrahydrofolate reduc-
tase associated with low red-cell folates: Implications for folate intake recommendations. Lancet
1997; 349(9065):1591-1593.
67. Christensen B, Arbour L, Tran P et al. Genetic polymorphisms in methylenetetrahydrofolate re-
ductase and methionine synthase, folate levels in red blood cells, and risk of neural tube defects.
Am J Med Genet 1999; 84(2):151-157.
68. Kaye JM, Stanton KG, McCann VJ et al. Homocysteine, folate, methylene tetrahydrofolate reduc-
tase genotype and vascular morbidity in diabetic subjects. Clin Sci (Lond) 2002; 102(6):631-637.
69. Castro R, Rivera I, Ravasco P et al. 5,10-Methylenetetrahydrofolate reductase 677C-->T and
1298A-->C mutations are genetic determinants of elevated homocysteine. QJM 2003; 96(4):297-303.
70. Zittoun J, Tonetti C, Bories D et al. Plasma homocysteine levels related to interactions between
folate status and methylenetetrahydrofolate reductase: A study in 52 healthy subjects. Metabolism
1998; 47(11):1413-1418.
71. Friso S, Choi SW, Girelli D et al. A common mutation in the 5,10-methylenetetrahydrofolate
Natl Acad Sci USA 2002; 99(8):5606-5611.
72. Molloy AM, Mills JL, Kirke PN et al. Whole-blood folate values in subjects with different
methylenetetrahydrofolate reductase genotypes: Differences between the radioassay and microbio-
logical assays. Clin Chem 1998; 44(1):186-188.
73. Wiltshire E, Thomas DW, Baghurst P et al. Reduced total plasma homocyst(e)ine in children and
adolescents with type 1 diabetes. J Pediatr 2001; 138(6):888-893.
74. Mager A, Lalezari S, Shohat T et al. Methylenetetrahydrofolate reductase genotypes and early-onset
coronary artery disease. Circulation 1999; 100(24):2406-2410.
75. Zychma MJ, Gumprecht J, Grzeszczak W et al. Methylenetetrahydrofolate reductase gene C677T
polymorphism, plasma homocysteine and folate in end-stage renal disease dialysis and non dialysis
patients. Nephron 2002; 92(1):235-239.
76. Rosenblatt DS, Cooper BA, Lue-Shing S et al. Folate distribution in cultured human cells. Studies
on 5,10-CH2-H4PteGlu reductase deficiency. J Clin Invest 1979; 63(5):1019-1025.
USA 1998; 95(22):13217-13220.
78. Quere I, Wutschert R, Zittoun J et al. Association of red-blood methylfolate but not plasma folate
with C677T MTHFR polymorphism in venous thromboembolic disease. Thromb Haemost 1998;
80(4):707-709.
79. Lievers KJ, Boers GH, Verhoef P et al. A second common variant in The methylenetetrahydrofolate
reductase (MTHFR) gene and its relationship to MTHFR enzyme activity, homocysteine, and car-
diovascular disease risk. J Mol Med 2001; 79(9):522-528.
80. Hanson NQ, Aras O, Yang F et al. C677T and A1298C polymorphisms of the methylene-
tetrahydrofolate reductase gene: Incidence and effect of combined genotypes on plasma fasting and
post-methionine load homocysteine in vascular disease. Clin Chem 2001; 47(4):661-666.
156(2):409-415.
Riboflavin and Methylenetetrahydrofolate Reductase 71
CHAPTER 6
Riboflavin and Methylenetetrahydrofolate

Reductase
Steinar Hustad, Jørn Schneede and Per Magne Ueland
Abstract
T
he flavoenzyme methylenetetrahydrofolate reductase (MTHFR) catalyzes the conversion
of 5, 10-methylenetetrahydrofolate to 5-methyltetrahydrofolate, which serves as
a methyl group donor in the conversion of homocysteine to methionine. In rats,
experimental riboflavin deficiency is associated with low MTHFR activity and reduced levels
of 5-methyltetrahydrofolate. In humans, reduced enzyme activity caused by the commonly
occurring 677C→T substitution of the MTHFR gene is associated with elevated plasma ho-
mocysteine. The mutant enzyme has lower affinity for its flavin cofactor than the wild-type
enzyme, and recent studies show that plasma homocysteine is inversely related to riboflavin in
subjects with the T-allele. This indicates that the metabolic effect of the 677C→T polymor-
phism is related to riboflavin status, which may have implications for future studies on the
relationship between this polymorphism and various clinical and biochemical endpoints.
Introduction
Riboflavin is a water-soluble vitamin, which serves as the precursor of flavin mononucle-
otide (FMN) and flavin adenine dinucleotide (FAD).1 FMN is formed by the phosphorylation
of riboflavin, and FAD is formed in a subsequent ATP-dependent reaction as most of the
FMN is adenylated.1 FMN and FAD are cofactors for more than 150 reduction-oxidation
enzymes, some of which are involved in the metabolism of folate, vitamin B6 and cobalamin 1,2
(Fig. 1).
The majority of flavoenzymes, including methylenetetrahydrofolate reductase (MTHFR),
are FAD-dependent.1,3 Mammalian MTHFR is a cytosolic homodimer, and each subunit con-
tains a catalytic N-terminal domain as well as a regulatory C-terminal domain,4 which binds
the allosteric inhibitor S-adenosylmethionine (AdoMet).3,4 The enzyme uses NADPH as a
cofactor in addition to FAD, and catalyzes the transformation of 5, 10–methylenetetrahydrofolate
to 5-methyltetrahydrofolate (5-methylTHF). The reaction is irreversible in vivo and is the only
source of 5-methylTHF, which serves as the methyl donor for the cobalamin-dependent con-
version of homocysteine to methionine.5 In most tissues, this provides the sole pathway for
homocysteine remethylation. Concentrations of homocysteine in tissues and blood also de-
pend upon its degradation through vitamin B6-dependent transsulfuration in the liver and
kidneys.5
Thermolabile MTHFR associated with lower catalytic activity was reported approximately
15 years ago.6 Later, this thermolability was shown to be caused by a 677C→T transition in
the MTHFR gene, resulting in an alanine to valine substitution in the enzyme.7 MTHFR in
lymphocytes from subjects with the TT genotype has approximately 30% of the catalytic
©2005 Eurekah.com.
Figure 1. Riboflavin is linked to the metabolism of B-vitamins involved in homocysteine remethylation and
transsulfuration. Riboflavin is transformed to FMN and FAD by the action of riboflavin kinase (EC
2.7.1.26) and FAD synthase (EC 2.7.7.2). FMN and FAD are cofactors for vitamin B6, cobalamin and
folate metabolizing enzymes. Vitamin B6 and cobalamin serve as cofactors for enzymes involved in ho-
mocysteine metabolism, while folate serves as a one-carbon donor for the remethylation of homocysteine.
Abbreviations: Cbl(I)= cob(I)alamin; Cbl(II)= cob(II)alamin; CBS= cystathionine β-synthase; CH2-THF=
5, 10-methylenetetrahydrofolate; CH3-THF= 5-methyltetrahydrofolate; CL= cystathionine lyase; Cys=
cysteine; FS= FAD synthase; Met= methionine; MS= methionine synthase; MSR= methionine synthase
reductase; MTHFR= 5, 10-methylenetetrahydrofolate reductase; PLP= pyridoxal-5´-phosphate; PMP=
pyridoxamine-5´-phosphate; PNP= pyridoxine-5´-phosphate; PPO= pyridoxine-5´-phosphate oxidase;
RK= riboflavin kinase; THF= tetrahydrofolate.
activity of the wild-type, while the CT genotype has 65% of the catalytic activity.7 The fre-
quency of the T-allele is about 0.3-0.4 in several populations of Asian and European descent,
but is much lower in sub-Saharan Africa and in some other areas.8
Guenther et al studied the biochemical properties of MTHFR in an E. coli model.9 They
found that the enzyme variant homologous to the human Ala222Val substitution had lower
binding affinity for FAD than the wild-type, and that folate stabilized the enzyme and in-
creased FAD affinity.9 Purified human MTHFR was later shown to have similar biochemical
properties.4 Folate and AdoMet stabilized the enzyme, particularly the mutant, and protected
against flavin loss. The stabilizing effect of FAD was studied with respect to loss of enzyme
activity after heating and after enzyme dilution, and in both cases the flavin cofactor was found
to have a protective effect.4
Animal Studies
In 1968, Narisawa et al investigated the relationship between riboflavin status and MTHFR
activity, and found lower enzyme activity in livers of riboflavin-deficient rats than in controls.10
Bates and Fuller confirmed these findings by demonstrating a dose-dependent relationship
between riboflavin status and MTHFR activity.11 They also reported that MTHFR was par-
ticularly sensitive to riboflavin deficiency, and that enzyme activity decreased relatively more
than intracellular concentrations of FAD when riboflavin was scarce. Low MTHFR activity in
riboflavin deficiency is associated with altered distribution of liver folates, and 5-methylTHF is
reduced relative to other folate forms.10-12 Redistribution of folates may explain why the
urinary excretion of formiminoglutamic acid after histidine loading is reduced in
riboflavin-deficient rats compared to controls,10,12 and possibly why homocysteine is elevated
in the skin of riboflavin-deficient rats.13
MTHFR activity has also been investigated in hypothyroid rats.14-16 Hypothyroidism is
associated with low intracellular FMN and FAD, which has been explained by low activity of
riboflavin kinase, the enzyme catalysing the conversion of riboflavin to FMN17,18 (Fig. 1). As
with alimentary riboflavin deficiency, MTHFR activity is low,15 and intracellular levels of
5-methylTHF are reduced relative to other folates.14-16 Histidine oxidation is increased,14-16
both at high and low intakes of methionine.14,15 Hypothyroidism is also associated with higher
levels of AdoMet, which may inhibit MTHFR activity.16
Human Studies
Assessment of Riboflavin Status
Glutathione reductase is a FAD-dependent enzyme, which is sensitive to riboflavin defi-
ciency.19 This probably explains why the erythrocyte glutathione reductase activation coeffi-
cient (EGRAC), which is the ratio between in vitro enzyme activity determined with and
without the addition of FAD, is useful for the assessment of riboflavin status.20,21 High levels
of EGRAC indicate low FAD saturation of the apoenzyme and biochemical riboflavin deficiency.
The method has been used as an indicator of riboflavin status in several clinical studies.21
In some human studies, concentrations of plasma riboflavin have been used for the assess-
ment of riboflavin status.22,23 This parameter appears to be a more sensitive indicator of ribo-
flavin status than plasma concentrations of flavin cofactors.24 In severe riboflavin deficiency,
plasma concentrations of FAD are probably lowered as well.24,25
Blood Levels of tHcy and Folate in MTHFR Deficiency

The role of MTHFR in homocysteine remethylation is illustrated by the finding that both
mild MTHFR deficiency, which is observed in subjects with the 677C→T transition, and
severe MTHFR deficiency are associated with elevated concentrations of plasma tHcy.7,8 High
plasma tHcy in subjects with the TT genotype is usually observed only under conditions of
impaired folate status,26 which probably reflects the role of folate as a substrate and as a
genotype-dependent regulator of enzyme stability. Moreover, the TT genotype is frequently
associated with low levels of plasma folate.27 5-methylTHF is the predominant folate species in
plasma,27 and low folate is probably related to impaired synthesis of 5-methylTHF. Published
data on erythrocyte folate may appear contradictory, and increased27 or decreased28 concentra-
tions have been reported in subjects with the TT genotype compared to the CT and CC
genotypes. Such apparent inconsistencies may be method dependent and reflect the ability of
different assays to detect various cellular folate species.29 This idea is supported by the finding
of formylated tetrahydrofolates in erythrocytes from subjects with the TT genotype, whereas
CC cells contain only 5-methylTHF.30
Riboflavin Intake and Plasma tHcy

The relationship between dietary riboflavin and plasma concentrations of tHcy has been
studied in a few cross-sectional studies.31-33 In individuals from the Framingham Offspring
cohort (n = 1960), plasma tHcy was 1.0 µmol/l higher in the lowest compared to the highest
quintile of riboflavin intake in a multivariate model adjusted for intakes of folate, cobalamin,
vitamin B6 and other possible determinants of tHcy.31 Samples were collected prior to the
implementation of mandatory folic acid fortification in 1998, and individuals who were regu-
lar users of B-vitamin supplements were excluded from the analysis.31 A strong inverse rela-
tionship between riboflavin intake and plasma tHcy was reported in another American study,
but the data were not adjusted for folate or other B-vitamins.32 Dietary intakes of folate, ribo-
flavin, vitamin B6 and cobalamin were inversely related to plasma tHcy in 2435 men and
Figure 2. Dose-response curves for the relationship between plasma riboflavin and plasma homocysteine
according to MTHFR 677C→T genotype. Solid lines indicate estimated dose-response curves, and shaded
areas represent 95% confidence intervals. The curves are obtained by additive Gaussian regression analysis,
and the model is adjusted for sex, age, serum folate, serum cobalamin, and serum creatinine. The figure is
modified from reference 22 with permission.
women from a Dutch population-based cohort,33 but only folate remained associated with
tHcy in a multivariate model, adjusted for B-vitamins and other determinants of plasma tHcy.
Individuals who used B-vitamin supplements were excluded from the study. None of the above
studies investigated the possible effect of MTHFR 677C→T genotype on the riboflavin-tHcy
relationship.31,32
Riboflavin and the MTHFR 677C→T Polymorphism As Determinants

of tHcy
An inverse relationship between plasma concentrations of riboflavin and tHcy was reported
by Hustad et al in a study of 423 Norwegian blood donors22 (Fig. 2). Plasma tHcy was 1.4
µmol/l higher in the lowest compared to the highest riboflavin quartile in a multiple regression
model adjusted for folate and other determinants of tHcy. The riboflavin-tHcy relationship
was modified by the MTHFR 677C→T polymorphism and was essentially confined to sub-
jects with the T allele (Fig. 2). The riboflavin-tHcy relationship was not significantly modified
by levels of serum folate.
Jacques et al studied 450 subjects from the Framingham Offspring cohort, selected accord-
ing to the MTHFR 677C→T polymorphism and equally distributed between the CC, CT and
TT genotypes.23 They found an inverse association between plasma concentrations of ribofla-
vin and plasma tHcy, but only in subjects with the TT genotype and plasma folate below the
median (12.5 nmol/l). In this group, tHcy was 2.9 µmol/l higher in the lowest compared to the
highest riboflavin tertile after adjustment for sex, age and folate. There was no relationship
between concentrations of flavin cofactors and tHcy.23
Thus, both the American and the Norwegian studies22,23 demonstrated an association be-
tween riboflavin and tHcy, which was dependent on the MTHFR 677C→T polymorphism. In
the Framingham Offspring cohort, the riboflavin-tHcy relationship was weaker, however. In
addition, plasma tHcy was less strongly associated with the MTHFR genotype, and concentra-
tions of serum folate were independent of genotype.22,23 Thus, the polymorphism apparently
had less metabolic effects in the American than in the Norwegian study.
This disparity might be explained by different intakes of vitamins or nutrients, which modu-
late MTHFR activity. Riboflavin intake was probably higher in the Americans, because grain
products have been riboflavin fortified in the USA since the 1940s23,34 and levels of fortifica-
tion have been rising over the years.34 Although samples were collected prior to the implemen-
tation of mandatory folic acid fortification in the USA, many breakfast cereals were fortified at
the time of the study,23,34 and folate status may have been better in the Americans. The idea
that vitamin intake differs between the two populations is further supported by the finding of
an overall correlation between concentrations of riboflavin and folate in the American study
(Spearman correlation coefficient = 0.31, P <0.001),23 but not in the Norwegian study.22 Such
a correlation might be related to common dietary sources for these vitamins.
In a recent study by McNulty et al of 286 healthy individuals from Northern Ireland,
EGRAC was used to determine riboflavin status.35 The authors demonstrated a significant
inverse association between riboflavin status and plasma concentrations of tHcy. As in previous
studies, this relationship was modified by MTHFR genotype, and in the lowest tertile of ribo-
flavin status, mean tHcy was 18.09 µmol/l in the TT group as compared to 10.15 µmol/l in
the CT and 8.32 µmol/l in the CC groups. When riboflavin status was higher, tHcy was
independent of genotype. The genotype-tHcy relationship was also dependent on folate, and
was only observed in subjects with concentrations of erythrocyte folate below the median.35
The authors did not assess the combined effects of riboflavin, folate and MTHFR genotype on
tHcy in a multivariate model, but they found no correlation between riboflavin and erythro-
cyte folate,35 which makes serious confounding from folate unlikely.
The relationship between plasma tHcy and riboflavin status has also been investigated in
end-stage renal disease.36 In a study of 54 nonvitamin supplemented patients with a mean age
of 54 years who were maintained on peritoneal dialysis, mean plasma tHcy was 33.0 µmol/l.
Ten patients had EGRAC equal to or greater than 1.52, indicating riboflavin deficiency.36 The
authors found a positive association between EGRAC and tHcy, which is consistent with high
tHcy when riboflavin status is low. Riboflavin was significantly related to plasma tHcy in mul-
tivariate models, which included folate and other B-vitamins. The riboflavin-tHcy relationship
was not studied in relation to the MTHFR 677C→T polymorphism, because of the relatively
low number of patients.36
Riboflavin Intervention Studies

The homocysteine lowering effect of 15 days of riboflavin (10 mg/d; n = 10) or vitamin B6
(20 mg/d; n = 10) supplementation was investigated in a small study of riboflavin and vitamin
B6-deficient Indian women.37 Mean tHcy was 12.1 and 14.7 µmol/l in the riboflavin and B6
groups, respectively, and tHcy decreased only in the B6 group.37 There was no placebo group,
and no data on MTHFR 677C→T genotype.
Another riboflavin intervention study was published by McKinley et al.38 In the first phase
of this study, 46 subjects with suboptimal riboflavin status (EGRAC ≥ 1.20) received low-dose
riboflavin (1.6 mg/d; n = 23) or placebo (n = 23) for 12 weeks. In the second phase of the study,
participants originally on placebo received 400 µg/d of folic acid for 6 weeks followed by a
combination of folic acid and riboflavin for 12 weeks. Folic acid supplementation lowered
tHcy, but no effect of riboflavin was observed in either phase, and apparently riboflavin status
was not a determinant of plasma tHcy in this study. A possible reason is that only five subjects
had the TT genotype. In addition, riboflavin status was only modestly impaired, and no sub-
ject had EGRAC higher than 1.40.38
Implications
MTHFR may be sensitive to riboflavin status,11 particularly in subjects with the 677C→T
substitution of the MTHFR gene. In subjects with the TT genotype, higher riboflavin intake
could be necessary for the formation of adequate amounts of 5-methyl-THF involved in ho-
mocysteine remethylation. Although the TT genotype comprises only around 10% of the popu-
lation in many countries,8 it is more prevalent in subjects with hyperhomocysteinemia. In a
Norwegian population-based study of men and women aged 40-67 years, 73% of individuals
with plasma tHcy equal to or higher than 40 µmol/l had the TT genotype, compared to 10%
of the controls.39 In men from Northern Ireland, the TT genotype occurred in 48, 35, and
23% of the top 5, 10, and 20% of individuals ranked by plasma tHcy levels.40 This indicates
that the MTHFR 677C→T polymorphism may be important for the development of
moderate hyperhomocysteinemia, and an effect of the enzyme variant associated with the TT
genotype might be partly attributed to riboflavin.
Much of the interest in the MTHFR 677C→T polymorphism stems from its association
with moderate hyperhomocysteinemia, which is a risk factor for occlusive arterial disease, venous
thrombosis, neural tube defects and pregnancy complications.5,8 It is still not clear whether
elevated tHcy is the cause of these conditions or if it is mainly a surrogate marker. In several
studies, the MTHFR 677C→T polymorphism itself has been shown to modify the risk of
certain diseases,8 including cardiovascular disease,41,42 neural tube defects27 and colon can-
cer,43,44 while other studies show no such relationship.8 Inconsistent reports on the MTHFR
677C→T polymorphism and disease risk could be explained through effect modification by
nutritional factors. This has been demonstrated for folate,42,44 whereas the role of riboflavin
has received less attention.45
Further research is warranted to investigate the importance of riboflavin as a regulator of
MTHFR activity, particularly with respect to its interaction with folate and its relationship to
various clinical endpoints.
References
1. Rivlin RS, Pinto JT. Riboflavin (Vitamin B2). In: Rucker RB, Suttie JW, McCormick DB et al,
eds. Handbook of Vitamins. 3rd ed. New York: Marcel Dekker, Inc 2001:255-273.
2. Sauberlich HE. Interactions of thiamin, riboflavin, and other B-vitamins. Ann NY Acad Sci 1980;
355:80-97.
3. Kutzbach C, Stokstad EL. Mammalian methylenetetrahydrofolate reductase. Partial purification,
properties, and inhibition by S-adenosylmethionine. Biochim Biophys Acta 1971; 250:459-477.
4. Yamada K, Chen ZT, Rozen R et al. Effects of common polymorphisms on the properties of
recombinant human methylenetetrahydrofolate reductase. Proc Nat Acad Sci USA 2001;
98:14853-14858.
5. Selhub J. Homocysteine metabolism. Annu Rev Nutr 1999; 19:217-246.
mutation at the methylenetetrahydrofolate reductase. Nat Genet 1995; 10:111-113.
polymorphism. Trends Pharmacol Sci 2001; 22:195-201.
10. Narisawa K, Tamura T, Tanno K et al. Tetrahydrofolate-dependent enzyme activities of the rat
liver in riboflavin deficiency. Tohoku J Exp Med 1968; 94:417-430.
11. Bates CJ, Fuller NJ. The effect of riboflavin deficiency on methylenetetrahydrofolate reductase
(NADPH) (EC 1.5.1.20) and folate metabolism in the rat. Br J Nutr 1986; 55:455-464.
12. Honda Y. Folate derivatives in the liver of riboflavin-deficient rats. Tohuko J Exp Med 1968;
95:79-86.
13. Lakshmi R, Lakshmi AV, Bamji MS. Mechanism of impaired skin collagen maturity in riboflavin
or pyridoxine deficiency. J Biosci 1990; 15:289-295.
14. Chan MM, Stokstad EL. Metabolic responses of folic acid and related compounds to thyroxine in
rats. Biochim Biophys Acta 1980; 632:244-253.
15. Stokstad EL, Chan MM, Watson JE et al. Nutritional interactions of vitamin B12, folic acid, and
thyroxine. Ann N Y Acad Sci 1980; 355:119-129.
16. Keating JN, Kusano G, Stokstad EL. Effect of thiouracil in modifying folate function in severe
vitamin B12 deficiency. Arch Biochem Biophys 1988; 267:119-124.
17. Rivlin RS, Langdon RG. Effects of thyroxine upon biosynthesis of flavin mononucleotide and
flavin adenine dinucleotide. Endocrinology 1969; 84:584-588.
18. Lee SS, McCormick DB. Thyroid hormone regulation of flavocoenzyme biosynthesis. Arch Biochem
Biophys 1985; 237:197-201.
19. Ross NS, Hansen TP. Riboflavin deficiency is associated with selective preservation of critical
flavoenzyme-dependent metabolic pathways. Biofactors 1992; 3:185-190.
20. Glatzle D, Körner WF, Christeller S et al. Method for the detection of a biochemical riboflavin
deficiency. Stimulation of NADPH2-dependent glutathione reductase from human erythrocytes by
FAD in vitro. Investigations on the vitamin B2 status in healthly people and geriatric patients. Int
Z Vitaminforsch 1970; 40:166-183.
21. Bates C. Riboflavin. Int J Vitam Nutr Res 1993; 63:274-277.

2000; 46:1065-1071.
homocysteine in the Framingham offspring cohort is influenced by folate status and the C677T
24. Hustad S, McKinley MC, McNulty H et al. Riboflavin, flavin mononucleotide, and flavin adenine
dinucleotide in human plasma and erythrocytes at baseline and after low-dose riboflavin supple-
mentation. Clin Chem 2002; 48:1571-1577.
25. Burch HB, Bessey OA, Lowry OH. Fluorometric measurements of riboflavin and its natural de-
rivatives in small quantities of blood serum and cells. J Biol Chem 1948; 175:457-470.
93:7-9.
27. van der Put NM, Steegers-Theunissen RP et al. Mutated methylenetetrahydrofolate reductase as a
risk factor for spina bifida. Lancet 1995; 346:1070-1071.
28. Molloy AM, Daly S, Mills JL et al. Thermolabile variant of 5,10-methylenetetrahydrofolate reduc-
tase associated with low red-cell folates: Implications for folate intake recommendations. Lancet
1997; 349:1591-1593.
29. Molloy AM, Mills JL, Kirke PN et al. Whole-blood folate values in subjects with different
methylenetetrahydrofolate reductase genotypes: Differences between the radioassay and microbio-
logical assays. Clin Chem 1998; 44:186-188.
USA 1998; 95:13217-13220.
31. Jacques PF, Bostom AG, Wilson PW et al. Determinants of plasma total homocysteine concentra-
tion in the Framingham Offspring cohort. Am J Clin Nutr 2001; 73:613-621.
32. Shimakawa T, Nieto FJ, Malinow MR et al. Vitamin intake: A possible determinant of plasma
homocyst(e)ine among middle-aged adults. Ann Epidemiol 1997; 7:285-293.
33. de Bree A, Verschuren WMM, Blom HJ et al. Association between B vitamin intake and plasma
homocysteine concentration in the general Dutch population aged 20-65 y. Amer J Clin Nutr
2001; 73:1027-1033.
34. Backstrand JR. The history and future of food fortification in the United States: A public health perspective.
Nutr Rev 2002; 60:15-26.
reductase is dependent on riboflavin status: Implications for riboflavin requirements. Am J Clin Nutr 2002;
76:436-441.
36. Skoupy S, Födinger M, Veitl M et al. Riboflavin is a determinant of total homocysteine plasma
concentrations in end-stage renal disease patients. J Amer Soc Nephrol 2002; 13:1331-1337.
37. Lakshmi AV, Ramalakshmi BA. Effect of pyridoxine or riboflavin supplementation on plasma ho-
mocysteine levels in women with oral lesions. Natl Med J India 1998; 11:171-172.
38. McKinley MC, McNulty H, McPartlin J et al. Effect of riboflavin supplementation on plasma
homocysteine in elderly people with low riboflavin status. Eur J Clin Nutr 2002; 56:850-856.
mediate hyperhomocysteinemia (≥ 40 µmol/liter)—the Hordaland homocysteine study. J Clin In-
vest 1996; 98:2174-2183.
40. Harmon DL, Woodside JV, Yarnell JWG et al. The common “thermolabile” variant of methylene
tetrahydrofolate reductase is a major determinant of mild hyperhomocysteinemia. Q J Med 1996;
89:571-577.
41. Brattström L, Wilcken DE, Öhrvik J et al. Common methylenetetrahydrofolate reductase gene
mutation leads to hyperhomocysteinemia but not to vascular disease: The result of a meta-analysis.
Circulation 1998; 98:2520-2526.
42. Klerk M, Verhoef P, Clarke R et al. MTHFR 677C-->T polymorphism and risk of coronary heart
disease—A meta-analysis. JAMA 2002; 288:2023-2031.
43. Chen J, Giovannucci E, Kelsey K et al. A methylenetetrahydrofolate reductase polymorphism and
the risk of colorectal cancer. Cancer Res 1996; 56:4862-4864.
44. Ma J, Stampfer MJ, Giovannucci E et al. Methylenetetrahydrofolate reductase polymorphism, di-
etary interactions, and risk of colorectal cancer. Cancer Res 1997; 57:1098-1102.
45. Verhoef P, Rimm EB, Hunter DJ et al. A common mutation in the methylenetetrahydrofolate
reductase gene and risk of coronary heart disease: Results among U.S. men. J Am Coll Cardiol
1998; 32:353-359.
CHAPTER 7
The Molecular Dynamics of Abnormal Folate

Metabolism and DNA Methylation:
Implications for Disease Susceptibility and Progression
S. Jill James
Introduction
N
ormal folate-dependent one-carbon metabolism is essential for (a) the synthesis and
balance of deoxynucleotide triphosphate (dNTP) DNA precursor pools required for
error-free DNA synthesis and repair;1 and (b) the establishment and maintenance of
stable DNA methylation patterns required for tissue-specific gene expression and chromatin
conformation.2 Both functions are negatively affected by inadequate folate intake and/or by
genetic polymorphisms in these pathways.
The activity of MTHFR is in dynamic equilibrium with one-carbon metabolism and pro-
vides the metabolic link between DNA synthesis and DNA methylation. MTHFR is uniquely
positioned to direct the distribution of folate derivatives to meet cellular metabolic require-
ments for active cell proliferation (DNA synthesis) or for differentiated cell function (DNA
methylation and normal gene expression) as diagrammed in Figure 1. The sensitivity of MTHFR
to alterations in S-adenosylmethionine (AdoMet) and S-adenosylhomocysteine (AdoHcy) lev-
els provides the perception and response to the one-carbon needs of the cell. Methionine cycle
saturation and the associated increase in AdoMet levels is effectively reversed by enhanced
binding of AdoMet to the regulatory region of MTHFR which down-regulates its activity.3 As
a result, reduced synthesis of 5-methylTHF attenuates methionine synthesis to restore AdoMet
levels and balance to the methionine cycle. The indirect result of the AdoMet-induced decrease
in MTHFR activity is a reciprocal increase in the availability of 5,10-methyleneTHF for
deoxynucleotide precursor synthesis for DNA synthesis and repair (Fig. 1). Conversely, a defi-
cit in AdoMet relieves MTHFR inhibition and results in enzyme activation and diversion of
5methylTHF towards methionine and AdoMet synthesis for DNA methylation.4,5 The extent
of MTHFR activation and diversion of folate one-carbon groups towards methionine
remethylation is accompanied by a proportional diversion of folate away from deoxynucleotide
and DNA synthesis. Thus, under normal physiologic conditions, the reciprocity of MTHFR
activity maintains one-carbon homeostasis for DNA synthesis and DNA methylation.
MTHFR Activity and Folate Methyl Group Dispersal for Normal

DNA Methylation and DNA Synthesis
MTHFR and DNA Methylation
The physiologic importance of regulated MTHFR activity for the balanced distribution of
folate one-carbon groups is underscored by the negative impact of reduced 5-methylTHF
©2005 Eurekah.com.
Abnormal Folate Metabolism and DNA Methylation 79
Figure 1. An overview of the interactive and interdependent pathways involved in one-carbon metabolism
with emphasis on their importance for DNA methylation and DNA synthesis. These two major functions
of one-carbon metabolism intersect at the folate-/B12-dependent methionine synthase reaction. This cen-
tral reaction generates metabolically active tetrahydrofolate (THF) for DNA nucleotide synthesis and, at
the same time, regenerates methionine from homocysteine for transmethylation reactions. Both DNA
synthesis and DNA methylation are adversely affected by cofactor micronutrient deficiencies and/or by
genetic polymorphisms in these pathways. It is important to note that an elevation in homocysteine results
in the reversal of the AdoHcy hydrolase reaction and an elevation in AdoHcy, a potent product inhibitor
of DNA methyltransferases.
synthesis on DNA methylation (Fig. 1). Reduced availability of 5-methylTHF for the me-
thionine synthase reaction is a double-edged sword that negatively affects DNA methyltransferase
activity in two ways. Genetic or nutritional factors that decrease levels of 5-methylTHF reduce
the synthesis and availability of S-adenosylmethionine (AdoMet), the primary methyl donor
for most cellular methyltransferases, and, at the same time, increase levels of homocysteine and
consequently S-adenosylhomocysteine (AdoHcy), a potent product inhibitor of
methyltransferase activity (discussed in detail in the next section). In a knockout mouse model
lacking one or both MTHFR alleles, reduced AdoMetand/or increased AdoHcy was associated
with tissue-specific DNA hypomethylation.6 In this study, heterozygous and homozygous
MTHFR knockout mice exhibited an allelic dose/response impact on DNA hypomethylation
in brain, ovaries, testes, and liver, however spleen was not affected. In humans, the homozy-
gous MTHFR 677TT genotype was associated with a significant decrease in mononuclear cell
DNA methylation, but only among individuals with low folate status.7 Recent evidence sug-
gests that nutritional folate deficiency alone can independently promote DNA hypomethylation
in humans. In two independent studies, controlled dietary folate depletion of postmenopausal
and elderly women over a period of several weeks was associated with reduced plasma levels of
5-methylTHF, increased homocysteine, and lymphocyte DNA hypomethylation.8,9 In the
postmenopausal women (age 49-63), folate repletion normalized both homocysteine and DNA
hypomethylation; however in the elderly women (age 60-85), folate repletion normalized folate
and homocysteine levels, but not DNA methylation, suggesting that normalization of DNA
methylation with folate repletion may be delayed in older women.9 Although not evaluated in
these studies, it is possible that a gene-nutrient interaction between folate status and MTHFR
genotype contributed individual sensitivity to folate repletion and DNA methylation.
MTHFR and DNA Synthesis

With sufficient folate availability, the MTHFR 677C→T genotypes are phenotypically in-
distinguishable due to the folate-mediated stabilization of the enzyme from loss of its FAD
cofactor.10 However, under conditions of marginal folate status, variations in the MTHFR
677C→T phenotype (the substitution of alanine 222 by valine) and enzyme activity will de-
termine whether available folates are diverted towards DNA synthesis (677TT) or DNA me-
thylation (MTHFR 677CC). With insufficient intracellular folate, the MTHFR 677CC geno-
type can successfully compete with thymidylate synthase for limited 5,10-methyleneTHF such
that the synthesis of 5-methylTHF exceeds the methylation of dUMP to dTMP. Depending
on the cell type and physiologic status of the cell, the substitution can have beneficial or delete-
rious effects. Because the dNTPs are the substrates for the DNA polymerases with a half-life of
seconds, the fidelity of DNA replication and repair synthesis is critically dependent on the
maintenance of the correct balance of dNTPs.1 In vivo studies with folate-/methyl-deficient
diets in rats and in vitro studies with folate-deficient CHO cells have confirmed that the ratio
of dUTP/dTTP is significantly elevated and that uracil is misincorporated into DNA at a rate
that exceeds its repairability.11,12 The reiterative misincorporation of uracil under these condi-
tions was shown to promote the formation of procarcinogenic DNA strand breaks and abasic
sites. In folate-deficient CHO cells, acute folate repletion in the presence of folate-deficient
DNA lesions promoted gene amplification, anchorage-independent growth, and subsequent
neoplastic transformation.12 Both in vitro and in vivo studies in humans have confirmed an
increase in uracil in DNA with chronic folate deficiency that could have deleterious effects on
DNA integrity.13-15
Uracil is a normal base in RNA and is present in DNA as 5-methyl-uracil (or thymine).
Whereas cytosine in DNA is enzymatically methylated by DNA methyltransferases in situ,
uracil is enzymatically methylated by thymidylate synthase in the cytoplasm and is incorpo-
rated into DNA as thymidine (5-methyl-uracil), a premethylated base. It is important to note
that the substitution of uracil for thymine at specific DNA sequences may be an
under-appreciated and genetically significant form of DNA hypomethylation. Recently, the
critical importance of the thymine methyl group in maintaining normal base stacking interac-
tions for proper DNA conformation16,17 and also for sequence specific DNA-protein binding
has received increased research attention.18 The thymine methyl group projects into the major
groove of the DNA where it has been shown to alter the sequence-specific binding of the
cAMP-responsive element (CRE),18 the AP1 binding site (TPA-responsive element),19 E. coli
RNA polymerase and lacI repressor.20 Although excessive uracil in DNA has been shown to
induce significant DNA lesions, research on its impact on DNA conformation and gene ex-
pression is warranted.
It is informative to note that the reactive increase in MTHFR activity in response to AdoMet
deficit and the consequent diversion of folates away from DNA synthesis is the functional
equivalent of the MTHFR 677CC genotype. Similarly, the decrease in MTHFR activity and
diversion of folates towards DNA synthesis in response to elevated AdoMet is the functional
equivalent of the MTHFR 677TT genotype. These reciprocal diversions in folate metabolites
may contribute to differences in disease risk associated with homozygous mutant or wild type
MTHFR 677 genotypes.
AdoMet/AdoHcy Dynamics and DNA Hypomethylation

The interactive pathways that maintain the normal flow of transmethylation and
transsulfuration metabolites are diagrammed (Fig. 1). The B12-dependent transfer of the me-
thyl group from 5-methylTHF to homocysteine via the methionine synthase reaction provides
the metabolic link between folate and methionine metabolism. Methionine is energetically
activated by the ATP-mediated transfer of adenosine by methionine adenosyltransferase (MAT).

The product, S-adenosylmethionine (AdoMet), is the primary methyl donor for a multitude of
cellular methyltransferase enzymes that catalyze the transfer of the methyl group from AdoMet
to a wide variety of macromolecules including DNA, RNA, proteins, membrane phospholip-
ids, neurotransmitters, and numerous small molecules. 21 Notably, the sequential
AdoMet-dependent methylation of guanidinoacetate to form creatine in the liver consumes
more AdoMet than all the other methyltransferase reactions combined.22 After transfer of its
methyl group, AdoMet is converted to S-adenosylhomocysteine (AdoHcy) within the active
site of the methyltransferase. Because AdoHcy binds to the active site with a higher affinity
than AdoMet, most methyltransferases are subject to potent product inhibition by AdoHcy.23
The transfer and metabolic degradation of AdoHcy by the AdoHcy hydrolase reaction is abso-
lutely essential for efficient methyltransferase activity. In the nucleus, it has been confirmed
that the DNA methyltransferase and AdoHcy hydrolase enzymes act in close proximity.24 In-
terestingly, recent x-ray crystallography studies have shown that the polypeptide folding pat-
tern of the catalytic domain of AdoHcy is practically identical to that reported for DNA
methyltransferase suggesting a facile transfer of AdoHcy between the two enzymes.25 Reduced
efficiency of the AdoHcy hydrolase enzyme would directly affect efficiency of methyltransferase
reactions by allowing excess AdoHcy to remain in the active site, precluding AdoMet binding
and further methyl transfer. Excessive accumulation of AdoHcy has been shown to block most
AdoMet-dependent methyltransferases, including DNA methyltransferases,23,26,27 although the ex-
tent of inhibition would be expected to vary with the specific Km for AdoMet and the Ki for
AdoHcy of each methyltransferase. Clarke et al have examined this question and shown that
methyltransferases vary in their sensitivity to AdoHcy inhibition and certain methyltransferases
are much more susceptible to AdoHcy inhibition than others.26
Under normal physiologic conditions, AdoHcy is hydrolyzed by AdoHcy hydrolase to ho-
mocysteine and adenosine. This reaction is the sole source of homocysteine in all cells. Al-
though metabolically counter-intuitive, the equilibrium dynamics of the AdoHcy hydrolase
reaction actually favor the reverse direction, the synthesis rather than the hydrolysis of
AdoHcy.22,28 Thus, the hydrolytic direction of AdoHcy hydrolase is absolutely dependent on
efficient removal of its two products, homocysteine and adenosine.22,29 Metabolic perturba-
tions that increase homocysteine and/or adenosine will lead to AdoHcy accumulation.30,31
Elevated homocysteine reverses AdoHcy hydrolase leading to increased AdoHcy synthesis whereas
elevated adenosine binds to the active site of the hydrolase and inhibits enzyme activity;32 both
perturbations will lead to AdoHcy accumulation and the potential adverse effects of cellular
methyltransferase inhibition.4,26,33
Intracellular AdoHcy levels reflect the cumulative balance between the combined activities
of the multiple methyltransferases, the rates of synthesis and hydrolysis, the efficiency of ho-
mocysteine and adenosine product removal, and, at high levels, the extent of export to the
plasma. The one-way cystathionine β-synthase (CBS) reaction is an important regulator of
AdoHcy levels by providing the sole catabolic exit route for homocysteine and its permanent
removal from the methionine cycle.34 In mammals, only liver, kidney, pancreas, and intestine
have been reported to express CBS activity. 4 Only liver and kidney cells express
betaine-homocysteine methyltransferase (BHMT), the folate-independent pathway for
remethylation of homocysteine to methionine.35 Therefore, the majority of cell types in the
body do not express either CBS or BHMT and thus are relatively limited in their capacity to
metabolize homocysteine. Nutritional or genetic insufficiencies that directly or indirectly re-
duce methionine synthase activity would render these cells highly sensitive to homocysteine/
AdoHcy accumulation and methyltransferase inhibition, depending on the tissue-specific Ki.
Because of the tissue-specific distribution of CBS and BHMT, it is likely that the primary
source for plasma homocysteine and AdoHcy are extra-hepatic cells lacking expression of CBS
and/or BHMT.
Because AdoMet and AdoHcy are the precursor and product, respectively, of methyltransferase
reactions, the AdoMet/AdoHcy ratio is often used as an indicator of DNA and/or cellular
methylation status. However, it is not clear from the ratio whether a decrease in AdoMet or
increase in AdoHcy (or both) is necessary to affect cellular methylation capacity. This question
was addressed using a combined genetic and dietary approach with CBS wild type and het-
erozygous mice fed either a control diet or a folate-/methyl-deficient diet.36 Intracellular levels
of AdoMet and AdoHcy were measured in liver, kidney, brain and testes and correlated with
tissue-specific DNA methylation. The results indicated that a decrease in AdoMet alone was
not sufficient to affect DNA methylation, whereas an increase in AdoHcy, alone or in combi-
nation with a decrease in AdoMet, was most consistently associated with reduced DNA methy-
lation.36 Similarly, in MTHFR knockout mice, increased AdoHcy with or without a decrease
in AdoMet was associated with DNA hypomethylation in brain, ovaries, testes, and liver.6
Despite a massive decrease in AdoMet in MAT1A knockout mice, AdoHcy was not elevated
and the DNA was not hypomethylated.37 Finally, Weir et al observed that the reduced AdoMet/
AdoHcy ratio in brain tissues from nitrous oxide-exposed pigs was driven by an increase in
AdoHcy rather than a decrease in AdoMet.38 These animal studies support the notion that an
increase in AdoHcy, secondary to genetic or dietary conditions that increase homocysteine
levels, is the primary effector of reduced cellular methylation capacity.
The relatively recent development of highly sensitive methodology to measure plasma AdoMet
and AdoHcy levels has facilitated the understanding of AdoMet/AdoHcy dynamics in hu-
mans. In healthy humans, moderate elevations in plasma homocysteine were paralleled by
similar elevations in AdoHcy.39 Further, plasma AdoHcy, but not AdoMet, was significantly
correlated with lymphocyte DNA hypomethylation. In a study of patients with peripheral
artery disease, Loehrer et al found that the reduced ratio of AdoMet/AdoHcy in plasma and
erythrocytes was due to elevated AdoHcy levels rather than decreased AdoMet.40 In patients
with chronic renal failure and hyperhomocysteinemia, Perna et al found a four- to eight-fold
increase in AdoHcy with minimal change in AdoMet associated with protein
hypomethylation.41,42 Finally, in a recent study, Gonzales et al reported that colorectal biopsies
had significant decreases in folate levels associated with two-fold higher AdoHcy levels and
DNA hypomethylation. 43 The lack of correlation between low AdoMet and DNA
hypomethylation in rodent and human studies would suggest that AdoMet is not a limiting
factor for the DNA methyltransferase, at least not at physiologic levels. However, low levels of
AdoMet have been convincingly associated with an upregulation of the MTHFR enzyme to
divert 5,10-methyleneTHF towards transmethylation and away from nucleotide synthesis. Thus,
low AdoMet levels may have a greater impact on down-regulating the rate of DNA synthesis
than on down-regulating cellular methyltransferases. In terms of AdoHcy levels, the results in
rodents and humans suggest that an elevation in intracellular AdoHcy may be more toxic than
homocysteine and that homocysteine may be a convenient marker for intracellular AdoHcy
levels and cellular methylation status. Because of the more facile membrane transport of ho-
mocysteine relative to AdoHcy, an under-appreciated function of homocysteine may be to
serve as an exportable form of the more toxic metabolite, AdoHcy.27
DNA Methyltransferases and the Histone/Chromatin Connection

DNA Methylation and Methyltransferases
DNA methylation is a physical modification of DNA that has a profound functional influ-
ence on transcriptional repression, chromatin structure, X-chromosome inactivation, allelic
imprinting, and the silencing of parasitic elements in DNA.44,45 Physically, it is the covalent
addition of a methyl group at position 5 of the cytosine ring by AdoMet-dependent DNA
methyltransferases. The enzymatic mechanism is unusual in that the DNA methyltransferases
flip the target cytosine out of the DNA helix into the active site of the enzyme.46 Only cy-
tosines within the context of CG dinucleotides are targets for DNA methyltransferase activity
and methylation. The distribution of CpG dinucleotides in the genome is not random and
tends to be concentrated in “CpG islands”, regions of DNA where the CG dinucleotide motif
occurs at a five-fold higher frequency than in the rest of the genome.47 CpG islands occur
predominantly within repetitive elements of heterochromatin where they are hypermethylated
and transcriptionally inactive. Proportionally fewer CpG islands occur in promoter regions of
housekeeping genes where they are predominantly hypomethylated and transcriptionally ac-
tive. In differentiated cells, CpG island methylation patterns occur in a tissue-specific manner
such that promoter regions of expressed genes tend to be hypomethylated in a given tissue
whereas CpG islands in promoters of repressed genes tend to be hypermethylated. Thus, DNA
hypermethylation within regulatory sequences such as promoters, enhancers, and repressors
represses transcription and silences gene expression.48 Methylation and silencing of repressor
elements, however, is an interesting exception whereby hypermethylation indirectly results in
gene activation.
In addition to gene expression, DNA methylation modifies the functional domains of chro-
matin involved in DNA replication, repair, recombination and chromosome segregation. In
gene-poor heterochromatin, hypermethylation of repetitive and transposable elements serves
as an effective defence against transcriptional activation of parasitic DNA that could pose a
significant threat to genomic integrity and cell survival.49 It has been proposed that DNA
methylation originally evolved as a mechanism to control invasion of genomic parasites and to
compartmentalize the genome into transcriptionally active and inactive domains, and subse-
quently evolved into a mechanism for gene-specific silencing.50 DNA methylation is com-
monly defined as an “epigenetic” phenomenon because it imparts heritable information to the
genome that is not encoded by the nucleotide base sequence. However, because the methyl
addition has such profound effects on DNA function, 5-methyldeoxycytosine is also referred
to as the fifth base in DNA. It should be noted, however, that 5-methylcytosine is also a highly
mutable site that can undergo spontaneous deamination to produce a C to T transition that
can become a heritable mutation if not efficiently repaired.51,52
The stable heritability of normal DNA methylation patterns is essential for the mainte-
nance of differentiated cell function. Loss or disruption of DNA methylation patterns is the
basis for many genetic diseases and a prelude to many of the chronic diseases of aging, espe-
cially cancer.53,54 The stable inheritance of DNA methylation between cell generations is pro-
vided by the “maintenance” DNA methyltransferase, Dnmt1. Dnmt1 has a strong preference
for hemimethylated double-stranded DNA relative to unmethylated double-stranded DNA.
At the replication fork, the newly synthesized daughter strand creates hemimethylated CpG
sites that are rapidly methylated by Dnmt1 such that the original epigenetic information en-
coded by the parental methylation patterns is recreated.55 Although expression of Dnmt1 is
highly correlated with proliferative status in most cells, it is also paradoxically highly expressed
in nonproliferative tissues such as heart and brain.56 Increased expression without increased
proliferation may reflect the requirement to reestablish and maintain methylation patterns
after DNA strand repair in cells exposed to increased oxidative stress and DNA damage. The
essentiality of Dnmt1 expression and activity for normal development during embryogenesis is
underscored by midgestation lethality in Dnmt1 knockout embryos associated with severe
reduction in DNA methylation density.57 Conditional deletion of Dnmt1 using the crelox
system in primary fibroblasts resulted in extensive genomic hypomethylation and p53- depen-
dent apoptosis within 6 days.58 These studies underscore the importance of Dnmt1 and epige-
netic inheritance for normal cell function in differentiated cells and viability during develop-
ment.
In addition to Dnmt1, two other catalytically active AdoMet-dependent DNA
methyltransferases have been identified. In contrast to the maintenance activity of Dnmt1,
Dnmt3a and 3b are “de novo” methyltransferases that methylate cytosines at double-stranded
unmethylated CpG sites.59 These de novo methyltransferases are thought to function
primarily during embryonic development to create and establish the adult tissue-specific
methylation patterns that are then “maintained” by Dnmt1.60 After fertilization, the paternal
and maternal gamete-specific methylation patterns undergo passive and active demethylation,
respectively, such that the only methylated genes remaining in the preimplantation embryo are
a few maternal “imprinted” genes.60,61 The phenomenon of genomic imprinting refers to dif-
ferential methylation of maternal and paternal alleles that results in allele-specific gene expres-
sion. While all three Dnmts share a common catalytic domain, they have diverse N-terminal
regions that appear to direct their functional diversity. Dnmt1 is expressed in a cell-cycle de-
pendent manner whereas Dnmt3a and 3b expression is independent of cell cycle stage.62 Dnmt3b
is highly expressed before embryonic day 10 and tends to be concentrated in peri-centromeric
regions, whereas Dnmt3a is expressed after day 10 during later stages of development.63 Inter-
estingly, knockout mice for Dnmt3b die at midgestation whereas Dnmt3a mice appear to
develop normally but die soon after birth.64 Thus, Dnmt3b methylation may be more impor-
tant for repression of parasitic repetitive elements during development and Dnmt3a methyla-
tion may be important for cell differentiation and tissue-specific gene expression.65
The Histone Connection

Recent studies elucidating the molecular link between DNA methylation, multi-protein
complexes, and chromatin remodeling have revolutionized the understanding of the mecha-
nisms by which DNA methylation influences transcription, replication, recombination, and
chromosome segregation.63 Dynamic changes in chromatin structure are mediated by
post-translational changes in the amino-terminal tails of histones H3 and H4.66,67 Specific
methyl-binding proteins bind to methylated cytosines in DNA in multi-protein complexes
that catalyze specific modifications of histone tails. For example, the methyl-binding protein,
MeCP2, binds to DNA in a complex with histone deacetylase (HDAC) to deacetylate lysine 9
of histone 3.68 The enzymatic removal of acetyl groups from histones alters the local charge
distribution such that the histone tails wrap more tightly around the DNA helix. This induces
a tightening or condensation of chromatin that effectively limits the ability of transcription,
replication, or recombination machinery to bind to DNA. Thus, by physically limiting acces-
sibility, histone hypoacetylation and chromatin condensation, mediated by methyl-binding
proteins, repress DNA methylation-dependent functions.69 In summary, promoter regions of
repressed genes tend to be hypermethylated with hypoacetylated histones whereas promoters
of actively expressed genes tend to be hypomethylated with hyperacetylated histone tails.
Based on these observations, it has become clear that the functional alterations associated
with DNA methylation occur within the context of chromatin infrastructure. The
post-translational modifications of the histone tails that protrude from the nucleosome are
intimately involved with DNA methylation in the remodeling of chromatin structure and
repression of gene expression.69,70 Recent studies have added a new layer of complexity to the
DNA methylation-chromatin connection with the identification of a direct link between Dnmts
and histone deacetylation that does not require association with methyl-binding proteins.71 It
is now well established that the noncatalytic domain of all three Dnmt proteins can associate
directly with histone deacetylases (HDACs) and can repress transcription through histone
deacetylation and chromatin condensation, and that this activity is independent of their
methyltransferase activity.72 The association of Dnmt1 with HDAC would be an effective
means of targeting deacetylase activity to the replication fork to insure propagation of repres-
sive chromatin by hypoacetylating the newly assembled nucleosomes in newly replicated DNA.
The story becomes more interesting albeit more complex with the recent discovery of a link
between DNA methylation and histone methylation.73 An additional layer of epigenetic con-
trol of chromatin structure and transcriptional repression is now known to occur by the methy-
lation of lysine 9 on histone H3 by a site-directed histone methyltransferase.74 An interaction
between histone H3 lysine 9 methylation and DNA methylation is supported by the fact that
both modifications are associated with transcriptional silencing. Specifically, the methyl group
on H3 lysine 9 provides a specific binding site for the heterochromatin protein, HP1, which in
turn is a specific binding site for Dnmt 3a.59 This provides an elegant mechanism for the
targeting of de novo Dnmts to DNA sequences containing HP1 bound to methylated H3
lysine 9.75 In addition to HP1, Dnmts can be recruited to promoter regions through associa-
tion with specific transcription factors.72
An unexpected connection between DNA methylation and histone methylation was re-
cently discovered in the filamentous fungus Neurospora crassa.76 Mutations in the histone
methyltransferase resulted in a complete loss of DNA methylation suggesting for the first time
that histone methylation may be an essential prerequisite for DNA methylation. In Arabidopsis,
an inactivating mutation in a SNF2-like ATPase caused a severe reduction in DNA methyla-
tion suggesting that DNA methyltransferase binding and DNA methylation may additionally
depend on prior chromatin alterations by SNF2-like ATPases.77 The challenge of these new
discoveries will be to unravel the temporal sequence in mammals between histone acetylation/
methylation and DNA methylation that mediates transient or permanent alterations in chro-
matin structure and gene expression. Whether histone methylation precedes DNA methyla-
tion or vice versa is likely to vary with the particular chromosomal region, replicative state, and
the permanent or transient status of transcriptional repression.
Relevant to folate status, histone methyltransferases have been shown to be inhibited by
local increases in AdoHcy levels.78 It will be of great interest to determine whether the estab-
lished effect of folate deficiency on DNA methylation can be attributable to AdoHcy inhibi-
tion of histone methyltransferases as well as DNA methyltransferases. These recent discoveries
have broadened the understanding of DNA methyltransferases as multi-functional proteins
that undergo complex protein-protein and DNA-protein interactions to alter local chromatin
infrastructure to bring about changes in gene expression.69
Folate Deficiency and Alterations in DNA Methylation during

Embryonic Development and Aging
Dynamic Change in DNA Methylation During Embryonic Development
Mammals undergo a critical window of development between fertilization and blastocyst
implantation that is influenced by both genetic and environmental factors. During the preim-
plantation period, a massive wave of global demethylation occurs such that the parental
gamete-specific methylation patterns are totally erased, and by the 16 cell stage, most CpG
sites are demethylated.79-81 This epigenetic reformatting appears to be necessary for the cre-
ation of a pluripotent state before specific cell lineages can be established. The process of estab-
lishing the adult methylation pattern involves an initial wave of de novo methylation at the
stage of pregastrulation, followed by gene-specific demethylations associated with activation
and transcription of tissue-specific genes.81-85 The adult DNA methylation patterns estab-
lished during early embryogenesis by the de novo methyltransferases are nonrandom and heri-
table and are the basis for tissue-specific gene expression, X-inactivation, and imprinting.84,85
Experimental studies in mice lacking DNA methyltransferase expression have confirmed that
the failure to establish the correct DNA methylation patterns results in embryonic lethality 57
and developmental malformations.86-90 Similarly, the targeted inactivation of histone
methyltransferases results in mid-gestation lethality and maldevelopment.91 Although not ex-
tensively studied to date, aberrant DNA and histone methylation patterns in early embryogen-
esis are likely to be an important mechanism contributing to the high incidence of develop-
mental and pregnancy failure in mammals and clearly warrant intensive investigation.
Folate Deficiency and DNA Methylation Dysregulation in Fetal Pathology

It is apparent from the preceding discussion that normal embryogenesis and fetal develop-
ment are dependent on rapid and dynamic changes in DNA methylation and that these changes
are programmed to occur in a defined and tightly regulated temporal sequence.92 Aberrations
in the establishment of normal DNA methylation patterns will disrupt protein/DNA
interactions as well as protein-protein interactions required for normal chromatin structure,

suppression of parasitic elements, and gene expression.93 Inappropriate activation or inactiva-
tion of genes during defined developmental windows will promote tissue-specific malforma-
tions and/or embryonic lethality.
Folate/methyl deficiency is well known to disrupt normal DNA methylation and gene ex-
pression,94 to induce replication fork delay,95 and to promote inappropriate apoptosis.96,97
Although it has been suggested that these aberrations may be causally related to neural tube
defects associated with folate deficiency, direct evidence is lacking. This is partly due to the fact
that animal models of dietary folate deficiency have been surprisingly unsuccessful in inducing
malformations. As a result, progress in understanding the biochemical and molecular basis for
birth defects with dietary folate deficiency has been slow. The MTHFR knockout mouse pro-
vided the first genetic model of folate deficiency to confirm an association with tissue-specific
DNA hypomethylation during gestation with low folate status and maldevelopment.6 The
folate receptor (folbp1) knockout mouse was also shown to exhibit abnormal DNA methyla-
tion during embryonic development.98 Recently, the first successful animal model of neural
tube defects induced by dietary folate deficiency was reported.99 Both genetic and dietary
models of folate deficiency will be important to unravel the temporal changes in DNA methy-
lation and histone methylation/acetylation during embryogenesis to determine whether aber-
rations in methylation are functionally involved in structural malformations and other mani-
festations of fetal pathology. It may be that DNA methylation during embryogenesis is more
important in silencing repetitive parasitic elements and that histone acetylation/methylation is
more important in regulating temporal aspects of tissue-specific gene expression. Depending
on the specific gene and chromosomal region, DNA methylation may serve as the final clamp
to permanently silence transcription of relevant genes.65 Thus, over the last decade it has be-
come apparent that DNA methylation does not act alone, but is an integral part of a cascade of
epigenetic events that are required for normal regulation of gene expression.
Alterations in DNA Methylation and Transmethylation Metabolites

with Advanced Age
Genome-wide DNA methylation density declines with age in rodents.100 Loss of methyl
groups occurs primarily in the repetitive sequences in the centromeric/pericentromeric hetero-
chromatin regions. Whether a similar global decline in DNA methylation occurs with age in
humans has been more difficult to ascertain, but appears to be rare, at least in human lympho-
cytes.101 In contrast, several human studies have documented a paradoxical increase in DNA
methylation within promoter CpG islands of specific genes. For example, an increase in me-
thylation of the estrogen receptor promoter with age was found in colon mucosal cells.102
Similarly, the MYOD gene showed an age-related increase in methylation in normal colon
mucosa (r = 0.7; P < 0.0001).103 In this study, age-related alterations in methylation were gene
specific since several other candidate genes were not affected. It is also highly likely that the
methylation dysregulation that occurs with age is tissue-specific. Available data indicate that
normal aging is associated with loss of methylation in heterochromatic regions that are nor-
mally heavily methylated and with gain of methylation in promoter CpG islands that are nor-
mally hypomethylated.104 Interestingly, the same global hypomethylation and promoter re-
gion hypermethylation observed with aging has been also been observed in rodent models of
folate/methyl deficiency and in preneoplastic and neoplastic tissues of humans and rodents.105,106
The fact that normal aging is associated with increases in plasma homocysteine and AdoHcy
levels raises the interesting possibility that dysregulation of DNA methylation with age may be
causally related to age-related alterations in the methionine cycle. Supporting this notion, an
age-related decrease in red cell AdoHcy hydrolase has been reported107 as well as a 65% de-
crease in the AdoMet/AdoHcy ratio.108 In a study of patients with Alzheimer’s disease (AD),
plasma AdoMet levels were found to be severely decreased compared to age-matched con-
trols.109 Further, decondensation of pericentromeric heterochromatin in lymphocytes of AD
patients compared to healthy elderly subjects is consistent with histone and/or DNA
hypomethylation in repetitive elements.110 Finally, site-specific hypomethylation in the amy-
loid precursor protein gene in AD patients has been reported.111 It is now well established that
homocysteine is a risk factor for age-related cognitive decline and AD dementia;112,113 how-
ever, whether homocysteine directly contributes to CNS pathology or whether pathology is
secondary to homocysteine-related vascular disease is still an open question.114 The relation-
ship between CNS pathology, homocysteine, AdoHcy, and alterations in DNA methylation is
a relatively unexplored area of research that warrants further investigation, especially in light of
the high levels of DNA methyltransferase expression in the brain.
DNA Methylation Instability and the Epigenetic Basis for Disease:

Modification by Diet
Genetic Disease
The recognition that DNA methylation dysregulation is causally related to several human
genetic diseases such as ICF syndrome (Immune deficiency-Centromeric instability-Facial
anomalies), Rett syndrome, and fragile X, has firmly established the critical importance of
normal methylation patterns for normal health and development. These diseases reflect the
loss of epigenetic control during development that results in constitutive abnormalities in me-
thylation patterns leading to severe neurodevelopmental and immunologic pathology.
ICF is a rare autosomal recessive disease characterized by immunoglobulin deficiency, mild
facial dysmorphism, neurodevelopmental delay, and recurrent infections.115 Diagnosis is con-
firmed by the presence of bizarre multi-branched chromosomes in PHA-stimulated lympho-
cytes. The multiradial chromosome configuration primarily affects chromosomes 1, 9, and 16
and is due to severe regional decondensation and significant hypomethylation in classical satel-
lite DNA.116 Normally these repetitive elements in pericentromeric heterochromatin are heavily
methylated in order to maintain a condensed and stable structure in this region. The genetic
basis for ICF was recently discovered to be due to mutations in conserved regions of the de
novo DNA methyltransferase 3B gene, making ICF the first identified genetic disease due to
defective methyltransferase activity.117,118 Subsequent studies in ICF patients have identified
hypomethylation in CpG islands of X-linked genes that are normally inactivated by methyla-
tion119 as well as defective methylation and silencing of viral sequences.120 These observations
provide insights into the function and regional specificity of DNA methyltransferase 3B as well
as implications for the mechanisms of chromosomal rearrangements. ICF is the only known
human disorder caused by a germline mutation in a DNA methyltransferase.
Rett syndrome is a neurodevelopmental disorder of early post-natal brain development that
predominantly affects females and was recently shown to be due to mutation in the X-linked
methyl CpG binding protein, MeCP2.121 The MeCP2 gene codes for a protein that binds to
methylated cytosines in a complex with histone deacetylase to mediate chromatin condensa-
tion and local transcriptional repression. Recurrent C to T transition mutations, deletions, and
frameshift mutations in the MeCp2 gene have been found in at least 80% of girls affected with
classic Rett syndrome.122 Mutations occur either in the methylated CpG binding domain or in
the transcriptional repression domain that recruits the histone deacetylase complex. Affected
girls appear to have normal development for 6-18 months followed by cognitive and behav-
ioral regression marked by unusual hand-wringing, ataxia, speech and growth retardation, au-
tistic tendencies, and mental retardation.123 An interesting but as yet unanswered question is
why the brain should be the primary site of Rett syndrome pathology. MeCp-2 is more abun-
dantly expressed in the brain than in any other tissue and may offer a partial explanation.124
Although a defect in gene silencing with MeCP2 deficiency would be a logical candidate for
neurodevelopmental abnormalities, this remains to be proven. The CNS specificity for MeCP-2
deficiency suggests the possibility that other genes affecting DNA methylation function may
be involved in the etiology of other neurodevelopmental defects.
The fragile X syndrome is due to a rare folate-sensitive fragile site on the X chromosome
and is a major genetic cause of mental retardation. A massive expansion of a CGG trinucle-
otide repeat in the promoter of the FMR1 gene creates an inappropriate CpG island that
subsequently becomes de novo methylated.125 Methylation and the resulting localized conden-
sation in the FMR1 promoter region leads to transcriptional inactivation and the phenotypic
expression of the fragile X syndrome.126,127 In addition to “rare” fragile sites such as the fragile
X, “common” fragile sites or gaps occur nonrandomly in all chromosomes under cell culture
conditions that induce folate depletion, deoxynucleotide imbalance, and/or block DNA poly-
merase progression.128-130 Common fragile sites occur in regions of chromosome instability as
evidenced by increased recombination, translocations, sister chromatid exchanges, and
intrachromosomal gene amplification and deletions at these sites.128,130-132 Pertinent to folate
deficiency, breakage at sites of chromosomal fragility often occur in regions of decondensed
and hypomethylated DNA that flank hypermethylated condensed regions. Further, agents that
induce DNA hypomethylation, such as 5-azacytidine, also induce fragile site expression.133
The coincidence of several common fragile sites with specific deletions, chemically-induced
lesions, chromosomal translocations, and viral insertions suggests that agents or conditions
that predispose to fragile sites in vivo may also predispose to certain cancers.132,134 It is tempt-
ing to speculate that dNTP imbalance, DNA hypomethylation, and stalling of the replication
fork associated with nutritional folate deficiency in humans may facilitate induction of com-
mon fragile sites in vivo.
Methyl Donor Supplementation and Modulation of Epigenetic Disease

The possibility that maternal diet could modulate the severity of genetic phenotype in
methylation-related developmental disorders was recently explored by Wolff et al using the
agouti viable yellow mouse strain.135 The agouti gene is under the control of an inserted viral
IAP promoter that is variably methylated and expressed in the offspring.136 Increased methyla-
tion of the IAP promoter is associated with a mottled “agouti” coat color whereas reduced
methylation is associated with phenotypic expression of a yellow coat color. Thus, coat color
serves as a convenient phenotypic marker of agouti gene methylation and expression in mice
that are genetically identical. By feeding pregnant agouti mice a diet enriched with the methyl
donors B12, folate, betaine, choline and methionine, the expression and methylation of the
agouti gene was altered in utero as confirmed by an increase in agouti coat color in the off-
spring.135,137 Although it must be emphasized that IAP elements are rare in humans and the
methyl content of the control chow diet was not analyzed, nonetheless these observations sug-
gest the possibility that phenotypic expression of genetic alterations affecting DNA methyla-
tion and human disease may be modifiable by diet. Diets containing folinic acid and betaine
have been used to treat infants with severe mutations affecting MTHFR activity.138 Similar
dietary interventions are currently being tested in children with Rett syndrome and Down
syndrome (unpublished). With in utero diagnosis of genetic defects affecting DNA methyla-
tion, it is possible that targeted alterations in maternal diet could lessen the severity of the
disease phenotype.
Chromosomal Aberrations with Chronic Folate Insufficiency

Folate and/or B12 deficiency in humans has been long associated with chromosomal insta-
bility and aberrations. Early cytogenetic studies of folate- or B12-deficient human lymphocytes
or bone marrow cells revealed multiple chromosomal breaks and gaps, decondensed chromo-
somes, premature centromeric division, and centromeric spreading.139,140 As discussed earlier,
chronic folate deficiency is manifested in a futile cycle of uracil misincorporation during DNA
replication and repair.13,141-143 The misincorporation of uracil for thymine per se is not a
premutagenic lesion because the DNA polymerase will insert the correct adenine base opposite
either thymine or uracil.141,144 However, the presence of an active uracil glycosylase leads to
repair-related site-specific abasic sites and single strand breaks that represent significant
premutagenic lesions.142 These sites facilitate viral integration, recombination, and/or chro-
mosomal breaks that result in the formation of micronuclei (small chromosome fragments).
The increased frequency of micronuclei and DNA strand breaks in folate- or B12- deficient
human lymphocytes has been shown to be reversible with folate repletion, implying a causal
role for folate deficiency in micronuclei formation.145,146 In a recent study, folate deficiency in
post-menopausal women was associated with an increased frequency of centromeric
kinetochorepositive micronuclei.147 In this metabolic study, folate repletion following the folate
depletion phase was associated with a significant decrease in kinetochorepositive, but not
kinetochorenegative, micronuclei.147 Micronuclei containing centromeric kinetochores are
surrogate markers for abnormal chromosome segregation; thus, these findings implicate a role
for folate deficiency as a risk factor for human aneuploidy.
Genetic Polymorphisms, DNA Hypomethylation, and Maternal Risk

of Down Syndrome
Down syndrome (DS) or trisomy 21 is a complex genetic and metabolic disorder due to the
presence of three copies of chromosome 21. The extra chromosome derives from the mother in
93% of cases and is due to failure of normal chromosome segregation during meiosis. Except
for advanced age at conception, maternal risk factors for having a child with DS are not known.
A recent preliminary study compared mothers of children with DS to age-matched control
mothers with no abnormal pregnancies and found a significant increase in plasma homocys-
teine levels and an increased frequency of the MTHFR 677 T allele associated with a 2.3-fold
increased risk of having a child with DS.148 Based on this evidence, it was hypothesized that
genetic or nutritional factors that compromise folate status could promote centromeric/
pericentromeric DNA hypomethylation and abnormal chromosome segregation. A subsequent
study with a much larger sample size confirmed that an increased maternal risk of DS was
associated with the MTHFR polymorphism and additionally found that a second polymor-
phism in the folate pathway, in the methionine synthase reductase gene (MTRR 66A→G),
also increased the likelihood of having a child with DS.149 Further, the combined presence of
both polymorphisms conferred a greater risk than either polymorphism alone. A recent Irish
study confirmed that the presence of both MTRR and MTHFR polymorphisms was associated
with an increased risk of having a child with DS and also observed an independent increase in
risk associated with MTRR, but not MTHFR. 150 Two other studies, one from France and the
other from Italy, did not confirm an increased risk of DS associated with MTHFR.151,152 How-
ever, both of these countries have a higher folate intake relative to other countries and a corre-
sponding lower frequency of NTDs. Thus, increased periconceptional folate intake could have
been a significant confounder in the latter two studies. If abnormal one-carbon metabolism is
found to be associated with abnormal chromosome segregation and aneuploidy, it is likely that
the MTHFR 677C→T polymorphism is but one of several possible predisposing factors de-
pending on maternal diet and other interactive genetic polymorphisms.
Recent mechanistic studies in PHA-stimulated human lymphocytes have confirmed earlier
in vivo and in vitro reports that premature centromeric division is induced in vitro under
conditions of folate and thymidine deficiency.139,140 Follow-up studies in folate-deficient lym-
phocytes demonstrated significant hypomethylation in centromeric/pericentromeric repeats
that was associated with increased endonuclease sensitivity in this region (unpublished obser-
vations, SJJ). These results are consistent with a decondensed chromatin structure in the cen-
tromeric region that could predispose to abnormal chromosome segregation.
Cancer, DNA Methylation Dysregulation, and Folate/Methyl Deficiency

Abnormal DNA metabolism and a variety of procarcinogenic lesions have been associated
with clinical and experimental models of folate/methyl donor deficiency. These aberrations
include dNTP pool imbalance,96,143,153 DNA strand breaks,142,154 fragile sites,155 uracil
misincorporation,11,13,14 deletions,156 abnormal recombination,157 micronucleus formation145
and DNA hypomethylation.154,158-160 Functionally, these lesions translate into inappropriate

gene expression, inefficient DNA repair,14 delayed S phase progression,95,161 megaloblastosis,162
heritable mutations,163 and carcinogenesis.143,164 Because prolonged severe folate deprivation
in vivo is lethal within weeks, the isolated role of folate deficiency as a complete carcinogen is
not clear. However, convincing evidence from clinical and experimental studies indicates that
moderate folate deficiency interacts with other risk factors and known carcinogens to acceler-
ate tumor progression in vivo.165-168
Maintenance of adequate folate status from dietary sources and/or by synthetic folic acid
supplementation has been associated with a protective effect and reduced incidence of a variety
of human cancers.143,169-172 It has been suggested that the protective effect of dietary fruits and
vegetables may be due, in part, to the presence of folate.173 It is important to note, however,
that the protective effect of folate may only apply during the reversible phase of tumor promo-
tion. Once irreversibly transformed, excessive folate supplementation may, in fact, exacerbate
neoplastic growth.174 Thus, while folate deficiency appears to promote neoplastic progression
during early stages of cancer, folate deficiency as antifolate therapy, arrests the growth of ad-
vanced cancers. Ironically, the same negative effects of folate deprivation on DNA metabolism
can either promote or regress tumors depending on the timing relative to the neoplastic stage.
Similarly, the cellular response to excess folate may also vary with the physiologic state of the
cell. Folate supplementation of normal cells appears to have a protective effect, whereas in
advanced cancer, excessive folate supplementation may accelerate cancer growth.174 Several
clinical and experimental studies support the notion that manipulation of intracellular folate
can be a double-edged sword depending on the pathophysiologic state of the cell.143,175,176
There are three reproducible aspects of methylation dysregulation that occur during human
cancer progression: global hypomethylation, regional hypermethylation, and increased DNA
methyltransferase activity.177 These alterations are often present in the premalignant cells and
thus may be mechanistically involved in facilitating neoplastic transformation. The majority of
methyl groups tend to be lost in heavily methylated CpG islands of repetitive parasitic ele-
ments within pericentromeric heterochromatic domains.178 It has been proposed that an early
evolutionary function of DNA methyltransferases was to suppress the expression and spread-
ing of parasitic genomic invaders and that this function was later adapted to tissue specific gene
silencing.179 In some tumor types, the extent of methylation loss parallels tumor grade and
thus may have utility as a prognostic marker.180
Paradoxically, although the total methylation density is reduced in preneoplastic and neo-
plastic cells, aberrant CpG island hypermethylation occurs within promoter regions of many
cancer-related genes. Inappropriate methylation and silencing of genes involved in cell cycle
regulation, DNA repair, apoptosis, and imprinting (H19) has been identified in a variety of
tumors.181 Using a candidate gene approach, aberrant promoter region hypermethylation and
silencing of several tumor suppressor genes have also been demonstrated to contribute to tu-
mor promotion.54 It is important to note that transcriptional silencing of tumor suppressor
genes by promoter hypermethylation is the equivalent of coding region mutation in terms of
functional inactivation and tumor promotion.
Mechanistic Studies of Hepatocarcinogenesis with Chronic Folate/Methyl

Deficiency
Because folate deficiency and alterations in DNA methylation in human cancers are cov-
ered in other chapters, the focus here will be on mechanistic studies with animal and cell
culture models of folate/methyl deficiency and tumorigenesis. All three aberrations in DNA
methylation (global hypomethylation, promoter hypermethylation, and increase in DNA
methyltransferase activity) have been shown to occur in rat models of hepatocarcinogenesis
with folate/methyl deficiency, making this model one of the few in vivo models in which
progressive changes in global and gene-specific methylation can be evaluated during tumor
progression. Thus, this model can provide insights into temporal alterations in
methylation-related tumorigenesis in vivo not possible in human studies.
Chronic dietary insufficiency of the lipotropic nutrients, choline and methionine, is repro-
ducibly hepatocarcinogenic in the rat and certain mouse strains with or without chemical
initiation.182-184 Both semi-synthetic and amino acid-defined diet formulations have been uti-
lized by investigators since the initial observations of tumorigenesis with lipotrope deprivation
by Copeland and Salmon in 1946.185 The diet preparations currently in use tend to vary widely
between laboratories in terms of (a) protein source and methionine content, (b) fat source and
content; (c) amino acid-defined vs. semi-purified diet formulation; and (d) additional deficien-
cies in the two lipotropic vitamins, folic acid and vitamin B12. The most commonly used
semi-purified diet is the choline-devoid, low methionine (0.18% from soy, casein and peanut
meal protein) formulation adapted by Lombardi et al.186 A methyl-deficient amino-acid
(AA)-defined diet has also been used to produce a more severe, albeit less physiologic form of
methyl insufficiency.182 Despite formulation differences, both semi-synthetic and AA-defined
diet preparations have been demonstrated to be hepatocarcinogenic with chronic feeding al-
though distinct differences exist in the onset and severity of tumorigenicity as well as in specific
histopathology.187
Because the metabolic pathways of choline, methionine, and folic acid are mutually inter-
dependent, a deficiency in one will alter the requirement and metabolic priorities of the others
in attempt to maintain normal one-carbon metabolism. Consequently, single deficiencies should
not be discussed in metabolic isolation. For example, chronic deficiencies in choline and me-
thionine increase the requirement for folate metabolites and reduce total folate levels in the
liver.188 Similarly, low dietary methionine increases the requirement for exogenous choline
such that in vivo choline deficiency is difficult to achieve without marginal dietary methionine
(0.18%).189 Thus, the carcinogenic effects of a “choline-deficient” diet have not been demon-
strated in the presence of adequate methionine (> 0.6%).190 These observations underscore the
metabolic interdependence of the lipotropic nutrients in maintaining normal one-carbon me-
tabolism and their interaction in promoting carcinogenesis with chronic deficiency.
In the folate-/methyl-deficient rat model of hepatocarcinogenesis, the methyl-deficient diet
is initiated at weaning and continued for 12-16 months for tumor development. DNA strand
breaks, uracil misincorporation, reduced DNA repair, abasic sites, apoptosis occur after 3-9
weeks in preneoplastic liver.96,141,154 Methyl-deficient diets induce a significant decrease in
AdoMet and increase in AdoHcy and global DNA hypomethylation as early as 3 weeks after
diet initiation.191,192 Paradoxically, the progressive decrease in DNA methylation is inversely
proportional to an increase in DNA methyltransferase activity and associated with gene-specific
regional hypermethylation.154 These methylation-related changes are exactly those shown to
occur during human tumorigenesis making this an ideal model system to gain mechanistic
insights into temporal changes in methylation during tumor progression. An unusual aspect of
the rat model is that preneoplastic lesions, methylation changes and tumor development only
occur in the male rat.154 It is possible that the estrogen-induced increase in MAT1 activity and
AdoMet levels protects female rats from alterations in DNA methylation and methyltransferase
activity (J.D. Finkelstein, personal communication). Alterations in DNA methylation are found
only in the liver, the target organ for tumorigenesis, further implicating a causal role for methy-
lation dysregulation in tumor development. DNA methylation is not altered in intestine or
colon with folate- or methyl-deficient diets in rodent models.193-195
The folate-/methyl-deficient model has been used to evaluate progressive alterations in pro-
moter region methylation of the p53 tumor suppressor gene in preneoplastic liver and tumor
tissue. Site-specific alterations in CpG methylation were found that varied with the stage of
carcinogenesis.196 In preneoplastic liver, the p53 promoter was hypomethylated relative to
control-fed rats and associated with increased p53 expression. In tumor tissue, a specific site in
the promoter was found to be de novo methylated and associated with 85% suppression of
expression in a reporter gene construct.196 Methylation changes within the p16 tumor suppres-
sor gene promoter were also evaluated in the folate-/methyl-deficient model.197 In contrast to
the p53 gene, the p16 gene became progressively more methylated with time on the folate-/
methyl-deficient diet with a progressive loss of expression during the preneoplastic period. The
difference may reflect the fact that the p53 gene does not contain a CpG island and may be
more sensitive to site-specific methylation whereas the p16 gene contains a CpG island that
becomes progressively methylated and repressed during tumor progression.
In addition to providing new information on progressive alterations in promoter region
methylation in vivo, the folate-/methyl-deficient model of hepatocarcinogenesis has provided
insights into one of the major unanswered paradoxes in DNA methylation and cancer research:
the mechanistic basis for global hypomethylation despite increased methyltransferase activity
and regional hypermethylation. Previous studies characterizing DNA methyltransferase affini-
ties and activity revealed that the enzyme binds with higher affinity to DNA strand breaks,
abasic sites, and uracil than it does to its cognate hemimethylated CpG sites, consistent with its
original function as a DNA repair enzyme.198 These DNA lesions are exactly the same lesions
that are induced in the folate-/methyl-deficient model of hepatocarcinogenesis141,170 and in
folate-deficient CHO cells12 and in lymphocytes.14 Although rare in normal cells, abasic sites,
strand breaks, gaps, and strand breaks are often present as chronic unrepaired lesions in
preneoplastic cells. In the folate-/methyl-deficient cancer model, the early appearance and ac-
cumulation of these lesions parallels DNA hypomethylation and inefficient repair.96,141,154
Thus, the high affinity binding of the DNA methyltransferase to these lesions may offer a
unifying explanation for the paradoxical alterations in DNA methylation in preneoplastic and
neoplastic cells: methyltransferase bound to DNA lesions would not be available at the replica-
tion fork to methylate the newly replicated strand and would promote passive
replication-dependent demethylation. After a second round of replication, the previously
hemimethylated CpG site would be permanently demethylated on both DNA strands and no
longer a target for methyltransferase activity. This scenario would lead to a heritable state of
global hypomethylation. The binding of the methyltransferase to lesions at a CpG site would
simultaneously increase the probability for inappropriate site-specific de novo cytosine methy-
lation on the opposite strand. Preliminary evidence supporting this hypothesis showed that the
methylation capacity of the DNA methyltransferase was reduced in the presence of DNA le-
sions induced by folate-/methyl-deficiency (unpublished observations, SJJ). Subsequent ex-
periments confirmed that once methylation is lost on both strands, an adaptive increase in
DNA methyltransferase activity is unable to restore the original methylation pattern even with
dietary repletion of methyl donors. These insights are only possible in a controlled experimen-
tal model with unlimited intervention capacity and imply that DNA damage is a necessary
prerequisite for the loss of normal methylation patterns. Further research will be required to
determine whether these insights from the rat model of folate/methyl deficiency and
hepatocarcinogenesis may apply to methylation dysregulation observed in human cancers. The
protective effect of the homozygous MTHFR TT genotype against a variety of cancers by
promoting an increase in 5,10-methyleneTHF and error-free DNA synthesis199,200 indirectly
supports the hypothesis that DNA lesions are a necessary prerequisite for DNA methylation
dysregulation and cancer promotion.
Summary
The association between chronic folate deficiency and the generation of abnormal DNA
methylation patterns is a relatively recent discovery. Continued research into this new and
provocative relationship has the potential to provide new mechanistic insights into the patho-
genesis of several genetic diseases, birth defects, neurologic disorders, aging, and cancer.
References
1. Das SK, Kunkel TA, Loeb LA. Effects of altered nucleotide concentrations on the fidelity of DNA
replication. Basic Life Sci 1985; 31:117-126.
2. Eto I, Krumdieck CL. Role of vitamin B12 and folate in carcinogenesis. In: Poirier LA, Newberne
PM, Pariza MW, eds. Essential Nutrients in Carcinogenesis, New York: Academic Press,
1986:13-330.
3. Kutzbach C, Stokstad ELR. Feedback inhibition of methylene-tetrahydrofolate reductase in rat liver
by S-adenosylmethionine. Biochim Biophys Acta 1967; 139:217-220.
4. Finkelstein JD. Pathways and regulation of homocysteine metabolism in mammals. Semin
Thromb Hemost 2000; 26:219-225.
5. Roje S, Chan SY, Kaplan F et al. Metabolic engineering in yeast demonstrates that
S-adenosylmethionine controls flux through the methylenetetrahydrofolate reductase reaction in vivo.
J Biol Chem 2002; 277:4056-4061.
6. Chen ZT, Karaplis AC, Ackerman SL et al. Mice deficient in methylenetetrahydrofolate reductase
exhibit hyperhomocysteinemia and decreased methylation capacity, with neuropathology and aortic
lipid deposition. Hum Mol Genet 2001; 10:433-443.
reductase gene affects genomic DNA methylation through an interaction with folate status. PNAS
2002; 99:5606-5611.
8. Jacob RA, Gretz DM, Taylor PC et al. Moderate folate depletion increases plasma homocysteine
and decreases lymphocyte DNA methylation in postmenopausal women. J Nutr 1998;
128:1204-1212.
9. Rampersaud G, Kauwell GP, Hutson AD et al. Genomic DNA methylation decreases in response
to miderate folate depletion in elderly women. Am J Clin Nutr 2000; 72:998-1003.
10. Yamada K, Chen ZT, Rozen R et al. Effects of common polymorphisms on the properties of
recombinant human methylenetetrahydrofolate reductase. PNAS 2001; 98:14853-14858.
11. Pogribny IP, Muskhelishvili L, Miller BJ et al. Presence and consequence of uracil in preneoplastic
DNA from folate/methyl-deficient rats. Carcinogenesis 1997; 18:2071-2076.
12. Melnyk S, Pogribna M, Miller BJ et al. Uracil misincorporation, DNA strand breaks, and gene
amplification are associated with tumorigenic cell transformation in folate deficient/repleted Chi-
nese hamster ovary cells. Cancer Lett 1999; 146:35-44.
13. Blount BC, Mack MM, Wehr CM et al. Folate deficiency causes uracil misincorporation into
human DNA and chromosome breakage: Implications for cancer and neuronal damage. PNAS
1997; 94:3290-3295.
14. Duthie SJ, Hawdon A. DNA instability (strand breakage, uracil misincorporation, and defective
repair) is increased by folic acid depletion in human lymphocytes in vitro. FASEB J 1998;
12:1491-1497.
15. Duthie SJ, Narayanan S, Blum S et al. Folate deficiency in vitro induces uracil misincorporation
and DNA hypomethylation and inhibits DNA excision repair in immortalized normal human co-
lon epithelial cells. Nutr Cancer 2000; 37:245-251.
16. Kypr J, Sagi J, Ebinger K et al. Thymine methyl groups stabilize the putative A-form of the syn-
thetic DNA poly(amino dA-dT). Biochemistry 1994; 33:3801-3806.
17. Richards RG, Sowers LC, Laszlo J et al. The occurrence and consequences of deoxyuridine in
DNA. Adv Enzyme Regul 1986; 22:157-185.
18. Verri A, Mazzarello P, Biamonti G et al. The specific binding of nuclear proteins to the cAMP
responsive element sequence is reduced by the misincorporation of uracil and increased by the
deamination of cytosine. Nucl Acids Res 1990; 18:5775-5780.
19. Risse G, Jooss K, Neuberg M et al. Asymmetrical recognition of the palindromic AP1 binding site
(TRE) by fos protein complexes. EMBO J 1989; 8:3825-3832.
20. Dubendorff JW, deHaseth PL, Rosendahl MS et al. DNA functional groups required for the for-
mation of open complexes between E. Coli RNA polymerase and the promoter: Identification via
base analog substitutions. J Biol Chem 1987; 262:892-898.
21. Chiang PK, Gordon RK, Tal J et al. S-adenosylmethionine and methylation. FASEB J 1996;
10:471-480.
22. Cantoni GL. The role of S-adenosylhomocysteine in the biological utilization of S- adenosylmethionine.
Prog Clin Biol Res 1985; 198:47-65.
23. Hoffman DR, Cornatzer WE, Duerre JA. Relationship between tissue levels of S-adenosylmethionine,
S-adenosylhomocysteine, and transmethylation reactions. Canad J Biochem 1979; 57:56-65.
24. Radomski N, Kaufmann C, Dreyer C. Nuclear accumulation of S-adenosylhomocysteine hydrolase
in transcriptionally active cells during development of Xenopus laevis. Mol Biol Cell 1999;
10:4283-4298,.
25. Hu Y, Komoto J, Gomi T et al. Chrystal structure of S-adenosylhomocysteine from rat liver.
Biochemistry 1999; 38:8323-8333.
26. Clarke S, Banfield K. S-Adenosylmethionine-dependent Methyltransferases. In: Carmel R, Jacobsen
DW, eds. Homocysteine in Health and Disease. Boston: Cambridge University Press, 2001:63-78.
27. James SJ, Melnyk S, Pogribna M et al. Elevation in S-adenosylhomocysteine and DNA
hypomethylation: Potential epigenetic mechanism for homocysteine-related pathology. J Nutr 2002;
132:2361S-2366S.
28. de la HG, Agostini S, Bozzi A et al. S-adenosylhomocysteinase: Mechanism of reversible and irre-
versible inactivation by ATP, cAMP, and 2'-deoxyadenosine. Biochemistry 1986; 25:8337-8342.
29. Finkelstein JD, Harris B. Methionine metabolism in mammals: S-adenosylhomocysteine hydrolase
in rat intestinal mucosa. Arch Biochem Biophys 1975; 171:282-286.
30. Hoffman DR, Marion DW, Cornatzer WE et al. S-adenosylmethionine and S-adenosylhomocysteine
metabolism in isolated liiver. J Biol Chem 1980; 22:10822-10827.
31. Finkelstein JD. The metabolism of homocysteine: Pathways and regulation. Eur J Pediatr 1998;
157:S40-S44.
32. Ueland PM. S-Adenosylhomocysteinase from mouse liver. Inactivation of the enzyme in the pres-
ence of metabolites. Int J Biochem 1982; 14:207-213.
33. Ueland PM. Pharmacological and biochemical aspects of S-adenosylhomocysteine and S-
adenosylhomocysteine hydrolase. Pharmacol Rev 1982; 34:223-253.
34. Finkelstein JD. The regulation of homocysteine metabolism. In: Graham I, Refsum H, Rosenberg
IH, et al, eds. Homocysteine Metabolism: From Basic Science to Clinical Medicine. Norwell, MA:
Kluwer Academic Publishers, 1997:3-9.
35. Sunden SL, Renduchintala MS, Park EI et al. Betaine-homocysteine methyltransferase expression
in porcine and human tissues and chromosomal localization of the human gene. Arch Biochem
Biophys 1997; 345:171-174.
36. Caudill MA, Wang JC, Stepan Melnyk et al. Intracellular S-adenosylhomocysteine concentrations
predict global DNA hypomethylation in tissues of methyl deficient cystathionine β-synthase het-
erozygous mice. J Nutr 2001; 131:2811-2818.
37. Lu SC, Alvarez L, Huang ZZ et al. Methionine adenosyltransferase 1A knockout mice are predis-
posed to liver injury and exhibit increased expression of genes involved in proliferation. PNAS
2001; 98:5560-5565.
38. Weir DG, Molloy AM, Keating JN. Correlation of the ratio of S-adenosylmethionine to
S-adenosylhomocysteine in the brain and cerebrospinal fluid of the pig: Implications for the deter-
mination of this methylation ratio in human brain. Clinz Sci 1992; 82:93-97.
39. Yi P, Melnyk S, Pogribna M et al. Increase in plasma homocysteine associated with parallel in-
creases in plasma S-adenosylhomocysteine and lymphocyte DNA hypomethylation. J Biol Chem
2000; 275:29318-29323.
40. Loehrer FM, Tschopl M, Angst CP et al. Disturbed ratio of erythrocyte and plasma S-adenosylmethioine/
S-adenosylhomcteine in peripheral arterial occlusive disease. Atherosclerosis 2001; 154:147-154.
41. Perna AF, Ingrosso D, Satta E et al. Metabolic consequences of hyperhomocysteinemia in uremia.
Am J Kidney Dis 2001; 38:S85-S90.
42. Perna AF, Castaldo P, De Santo NG et al. Plasma proteins containing damaged L-aspartyl residues
are increased in uremia: Implications for mechanism. Kidney Int 2001; 59:54-56.
43. Gonzales MP, Alonso-Aperte E, Vilches M et al. Impaired methionine metabolism in patients un-
dergoing colorectal carcinogenesis surgery: A pilot study in Spain. 3rd International Conference on
Homocysteine. Sorrento, Italy: 2001.
44. Martienssen RA, Richards EJ. DNA methylation in eukaryotes. Curr Opin Genet Dev 1995;
5:234-242.
45. Costello JF, Plass C. Methylation matters. J Med Genet 2001; 38:285-303.
46. Klimasauskas S, Kumar S, Roberts RJ et al. Hhal Methyltransferase flips its target base out of the
DNA helix. Cell 1994; 76:357-369.
47. Robertson KD, Jones PA. DNA methylation: Past, present and future directions. Carcinogenesis
2000; 21:461-467.
48. Herman JG, Baylin SB. Promoter-region hypermethylation and gene silencing in human cancer.
Curr Top Microbiol Immunol 2000; 249:35-54.
49. Yoder JA, Walsh CP, Bestor TH. Cytosine methylation and the ecology of intragenomic parasites.
Trends Genet 1997; 13:335-340.
50. Bestor TH. DNA methylation: Evolution of a bacterial immune function into a regulator of gene
expression and genome structure in higher eukaryotes. Phil Trans Roy Soc Lond 1990; 326:170-187.
51. Gonzalgo ML, Jones PA. Mutagenic and epigenetic effects of DNA methylation. Mutat Res Rev
1997; 386:107-118.
52. Jones PA, Rideout WM, III Shen J-C et al. Methylation, mutation and cancer. Bio Essays 1992;
14:33-36.
53. Issa JP. Aging. DNA methylation and cancer. Crit Rev Oncol Hematol 1999; 32:31-43.
54. Baylin SB, Belinsky SA, Herman JG. Aberrant methylation of gene promoters in cancer - Con-
cepts, misconcepts, and promise. JNCI 2000; 92:1460-1461.
55. Bestor TH, Tycko B. Creation of genomic methylation patterns. Nature 1996; 12:363-367.
56. Brooks PJ, Marietta C, Goldman D. DNA mismatch repair and DNA methylation in adult brain
neurons. J Neurosci 1996; 16:939-945.
57. Li E, Bestor TH, Jaenisch R. Targeted mutation of the DNA methyltransferase gene results in
embryonic lethality. Cell 1992; 69:915-926.
58. Jackson-Grusby L, Beard C, Possemato R et al. Loss of genomic methylation causes p53-dependent
apoptosis and epigenetic deregulation. Nat Genet 2001; 27:31-39.
59. Bachman KE, Rountree MR, Baylin SB. Dnmt3a and Dnmt3b are transcriptional repressors that
exhibit unique localization properties to heterochromatin. J Biol Chem 2001; 276:32282-32287.
60. Li E. Chromatin modification and epigenetic reprogramming in mammalian development. Nat
Rev Genet 2002; 3:662-673.
61. Tycko B, Trasler J, Bestor T. Genomic imprinting: Gametic mechanisms and somatic consequences.
J Androl 1997; 18:480-486.
62. Robertson KD, Keyomarsi K, Gonzales FA et al. Differential mRNA expression of the human
DNA methyltransferases (DNMTs) 1, 3a and 3b during the G(0)/G(1) to S phase transition in
normal and tumor cells. Nucleic Acids Res 2000; 28:2108-2113.
63. Robertson KD. DNA methylation and chromatin - unraveling the tangled web. Oncogene 2002;
21:5361-5379.
64. Watanabe D, Suetake I, Tada T et al. Stage- and cell-specific expression of Dnmt3a and Dnmt3b
during embryogenesis. Mech Dev 2002; 118:187-190.
65. Urnov F. Methylation and the Genome: The power of the small amendment. J Nutr 2002;
132:2450S-2456S.
66. Rice JC, Allis CD. Histone methylation versus histone acetylation: New insights into epigenetic
regulation. Curr Opin Cell Biol 2001; 13:263-273.
67. Berger SL. Histone modifications in transcriptional regulation. Curr Opin Genet Dev 2002;
12:142-148.
68. Nan X, Ng HH, Johnson CA et al. Transcriptional repression by the methyl CpG-binding protein
MeCP2 involves a histone deacetylase complex. Nature 1998; 393:386-389.
69. Geiman TM, Robertson KD. Chromatin remodeling, histone modifications, and DNA methyla-
tion - How does it all fit together? J Cell Biochem 2002; 87:117-125.
70. Curradi M, Izzo A, Badaracco G et al. Molecular mechanisms of gene silencing mediated by DNA
methylation. Mol Cell Biol 2002; 22:3157-3173.
71. Fuks F, Burgers WA, Brehm A et al. DNA methyltransferase Dnmt1 associates with histone
deacetylase activity. Nat Genet 2000; 24:88-91.
72. Burgers WA, Fuks F, Kouzarides T. DNA methyltransferases get connected to chromatin. Trends
Genet 2002; 18:275-277.
73. Lachner M, Jenuwein T. The many faces of histone lysine methylation. Current Opinion in Cell
Biology 2002; 14:286-298.
74. Nakayama J, Rice JC, Strahl BD et al. Role of histone H3 lysine 9 methylation in epigenetic
control of heterochromatin assembly. Science 2001; 292:110-113.
75. Lachner M, O’Carroll D, Rea S et al. Methylation of histone H3 lysine 9 creates a binding site for
HP1 proteins. Nature 2001; 410:116-120.
76. Tamaru H, Selker EU. A histone H3 methyltransferase controls DNA methylation in Neurospora
crassa. Nature 2001; 414:277-283.
77. Jackson JP, Lindroth AM, Cao X et al. Control of CpNpG DNA methylation by the KRYPTONITE
histone H3 methyltransferase. Nature 2002; 416:556-560.
78. Hoffman DR, Cornatzer WE, Duerre JA. Relationship between tissue levels of S-adenosylmethionine,
S- adenylhomocysteine, and transmethylation reactions. Can J Biochem 1979; 57:56-65.
79. Monk M, Boubelik M, Lehnert S. Temporal and regional changes in DNA methylation in embry-
onic, extraembryonic, and germ cell lineages during mouse embryo development. Development
1987; 99:371-382.
80. Turker MS, Bestor TH. Formation of methylation patterns in the mammalian genome. Mutat Res
Rev 1997; 386:119-130.
81. Heby O. DNA methylation and polyamines in embryonic development and cancer. Int J Dev Biol
1995; 39:737-757.
82. Cedar H, Razin A. DNA methylation and development. Biochim Biophys Acta 1990; 1049:1-8.
83. Razin A, Cedar H. DNA methylation and embryogenesis. In: Jost JP, Saluz HP, eds. DNA Methy-
lation: Biology and Biological Significance. Basal, Switzerland: Birkhauser Verlag, 1993:343-357.
84. Kafri T, Gao X, Razin A. Mechanistic aspects of genome-wide demethylation in the preimplanta-
tion mouse embryo. PNAS 1993; 90:10558-10562.
85. Kafri T, Ariel M, Brandeis M et al. Developmental pattern of gene-specific DNA methylation in
the mouse embryo and germ line. Genes Dev 1992; 6:705-714.
86. Zagris N, Podimatas T. 5-Azacytidine changes gene expression and causes developmental arrest of
early chick embryo. Int J Dev Biol 1994; 38:741-744.
87. Martin CC, Laforest L, Akimenko MA et al. A role for DNA methylation in gastrulation and
somite patterning. Dev Biol 1999; 206:189-205.
88. Kakutani T, Jeddeloh JA, Flowers SK et al. Developmental abnormalities and epimutations associ-
ated with DNA hypomethylation mutations. PNAS 1996; 93:12406-12411.
89. Matsuda M, Yasutomi M. Inhibition of cephalic neural tube closure by 5-azacytidine in neurulating
rat embryos in vitro. Anat Embryol (Berl) 1992; 185:217-223.
90. Alonso-Aperte E, Ubeda N, Achon M et al. Impaired methionine synthesis and hypomethylation
in rats exposed to valproate during gestation. Neurology 1999; 52:750-756.
91. Tachibana M, Sugimoto K, Nozaki M et al. G9a histone methyltransferase plays a dominant role
in euchromatic histone H3 lysine 9 methylation and is essential for early embryogenesis. Genes
and Development 2002; 16:779-1791.
92. Razin A, Shemer R. DNA methylation in early development. Hum Mol Genet 1995; 4:1751-1755.
93. Bestor TH. The DNA methyltransferases of mammals. Hum Mol Genet 2000; 9:2395-2402.
94. Jhaveri MS, Wagner C, Trepel JB. Impact of extracellular folate levels on global gene expression.
Mol Pharmacol 2001; 60:1288-1295.
95. Wickramasinghe SN, Hoffbrand AV. Reduced rate of replication fork movement in megaloblastic
anemia. J Clin Invest 1980; 65:26-30.
96. James SJ, Miller BJ, Basnakian AG et al. Apoptosis and proliferation under conditions of
deoxynucleotide pool imbalance in liver of folate/methyl deficient rats. Carcinogenesis 1997;
18:287-293.
97. Chern CL, Huang RF, Chen YH et al. Folate deficiency-induced oxidative stress and apoptosis are
mediated via homocysteine-dependent overproduction of hydrogen peroxide and enhanced activa-
tion of NF-kappaB in human Hep G2 cells. Biomed Pharmacother 2001; 55:434-442.
98. Finnell RH, Spiegelstein O, Wlodarczyk B et al. DNA methylation in Folbp1 knockout mice
supplemented with folic acid during gestation. J Nutr 2002; 132:2457S-2461S.
99. Burgoon JM, Selhub J, Nadeau M et al. Investigation of the effects of folate deficiency on embry-
onic development through the establishment of a folate deficient mouse model. Teratology 2002;
65:219-227.
100. Wilson VL, Smith RA, Ma S et al. Genomic 5-methyldeoxycytidine decreases with age. J Biol
Chem 1987; 262:9948-9951.
101. Tra J, Kondo T, Lu Q et al. Infrequent occurrence of age-dependent changes in CpG island
methylation as detected by restriction landmark genome scanning. Mech Ageing Dev 2002;
123:1487-1503.
102. Issa J.-PJ, Ottaviano YL, Celano P et al. Methylation of the oestrogen receptor CpG island links
ageing and neoplasia in human colon. Nat Genet 1994; 7:536-540.
103. Ahuja N, Li Q, Mohan AL et al. Aging and DNA methylation in colorectal mucosa and cancer.
Cancer Res 1998; 58:5489-5494.
104. Issa JP. Aging, DNA methylation and cancer. Crit Rev Oncol Hematol 1999; 32:31-43.
105. Pogribny IP, Miller BJ, James SJ. Alterations in hepatic p53 gene methylation patterns during
tumor progression with folate/methyl deficiency in the rat. Cancer Lett 1997; 115:31-38.
106. Pogribny IP, James SJ. Reduction of p53 gene expression in human primary hepatocellular carci-
noma is associated with promoter region methylation without coding region mutation. Cancer Lett
2002; 176:169-174.
107. Bozzi A, Furciniti-La Chiusa B, Strom R et al. S-adenosylhomocysteine hydrolase and adenosine
deaminase activities in human red cell ageing. Clin Chim Acta 1990; 189:81-86.
108. Varela-Moreiras G, Perez-Olleros L, Garcia-Cuevas M et al. Effects of ageing on folate metabolism
in rats fed a long-term folate deficient diet. Int J Vitam Nutr Res 1994; 64:294-299.
109. Morrison LD, Smith DD, Kish SJ. Brain S-adenosylmethionine levels are severly decreased in
Alzheimer’s disease. J Neurochem 1996; 67:1328-1331.
110. Payao SL, Smith MD, Bertolucci PH. Differential chromosome sensitivity to 5-azacytidine in
Alzheimer’s disease. Gerontology 1998; 44:267-271.
111. West RL, Lee JM, Maroun LE. Hypomethylation of the amyloid precursor protein gene in the
brain of an Alzheimer’s disease patient. J Mol Neurosci 1995; 6:141-146.
112. Seshadri S, Beiser A, Selhub J et al. Plasma homocysteine as a risk factor for dementia and
Alzheimer’s disease. N Engl J Med 2002; 346:476-483.
113. Clarke R, Smith AD, Jobst KA et al. Folate, vitamin B12, and serum total homocysteine levels in
confirmed Alzheimer disease. Arch Neurol 1998; 55:1449-1455.
114. Miller JW, Green R, Mungas DM et al. Homocysteine, vitamin B6, and vascular disease in AD
patients. Neurology 2002; 58:1471-1475.
115. Clinical Synopsis, ICF Syndrome. OMIM (Online Mendelian Inheritance in Man) 1998.
116. Hassan KMA, Norwood T, Gimelli G et al. Satellite 2 methylation patterns in normal and ICF
syndrome cells and association of hypomethylation with advanced replication. Hum Genet 2001;
109:452-462.
117. Xu GL, Bestor TH, Bourc’hisD et al. Chromosome instability and immunodeficiency syndrome
caused by mutations in a DNA methyltransferase gene. Nature 1999; 402:187-191.
118. Hansen RS, Wijmenga C, Luo P et al. The DNMT3B DNA methyltransferase gene is mutated in
the ICF immunodeficiency syndrome. PNAS 1999; 96:14412-14417.
119. Hansen RS, Stöger R, Wijmenga C et al. M. Escape from gene silencing in ICF syndrome: Evi-
dence for advanced replication time as a major determinant. Hum Mol Genet 2000; 9:2575-2587.
120. Tao Q, Huang H, Geiman TM et al. Defective de novo methylation of viral and cellular DNA
sequences in ICF syndrome cells. Hum Mol Genet 2002; 11:2091-2102,.
121. Amir RE, Van dV, I Wan M et al. Rett syndrome is caused by mutations in X-linked MECP2,
encoding methyl-CpG-binding protein 2. Nat Genet 1999; 23:185-188.
122. Wan M, Lee SS, Zhang X et al. Rett syndrome and beyond: Recurrent spontaneous and familial
MECP2 mutations at CpG hotspots. Am J Hum Genet 1999; 65:1520-1529.
123. Amir RE, Zoghbi HY. Rett syndrome: Methyl-CpG-binding protein 2 mutations and
phenotype-genotype correlations. Am J Med Genet 2000; 97:147-152.
124. Willard HF, Hendrich BD. Breaking the silence in Rett syndrome. Nat Genet 1999; 23:127-128.
125. Knight SJ, Flannery AV, Hirst MC et al. Trinucleotide repeat amplification and hypermethylation
of a CpG island in FRAXE mental retardation. Cell 1993; 74:127-134.
126. Sandberg G, Schalling M. Effect of in vitro promoter methylation and CGG repeat expansion on
FMR-1 expression. Nucleic Acids Res 1997; 25:2883-2887.
127. Godde JS, Kass SU, Hirst MC et al. Nucleosome assembly on methylated CGG triplet repeats in
the fragile X mental retardation gene 1 promoter. J Biol Chem 1996; 271:24325-24328.
128. Glover TW. Instability at chromosomal fragile sites. Recent Results Cancer Res 1998; 154:185-199.
129. Glover TW, Stein CK. Chromosome breakage and recombination at fragile sites. Am J Hum Genet
1988; 43:265-273.
130. Sutherland GR, Richards RI. The molecular basis of fragile sites in human chromosomes. Curr
Opin Genet Dev 1995; 5:323-327.
131. Coquelle A, Pipiras E, Toledo F et al. Expression of fragile sites triggers intrachromosomal mam-
malian gene amplification and sets boundaries to early amplicons. Cell 1997; 89:215-225.
132. Popescu NC. Chromosome fragility and instability in human cancer. Crit Rev Oncogenesis 1994;
5:121-140.
133. Djalali M, Adolph S, Steinbach P et al. Fragile sites induced by 5-azacytidine and 5-azadeoxycytidine
in the murine genome. Hereditas 1990; 112:77-81.
134. Mangelsdorf M, Ried K, Woollatt E et al. Chromosomal fragile site FRA16D and DNA instability
in cancer. Cancer Res 2000; 60:1683-1689.
135. Wolff GL, Kodell RL, Moore SR et al. Maternal epigenetics and methyl supplements affect agouti
gene expression in Avy/a mice. FASEB J 1998; 12:949-957.
136. Michaud EJ, van Vugt MJ, Bultman SJ et al. Differential expression of a new dominant agouti
allele (Aiapy) is correlated with methylation state and is influenced by parental lineage. Genes Dev
1994; 8:1463-1472.
137. Cooney CA, Dave AA, Wolff GL. Maternal methyl supplements in mice affect epigenetic variation
and DNA methylation of offspring. J Nutr 2002; 132:2393S-2400S.
138. Al Essa MA, Al Amir A, Rashed M et al. Clinical, fluorine-18 labeled 2-fluoro-2-deoxyglucose
positron emission tomography of the brain, MR spectroscopy, and therapeutic attempts in
methylenetetrahydrofolate reductase deficiency. Brain Dev 1999; 21:345-349.
139. Heath Jr CW. Cytogenetic observations in vitamin B12 and folate deficiency. Blood 1966;
27:800-815.
140. Fuster C, Miro R, Barrios L et al. Induction of premature centromere division affecting all chro-
mosomes under culture conditions of fragile site expression. Cancer Genet Cytogenet 1992;
58:152-154.
141. Pogribny IP, Muskhelishvili L, Miller BJ et al. Presence and consequence of uracil in preneoplastic
DNA from folate/methyl deficient rats. Carcinogenesis 1997; 18:2071-2076.
143. Krumdieck CL. Role of folate deficiency in carcinogenesis. In: Butterworth CE, Hutchenson ML,
eds. Nutritional factors in the induction and maintenance of malignancy, New York: Academic
Press, 1983:225-245.
144. Kunz BA. Genetic effects of deoxyribonucleotide pool imbalances. Env Mut 1982; 4:695-725.
145. Everson RB, Wehr C, Erexson GL et al. Association of marginal folate depletion with increased
human chromosomal damage in vivo: Demonstration by analysis of micronucleated erythrocytes.
JNCI 1988; 80:525-529.
146. MacGregor JT. Dietary factors affecting spontaneous chromosomal damage in man. Prog Clin Biol
Res 1990; 347:139-153.
147. Titenko-Holland N, Jacob RA, Shang N et al. Micronuclei in lymphocytes and exfoliated buccal
cells of postmenopausal women with dietary changes in folate. Mutat Res Genet Toxicol Environ Mu-
tagen 1998; 417:101-114.
148. James SJ, Pogribna M, Pogribny IP et al. Abnormal folate metabolism and mutation in the
methylenetetrahyrofolate reductase (MTHFR) gene may be maternal risk factors for Down Syn-
drome. Am J Clin Nutr 1999; 70:495-501.
149. Hobbs CA, Sherman SL, Yi P et al. Polymorphisms in genes involved in folate metabolism as
maternal risk factors for Down syndrome. Am J Hum Genet 2000; 67:623-630.
150. O’Leary VB, Parle-McDermott A, Molloy AM et al. MTRR and MTHFR polymorphism: Link to
Down syndrome? Am J Med Genet 2002; 107:151-155.
151. Chadefaux-Vekemans B, Coude M, Muller F et al. Methylenetetrahydrofolate reductase polymor-
phism in the etiology of Down syndrome. Pediatr Res 2002; 51:766-767.
152. Stuppia L, Gatta V, Gaspari AR et al. C677T mutation in the 5,10-MTHFR gene and risk of
Down syndrome in Italy. Eur J Hum Genet 2002; 10:388-390.
153. James SJ, Cross DR, Miller BJ. Alterations in nucleotide pools in rats fed diets deficient in cho-
line, methionine and/or folic acid. Carcinogenesis 1992; 13:2471-2474.
154. Pogribny IP, Basnakian AG, Miller BJ et al. Breaks in genomic DNA and within the p53 gene are
associated with hypomethylation in livers of folate/methyl deficient rats. Cancer Res 1995;
55:1894-1901.
155. Fenech M. The role of folic acid and Vitamin B12 in genomic stability of human cells. Mutation
Research 2001; 475:57-67.
156. Branda RF, Lafayette AR, O’Neill JP et al. Effect of folate deficiency on mutations at the hprt
locus in Chinese hamster ovary cells exposed to monofunctional alkylating agents. Cancer Res
1997; 57:2586-2588.
157. Mishina Y, Ayusawa D, Seno T et al. Thymidylate stress induces homologous recombination activ-
ity in mammalian cells. Mut Res 1991; 246:215-220.
158. Balaghi M, Wagner C. DNA Methylation in Folate Deficiency - Use of CpG Methylase. Biochem
Biophys Res Commun 1993; 193:1184-1190.
128:1204-1212.
160. Cravo M, Mason J, Saloman RN et al. Folate deficiency in rats causes hypomethylation of DNA.
FASEB J 1991; 5:A914.
161. James SJ, Miller BJ, McGarrity LJ et al. The effect of folic acid and/or methionine deficiency on
deoxyribonucleotide pools and cell cycle distribution in mitogen-stimulated rat lymphocytes. Cell
Proliferation 1994; 27:395-406.
162. Koury MJ, Horne DW, Brown ZA et al. Apoptosis of late-stage erythroblasts in megaloblastic
anemia: Association with DNA damage and macrocyte production. Blood 1997; 89:4617-4623.
163. James SJ, Basnakian AG, Miller BJ. In vitro folate deficiency induces deoxynucleotide pool imbal-
ance, apoptosis, and mutagenesis in Chinese hamster ovary cells. Cancer Res 1994; 54:5075-5080.
164. Glynn SA, Albanes, D. Folate and cancer: A review of the literature. Nutr Cancer 1994; 22:101-119.
165. Giovannucci E, Rimm EB, Ascherio A et al. Alcohol, low-methionine-low-folate diets, and risk of
colon cancer in men. JNCI 1995; 87:265-273.
166. Cravo ML, Mason JB, Dayal Y et al. Folate deficiency enhances the development of colonic neo-
plasia in dimethylhydrazine-treated rats. Cancer Res 1992; 52:5002-5006.
167. Wilson MJ, Shivapurkar N, Poirier LA. Hypomethylation of hepatic nuclear DNA in rats fed with
a carcinogennic methyl-deficient diet. Biochem J 1984; 218:987-990.
168. Newberne PM, Rogers AE. Labile methyl groups and the promotion of cancer. Ann Rev Nutr
1986; 6:407-432.
169. Lashner BA, Provencher KS, Seidner DL et al. The effect of folic acid supplementation on the risk
for cancer or dysplasia in ulcerative colitis. Gastroenterology 1997; 112:29-32.
170. Choi SW, Mason JB. Folate and carcinogenesis: An integrated scheme. J Nutr 2000; 130:129-132.
171. Khosraviani K, Weir HP, Hamilton P et al. Effect of folate supplementation on mucosal cell pro-
liferation in high risk patients for colon cancer. Gut 2002; 51:195-199.
172. Ryan BM, Weir DG. Relevance of folate metabolism in the pathogenesis of colorectal cancer. J
Lab Clin Med 2001; 138:164-176.
173. Ames BN. Micronutrients prevent cancer and delay aging. Toxicol Lett 1998; 103:5-18.
174. Baggott JE, Vaughn WH, Juliana M et al. Effects of folate deficiency and supplementation on
methylnitrosourea-induced rat tumors. JNCI 1992; 84:1740-1744.
175. Sibani S, Melnyk S, Pogribny IP et al. Studies of methionine cycle intermediates (SAM, SAH),
DNA methylation and the impact of folate deficiency on tumor numbers in Min mice. Carcino-
genesis 2002; 23:61-65.
176. Song J, Medline A, Mason JB et al. Effects of dietary folate on intestinal tumorigenesis in the
Apcfs mouse. Cancer Res 2000; 60:5434-5440.
177. Baylin SB, Herman JG, Graff JR et al. Alterations in DNA methylation: A fundamental aspect of
neoplasia. Adv Cancer Res 1997; 72:141-196.
178. Saito Y, Kanai Y, Sakamoto M et al. Expression of mRNA for DNA methyltransferases and
methyl-CpG-binding proteins and DNA methylation status on CpG islands and pericentromeric
satellite regions during human hepatocarcinogenesis. Hepatol 2001; 33:561-568.
179. Bestor TH, Tycko B. Creation of genomic methylation patterns. Nat Genet 1996; 12:363-367.
180. Soares J, Pinto AE, Cunha CV et al. Global DNA hypomethylation in breast carcinoma - Correla-
tion with prognostic factors and tumor progression. Cancer 1999; 85:112-118.
181. Robertson KD. DNA methylation, methyltransferases, and cancer. Oncogene 2001; 20:3139-3155.
182. Mikol YB, Hoover KL, Creasia D et al. Hepatocarcinogenesis in rats fed methyl-deficient, amino
acid-defined diets. Carcinogenesis 1983; 4:1619-1629.
183. Ghoshal AK, Farber E. The induction of liver cancer by a dietary deficiency of choline and me-
thionine without added carcinogens. Cancer Res 1984; 5:1367-1370.
184. Rogers AE, Zeisel SH, Groopman J. Diet and Carcinogenesis. Carcinogenesis 1993; 14:2205-2217.
185. Copeland DH, Salmon WD. The occurrance of neoplasms in the liver, lungs, and other tissues of
the rat as a result of prolonged choline deficiency. Am J Pathol 1946; 22:1059-1079.
186. Lombardi B, Chandar N, Locker J. Nutritional model of carcinogenesis: Rats fed a choline devoid
diet. Dig Dis Sci 1991; 36:979-984.
187. Nakae D, Yoshiji H, Mizumoto Y et al. High incidence of hepatocellular carcinomas induced by a
choline deficient L-amino acid defined diet in rats. Cancer Res 1992; 52:5042-5045.
188. Rogers AE. Methyl donors in the diet and responses to chemical carcinogens. Am J Clin Nutr
61(Suppl):659S-665S 1995.
189. Horne DE, Cook RJ, Wagner C. Effect of dietary methyl group deficiency on folate metabolism
in rats. J.Nutr 1989; 119:618-621.
190. Aoyama Y, Yasui H, Ashida K. Effect of dietary protein and amino acids in a choline-deficient diet
on lipid accumulation in rat liver. J Nutr 1971; 101:730-746.
191. Shivapurkar N, Poirier LA. Tissue levels of S-adenosylhomocysteine in rats fed methyl-deficient
amino acid-defined diets for one to five weeks. Carcinogenesis 1983; 4:1051-1057.
192. Henning SM, McKee RW, Swendseid ME. Hepatic content of S-adenosylmethionine,
S-adenosylhomocysteine and glutathione in rats receiving treatments modulating methyl donor avail-
ability. J Nutr 1989; 119:1478-1482.
193. Duthie SJ, Narayanan S, Brand GM et al. DNA stability and genomic methylation status in
colonocytes isolated from methyl-donor-deficient rats. Eur J Nutr 2000; 39:106-111.
194. Trasler J, Deng L, Melnyk S et al. Impact of dnmt1 deficiency, with and without low folate diets,
on tumor numbers and methylation in min mice. Carcinogenesis 2003; 24:39-45.
195. Song J, Sohn KJ, Medline A et al. Chemopreventive effects of dietary folate on intestinal polyps in
Apc+/-Msh2-/- mice. Cancer Res 2000; 60:3191-3199.
196. Pogribny IP, Pogribna M, Christman JK et al. Single-site methylation within the p53 promoter
region reduces gene expression in a reporter gene construct: Possible in vivo relevance during tum-
origenesis. Cancer Res 2000; 60:588-594.
197. Pogribny IP, James SJ. De novo methylation of the p16 gene in early preneoplastic liver and
tumors induced by folate/emthyl deficiency in rats. Cancer Lett 2002; 176:169-174.
198. Klimasauskas S, Roberts RJ. M.HhaI binds tightly to substrates containing mismatches at the tar-
get site. Nucl Acids Res 1995; 23:1388-1395.
200. Skibola CF, Smith MT, Kane E et al. Polymorphisms in the methylenetetrahydrofolate reductase
gene are associated with susceptibility to acute leukemia in adults. PNAS 1999; 96:12810-12815.
CHAPTER 8
677C→T Polymorphism and Risk of Arterial
Occlusive Disease
Mariska Klerk and Petra Verhoef
Introduction
M
oderately elevated plasma levels of homocysteine are associated with increased risk
of arterial occlusive disease (AOD), but whether this association is causal is uncertain.
Retrospective studies on AOD have consistently reported that cases have higher
homocysteine levels than age-matched controls.1 However, the weaker associations between
homocysteine and risk of AOD in prospective studies,2,3 have led to uncertainty about the
relevance of moderately elevated homocysteine levels for cardiovascular risk. Some researchers
have argued that elevated homocysteine levels might only be a marker of subclinical atheroscle-
rotic disease, including impaired renal function, or a consequence of the effects of other known
cardiovascular risk factors rather than being a cause of cardiovascular disease (CVD).4
Investigation of the association between a polymorphism that causes high levels of ho-
mocysteine—such as the MTHFR 677C→T polymorphism—and risk of AOD could poten-
tially reveal whether homocysteine plays a causal role in the occurrence of AOD risk. The
advantage of studying the relationship between a polymorphism—instead of plasma concen-
trations of homocysteine—and risk of AOD is that the metabolic effects of a particular geno-
type start early in life. Therefore, the level of the risk factor—in this case homocysteine—
cannot have been caused solely by the presence of the disease. Comparison of AOD risk in
individuals with none, one, or two copies of the allele that results in increased homocysteine
concentrations can be interpreted as an experiment of nature in which individuals have been
randomly allocated to high or low homocysteine concentrations. Furthermore, due to so-called
Mendelian randomization, bias and confounding are theoretically absent, even in case-control
studies.5
As described in an earlier chapter, methylenetetrahydrofolate reductase (MTHFR) is an
enzyme that converts 5,10-methylenetetrahydrofolate to 5-methyltetrahydrofolate, which is
required for the conversion of homocysteine to methionine. Individuals who have a C→T
substitution at base 677 of the cDNA (amino acid change A222V) have reduced enzyme activ-
ity.6 The prevalence of the TT genotype varies between 3%7 and 30%8 in healthy controls of
different populations. Subjects with the TT genotype have on average 2.5 µmol/L higher ho-
mocysteine levels than those with the CC genotype.9 In addition, several studies have shown
that TT subjects have lower circulating folate levels than CC subjects.10-13
Individual studies investigating the association between the MTHFR 677C→T polymor-
phism and AOD risk have shown inconsistent results, and have lacked the statistical power to
estimate a small relative risk. In addition, they have not been able to test the hypothesis that the
©2005 Eurekah.com.
MTHFR 677C→T Polymorphism and Arterial Disease 101
MTHFR 677TT genotype might emerge as a risk factor for AOD mainly in populations with
low-normal folate intake.14
In this chapter we will give an overview of several meta-analyses that investigated the asso-
ciation between the MTHFR 677C→T polymorphism and risk of coronary heart disease and
stroke. Possible effect modification by folate status and other factors, and various types of bias
will be taken into account. The association between the MTHFR 677C→T polymorphism
and venous thrombosis will be discussed in the next chapter. The number of studies including
peripheral atherosclerosis is limited and will therefore not be discussed.
The MTHFR 677C→T Polymorphism and Risk of CHD

In 1998, Brattstrom et al9 published the results of a meta-analysis of 23 available studies
that investigated the association between the MTHFR 677C→T polymorphism and cardio-
vascular disease (CVD) risk. The study comprised 5869 CVD cases (mostly coronary heart
disease) and 6644 controls. Data on plasma homocysteine concentrations were available for 13
studies. The summary OR of CVD was 1.12 (95%Cl 0.92 to 1.37). The summary OR of 17
studies that included only patients with coronary heart disease (CHD) was 1.11 (95%Cl 0.91
to 1.37). The authors concluded that the MTHFR 677C→T polymorphism did not increase
CVD risk, and suggested that the increase in homocysteine found in cardiovascular patients
was just an epiphenomenon. However, based on results of several meta-analyses of prospective
studies, an increase in homocysteine of 2.6 µmol/L (which was the observed difference in
homocysteine between the CC and TT genotype) would be expected to be associated with a
10-15% increase in CHD risk.2,3,15,16 Therefore, the conclusion of Brattstrom et al was prob-
ably not justified, as the estimate they found was similar to what would be expected. The
meta-analysis included too few subjects to provide conclusive evidence for or against an asso-
ciation of this polymorphism and CVD risk. Therefore, the outcome of this meta-analysis
cannot be considered as evidence against a causal role of homocysteine.
Recently we conducted a larger meta-analysis by pooling individual participant data from
all case-control studies with data on the MTHFR 677C→T polymorphism and risk of CHD.17
We aimed to have sufficient statistical power to estimate an odds ratio of CHD for the TT vs.
the CC genotype in the range of 1.10-1.15. In addition, we wanted to explore possible effect
modification by folate status or CHD risk factors. Eligible studies were identified by searching
the electronic literature (Medline and Current Contents) for relevant reports published before
June 2001 (using the search-terms MTHFR and coronary heart disease), by hand-searching
reference lists of original studies and review articles (including meta-analyses) on this topic,
and by personal contact with investigators in the field. Studies were included if they had data
on the MTHFR 677C→T genotype, had a case-control design (retrospective or nested
case-control), and involved CHD as an endpoint (angiographically-confirmed stenosis or myo-
cardial infarction). In addition, we collected data on plasma levels of homocysteine, folate, and
other cardiovascular risk factors, if available. Among a total of 53 published studies that exam-
ined the relation between the MTHFR 677C→T polymorphism and risk of CHD risk, 6
studies were not included, because they did not have a proper case-control design.18-22 or stud-
ied cardiovascular mortality23 only. Data on 13 other studies were unavailable because the inves-
tigators were unable or unwilling to collaborate.24-36 Eventually, data were available for these
analyses from 40 studies (34 published7,8,12,13,37-65 and 6 studies that had not previously reported
on the MTHFR 677C→T polymorphism66-69), which involved 11,162 cases and 12,758 con-
trols (see Fig. 1). About half the data came from studies of European populations and about a
quarter from those of North American populations. Figure 1 shows the OR and 95% confidence
interval of CHD for the TT genotype compared with the CC genotype in individual studies, and
a summary estimate for the combined analysis of all studies. Overall, individuals with the TT
genotype had a significantly higher risk of CHD compared with individuals with the CC geno-
type (OR 1.16 ; 95% Cl 1.05 to 1.28). There was a trend towards an increased risk, for the CT
genotype compared with the CC genotype (OR 1.04; 95% Cl 0.98 to 1.10).
Figure 1. The odds ratio (and 95% Cl) of CHD for MTHFR 677 TT vs CC genotype by region of origin.
The size of the squares is inversely proportional to the variance of the log OR and the horizontal lines
represent the 95% Cl. Studies are ordered by the number of cases in each region. The combined OR and
the sub-totals for each region and their 95% Cl are indicated by the diamonds. Figure used with permission
from: Klerk M, Verhoef P, Clark R et al. MTHFR 677C>T polymorphism and risk of coronary heart disease:
a meta-analysis. JAMA 2002; 288:2023-2031. © 2002, American Medical Association. All rights reserved.
Our results were confirmed by Wald et al16 who performed a meta-analysis based on pub-
lished OR’s of 46 studies on the MTHFR 677C→T polymorphism and risk for CHD. The
study comprised 12,193 cases and 11,945 controls. Data on homocysteine concentrations were
available from 33 studies. They found an OR of 1.21 (95%Cl 1.06 to 1.39) for the TT geno-
type and 1.06 (95%Cl 0.99-1.13) for the CT genotype compared with the CC genotype.
In summary, the risk estimates of CHD for the TT versus the CC genotype in the three
meta-analyses are quite similar; this is a logical conclusion since the analyses included essen-
tially the same studies. They are also in line with the expected risk based on homocysteine
differences between CC and TT. In the report of Wald et al16 and in our meta-analysis,17 the
association was statistically significant.
The MTHFR 677C→T Polymorphism and Risk of Stroke

There are fewer publications on the association between the MTHFR 677C→T polymor-
phism and risk of stroke than publications on CHD. The meta-analysis of Wald et al16 in-
cluded 7 studies (1,217 events).30,70-75 Stroke was defined as cerebral infarction, cerebral
haemorrhage or transient ischaemic attack (TIA) confirmed by neuroimaging. The pooled OR
of stroke was 1.31 (95%Cl 0.80 to 2.15) for the TT genotype and 1.15 (95%Cl 0.93-1.42) for
the CT genotype compared with the CC genotype. Thus, the risk estimate for stroke for the
TT versus the CC genotype seems to be slightly stronger than for CHD. However, the confi-
dence intervals are quite wide. The summary estimate seems to be driven mainly by two studies
with quite a strong OR (3.1973 and 4.50,74 respectively). The other 5 studies did not find any
association between the TT genotype and stroke risk.
In another meta-analysis, Kelly et al76 included 19 studies (2,788 cases and 3,962 controls).
The pooled estimate of the risk for stroke associated with the TT genotype was 1.23 (95% Cl
0.96 to 1.58). This finding confirms the result of Wald et al that the TT genotype is associated
with increased risk of stroke. However, they used CC + CT genotypes combined as the refer-
ence group, and they mixed crude and adjusted OR’s. Therefore, the result is not completely
comparable with the result of Wald et al.
In summary, the association between the MTHFR 677TT genotype and stroke was slightly
stronger than that for CHD, although the risk estimates were driven mainly by a few studies.
Generally, however, there seems to be a positive association between the MTHFR 677TT geno-
type and risk of stroke.
Evidence for a Causal Role of Homocysteine in Occurrence of AOD

The results of the meta-analyses on the MTHFR 677C→T polymorphism and AOD are
summarized in Table 1. The findings show that the MTHFR 677TT genotype is not a strong
risk factor for CHD, nor for stroke. However, the results of these meta-analyses support the
hypothesis that impaired folate metabolism leading to high homocysteine levels is causally
involved in the occurrence of AOD. From Table 1, it can also be concluded that the magnitude
of the risk associated with the TT genotype is consistent with the magnitude of risk associated
with high homocysteine levels as observed in two recent meta-analyses of prospective studies
on homocysteine and ischemic heart disease (IHD) or stroke (Table 1). All together these
results suggest that a 25% (≈3 µmol/L) reduction in homocysteine, which is about the maxi-
mal effect that can be achieved by folic acid supplementation,77 would decrease risk of CHD/
IHD by 10-15% and risk of stroke by 20-25%.
Possible Explanations for Study Heterogeneity

In the various meta-analyses, there was quite some heterogeneity between individual studies
in the observed risk estimates of CHD for the MTHFR 677TT genotype compared with the
CC genotype. In the following sections, we discuss several factors that might explain why some
studies found clear positive associations between the TT genotype and CHD risk, whereas
others did not.
Table 1. Overview of results of the meta-analyses on the MTHFR 677C→T

polymorphism and homocysteine in relation to CHD or stroke
Difference in
Homocysteine Risk Reduction per
Summary OR between TT and 25% Reduction
No. of No. of for TT vs CC Genotype in Homocysteine
Studies Cases CC (95%Cl) (µmol/L) (%)**
CHD (IHD)
MTHFR
Brattstrom et al 17 4,694 1.11 (0.91 – 1.37) 2.6 11
Klerk et al 40 11,162 1.16 (1.05 – 1.28) 1.8* / 2.8 22* / 15
Wald et al 46 12,193 1.21 (1.06 – 1.39) 2.7 19
Homocysteine
HSC 11 1,968 11*
Wald et al 16 3,144 15
Stroke
MTHFR
Wald et al 7 1,217 1.31 (0.80 – 2.15) 2.7 26
Homocysteine
HSC 8 463 19*
Wald et al 8 676 24
HSC = Homocysteine Studies Collaboration; * based on log-transformed homocysteine concentration;
** 25% reduction in homocysteine ≈ 3 µmol/L. For the meta-analyses on MTHFR, the OR per 25%
reduction in homocysteine was calculated by the following equation: 1/(ORTTvsCC3/d); where d = the
observed difference in homocysteine concentration between CC and TT.
Effect Modification by Folate Status

In our meta-analysis, continent-specific analyses showed that CHD risk was significantly
increased for individuals with the TT genotype, compared with those with the CC genotype,
in Europe (OR 1.14; 95% Cl 1.01 to 1.28), but not in North America (OR 0.87; 95% Cl 0.73
to 1.05).17 A difference in nutritional intake of folate, the substrate of MTHFR, is likely to
account for part of this heterogeneity.3,42,78 Several studies have shown that the MTHFR
677C→T polymorphism is only associated with high homocysteine levels or increased CHD
risk when folate status is low.12,13,40,42,65,79,80 Hence, at higher dietary intakes of folate, the
MTHFR 677C→T genotype will not be associated with increased plasma homocysteine levels,
and thus the difference in homocysteine concentration between genotypes will be smaller than
that with low folate intake.
The average use of vitamin supplements has been consistently higher for several years in
North America (25-40%)55,81-84 than in Europe (5-15%).66 Furthermore,85 even though the
North American studies were carried out before the required food fortification with folate in
1998, fortification of some breakfast cereals had been introduced several years before this time.
Therefore, we expected that the difference in both homocysteine and folate concentrations
between the TT and CC genotype would be smaller in North American populations than in
European populations. Our meta-analysis revealed that this was indeed the case. In North
American studies, the difference between the TT and CC genotype was 1.3 µmol/L for ho-
mocysteine and 1.7 nmol/L for folate concentration, while in European studies these differ-
ences were 2.5 µmol/L for homocysteine and 2.1 nmol/L for folate concentration, respectively.
Subsequently, we explored possible effect modification by folate status by calculating ORs
of CHD within strata of the MTHFR 677C→T genotype and folate status (Table 2). The data
Table 2. ORs (95% Cl) of coronary heart disease by strata of MTHFR 677C→T
polymorphism and folate status
MTHFR 677C→T
CC CT TT
Cases/controls (N) 1543 / 2180 1355 / 1847 364 / 445

High folate status 1.00† 0.91 (0.78 – 1.06) 0.99 (0.77 – 1.29)
Low folate status 1.24 (1.06 – 1.44) 1.32 (1.13 – 1.54) 1.44 (1.12 – 1.83)
*Studies that had data on folate status were: 4, 11, 12, 37, 38, 42, 43, 44, 46, 52, 54, 55, 57, 58, 61,
62, 67, Ferrer 2 and Hopkins (unpublished). Folate status was defined as below or above the median
serum/plasma folate per continent. † Reference category. Table used with permission from: Klerk M,
Verhoef P, Clark R et al. MTHFR C677C>T polymorphism and risk of coronary heart disease: a
meta-analysis. JAMA 2002; 288:2023-2031. ©2002, American Medical Association. All rights
reserved.
suggest a trend of increasing CHD risk with increasing numbers of T alleles. However, we
think that these results should be interpreted with caution, since they are based on only part of
the data, and there might be misclassification of folate status because of the different assays
among the studies. Therefore, the absolute estimates might not be completely valid. Nonethe-
less, these results, in combination with other clinical and biochemical observations, suggest
that the MTHFR genotype may be a risk factor predominantly when folate status is low.
Effect Modification by CVD Risk Factors

Some researchers have suggested that the difference in strength of the associations between
studies could also be explained through effect modification by factors other than folate sta-
tus.63,78,86,87 Two studies showed a synergistic effect of the MTHFR 677TT and classical CHD
risk factors.63,86 Another study, however, showed that the risk of CHD conferred by the MTHFR
677TT genotype was larger in cases without than with other CHD risk factors.87 One study
suggested that the polymorphism is only a risk factor in young people with few or no other risk
factors, whereas among older people the effect is presumably swamped by other risk factors.20
Furthermore, it has been suggested that the strong positive association between the MTHFR
677C→T polymorphism and CHD found in Japanese studies might be explained by the gen-
erally lower cholesterol levels and lower BMI in these populations compared with Western
countries.78
In our pooled meta-analysis, we explored whether differences in CHD risk profile between
European and North American populations might have contributed to the observed heteroge-
neity with respect to the association between the MTHFR 677C→T polymorphism and CHD
risk among the studies for which these data were available. We hypothesized that a certain
factor could only explain the difference in the association between 677C→T polymorphism
and CHD risk between Europe and North America if three criteria were fulfilled: (1) there
should be a difference between European and North American studies with regard to preva-
lence or mean level of that factor, (2) the factor should be an effect modifier of the association
between the MTHFR 677C→T polymorphism and CHD risk, and (3) the direction of the
effect modification and the difference under 1 should, in combination, be a logical explanation
of observed difference in association between the MTHFR 677C→T polymorphism and CHD
risk between the continents.
Table 3 shows the ORs of CHD for the TT versus the CC genotype within strata of CHD
risk factors. TT subjects without hypercholesterolemia had increased risk of CHD, whereas
TT subjects with hypercholesterolemia did not. However, since the prevalence of hypercholes-
terolemia was about 1.5 times higher in Europe compared with North America, the observed
Table 3. OR of CHD for TT vs. CC within strata of CHD risk factors
Cases/Controls OR 95% Confidence Interval
Age (years)
≤ 55 4233 / 6203 1.22 1.06 – 1.41
> 55 4771 / 4180 1.13 0.98 – 1.31
BMI (kg/m2)
≤ 25 1480 / 2216 1.09 0.87 – 1.37
> 25 2582 / 2267 1.05 0.86 – 1.28
Sex
Men 7926 / 7396 1.17 1.05 – 1.31
Women 1704 / 3310 1.24 1.01 – 1.52
Hypertension
No 4114 / 6809 1.12 0.98 – 1.28
Yes 3088 / 1555 1.14 0.92 – 1.42
Hypercholesterolemia
No 4600 / 6761 1.22 1.07 – 1.38
Yes 1877 / 1313 1.02 0.78 – 1.32
Diabetes
No 5829 / 7765 1.08 0.96 – 1.22
Yes 1030 / 379 0.90 0.59 – 1.36
Smoking
No 3959 / 5838 1.09 0.95 – 1.26
Yes 2518 / 2198 1.14 0.93 – 1.39
Alcohol
No 841 / 1080 0.78 0.56 – 1.09
Yes 1735 / 2736 1.01 0.82 – 1.26
interaction between the MTHFR 677C→T polymorphism and hypercholesterolemia could

not explain the higher OR of CHD observed in Europe. A slight positive association between
the TT genotype and CHD risk was observed in subjects without diabetes, whereas there was
a slight inverse association in subjects with diabetes. In addition, there was a slight inverse
association between the TT genotype and CHD in alcohol abstainers, whereas no association
could be observed in alcohol users. However, interaction between the MTHFR 677C→T poly-
morphism and diabetes and alcohol consumption could not explain heterogeneity, because
there were no clear differences in the prevalences of these risk factors between Europe and
North America (data not shown). For the other CHD risk factors, no interaction with the
MTHFR 677C→T polymorphism was observed.
Figure 2 further illustrates the issues discussed in the previous paragraph. It shows plots of
study-specific means or frequencies of CHD risk factors in controls against the corresponding
OR of CHD for TT versus CC genotype. In this figure, European and North American studies
can be distinguished. For most CHD risk factors, European and North American studies were
about equally distributed over the range of the risk factor. However, there were more European
studies with a higher prevalence of hypercholesterolemia and smoking (> 25%) compared with
North American studies. None of the CHD risk factors, including hypercholesterolemia and
smoking, did clearly correlate with the OR of CHD for TT versus CC genotype (r<|0.5| and
P>0.10). Therefore, these plots confirm that it is unlikely that the difference in association
between the MTHFR 677TT genotype and CHD risk between European and North Ameri-
can studies would be explained by differences in CHD risk profile.
Figure 2. Study specific means frequencies of CHD risk factors versus the OR of CHD for TT versus CC
genotype.
Effect Modification by Vitamin B2 Status

Recently, several studies suggested that riboflavin (vitamin B2) status may also modify the
association between the MTHFR 677TT genotype and homocysteine concentrations.88-90 Ri-
boflavin is the precursor of flavin adenine dinucleotide (FAD), which is the cofactor of MTHFR.
In the United States, products made from refined flour have been enriched with riboflavin
since the 1940s. Therefore, effect modification by dietary intake of riboflavin, in addition to
folate intake, may account for some of the heterogeneity between European and North Ameri-
can populations.90,91 In addition, a trial by McNulty et al recently showed that riboflavin
supplementation significantly decreased homocysteine concentration in individuals with the
MTHFR 677TT genotype, but not in individuals with the CT and CC genotype.92
Bias
Confounding
Although confounding is generally not expected in analyses of an association of a genotype
with disease, there may be some imbalance in the distribution of cardiovascular risk factors
over the MTHFR genotypes. In our pooled meta-analysis,17 adjustment for the possible con-
founders in a subset of 22 studies with available data did not substantially change the risk of
CHD for the TT compared with CC genotype (crude: OR 1.15, 95%Cl 1.02-1.30; adjusted:
OR=1.21, 95%Cl 1.06-1.38). In addition, the study of Wald et al16 showed that there were no
significant differences in cardiovascular risk profile between TT and CC subjects. They further
calculated that even if there were confounding by another risk factor, it could at most account
for half the excess risk observed for TT subjects. However, the possibility of residual confound-
ing, i.e., by factors that were either not measured or incompletely measured, cannot be com-
pletely excluded.
Population Stratification
Another potential source of bias might be the inclusion of individuals from heterogeneous
ethnic backgrounds. The MTHFR variant frequencies can be quite different between popula-
tions. For example, the prevalence of the TT genotype is much lower in Blacks (~1%) than in
Whites.93 If the distribution of individuals with a specific ethnic background is unequal be-
tween cases and controls (so-called population stratification), this may bias an association be-
tween a genotype and risk of CHD.94 This factor might be particularly important for Ameri-
can population studies, since these generally include very mixed populations. Therefore,
population stratification could also explain the discrepancy between American and European
studies.
Conclusions
The MTHFR 677TT genotype is a modest risk factor for AOD, predominantly when folate
status is inadequate. However, the results of several meta-analyses support the hypothesis that
impaired homocysteine metabolism plays a causal role in the occurrence of AOD.
Several large trials are currently underway to assess whether homocysteine lowering by
supplementation with folic acid and other B-vitamins can reduce the risk of CHD.95 Nei-
ther the meta-analyses nor these trials can solve the issue of whether high homocysteine
levels per se or the accompanying low folate levels, which may operate via other mechanisms,
are the cause of AOD. However, these meta-analyses already provide some indirect evidence
that increasing population mean levels of folate would reduce the incidence of AOD, con-
sidering that the MTHFR genotype does not seem to have an adverse effect on CHD risk in
the setting of normal folate status. Hence, there is little clinical value of screening for MTHFR
677C→T genotype in the general population for prediction of AOD risk, provided folate
status is adequate.
References
1. Boushey CJ, Beresford SA, Omenn GS et al. A quantitative assessment of plasma homocysteine as
a risk factor for vascular disease. Probable benefits of increasing folic acid intakes. JAMA 1995;
274:1049-1057.
2. Danesh J, Lewington S. Plasma homocysteine and coronary heart disease: Systematic review of
published epidemiological studies. J Cardiovasc Risk 1998; 5:229-232.
3. Ueland PM, Refsum H, Beresford SAA et al. The controversy over homocysteine and cardiovascu-
lar risk. Am J Clin Nutr 2000; 72:324-332.
4. Brattstrom L, Wilcken DEL. Homocysteine and cardiovascular disease: Cause or effect? Am J Clin
Nutr 2000; 72:315-323.
5. Clayton D, McKeigue PM. Epidemiological methods for studying genes and environmental factors
in complex diseases. Lancet 2001; 358:1356-1360.
mutation in methylenetetrahydrofolate reductase [letter]. Nat Genet 1995; 10:111-113.
7. Chambers JC, Ireland H, Thompson E et al. Methylenetetrahydrofolate reductase 677C→T muta-
tion and coronary heart disease risk in UK Indian Asians. Arteriol Thromb Vasc Biol 2000;
20:2448-2452.
8. Gemmati D, Serino ML, Trivellato C et al. C677T substitution in the methylenetetrahydrofolate
reductase gene as a risk factor for venous thrombosis and arterial disease in selected patients.
Haematologica 1999; 84:824-828.
9. Brattstrom L, Wilcken DE, Ohrvik J et al. Common methylenetetrahydrofolate reductase gene
Circulation 1998; 98:2520-2526.
10. McQuillan BM, Beilby JP, Nidorf M et al. Hyperhomocysteinemia but not the C677T mutation
of methylenetetrahydrofolate reductase is an independent risk determinant of carotid wall thicken-
ing—The Perth carotid ultrasound disease assessment study (CUDAS). Circulation 1999;
99:2383-2388.
11. Deloughery TG, Evans A, Sadeghi A et al. Common mutation in methylenetetrahydrofolate reduc-
tase. Correlation with homocysteine metabolism and late-onset vascular disease. Circulation 1996;
94:3074-3078.
12. Ma J, Stampfer MJ, Hennekens CH et al. Methylenetetrahydrofolate reductase polymorphism, plasma
folate, homocysteine, and risk of myocardial infarction in US physicians. Circulation 1996;
94:2410-2416.
13. Schwartz SM, Siscovick DS, Malinow MR et al. Myocardial infarction in young women in relation
to plasma total homocysteine, folate, and a common variant in the methylenetetrahydrofolate re-
ductase gene. Circulation 1997; 96:412-417.
14. Blom HJ, Verhoef P. Hyperhomocysteinemia, MTHFR, and risk of vascular disease. Circulation
2000; 101:E171.
15. Homocysteine Studies Collaboration: Homocysteine and risk of ischemic heart disease and stroke:
6000 events in 30 observational studies. JAMA 2002; 288:2015-2022.
16. Wald DS, Law M, Morris JK. Homocysteine and cardiovascular disease: Evidence on causality
from a meta-analysis. BMJ 2002; 325:1202.
17. Klerk M, Verhoef P, Clarke R et al. the MTHFR Studies Collaboration: MTHFR 677C→T poly-
morphism and risk of coronary heart disease: A meta-analysis. JAMA 2002; 288:2023-2031.
18. Araujo F, Lopes M, Goncalves L et al. Hyperhomocysteinemia, MTHFR C677T genotype and low
folate levels: A risk combination for acute coronary disease in a Portuguese population. Thromb
Haemost 2000; 83:517-518.
19. Chao CL, Tsai HH, Lee CM et al. The graded effect of hyperhomocysteinemia on the severity and
extent of coronary atherosclerosis. Atherosclerosis 1999; 147:379-386.
20. Mager A, Lalezari S, Shohat T et al. Methylenetetrahydrofolate reductase genotypes and early-onset
coronary artery disease. Circulation 1999; 100:2406-2410.
21. Raslova K, Smolkova B, Vohnout B et al. Risk factors for atherosclerosis in survivors of myocardial
infarction and their spouses: Comparison to controls without personal and family history of ath-
erosclerosis. Metabolism 2001; 50:24-29.
22. Thuillier L, Chadefaux-Vekemans B, Bonnefont JP et al. Does the polymorphism 677C→T of the
5,10-methylenetetrahydrofolate reductase gene contribute to homocysteine-related vascular disease?
J Inherit Metab Dis 1998; 21:812-822.
23. Roest M, van-der-Schouw YT, Grobbee DE et al. Methylenetetrahydrofolate reductase 677 C/T
genotype and cardiovascular disease mortality in postmenopausal women. Am J Epidemiol
2001;153:673-679.
24. Brugada R, Marian AJ. A common mutation in methylenetetrahydrofolate reductase gene is not a
major risk of coronary artery disease or myocardial infarction. Atherosclerosis 1997; 128:107-112.
25. Chen TY, Chen JH, Tsao CJ. Methylenetetrahydrofolate reductase gene polymorphism and coro-
nary artery disease in Taiwan Chinese. Haematologica 2000; 85:445-446.
26. Goracy I, Goracy J, Suliga M e al. Polimorfizm C677T genu reduktazy metylenotetrahydrofolianu
(MTHFR) u chorych po przebytym zawale miesnia sercowego. [C677T gene polymorphism of
methylenetetrahydrofolate reductase (MTHFR) in patients with myocardial infarction]. Pol Arch
Med Wewn 1999; 102:849-854.
27. Hsu LA, Ko YL, Wang SM et al. The C677T mutation of the methylenetetrahydrofolate late
reductase gene is not associated with the risk of coronary artery disease or venous thrombosis
among Chinese in Taiwan. Hum Hered 2000; 51:41-45.
28. Izumi M, Iwai N, Ohmichi N et al. Molecular variant of 5,10-methylenetetrahydrofolate reductase
is a risk factor of ischemic heart disease in the Japanese population [letter]. Atherosclerosis 1996;
121:293-294.
29. Kim CH, Hwang KY, Choi TM et al. The methylenetetrahydrofolate reductase gene polymor-
phism in Koreans with coronary artery disease. Int J Cardiol 2001; 78:13-17.
30. Kostulas K, Crisby M, Huang WX et al. A methylenetetrahydrofolate reductase gene polymor-
phism in ischaemic stroke and in carotid artery stenosis. Eur J Clin Invest 1998; 28:285-289.
31. Malik NM, Syrris P, Schwartzman R et al. Methylenetetrahydrofolate reductase polymorphism
(C-677T) and coronary artery disease. Clin Sci 1998; 95:311-315.
32. Narang R, Callaghan G, Haider AW et al. Methylenetetrahydrofolate reductase mutation and coro-
nary artery disease [letter]. Circulation 1996; 94:2322-2323.
33. Pinto X, Vilaseca MA, Garcia GN et al. Homocysteine and the MTHFR 677C→T allele in pre-
mature coronary artery disease. Case control and family studies. Eur J Clinl Invest 2001; 31:24-30.
34. Reinhardt D, Sigusch HH, Vogt SF et al. Absence of association between a common mutation in
the methylenetetrahydrofolate reductase gene and the risk of coronary artery disease. Eur J Clin
Invest 1998; 28:20-23.
35. Wilcken DE, Wang XL, Sim AS et al. Distribution in healthy and coronary populations of the
methylenetetrahydrofolate reductase (MTHFR) C677T mutation. Arterioscler Thromb Vasc Biol
1996; 16:878-882.
36. Zheng YZ, Tong J, Do XP et al. Prevalence of methylenetetrahydrofolate reductase C677T and its
association with arterial and venous thrombosis in the Chinese population. Br J Haematol 2000;
109:870-874.
37. Meisel C, Cascorbi I, Gerloff T et al. Identification of six methylenetetrahydrofolate reductase
(MTHFR) genotypes resulting from common polymorphisms: Impact on plasma homocysteine lev-
els and development of coronary artery disease. Atherosclerosis 2001; 154:651-658.
38. Abbate R, Sardi I, Pepe G et al. The high prevalence of thermolabile 5-10 methylenetetrahydrofolate
reductase (MTHFR) in Italians is not associated to an increased risk for coronary artery disease
(CAD). Thromb Haemost 1998; 79:727-730.
39. Ardissino D, Mannucci PM, Merlini PA et al. Prothrombotic genetic risk factors in young survi-
vors of myocardial infarction. Blood 1999; 94:46-51.
40. Girelli D, Friso S, Trabetti E et al. Methylenetetrahydrofolate reductase C677T mutation, plasma
homocysteine, and folate in subjects from northern Italy with or without angiographically docu-
mented severe coronary atherosclerotic disease: Evidence for an important genetic-environmental
interaction. Blood 1998; 91:4158-4163.
41. Kluijtmans LA, Kastelein JJ, Lindemans J et al. Thermolabile methylenetetrahydrofolate reductase
in coronary artery disease. Circulation 1997; 96:2573-2577.
42. Verhoef P, Kok FJ, Kluijtmans LA et al. The 677C→T mutation in the methylenetetrahydrofolate
reductase gene: Associations with plasma total homocysteine levels and risk of coronary atheroscle-
rotic disease. Atherosclerosis 1997; 132:105-113.
43. Verhoeff BJ, Trip MD, Prins MH et al. The effect of a common methylenetetrahydrofolate reduc-
tase mutation on levels of homocysteine, folate, vitamin B12 and on the risk of premature athero-
sclerosis. Atherosclerosis 1998; 141:161-166.
44. Szczeklik A, Sanak M, Jankowski M et al. Mutation A1298C of methylenetetrahydrofolate reduc-
tase: Risk for early coronary disease not associated with hyperhomocysteinemia. Am J Med Genet
2001; 101:36-39.
45. Ferrer AC, Palmeiro A, Morais J et al. The mutation C677T in the methylene tetrahydrofolate
reductase gene as a risk factor for myocardial infarction in the Portuguese population [letter]. Thromb
Haemost 1998; 80:521-522.
46. Fernandez-Arcas N, Dieguez-Lucena JL, Munoz-Moran E et al. The genotype interactions of
methylenetetrahydrofolate reductase and renin-angiotensin system genes are associated with myo-
cardial infarction. Atherosclerosis 1999; 145:293-300.
47. Thogersen AM, Nilsson TK, Dahlen G et al. Homozygosity for the C677→T mutation of
5,10-methylenetetrahydrofolate reductase and total plasma homocyst(e) ine are not associated with
greater than normal risk of a first myocardial infarction in northern Sweden. Coron Artery Dis
2001; 12:85-90.
48. Todesco L, Angst C, Litynski P et al. Methylenetetrahydrofolate reductase polymorphism, plasma
homocysteine and age. Eur J Clin Invest 1999; 29:1003-1009.
49. Adams M, Smith PD, Martin D et al. Genetic analysis of thermolabile methylenetetrahydrofolate
reductase as a risk factor for myocardial infarction. QJM 1996; 89:437-444.
50. Fowkes FG, Lee AJ, Hau CM et al. Methylene tetrahydrofolate reductase (MTHFR) and nitric
oxide synthase (ecNOS) genes and risks of peripheral arterial disease and coronary heart disease:
Edinburgh Artery Study. Atherosclerosis 2000; 150:179-185.
51. McDowell IF, Clark ZE, Bowen DJ et al. Heteroduplex analysis for C677T genotyping of methyl-
ene tetrahydrofolate: A case-control study in men with angina. Netherl J Med 1998; 52:S1-S61.
52. Christensen B, Frosst P, Lussier CS et al. Correlation of a common mutation in the
coronary artery disease. Arterioscler Thromb Vasc Biol 1997; 17:569-573.
53. Anderson JL, King GJ, Thomson MJ et al. A mutation in the methylenetetrahydrofolate reductase
gene is not associated with increased risk for coronary artery disease or myocardial infarction. J Am
Coll Cardiol 1997; 30:1206-1211.
54. Folsom AR, Nieto FJ, McGovern PG et al. Prospective study of coronary heart disease incidence
in relation to fasting total homocysteine, related genetic polymorphisms, and B vitamins: The ath-
erosclerosis risk in communities (ARIC) study. Circulation 1998; 98:204-210.
genotypes. Arterioscler Thromb Vasc Biol 1997; 17:1157-1162.
56. Schmitz C, Lindpaintner K, Verhoef P et al. Genetic polymorphism of methylenetetrahydrofolate
reductase and myocardial infarction. A case-control study. Circulation 1996; 94:1812-1814.
57. Tsai MY, Welge BG, Hanson NQ et al. Genetic causes of mild hyperhomocysteinemia in patients
with premature occlusive coronary artery diseases. Atherosclerosis 1999; 143:163-170.
58. Verhoef P, Rimm EB, Hunter DJ et al. A common mutation in the methylenetetrahydrofolate
reductase gene and risk of coronary heart disease: Results among U.S. men. J Am Coll Cardiol
1998; 32:353-359.
59. van Bockxmeer F, Mamotte CD, Vasikaran SD et al. Methylenetetrahydrofolate reductase gene
and coronary artery disease. Circulation 1997; 95:21-23.
60. Morita H, Taguchi J, Kurihara H et al. Genetic polymorphism of 5,10-methylenetetrahydrofolate
reductase (MTHFR) as a risk factor for coronary artery disease. Circulation 1997; 95:2032-2036.
61. Nakai K, Fusazaki T, Suzuki T et al. Genetic polymorphism of 5,10-methylenetetrahydrofolate
increases risk of myocardial infarction and is correlated to elevated levels of homocysteine in the
Japanese general population. Coron Artery Dis 2000; 11:47-51.
62. Ou T, YamakawaKobayashi K, Arinami T et al. Methylenetetrahydrofolate reductase and
apolipoprotein E polymorphisms are independent risk factors for coronary heart disease in Japa-
nese: A case-control study. Atherosclerosis 1998; 137:23-28.
63. Inbal A, Freimark D, Modan B et al. Synergistic effects of prothrombotic polymorphisms and
atherogenic factors on the risk of myocardial infarction in young males. Blood 1999; 93:2186-2190.
64. Gulec S, Aras O, Akar E et al. Methylenetetrahydrofolate reductase gene polymorphism and risk of
premature myocardial infarction. Clin Cardiol 2001; 24:281-284.
65. Tokgozoglu SL, Alikasifoglu M, Unsal et al. Methylene tetrahydrofolate reductase genotype and
the risk and extent of coronary artery disease in a population with low plasma folate. Heart 1999;
81:518-522.
66. Graham IM, Daly LE, Refsum HM et al. Plasma homocysteine as a risk factor for vascular disease.
The European Concerted Action Project. JAMA 1997; 277:1775-1781.
67. Wu LL, Wu J, Hunt SC et al. Plasma homocyst(e)ine as a risk factor for early familial coronary
artery disease. Clin Chem 1994; 40:552-561.
68. Silberberg JS, Crooks RL, Wlodarczyk JH et al. Association between plasma folate and coronary
disease independent of homocysteine. Am J Cardiol 2001; 87:1003-1004.
69. Tanis BC, van-den-Bosch MAAJ, Kemmeren JM et al. Oral contraceptives and the risk of myocar-
dial infarction. N Engl J Med 2001; 345:1787-1793.
70. Eikelboom JW, Hankey GJ, Anand SS et al. Association between high homocyst(e)ine and is-
chemic stroke due to large- and small-artery disease but not other etiologic subtypes of ischemic
stroke. Stroke 2000; 31:1069-1075.
71. Lalouschek W, Aull S, Serles W et al. Genetic and nongenetic factors influencing plasma homocys-
teine levels in patients with ischemic cerebrovascular disease and in healthy control subjects. J Lab
Clin Med 1999; 133:575-582.
72. Markus H, Kapozsta Z, Ditrich R et al. Increased common carotid intima-media thickness in UK
African Caribbeans and its relation to chronic inflammation and vascular candidate gene polymor-
phisms. Stroke 2001; 32:2465-2471.
73. Morita H, Kurihara H, Tsubaki S et al. Methylenetetrahydrofolate reductase gene polymorphism
and ischemic stroke in Japanese. Arteriolscler Thromb Vasc Biol 1998; 18:1465-1469.
74. Press RD, Beamer N, Evans A et al. Role of a common mutation in the homocysteine regulatory
enzyme methylenetetrahydrofolate reductase in ischemic stroke. Diagnost Mol Pathol 1999; 8:54-58.
75. Salooja N, Catto A, Carter A et al. Methylene tetrahydrofolate reductase C677T genotype and
stroke. Clin Laboratory Haematol 1998; 20:357-361.
76. Kelly PJ, Rosand J, Kistler JP et al. Homocysteine, MTHFR 677C→T polymorphism, and risk of
ischemic stroke: Results of a meta-analysis. Neurology 2002; 59:529-536.
77. Lowering blood homocysteine with folic acid based supplements: Meta-analysis of randomised tri-
als. Homocysteine Lowering Trialists’ Collaboration. BMJ 1998; 316:894-898.
78. Jee SH, Beaty TH, Suh I et al. The methylenetetrahydrofolate reductase gene is associated with
increased cardiovascular risk in Japan, but not in other populations. Atherosclerosis 2000;
153:161-168.
93:7-9.
80. Harmon DL, Woodside JV, Yarnell JW et al. The common ‘thermolabile’ variant of methylene
tetrahydrofolate reductase is a major determinant of mild hyperhomocysteinaemia. QJM 1996;
89:571-577.
82. Rimm EB, Willett WC, Hu FB et al. Folate and vitamin B6 from diet and supplements in relation
to risk of coronary heart disease among women [see comments]. JAMA 1998; 279:359-364.
83. Tucker KL, Selhub J, Wilson PW et al. Dietary intake pattern relates to plasma folate and ho-
mocysteine concentrations in the Framingham Heart Study. J Nutr 1996; 126:3025-3031.
84. Jacques PF, Bostom AG, Wilson PW et al. Determinants of plasma total homocysteine concentra-
tion in the Framingham Offspring cohort. Am J Clin Nutr 2001; 73:613-621.
85. Amorim-Cruz JA, Moreiras O, Brzozowska A. Longitudinal changes in the intake of vitamins and
minerals of elderly Europeans. SENECA Investigators. Eur J Clin Nutr 1996; 50(Suppl 2):S77-S85.
86. Gardemann A, Weidemann H, Philipp M et al. The TT genotype of the methylenetetrahydrofolate
reductase C677T gene polymorphism is associated with the extent of coronary atherosclerosis in
patients at high risk for coronary artery disease. Eur Heart J 1999; 20:584-592.
87. Arruda VR, von-Zuben PM, Chiaparini LC et al. The mutation Ala677→Val in the methylene
tetrahydrofolate reductase gene: A risk factor for arterial disease and venous thrombosis. Thromb
Haemost 1997; 77:818-821.
2000; 46:1065-1071.
Nutr 2002; 76:436-441.
91. Rozen R. Methylenetetrahydrofolate reductase: A link between folate and riboflavin? Am J Clin
Nut 2002; 76:301-302.
92. McNulty H, Dowey LC, Scott JM et al. Riboflavin supplementation lowers plasma homocysteine
in individuals homozygous for the MTHFRC677T polymorphism (abstract). Fourth International
Conference on Homocysteine Metabolism. Basel, Switzerland: 2003.
94. Fletcher O, Kessling AM. MTHFR association with arteriosclerotic vascular disease? Hum Genet
1998; 103:11-21.
95. Clarke R, Collins R. Can dietary supplements with folic acid or vitamin B6 reduce cardiovascular
risk? Design of clinical trials to test the homocysteine hypothesis of vascular disease. J Cardiovasc
Risk 1998; 5:249-255.
Methylenetetrahydrofolate Reductase and Venous Thrombosis 113
CHAPTER 9
and Venous Thrombosis
Miranda B.A.J. Keijzer and Martin den Heijer
V
enous thrombosis is a common disease which is associated with significant morbidity
and potentially lethal complications.1,2 Elevated homocysteine concentration
(hyperhomocysteinemia) is an independent risk factor for venous thrombosis.3-7 Both
genetic mutations and environmental factors, such as nutritional deficiencies, may underlie
hyperhomocysteinemia. The most important genetic determinant of elevated homocysteine
concentrations is the 677C→T polymorphism in the gene coding for the methylenetetrahydrofolate
reductase enzyme (MTHFR). MTHFR catalyses the conversion of 5,10-methylenetetrahydrofolate
to 5-methyltetrahydrofolate, which is required for the remethylation of homocysteine to me-
thionine. In 1988, Kang et al8-10 described a thermolabile variant of this enzyme and in 1995,
Frosst et al11 discovered that the molecular basis of this variant was a C to T substitution at
nucleotide 677 of the MTHFR gene. The MTHFR 677TT genotype appeared to be associated
with decreased enzyme activity and to have a profound effect on homocysteine concentra-
tions, particularly in the presence of low folate intake, with about 25% higher homocysteine
levels in subjects with the MTHFR 677TT genotype compared to subjects with the MTHFR
677CC genotype.11-14 In this chapter, we review the importance of this and other polymor-
phisms in the MTHFR gene with respect to venous thrombosis.
The first papers on the relationship between MTHFR and venous thrombosis were pub-
lished in 1997.15-20 Arruda et al15 investigated 127 patients with venous thrombosis and 296
control subjects, and found the MTHFR 677TT genotype to be associated with a 2.9-fold
(95%Cl 1.2 to 7.0) increased risk for venous thrombosis. The five other papers published in
1997 could not confirm the results of Arruda et al.15 Cattaneo et al19 for example, found that
20.8% of 77 patients with deep-vein thrombosis were homozygous for the 677C→T polymor-
phism compared with 22.7% of 154 control subjects (OR 0.8 (95%Cl 0.4 to 1.9)).
In subsequent years, the MTHFR 677C→T polymorphism received a lot of attention and
many papers were written on its possible association with venous thrombosis. Table 1 shows
the studies that addressed the role of the MTHFR 677TT variant in the risk of venous throm-
bosis.15-17,21-56 These studies differ in their conclusion as to whether this polymorphism is a
risk factor for venous thrombosis.
A few studies concluded that the 677TT genotype is associated with an increased risk of
venous thrombosis.15,28,30,34,37,41,42,57,58 Margaglione et al28 in their study of 277 patients with
deep vein thrombosis and 431 healthy subjects, found the MTHFR 677TT variant to be a risk
factor for venous thrombosis, mainly among individuals with a high-risk profile (OR 1.7 (95%Cl
1.1 to 2.5)). The study by Grandone et al57 investigated women during pregnancy and the
postpartum period, and found an odds ratio of 2.1 (95%Cl 1.0 to 4.5) for the association
between the MTHFR 677TT genotype and venous thromboembolism. In 1999, Gemmati et
al34 found an odds ratio of 1.82 (95%Cl 1.15 to 2.86), with a slightly higher odds ratio of 2.11
©2005 Eurekah.com.
114
Table 1. MTHFR 677TT genotype and the associated risk of venous thrombosis
CRUDE OR CRUDE OR
Article CASES CONTROLS (95%Cl) (95%Cl)
TT/ TT/ TT vs
TT CT CC Total TT CT CC Total CT/CC TT vs CC
Arruda et al, 199715 14 52 61 14/127 12 114 170 12/296 2.72 (1.22-6.04) 3.25 (1.43-7.42)
Tosetto et al,199716 8 36 21 8/65 17 71 42 17/130 0.94 (0.39-2.30) 0.94 (0.35-2.53)
Salden et al, 199717 26 102 88 26/216 18 75 71 18/164 1.10 (0.58-2.07) 1.17 (0.59-2.29)
Akar et al, 199821 7 20 25 7/52 6 37 63 6/106 2.38 (0.76-7.43) 2.94 (0.90-9.61)
Brown et al, 199822 55 242 261 55/558 67 209 224 67/500 0.74 (0.50-1.07) 0.70 (0.47-1.05)
De Stefano et al, 199823 37 101 56 37/194 35 98 65 35/198 1.08 (0.65-1.78) 1.23 (0.68-2.20)
Vargas et al, 199824 7 — — 7/57 20 — — 20/200 1.23 (0.49-3.05)
Kluijtmans et al, 199825 47 213 211 47/471 47 203 224 47/474 1.01 (0.66-1.54) 1.06 (0.68-1.66)
Dilley et al, 199826 1 15 77 1/93 4 29 152 4/185 0.50 (0.05-4.51) 0.49 (0.05-4.49)
Quere et al, 199827 21 60 39 21/120 19 53 48 19/120 1.11 (0.57-2.16) 1.36 (0.64-2.88)
Margaglione et al, 199828 60 107 55 60/222 78 223 130 78/431 1.49 (1.03-2.17) 1.82 (1.15-2.88)
Gaustadnes et al, 199929 17 80 99 17/196 90 449 545 90/1084 1.04 (0.61-1.79) 1.04 (0.59-1.82)
Gemmati et al, 199930 41 86 53 41/180 30 102 68 30/200 1.52 (0.91-2.53) 1.75 (0.97-3.17)
Franco et al, 199931 37 82 71 37/190 31 78 81 31/190 1.19 (0.71-2.00) 1.36 (0.77-2.42)
Hessner et al, 199932 41 143 146 41/330 19 64 109 19/192 1.26 (0.71-2.23) 1.61 (0.89-2.93)
Alhenc-Gelas et al, 199933 26 88 91 26/205 49 193 156 49/398 1.03 (0.62-1.71) 0.91 (0.53-1.56)
Gemmati et al, 199934 62 107 51 62/220 39 116 65 39/220 1.59 (1.02-2.47) 2.03 (1.18-3.49)
Ordóñez et al, 199935 13 47 47 13/107 20 88 92 20/200 1.21 (0.58-2.54) 1.27 (0.58-2.78)
Cattaneo et al, 199936 23 57 32 23/112 84 181 138 84/403 0.99 (0.59-1.64) 1.18 (0.65-2.15)
Salomon et al, 199937 37 — — 37/162 46 — — 48/336 1.60 (1.00-2.55)
Table 1. Continued
CRUDE OR CRUDE OR
Article CASES CONTROLS (95%Cl) (95%Cl)
TT/ TT/ TT vs
TT CT CC Total TT CT CC Total CT/CC TT vs CC
Angchaisuksiri et al, 200038 0 6 34 0/40 7 128 365 7/500 — —

Isotalo et al, 200039 12 28 25 12/65 8 35 21 8/64 1.48 (0.57-3.85) 1.26 (0.43-3.66)
Akar et al, 200040 5 33 30 5/68 7 19 40 7/66 0.69 (0.21-2.29) 0.95 (0.28-3.30)
Fujimura et al, 200041 10 30 32 10/72 6 43 36 6/85 1.97 (0.68-5.68) 1.88 (0.61-5.74)
Zheng et al, 200042 10 31 12 10/53 15 45 62 15/122 1.53 (0.65-3.64) 3.44 (1.25-9.47)
Toydemir et al, 200043 3 12 15 3/30 10 51 39 10/100 1.00 (0.26-3.87) 0.78 (0.19-3.23)
Lin et al, 200044 9 50 53 9/112 8 41 76 8/125 1.26 (0.47-3.37) 1.61 (0.58-4.45)
Methylenetetrahydrofolate Reductase and Venous Thrombosis
Hsu et al, 200145 7 40 60 7/107 8 44 55 8/107 0.88 (0.31-2.50) 0.80 (0.27-2.36)

Hanson et al, 200146 16 63 58 16/137 41 158 130 41/329 0.94 (0.51-1.73) 0.87 (0.45-1.68)
Hsu et al, 200147 7 28 48 7/83 6 41 57 6/104 1.46 (0.47-4.52) 1.39 (0.44-4.40)
Iglesias et al, 200148 26 103 61 26/190 26 97 77 26/200 1.05 (0.59-1.88) 1.26 (0.67-2.39)
Ray et al49 19 61 49 19/129 13 44 72 13/129 1.58 (0.74-3.40) 2.20 (0.98-4.94)
Keijzer et al, 200250 14 90 67 14/171 42 186 233 42/461 0.90 (0.48-1.69) 1.16 (0.60-2.25)
Grossmann et al, 200251 40 127 133 40/300 59 189 162 59/410 0.93 (0.60-1.42) 0.83 (0.52-1.31)
Hainaut et al, 200252 13 45 55 13/113 12 44 44 12/100 0.96 (0.42-2.20) 0.87(0.36-2.09)
Quere et al, 200253 42 111 85 42/238 37 116 85 37/238 1.14 (0.70-1.83) 1.14 (0.67-1.94)
Morelli et al, 200254 9 38 44 9/91 5 46 39 5/90 1.78 (0.57-5.52) 1.60 (0.49-5.17)
Domagala et al, 200255 14 44 88 14/146 6 37 57 6/100 1.60 (0.59-4.30) 1.51 (0.55-4.16)
Zalavras et al, 200256 24 82 70 24/176 30 153 117 30/300 1.36 (0.77-2.41) 1.34 (0.72-2.47)
Total 1.17 (1.06-1.29) 1.23 (1.10-1.38)
115
(95%Cl 1.17 to 3.78), when only idiopathic venous thromboembolism was considered. An-
other study by Gemmati et al30 found an odds ratio of 1.67 (95%Cl 0.99 to 2.81) for the risk
of venous thrombosis which became statistically significant after excluding subjects with
thrombophilic risk factors (OR 2.26 (95%Cl 1.28 to 3.99)). In the same year, Salomon et al37
studied a group of 162 patients with venous thromboembolism and 336 patients without a
history of venous thromboembolism as control subjects, and found the 677TT genotype to
yield a 2.1-fold (95%Cl 1.2 to 3.7) increased risk of venous thrombosis. An even higher odds
ratio of 3.4 (95%Cl 1.3 to 9.5) was found by Zheng et al42 in 53 Chinese patients and 122
control subjects. A similar effect was observed by Couturaud et al58 who found an odds ratio of
2.9 (95%Cl 1.0 to 8.6) in 366 patients with venous thromboembolism and 105 controls. The
authors concluded that the MTHFR 677C→T polymorphism was a risk factor for venous
thromboembolism in subjects with spontaneous events and without other common genetic
risk factors. Another study by Fujimura et al41 also found the MTHFR 677C→T polymor-
phism to confer an increased risk of deep vein thrombosis in a subgroup of subjects with
thrombophilia.
However, most studies, including the largest one, concluded that the homozygous MTHFR
677TT mutation is (probably) not associated with an increased risk of venous thrombosis (see
Table 1 and ref. 59, 60). Brown et al22 in a large study of 558 patients and 500 control subjects
found that homozygosity for the MTHFR 677C→T polymorphism yielded an odds ratio of
0.71 (95%Cl 0.48 to 1.03), and concluded that it was not a risk factor for venous thromboem-
bolism. The results of another large study by Kluijtmans et al25 in 471 patients and 474 con-
trols, in which a 1.01-fold (95%Cl 0.7 to 1.5) increased thrombotic risk associated with the
677TT genotype was found, are in line with those of Brown et al.22 Recently, we conducted a
study in 185 patients with recurrent venous thrombosis and 500 control subjects and found
the MTHFR 677TT genotype to be associated with a 1.4-fold (95%Cl 0.7 to 2.8) risk of
venous thrombosis.50
Meta-Analyses
There are several meta-analyses on the MTHFR 677TT genotype and its association with
venous thrombosis. Kluijtmans et al25 presented a summary estimate on 810 patients with
venous thrombosis and 870 control subjects, and found an odds ratio of 1.0 (95%Cl 0.8 to
1.4) for the risk of venous thrombosis associated with the MTHFR 677TT genotype. In 2000,
De Stefano et al61 also published a pooled odds ratio of 1.0 (95%Cl 0.9 to 1.2) for the risk of
venous thrombosis associated with the MTHFR 677TT genotype, after reviewing nine
case-control studies, involving 2225 patients with venous thromboembolism and 2994 control
subjects. In the same year, Brattström et al62 in their meta-analysis which included 15 studies,
found an odds ratio of 1.093 (95%Cl 0.907 to 1.316) for the risk of venous thromboembolism
associated with the MTHFR 677TT genotype. A very similar estimate was found by D’Angelo
et al63 who also analysed fifteen studies, pooling data of 3190 patients with venous throm-
boembolism and 4438 control subjects, yielding an odds ratio of 1.11 (95%Cl 0.98 to 1.27)
for the 677TT genotype.
The two largest meta-analyses on MTHFR and venous thrombosis were published recently,
by Wald et al64 and Ray et al.65 Wald et al64 included 26 studies in their analysis and reported
the MTHFR 677TT genotype to be associated with a 1.29-fold (95%Cl 1.08 to 1.54) in-
creased thrombotic risk. The largest meta-analysis on venous thrombosis and the MTHFR
677C→T polymorphism was performed by Ray et al.65 They concluded that the MTHFR
677TT genotype was weakly associated with an increased thrombotic risk (pooled OR= 1.2
(95%Cl 1.1 to 1.4)) after analysing the data of 31 studies, including a total of 4901 cases and
7886 controls.
For this chapter, we performed a meta-analysis in which we included 40 studies on MTHFR
677TT genotype and venous thrombosis (Table 1). Some studies were not included because
their data were not clear on MTHFR status or its association with venous thrombosis, the
study groups were not representative, or if there was a suspicion that data had been used in
more than one study.18,19,57-60,66-70 For each study, we calculated unmatched odds ratios and
their 95% confidence intervals, using Woolf ’s method. For calculating a pooled odds ratio, we
used a Mantel-Haenszel method and for the confidence intervals, we used the method of Rob-
ins.71 The reference category for investigating the risk associated with the MTHFR 677TT
genotype has not been well established. For this reason we performed two analyses, the first
using 677CC as a reference group and the second with 677CC/677CT as a reference group
(Table 1). The first analysis yielded an odds ratio of 1.23 (95%Cl 1.10 to 1.38) for the MTHFR
677TT mutation, suggesting that this mutation is associated with a slight but significantly
increased thrombotic risk. The second yielded a somewhat lower although still significant odds
ratio of 1.17 (95%Cl 1.06 to 1.29). This is not surprising given the fact that we found the
pooled odds ratio for the 677CT genotype to be 1.09 (95%Cl 1.01 to 1.17).
MTHFR 677C→T and Interaction with Other Risk Factors

for Venous Thrombosis
A second question concerning MTHFR 677C→T is whether there is a synergistic effect of
this mutation with other risk factors for venous thrombosis.
Factor V Leiden
In 1997, Cattaneo et al19 were the first to report a risk increase for venous thrombosis
associated with an interaction between MTHFR 677TT genotype and factor V Leiden. They
found this combination to be associated with a 17.3-fold (95%Cl 2.0 to 152.9) increased
thrombotic risk, while the MTHFR 677TT genotype and factor V Leiden mutation separately
yielded odds ratios of only 0.7 and 6.3, respectively. Tosetto et al70 investigated the MTHFR
677TT variant in subjects with factor V Leiden and found the 677TT genotype to be more
prevalent in subjects with venous thrombosis compared to those without. In 1999, Cattaneo
et al36 in a study of 118 patients with a first episode of deep vein thrombosis and 461 healthy
controls, found the concurrent presence of the MTHFR 677TT genotype and factor V Leiden
to yield an odds ratio of 17.0 (95%Cl 5.3 to 54.8) which was apparently higher than the
expected joint effects (additive (OR 6.7) or multiplicative (OR 5.9)). Recently, we found a
similar odds ratio of 18.7 (95%Cl 3.3 to 108) for this combination in a study of 185 patients
with recurrent venous thrombosis and 500 control subjects.50 Salomon et al37 in 1999, found
an even higher odds ratio of 35.0 (95%Cl 14.5 to 84.7) for the joint presence of the MTHFR
677TT genotype and factor V Leiden, in 162 patients with venous thromboembolism and 336
patients with no such history (OR 16.3 for isolated factor V Leiden). However, Akar et
al21,40concluded that there may be a synergistic effect between the MTHFR 677C→T poly-
morphism and factor V Leiden for the risk of deep vein thrombosis. In 2001, D‘Angelo et al63
analysed the data of 8 studies and found a pooled odds ratio of 1.00 (95%Cl 0.83 to 1.21) for
the MTHFR 677TT genotype, 5.63 (95%Cl 4.30 to 7.38) for isolated factor V Leiden and
10.3 (95%Cl 4.72 to 23.6) for the combination, suggesting a synergistic effect between these
two factors.
In contrast, other studies did not find an apparent interaction between MTHFR 677TT
and factor V Leiden for the risk of venous thrombosis. In 1997, Ocal et al20 did not find the
presence of MTHFR 677C→T to increase the thrombotic risk in carriers of factor V Leiden.
One year later, Brown et al22 who performed one of the largest studies on the MTHFR 677C→T
polymorphism and venous thrombosis, concluded that the risk for venous thrombosis associ-
ated with the presence of the MTHFR 677TT genotype (OR 0.71 (95%Cl 0.48 to 1.03)) was
not increased when coinherited with factor V Leiden (OR 1.52 (0.18 to 12.9). In that same
year, Kluijtmans et al25 published their summary estimate on 810 patients with venous throm-
bosis and 870 control subjects, and concluded that the MTHFR 677C→T polymorphism did
not modify the thrombotic risk associated with factor V Leiden (factor V Leiden OR 8.2
(95%Cl 5.4 to 12.6), factor V Leiden and MTHFR 677TT OR 7.2 (95%Cl 2.5 to 20.7)). In
1999, Rintelen et al68 investigated the influence of the MTHFR 677C→T polymorphism on
the risk of venous thrombosis in 81 patients with venous thrombosis and factor V Leiden, 111
of their family members and 77 controls without factor V Leiden. They found that the preva-
lence of venous thrombosis in family members was not higher in subjects with both the factor
V Leiden and MTHFR 677TT genotype (33%) compared to those with isolated factor V
Leiden (31.9%). Gaustadnes et al29 in their study also concluded that the MTHFR 677TT
genotype did not confer an additive effect in subjects with factor V Leiden. Another study of
205 patients with venous thromboembolism and 398 healthy subjects, by Alhenc-Gelas et al33
reached a similar conclusion, based on finding an odds ratio of 6.46 (95%Cl 3.33 to 12.55) for
isolated factor V Leiden and an odds ratio of 9.33 (95%Cl 1.03 to 84.23) for the combination
with the 677TT genotype. Hessner et al69 did not find the MTHFR 677TT genotype to be
more prevalent in subjects with factor V Leiden and venous thrombosis compared to subjects
with factor V Leiden without venous thrombosis. Toydemir et al43 including subjects with
Behçet’s disease and venous thrombosis, concluded that the MTHFR 677C→T polymorphism
did not increase the risk of deep venous thrombosis associated with factor V Leiden. Recently,
Grossman et al51 in their study of 300 patients with deep venous thrombosis and 410 healthy
blood donors, also found that the MTHFR 677TT genotype was not a risk factor for venous
thromboembolism, either alone (OR= 0.92 (95%Cl 0.54 to 1.41) or in combination with
factor V Leiden.
Whether there is a synergistic effect between the MTHFR 677TT genotype and factor V
Leiden is still inconclusive and larger studies are needed.
Prothrombin 20210G→A
Data on the interaction between the MTHFR 677C→T polymorphism and the prothrom-
bin 20210G→A mutation are sparse, and the overall finding is that there is no synergistic
effect accompanying this combination. One of the first and largest studies that investigated the
risk of venous thrombosis associated with this combination was performed by Brown et al.22
They found that the presence of the MTHFR 677TT genotype did not increase the throm-
botic risk when coinherited with the prothrombin 20210G→A mutation, with an odds ratio
of 0.92 (95%Cl 0.15 to 5.78) for the combination compared to 0.71 (95% Cl 0.48 to 1.03)
for the isolated MTHFR 677TT genotype. Similarly, the study by Cattaneo et al36 did not find
the 677TT genotype to increase the risk of venous thrombosis in subjects with the prothrom-
bin 20210G→A mutation, with odds ratios of 10.4 (95%Cl 4.3 to 25.2) for the prothrombin
mutation alone and 3.8 (95%Cl 0.5 to 27.5) for the combination. Salomon et al37 found an
odds ratio of 7.7 (95%Cl 3.0 to 19.6) for the combination of MTHFR 677TT genotype and
the prothrombin mutation and odds ratios of 3.6 (95%Cl 1.8 to 7.3) and 2.1 (95%Cl 1.2 to
3.7) for the isolated prothrombin 20210G→A and MTHFR 677TT genotype, respectively.
No conclusions could be drawn from the study of Iglesias et al48 who found the combination
of MTHFR 677TT genotype and the prothrombin 20210G→A mutation to be present in
only 1 out of 190 patients and none of the 200 control subjects. Madonna et al72 found the
combination of the MTHFR 677TT genotype and the prothrombin 20210G→A mutation to
be more prevalent in patients with venous thrombosis compared to controls, and Grossman et
al51 found that the concomitant presence of MTHFR 677TT and the prothrombin 20210G→A
mutation yielded no increase in risk (OR 1.37 (95%Cl 0.19 to 9.71) compared to the risk
conferred by the isolated prothrombin mutation (OR 2.42 (95%Cl 1.22 to 4.79)).
D‘Angelo et al63 published the only summary estimate, including three studies with a total
of 818 patients and 1278 controls, which yielded an odds ratio of 4.25 (95%Cl 1.41 to 13.7)
for the combination of MTHFR 677TT genotype and the prothrombin mutation. MTHFR
677TT genotype itself yielded an odds ratio of 0.96 (95%Cl 0.74 to 1.25) and the isolated
prothrombin mutation yielded an odds ratio of 2.31 (95%Cl 1.48 to 3.59).
Folate Concentrations, Hyperhomocysteinemia and Association

with Venous Thrombosis
It is well known that the MTHFR 677TT genotype is associated with elevated homocys-
teine concentrations.12,13 This association is stronger in the presence of a low folate intake.13
This suggests that the influence of MTHFR on thrombosis is stronger in a population with low
folate intake. Klerk et al73 for example, found an increased risk of coronary heart disease for the
677TT genotype in Europe but not in North America, which was explained by an interaction
between MTHFR and folate intake. Subsequently, they found in their analysis that the MTHFR
677TT genotype conferred an increased risk of coronary heart disease only when folate intake
was low. We found only a few studies that took folate or homocysteine status into account
when investigating the MTHFR 677C→T polymorphism and the associated risk of venous
thrombosis.18,19,24,29,34,48,50,52
Cattaneo et al19 found that the increased thrombotic risk associated with the concurrent
presence of factor V Leiden and the MTHFR 677TT genotype, (OR 13.7), was higher when
they adjusted for folate and vitamin B12 serum concentrations (OR 17.3). Legnani et al18 in
their study of subjects with an inherited thrombophilic condition and venous thrombosis,
found no difference with respect to folate, vitamin B12 or homocysteine levels between groups.
The MTHFR 677TT genotype was associated with higher homocysteine concentrations. Vargas
et al24 observed that the MTHFR 677TT genotype in patients was associated with higher
homocysteine levels, that patients with the 677TT genotype showed a trend towards low folate
concentrations, and that the MTHFR 677TT genotype did not increase the risk for venous
thrombosis. Kluijtmans et al25 found homocysteine concentrations to be higher in subjects
with the MTHFR 677TT variant compared to the MTHFR 677CT and MTHFR 677CC
groups, for both cases and controls. Gematti et al34 found a strong association between the
MTHFR 677TT genotype, low folate concentrations and elevated homocysteine concentra-
tions in patients with venous thromboembolism and in controls. Gaustadnes et al29 calculated
odds ratios for the presence of MTHFR 677C→T variants for quartiles of homocysteine con-
centrations in patients with venous thrombosis and found that the odds ratio for the MTHFR
677TT genotype was significantly higher in the top quartile. Iglesias et al48 found that ho-
mocysteine concentrations did not differ significantly between patients with venous thrombo-
sis and controls. For both groups, they found homocysteine concentrations to differ according
to MTHFR genotype with the highest levels found for subjects with the MTHFR 677TT
variant. Hainaut et al52 found that homocysteine concentrations were similar in patients with
venous thromboembolic disease and in control subjects. Ray et al49 found plasma homocys-
teine concentrations to differ significantly between women with venous thromboembolism
and control subjects, with higher homocysteine levels for those with venous thromboembo-
lism. They found no significant interaction between hyperhomocysteinemia and the MTHFR
677C→T polymorphism.
Recently, we found that the MTHFR 677TT genotype was associated with a 1.4-fold (95%Cl
0.7 to 2.8) increased risk of venous thrombosis.50 Subsequently we performed a subgroup
analysis on the effect of MTHFR in subjects with low folate concentration and found similar
odds ratios. However, it is important to realize that plasma folate concentration is not a valid
marker for folate intake, because it is also dependent on folate metabolism (including MTHFR
activity). There are no studies that account for folate intake.
MTHFR 1298A→C
Another polymorphism in the MTHFR gene, the MTHFR 1298A→C.74,75change, also
decreases enzyme activity although it is (probably) not associated with elevated homocysteine
concentrations.46,74,75 There are only a few studies on this mutation and its association with
venous thrombosis. Franco et al31 assessed the effect of the MTHFR 1298A→C polymor-
phism, alone and in combination with factor V Leiden or the prothrombin 20210G→A
mutation, on the risk of venous thrombosis in 190 patients and 190 control subjects. They did
not find the 1298A→C polymorphism to be associated with an increased thrombotic risk,
with an odds ratio of 0.83 (95%Cl 0.33 to 2.08) for the 1298CC genotype. No apparent
interaction with factor V Leiden or the prothrombin 20210G→A mutation was observed.
Akar et al40 who studied 68 patients with deep vein thrombosis and 66 controls, also did not
find the isolated MTHFR 1298 A→C polymorphism to be a risk factor for venous thrombosis
(OR 1.26 (95%Cl 0.46 to 1.5)) for the CC genotype; the concomitant presence of factor V
Leiden and the CC genotype was associated with an odds ratio of 5.39 (95%Cl 0.7 to 9.2).
Isotalo et al39 found no significant difference in prevalence of the MTHFR 1298 C allele in 65
patients with venous thrombosis and 64 healthy volunteers. Similarly, Hanson et al46 found
that the prevalence of the 1298CC genotype was not significantly different between 137 indi-
viduals with deep venous thrombosis (10.9%) and 329 control subjects (7.9%). Ray et al49
who included only women, found an odds ratio of 1.0 (95%Cl 0.6 to 1.7) for the MTHFR
1298CC genotype. Recently, we found the MTHFR 1298CC genotype to be associated with a
similar, 0.9-fold (95%Cl 0.5 to 1.7), risk of venous thrombosis.50 The combination with factor
V Leiden yielded a lower odds ratio (OR 3.3 (95%Cl 0.9 to 11.5)) than the odds ratio for
factor V Leiden itself (OR 5.1 (95%Cl 3.0 to 8.6)). A similar result was found by Domagala
et al55 who found an odds ratio of 1.03 (95%Cl 0.3 to 3.7) for the 1298CC variant in 146
patients with venous thromboembolic disease and 100 controls.
Compound Heterozygosity
Van der Put et al75 in a study of neural tube defects, found that subjects who were com-
pound heterozygotes for the MTHFR 1298 A→C and MTHFR 677C→T polymorphisms
had a diminished MTHFR activity and elevated homocysteine concentrations. With regard to
venous thrombosis, very little is known about the combination of these MTHFR genotypes.
Domagala et al55 did not find homocysteine concentrations to be higher in subjects with both
the heterozygous MTHFR 1298AC and MTHFR 677CT genotypes. Furthermore, they did
not find this combination to yield an increased risk of venous thrombosis (OR 0.7 (95%Cl 0.3
to 1.5)). Hanson et al46 found a slight increase in homocysteine levels for this combination in
772 subjects with coronary artery disease, 329 subjects with deep-vein thrombosis and 329
control subjects.
Franco et al31 investigated the thrombotic risk associated with the concurrent presence of
both the MTHFR 677CTand MTHFR 1298 AC genotypes. They did not find this combina-
tion to increase the risk of venous thrombosis (odds ratio of 1.44 (95%Cl 0.71 to 2.92)). Akar
et al40 found compound heterozygosity to yield an odds ratio of 2.5 (95%Cl 0.8 to 8.4) in
subjects without factor V Leiden and an odds ratio of 24 (95%Cl 2.3 to 240.3) in subjects with
factor V Leiden; isolated factor V Leiden yielded an odds ratio of 6.0 (95%Cl 0.4 to 81).
Isotalo et al39 studied both the MTHFR 677C→T polymorphism and the MTHFR 1298A→C
polymorphism in 65 patients with venous thrombosis and 64 healthy volunteers and found the
prevalence of several combinations of genotypes to differ significantly between patients and
controls. Recently, we investigated the joint effect of the MTHFR 677CT and MTHFR 1298AC
genotype in 185 subjects with recurrent venous thrombosis and 500 control subjects and found
that this combination was associated with a 2.1-fold (95%Cl 1.0 to 4.1) increased risk of
venous thrombosis.50
Conclusion
The MTHFR 677TT genotype is a major determinant of homocysteine concentrations.
Many studies have concluded that the MTHFR 677TT genotype is not a risk factor for venous
thrombosis, but these findings are mainly due to small sample sizes, since the two largest
meta-analyses published on this topic found the MTHFR TT genotype to be associated with a
small increase in thrombotic risk. Similarly, in the current meta-analysis, we found a moderate
but significant risk enhancement for venous thrombosis due to the MTHFR 677TT genotype.
Regarding additional questions on the interaction with other thrombophilic factors and on
the effect-modification by folate status, studies are to small to draw firm conclusions. The same
is true for the second polymorphism in the MTHFR gene (1298A→C).
References
1. White RH. The Epidemiology of Venous Thromboembolism. Circulation 2003;107 [Suppl I]:I-4-I-8.
2. Clive Kearon. Natural History of Venous Thromboembolism Circulation 2003 107 [Suppl
I]:I-22-I-30
3. Falcon CR, Cattaneo M, Panzeri D et al. High prevalence of hyperhomocyst(e)inemia in patients
with juvenile venous thrombosis Arterioscler Thromb 1994; 14:1080-1083.
4. den Heijer M, Blom HJ, Gerrits WBJ et al. Is hyperhomocysteinaemia a risk factor for recurrent
venous thrombosis? Lancet 1995; 345:882-885.
5. Ridker PM, Hennekens CH, Selhub J et al. Interrelation of hyperhomocyst(e)inemia, factor V
Leiden, and risk of future venous thromboembolism. Circulation 1997; 95:1777-1782.
6. den Heijer M, Koster T, Blom HJ et al. Hyperhomocysteinemia as a risk factor for deep-vein
thrombosis. N Engl J Med 1996; 334:759-762.
7. den Heijer M, Rosendaal FR, Blom HJ et al. Hyperhomocysteinemia and venous thrombosis: A
meta-analysis. Thromb Haemost 1998; 80:874-877.
8. Kang S-S, Wong PWK, Zhou J et al. Thermolabile methylenetetrahydrofolate reductase in patients
with coronary artery disease. Metabolism 1988; 37:611-613.
9. Kang S-S, Zhou J, Wong PWK et al. Intermediate homocysteinemia: A thermolabile variant of
10. Kang SS, Wong PWK, Susmano A et al. Thermolabile methylenetetrahydrofolate reductase: An
inherited risk factor for coronary artery disease. Am J Hum Genet 1991; 48:536-545.
mutation in methylenetetrahydofolate reductase. Nat Genet 1995; 10:111-113.
12. Brattström L, Wilcken DEL, Öhrvik J et al. Common methylenetetrahydrofolate reductase gene
Circulation 1998; 98:2520-2526.
93:7-9.
14. Kluijtmans LAJ, van den heuvel LPWJ, Boers GHJ et al. Molecular genetic analysis in mild
hyperhomocysteinemia: A common mutation in the methylenetetrahydrofolate reductase gene is a
genetic risk factor for cardiovascular disease. Am J Hum Genet 1996; 58:35-41.
15. Arruda VR, von Zuben PM, Chiaparini LC et al. The mutation Ala677?Val in the methylene
tetrahydrofolate reductase gene: A risk factor for arterial disease and venous thrombosis. Thromb
Haemost 1997; 77:818-821.
16. Tosetto A, Missiaglia E, Frezzato M et al. The VITA project: C677T mutation in the
methylene-tetrahydrofolate reductase gene and risk of venous thromboembolism. Br J Haematol
1997; 97:804-806.
17. Salden A, Keeney S, Hay CRM et al. The C677T MTHFR variant and the risk of venous throm-
bosis. Br J Haematol 1997; 99:472.
18. Legnani C, Palareti G, Grauso F et al. Hyperhomocyst(e)inemia and a common methylenetetrahydrofolate
reductase mutation (Ala223Val MTHFR) in patients with inherited thrombophilic coagulation de-
fects. Arterioscler Thromb Vasc Biol 1997; 17:2924-2929.
19. Cattaneo M, Tsai MY, Bucciarelli P et al. A common mutation in the methylenetetrahydrofolate
reductase gene (C677T) increases the risk for deep-vein thrombosis in patients with mutant factor
V (factor V:Q506). Arterioscler Thromb Vasc Biol 1997; 17:1662-1666.
20. Ocal IT, Sadeghi A, Press RD. Risk of venous thrombosis in carriers of a common mutation in the
homocysteine regulatory enzyme methylenetetrahydrofolate reductase. Mol Diagn 1997; 2:61-68.
21. Akar N, Akar E, Misirlioglu M et al. Search for genetic factors favoring thrombosis in Turkish
population. Thromb Res 1998; 92:79-82.
22. Brown K, Luddington R, Baglin T. Effect of the MTHFRC677T variant on risk of venous throm-
boembolism: Interaction with factor V Leiden and prothrombin (F2G20210A) mutations. Br J
Haematol 1998; 103:42-44.
23. De Stefano V, Chiusolo P, Paciaroni K et al. Prevalence of the 677C to T mutation in the
methylenetetrahydrofolate reductase gene in Italian patients with venous thrombotic disease. Thromb
Haemost 1998; 79:686-687.
24. Vargas M, Soto I, Pinto CR et al. Mild hyperhomocysteinemia and MTHFR C677T do not in-
crease the risk for venous thrombosis in a Spanish population. Blood Coag Fibrinol 1998; 9:555-556.
25. Kluijtmans LAJ, den Heijer M, Reitsma PH et al. Thermolabile methylenetetrahydrofolate reduc-
tase and factor V Leiden in the risk of deep-vein thrombosis. Thromb Haemost 1998; 79:254-258.
26. Dilley A, Austin H, Hooper WC et al. Relation of three genetic traits to venous thrombosis in an
African-American population. Am J Epidemiol 1998; 147:30-35.
27. Quéré I, Chassé JF, Dupuy E et al. Homocystéine, 5-10-méthylènetétrahydrofolate réductase et
thrombose veineuse profonde. Enquête auprès de 120 patients en médecine interne. Rev Méd In-
terne 1998; 19:29-33.
28. Margaglione M, D’ Andrea G, d’ Addedda M et al. The methylenetetrahydrofolate reductase TT677
genotype is associated with venous thrombosis independently of the coexistence of the FV Leiden
and the prothrombin A20210 mutation. Thromb Haemost 1998; 79:907-911.
29. Gaustadnes M, Rüdiger N, Møller J et al. Thrombophilic predisposition in stroke and venous
thromboembolism in Danish patients. Blood Coag Fibrinol 1999; 10:251-259.
30. Gemmati D, Serino ML, Trivellato C et al. C677T substitution in the methylenetetrahydrofolate
reductase gene as a risk factor for venous thrombosis and arterial disease in selected patients.
Haematologica 1999; 84:824-828.
31. Franco RF, Morelli V, Lourenço D et al. A second mutation in the methylenetetrahydrofolate
reductase gene and the risk of venous thrombotic disease. Br J Haematol 1999; 105:556-559.
32. Hessner MJ, Luhm RA, Pearson SL et al. Prevalence of prothrombin G20210A, Factor V G1691A
(Leiden), and methylenetetrahydrofolate reductase (MTHFR) C677T in seven different popula-
tions determined by multiplex allele-specific PCR. Thromb Haemost 1999; 81:733-738.
33. Alhenc-Gelas M, Arnaud E, Nicaud V et al. Venous thromboembolic disease and the prothrombin,
methylene tetrahydrofolate reductase and factor V genes. Thromb Haemost 1999; 81:506-510.
34. Gemmati D, Previati M, Serino ML et al. Low folate levels and thermolabile methylenetetrahydrofolate
reductase as primary determinant of mild hyperhomocystinemia in normal and thromboembolic subjects.
Arterioscler Thromb Vasc Biol 1999; 19:1761-1767.
35. Ordóñez AJG, Alvarez CRF, Rodrígues JMM et al Genetic polymorphism of methylenetetrahydrofolate
reductase and venous thromboembolism: A case-control study. Haematologica 1999; 84:190-191.
36. Cattaneo M, Chantarangkul V, Taioli E et al. The G20210A mutation of the prothrombin gene in
patients with previous first episodes of deep-vein thrombosis: Prevalence and association with fac-
tor V G1691A, methylenetetrahydrofolate reductase C677T and plasma prothrombin levels. Thromb
Res 1999; 93:1-8.
37. Salomon O, Steinberg DM, Zivelin A et al. Single and combined prothrombotic factors in patients
with idiopathic venous thromboembolism. Prevalence and risk assessment. Arterioscler Thromb
Vasc Biol 1999; 19:511-518.
38. Angchaisuksiri P, Pingsuthiwong S, Sura T et al. Prevalence of the C677T methylenetetrahydrofolate
reductase mutation in Thai patients with deep vein thrombosis. Acta Haematol 2000; 103:191-196.
39. Isotalo PA, Donnelly JG. Prevalence of methylenetetrahydrofolate reductase mutations in patients
with venous thrombosis. Mol Diagn 2000; 5:59-66.
40. Akar N, Akar E, Akçay R et al. Effect of metylenetetrahydrofolate reductase 677 C-T, 1298 A-C,
and 1317 T-C on factor V 1691 mutation in Turkish deep vein thrombosis patients. Thromb Res
2000; 97:163-167.
41. Fujimura H, Kawasaki T, Sakata T et al. Common C677T polymorphism in the methylenetetrahydrofolate
reductase gene increases the risk for deep vein thrombosis in patients with predisposition of thrombophilia.
Thromb Res 2000; 98:1-8.
42. Zheng Y-Z, Tong J, Do X-P et al. Prevalence of methylenetetrahydrofolate reductase C677T and its asso-
ciation with arterial and venous thrombosis in the Chinese population. Br J Haematol 2000; 109:870-874.
43. Toydemir PB, Elhan AH, Tükün A et al. Effects of factor V gene G1691A, methylenetetrahydrofolate
reductase gene C677T, and prothrombin gene G20210A mutations on deep venous thrombogenesis in
Behçet’s disease. J Rheumatol 2000; 27:2849-2854.
44. Lin J-S, Shen M-C, Tsai W et al. The prevalence of C677T mutation in the methylenetetrahydrofolate
reductase gene and its association with venous thrombophilia in Taiwanese Chinese. Thromb Res
2000; 97:89-94.
45. Hsu L-A, Ko Y-L, Wang S-M et al. The C677T mutation of the methylenetetrahydrofolate reduc-
tase gene is not associated with the risk of coronary artery disease or venous thrombosis among
Chinese in Taiwan. Hum Hered 2001; 51:41-45.
46. Hanson NQ, Aras Ö, Yang F et al. C677T and A1298C polymorphisms of the methylenetetrahydrofolate
reductase gene: Incidence and effect of combined genotypes on plasma fasting and post-methionine
load homocysteine in vascular disease. Clin Chem 2001; 47:661-666.
47. Hsu T-S, Hsu L-A, Chang C-J et al. Importance of hyperhomocysteinemia as a risk factor for
venous thromboembolism in a Taiwanese population. A case-control study. Thromb Res 2001;
102:387-395.
48. Iglesias Varela ML, Adamczuk YP, Forastiero RR et al. Major and potential prothrombotic geno-
types in a cohort of patients with venous thromboembolism. Thromb Res 2001; 104:317-324.
49. Ray JG, Langman LJ, Vermeulen MJ et al. Genetics University of Toronto Thrombophilia Study
in Women (GUTTSI): Genetic and other risk factors for venous thromboembolism in women.
Curr Control Trials Cardiovasc Med 2001; 2:141-149.
50. Keijzer MBAJ, den Heijer M, Blom HJ et al. Interaction between hyperhomocysteinemia, mutated
methylenetetrahydrofolatereductase (MTHFR) and inherited thrombophilic factors in recurrent
venous thrombosis. Thromb Haemost 2002; 88:723-728.
51. Grossmann R, Schwender S, Geisen U et al. CBS 844ins68, MTHFR TT677 and EPCR 4031ins23
genotypes in patients with deep-vein thrombosis. Thromb Res 2002; 107:13-15.
52. Hainaut P, Jaumotte C, Verhelst D et al. Hyperhomocysteinemia and venous thromboembolism:
A risk factor more prevalent in the elderly and in idiopathic cases. Thromb Res 2002; 106:121-125.
53. Quéré I, Perneger TV, Zittoun J et al. Red blood cell methylfolate and plasma homocysteine as
risk factors for venous thromboembolism: A matched case-control study. Lancet 2002; 359:747-752.
54. Morelli VM, Lourenço DM, D’Almeida V et al. Hyperhomocysteinemia increases the risk of venous
thrombosis independent of the C677T mutation of the methylenetetrahydrofolate reductase gene
in selected Brazilian patients. Blood Coagul Fibrinolysis 2002; 13:271-275.
55. Domagala TB, Adamek L, Nizankowska E et al. Mutations C677T and A1298C of the
5,10-methylenetetrahydrofolate reductase gene and fasting plasma homocysteine levels are not asso-
ciated with the increased risk of venous thromboembolic disease. Blood Coagul Fibrinolysis 2002;
13:423-431.
56. Zalavras ChG, Giotopoulou S, Dokou E et al. Lack of association between the C677T mutation in
the 5,10-methylenetetrahydrofolate reductase gene and venous thromboembolism in Northwestern
Greece. Int Angiol 2002; 21:268-271.
57. Grandone E, Margaglione M, Colaizzo D et al. Genetic susceptibility to pregnancy-related venous
thromboembolism: Roles of factor V Leiden, prothrombin G20210A, and methylenetetrahydrofolate
reductase C677T mutations. Am J Obstet Gynecol 1998; 179:1324-1328.
58. Couturaud F, Oger E, Abalain JH et al. Methylenetetrahydrofolate reductase C677T genotype and
venous thromboembolic disease. Respiration 2000; 67:657-661.
59. Philipp CS, Dilley A, Saidi P et al. Deletion Polymorphism in the angiotensin-converting enzyme
gene as a thrombophilic risk factor after hip arthroplasty. Thromb Haemost 1998; 80:869-73.
60. Gerhardt A, Scharf RE, Beckmann MW et al. Prothrombin and factor V mutations in women
with a history of thrombosis during pregnancy and the puerperium. N Eng J Med 2000;
342:374-380.
61. De Stefano V, Casorelli I, Rossi E et al. Interaction between hyperhomocysteinemia and inherited
thrombophilic factors in venous thromboembolism. Semin Thromb Hemost 2000; 26:305-311.
62. Brattström L, Wilcken DEL. Homocysteine and cardiovascular disease: Cause or effect? Am J Clin
Nutr 2000; 72:315-323.
63. Carmel R, Jacobsen DW. Homocysteine in health and disease. Cambridge University Press 2001.
from meta-analysis. BMJ 2002; 325:1202-1206.
65. Ray JG, Shmorgun D, Chan WS. Common C677T polymorphism of the methylenetetrahydrofolate
reductase gene and the risk of venous thromboembolism: Meta-analysis of 31 studies. Pathophysiol
Haemost Thromb 2002; 32:51-58.
66. de Franchis R, Fermo I, Mazzola G et al. Contribution of the cystathionine β-synthase gene
(844ins68) polymorphism to the risk of early-onset venous and arterial occlusive disease and of
fasting hyperhomocysteinemia. Thromb Haemost 2000; 84:576-582.
67. D’Angelo, Coppola A, Madonna P et al. The role of vitamin B12 in fasting hyperhomocysteinemia
Haemost 2000; 83:563-570.
68. Rintelen C, Mannhalter C, Lechner K et al. No evidence for an increased risk of venous thrombo-
sis in patients with factor V Leiden by the homozygous 677 C to T mutation in the
methylenetetrahydrofolate-reductase gene. Blood Coagul Fibrinolysis 1999; 10:101-105.
69. Hessner MJ, Dinauer DM, Luhm RA et al. Contribution of the glycoprotein Ia 807TT, methylene
tetrahydrofolate reductase 677TT and prothrombin 20210GA genotypes to prothrombotic risk
among factor V 1691A (Leiden) carriers. Br J Hematol 1999; 106:237-239.
70. Tosetto A, Rodeghiero F, Martinelli I et al. Additional genetic risk factors for venous thromboem-
bolism in carriers of factor V Leiden mutation. Br J Hematol 1998; 103:871-876.
71. Rothman KJ. Modern epidemiology. Boston/Toronto 1986.
72. Madonna P, Piemontino U, De Stefano V et al. Coexistence of thrombophilic gene polymor-
phisms among 559 unrelated consecutive patients with a history of thrombosis. Thromb Res 2001;
101:317-319.
73. Klerk M, Verhoef P, Clarke R et al. MTHFR 677C→T polymorphism and risk of coronary heart
disease. A meta-analysis. JAMA 2002; 288:2023-2031.
reductase (MTHFR) associated with decreased enzyme activity. Mol Genet Metab 1998; 64:169-172.
75. van der Put NMJ, Gabreëls F, Stevens EMB et al. A second common mutation in the
Hum Genet 1998; 62:1044-1051.
Congenital Malformations and SNPs in 5,10 MTHFR 125
CHAPTER 10
Neural Tube Defects, Other Congenital

Malformations and Single Nucleotide
Polymorphisms in the 5,10
(MTHFR) Gene:
A Meta-Analysis
Stein Emil Vollset and Lorenzo D. Botto
Abstract
W
e reviewed, and provide a meta-analysis of, more than 40 published case-control
studies on associations between single nucleotide polymorphisms (SNPs) of the
5,10-methylenetetrahydrofolatereductase (MTHFR) gene and congenital malforma-
tions. Most studies focused on neural tube defects (NTDs) and the 677C→T polymorphism.
Based on 21 reports with data from 1,657 cases of NTDs, we found that compared with the
CC genotype (no T allele), the TT genotype (that is, homozygosity for MTHFR 677C→T
SNP) was associated with an odds ratio of 1.76 (95% confidence interval [CI] 1.45-2.14) for
NTDs, whereas for the CT genotype (heterozygosity for the 677C→T SNP) the correspond-
ing odds ratio was 1.26 (95% CI: 1.09-1.45). Similar or slightly higher risk estimates were
associated with maternal genotype, whereas only a weak relationship was demonstrated with
paternal genotype. Studies of the 677C→T SNP in relation to risk for Down syndrome (5
studies), and orofacial clefts (4 studies) provide conflicting evidence, and their summary odds
ratios are close to one, consistent with a minor role (if any) for the 677C→T SNP in these
conditions. Two studies report an increased risk for cardiac malformations associated with TT
genotype of the fetus or child, though small sample sizes and methodologic issues limit the
interpretation of these findings. Thus far, no reports have been published on the other malfor-
mations, such as urinary tract and limb reduction defects, which have been found in some
studies to be inversely associated with preconceptional use of folic acid or multivitamins. Five
case-control studies of the second functional SNP of the MTHFR gene, the 1298A→C SNP,
suggest a 29% and 20% risk increase for NTD associated with the AC and CC genotypes,
respectively. This slight positive association, however, is driven by one study whose results do
not agree with the others. Combining the only two studies that report complete data for the
two SNPs (677C→T and 1298A→C), the highest risk for NTDs is observed for individuals
with combined heterozygosity of 677C→T and 1298A→C SNPs. Few data are available on
©2005 Eurekah.com.
interactions of these two SNPs with folate status or other possible effect modifiers such as diet,
alcohol use, methionine intake, and smoking habits.
Background and Scope of the Chapter

We are in the early genomic era of medicine.1 The first draft sequence of the roughly 3.2
billion base pairs constituting the human genome has been completed.2,3 Single nucleotide
polymorphisms (SNPs) represent common variations at single base pair locations, and 1.4-2.1
million SNPs were mapped by 2001.3,4 Of these SNPs, it is estimated that about 60,000, or
approximately 2 per gene, could have functional consequences in the encoded protein.4 The
coding sequence is known for many genes, including the gene coding for 5,10-methylene-
tetrahydrofolate reductase (MTHFR),5 which is located on the short arm of chromosome 1 (1p36.3).
While more than 20 rare mutations of the MTHFR gene are known to cause severe disease,6
two common SNPs, 677C→T and 1298A→C, appear to have milder functional effects and
have received considerable attention as possible modulators of the risk of cancers, cardiovascu-
lar disease, birth defects, and other health outcomes.7 These two common SNPs are associated
with variant protein sequences and changes in enzymatic activity. For example, homozygosity
for the 677C→T SNP (the TT genotype) is associated with a 70% lower MTHFR enzyme
activity, whereas heterozygosity (CT genotype) is associated with a 30% decrease in activity.8
Studying such functional polymorphisms of genes involved in metabolic pathways can pro-
vide valuable insights not only into the genetic but also the environmental determinants of
health and corroborate studies of such environmental factors. It has been noted that, in
well-designed studies, one may exploit the fact that “Mendelian randomization” ensures that
comparison of individuals with different genotypes is free of systematic group differences, just
as is achieved in a randomized clinical trial.9 Specifically, finding systematic case-control differ-
ences in the distribution of functional polymorphisms of MTHFR strengthens the case that
alterations in folate/homocysteine balance are involved in disease etiology. Just as in the ran-
domized trial, however, small sample size can limit group comparability that is ensured when
sufficiently large numbers of individuals are allocated at random.
Associations between SNPs in the MTHFR gene and birth defects were reviewed by Botto
and Yang10 in 2000. This review will update that information and add evidence that has been
published since.
Identification of Studies, Data Extraction and Statistical Approach

Studies were identified by searching the PubMed database, references in earlier reviews, and
relevant papers. Several research groups with ongoing data collection have reported updated
data and analyses. Examples include successive studies from the Netherlands,11,12 Germany,13-15
and the United States.16,17 In such instances, we have reported the incremental data separately
except when the numbers were too low to provide a meaningful separate estimate.16 When in
doubt, we communicated with the study authors.
All odds ratios were recalculated based on tabular data provided in the articles. Study-specific
and combined estimates were calculated using conditional logistic regression (PROC PHREG
with DISCRETE ties option in SAS version 8.2 for Windows) in which study was a stratifica-
tion factor. Because the data from California were reported by ethnicity, we considered each
ethnic group as a stratum in computing the overall risk estimates.
In our tables and meta-analyses we compared genotypes, not allele frequencies. For ex-
ample, for the 677C→T SNP, we used the CC genotype as the reference group, and generated
separate odds ratios (ORs) and confidence intervals (CIs) for the CT and TT genotypes. We
also examined the data for evidence of linear trend by assigning scores of 0, 1, and 2 to the CC,
CT and TT genotypes, respectively, and computing a p-value for trend. Some studies did not
report the complete genotype distribution, but rather combined some of the genotypes (for
example, CC with CT or TT with CT). In these instances, we present what summary measure
was allowed by such incomplete information.
Main Findings
A summary of the findings of the meta-analysis is presented in Table 1. The case-control
distribution and risk estimates for individual studies are presented in Tables 2 through 8. Risk
estimates are also summarized graphically in Figures 1 through 3. Risk estimates associated
with the joint genotype that includes 677C→T and 1298A→C SNPs are summarized in Fig-
ure 4. The following paragraphs describe these findings in more detail.
The 677C→T Polymorphism and Neural Tube Defects

In 1995, van der Put and colleagues reported an increased prevalence of the TT genotype of
the 677C→T SNP in the MTHFR gene in children with spina bifida and their parents com-
pared with controls recruited from a general practice in The Hague.11 They estimated that the
relative risk for spina bifida was 2.2 if the father carried the TT genotype, 3.7 if the mother
carried the TT genotype, and 2.9 if the child itself carried the TT genotype.
Since then, several other case-control studies have examined the risk for an NTD-affected
fetus or baby in relation to the 677C→T SNP genotype in fetuses or patients (21 studies,11-15,17-32
Table 2), in mothers (16 studies,11,12,20,25-37 Table 3) and in fathers (9 studies,11,12,20,25,26,30,31,33,35
Table 4). When combining these studies, and comparing the TT with the CC genotype, the
estimated relative risk for NTD-affected pregnancy or child was 1.76 (95% CI 1.45-2.14),
1.92 (95% CI 1.51-2.45), and 1.27 (95% CI 0.91-1.77) when the genotype was carried by the
infant, mother, or father, respectively (Figs. 1, 2). The corresponding relative risk estimates
comparing CT with CC genotype were 1.26 (95% CI 1.09-1.45), 1.24(95% CI 1.04-1.47),
1.21 (95% CI 0.97-1.50), respectively.
The 677C→T Polymorphism and Orofacial Clefts

Four case-control studies have provided data38-41 on the 677C→T polymorphism in pa-
tients with cleft lip with or without cleft palate (CLP) and controls (Table 5). Only one of the
studies assessed the association between cleft palate without cleft lip (CP) and the 677C→T
SNP.39 Of the four studies, only one39 with 93 affected children from Ireland found a statisti-
cally significant association between the 677C→T polymorphism and all orofacial clefts. The
association was strongest and statistically significant only for CP (OR 3.2, 95% CI 1.3-7.9,
P=0.02, comparing TT with CC/CT).
The other studies of CLP, including a large (310 cases) population-based case-control study
from California, consistently reported odds ratios close to one for all genotype comparisons
(TT vs. CC, CT vs. CC). For the comparison of TT with the combined group of CC and CT,
for which the Irish study found an odds ratio of 1.65 (95% CI 0.81-3.34), the other studies,
combined, found an odds ratio of 0.98 (95% CI 0.68-1.40). Combining all studies for the
comparison of TT vs. CT/CC, we computed an overall estimate for the odds ratio of 1.08
(95%CI 0.78-1.50, P=0.65). Of the two studies that evaluated maternal genotype,40,41 a sig-
nificantly increased risk for CLP-affected offspring associated with both CT and TT was found
in one study41 but not the other.
The 677C→T Polymorphism and Down Syndrome

Two case-control studies from one research group in the United States appeared in 199942
and 2000,43 showing a strong association between the TT variant of the 677C→T polymor-
phism and Down syndrome (Table 6 and Fig. 3). However, three European studies, from
France, Ireland, and Italy, found no evidence of such an association.44-46 In addition, all three
studies found no evidence of dose-response relation amongst genotypes, and observed the low-
est risks associated with the TT genotype. Combining all studies, the estimated odds ratios
comparing TT with CC, and CT with CC, were 1.17 (95% CI 0.76-1.79) and 1.39 (95% CI
1.04-1.85), respectively, with a non significant test for trend (P=0.15). A recent study with
data on 202 cases (including some from a previous study43) and their parents showed preferen-
tial transmission of the T allele to Down cases.47
128
Table 1. Summary of results of meta-analyses of MTHFR polymorphisms and birth defects
Polymorphism Number of Odds Ratio Odds Ratio (Other Contrasts)
Cases / Controls Studies CT vs CC TT vs CC P-Trend TT vs CC/CT CT/TT vs CC Dose Response
677C→T
CLP – infants 517/1449 4 0.91 0.93 0.57 1.08 No
Down – mothers 404/601 5 1.39 1.17 0.15 0.97 No
NTD – infants 1657/3143 21 1.26 1.76 < 0.0001 Yes
NTD – mothers 989/1592 13 1.24 1.92 <0.0001 Yes
NTD – fathers 624/1305 9 1.21 1.27 0.07 Weak
Heart – infants 140/321 2 0.004 1.83
1298A→C AC vs AA CC vs AA CC vs AA/AC AC/CC vs AA
NTD – infants 750/1080 6 1.29 1.20 0.053 1.06 1.28 No
NTD – mothers 510/1354 5 1.06 1.48 0.21 Weak

Table 2. Fetus or child MTHFR 677C→T SNP and risk of NTD
Number of Cases Number of Controls CT versus CC TT versus CC P-Trend Country Ref.
Study Year CC CT TT All CC CT TT All OR (95% CI) OR (95% CI)
van der Put 1995 22 26 7 55 111 86 10 207 1.52 (0.81-2.87) 3.51 (1.21-10.2) 0.02 Netherlands 11
Whitehead 1995 35 32 15 82 50 43 6 99 1.06 (0.57-1.99) 3.55 (1.26-10.0) 0.05 Ireland 18
Ou 1996 17 15 9 41 68 36 5 109 1.66 (0.75-3.70) 7.09 (2.11-23.8) 0.002 USA 19
Papapetrou 1996 16 20 5 41 81 94 24 199 1.08 (0.52-2.21) 1.05 (0.35-3.17) 0.87 U.K. 20
Bjorke-Monsen 1997 12 15 1 28 42 22 4 68 2.36 (0.95-5.89) 0.88 (0.09-8.51) 0.22 Norway 21
Mornet 1997 19 21 3 43 50 70 13 133 0.79 (0.39-1.62) 0.61 (0.16-2.37) 0.39 France 22
Koch 1998 58 60 19 137 75 61 17 153 1.27 (0.78-2.08) 1.44 (0.69-3.02) 0.24 Germany 13
Congenital Malformations and SNPs in 5,10 MTHFR
Shaw - combined 1998 73 100 41 214 218 213 72 503 1.34 (0.93-1.92) 1.47 (0.91-2.37) 0.07 California 23
Whites 1998 33 42 13 88 115 122 32 269 1.20 (0.71-2.02) 1.41 (0.67-3.00) 0.33
Hispanics 1998 26 52 27 105 63 71 35 169 1.77 (0.99-3.16) 1.87 (0.95-3.67) 0.06
de Franchis 1998 62 89 52 203 173 313 97 583 0.79 (0.55-1.15) 1.50 (0.96-2.33) 0.15 Italy 24
Boduroglu 1999 39 48 4 91 47 39 7 93 1.48 (0.81-2.69) 0.69 (0.19-2.52) 0.61 Turkey 25
Shields 1999 66 87 36 189 64 65 14 143 1.30 (0.81-2.08) 2.49 (1.23-5.03) 0.01 Ireland 26
Christensen 1999 19 26 11 56 42 44 11 97 1.30 (0.63-2.69) 2.20 (0.82-5.93) 0.13 Canada 27
Stegmann 1999 4 5 2 11 15 3 3 21 5.85 (1.00-34.2) 2.43 (0.31-19.08) 0.18 Germany 15
Barber 2000 3 17 4 24 19 60 14 93 1.79 (0.47-6.73) 1.80 (0.35-9.31) 0.48 Texas 28
Yu 2000 2 4 7 13 5 16 3 24 0.63 (0.09-4.42) 5.49 (0.68-44.2) 0.06 China 29
Volcik 2000 55 114 65 234 31 56 25 112 1.15 (0.67-1.98) 1.46 (0.77-2.77) 0.24 Texas/Calif 30
Johanning 2000 43 30 9 82 63 11 2 76 3.96 (1.80-8.72) 6.52 (1.35-31.5) 0.0002 Alabama 17
Richter 2001 11 18 7 36 23 29 7 59 1.29 (0.51-3.26) 2.07 (0.59-7.34) 0.27 Germany 14
Garcia Fragoso 2002 10 18 3 31 41 50 9 100 1.47 (0.62-3.52) 1.36 (0.31-5.95) 0.47 Puerto Rico 31
Cunha 2002 12 3 0 15 30 35 10 75 0.22 (0.06-0.84) 0.00 (0.00-1.35) 0.01 Brazil 32
All 590 764 303 1657 1342 1422 379 3143 1.26 (1.09-1.45) 1.76 (1.45-2.14) <0.0001
129
130
Table 3. Mother MTHFR 677C→T SNP and risk of NTD
Papapetrou 1996 16 15 5 36 81 94 24 199 0.81 (0.38-1.73) 1.05 (0.35-3.17) 0.88 U.K. 20
Speer 1997 25 31 9 65 30 29 6 65 1.28 (0.62-2.66) 1.79 (0.56-5.70) 0.30 U.S.A. 33
Molloy 1998 34 35 13 82 118 121 21 260 1.00 (0.59-1.71) 2.14 (0.97-4.71) 0.16 Ireland 34
Shields 1999 80 108 30 218 114 108 20 242 1.42 (0.96-2.10) 2.13 (1.13-4.02) 0.01 Ireland 26
Christensen 1999 24 27 11 62 44 36 10 90 1.37 (0.68-2.77) 2.01 (0.75-5.39) 0.15 Canada 27
Davalos 2000 19 35 14 68 31 51 19 101 1.12 (0.55-2.28) 1.20 (0.49-2.93) 0.68 Mexico 35
Barber 2000 23 50 14 87 16 55 16 87 0.63 (0.30-1.33) 0.61 (0.23-1.59) 0.28 U.S.A. 28
Lucock 2000 8 9 2 19 11 17 3 31 0.73 (0.22-2.45) 0.92 (0.13-6.70) 0.76 U.K. 36
Yu 2000 2 25 15 42 5 16 3 24 3.82 (0.67-21.8) 12.01 (1.56-92.3) 0.02 China 29
Volcik 2000 41 112 61 214 31 56 25 112 1.51 (0.86-2.66) 1.84 (0.95-3.56) 0.07 U.S.A. 30
Martinez 2001 11 12 15 38 12 16 3 31 0.82 (0.27-2.47) 5.33 (1.22-23.3) 0.04 Mexico 37
GarciaFragoso 2002 7 23 7 37 41 50 9 100 2.68 (1.05-6.84) 4.50 (1.27-16.0) 0.01 Puerto Rico 31
Cunha 2002 8 11 2 21 30 35 10 75 1.18 (0.42-3.29) 0.75 (0.14-4.11) 0.91 Brazil 32
All 368 567 226 1161 816 885 212 1913 1.24 (1.04-1.47) 1.92 (1.51-2.45) <0.0001
Table 4. Father MTHFR 677C→T SNP and risk of NTD
Papapetrou 1996 8 14 4 26 81 94 24 199 1.51 (0.60-3.76) 1.68 (0.47-6.06) 0.34 U.K. 20
Speer 1997 24 27 4 55 26 24 5 55 1.22 (0.56-2.65) 0.87 (0.21-3.59) 0.88 U.S.A. 33

Shields 1999 83 109 26 218 114 108 20 242 1.39 (0.94-2.04) 1.78 (0.93-3.41) 0.04 Ireland 26
Davalos 2000 15 15 9 39 31 51 19 101 0.61 (0.26-1.41) 0.98 (0.36-2.66) 0.80 Mexico 35
Volcik 2000 31 46 22 99 31 56 25 112 0.82 (0.44-1.55) 0.88 (0.41-1.88) 0.71 U.S.A. 30
Garcia Fragoso 2002 16 18 2 36 41 50 9 100 0.92 (0.42-2.03) 0.57 (0.11-2.93) 0.57 Puerto Rico 31
All 241 304 79 624 576 584 145 1305 1.21 (0.97-1.50) 1.27 (0.91-1.77) 0.07
131
Figure 1. Findings from published studies of MTHFR 677C→T SNP (case genotype) and risk for neural
tube defects.
Figure 2. Findings from published studies of MTHFR 677C→T SNP and risk for neural tube defects
(maternal genotype).
Table 5. Fetus or child MTHFR 677C→T SNP and risk of cleft lip with or without cleft palate (CLP)
Number of Cases Number of Controls CT versus CC TT versus CC P-Trend CT versus CC/CT P Ref.
Study Year CC CT TT All CC CT TT All OR (95% CI) OR (95% CI) OR (95% CI)
Shaw combined 1998 143 127 40 310 155 178 49 382 0.74 (0.53-1.02) 0.84 (0.52-1.35) 0.19 0.98 (0.63-1.54) 0.93 38
Shaw, Hispanics 31 45 12 88 35 47 21 103 1.08 (0.58-2.03) 0.65 (0.28-1.52) 0.43 0.62 (0.29-1.34) 0.22
Shaw, White 93 73 25 191 84 116 25 225 0.57 (0.38-0.86) 0.90 (0.48-1.69) 0.16 1.20 (0.67-2.17) 0.54
Shaw, Black 6 1 1 8 8 3 2 13 0.46 (0.04-5.32) 0.68 (0.05-8.81) 0.63 0.80 (0.06-9.89) 0.86
Shaw, Other 13 8 2 23 28 12 1 41 1.43 (0.48-4.29) 4.20 (0.36-49.6) 0.24 3.72 (0.33-42.6) 0.29
Mills 1999 56* 10 66 765* 83 848 1.65 (0.81-3.34) 0.17 39
Gaspar 2001 30 39 8 77 49 49 15 113 1.30 (0.70-2.41) 0.87 (0.33-2.30) 0.88 0.76 (0.31-1.88) 0.55 40
Martinelli 2002 22 30 12 64 46 43 17 106 1.46 (0.73-2.90) 1.47 (0.60-3.60) 0.31 1.21 (0.54-2.72) 0.65 41
All 447* 70 517 1285* 164 1449 1.08 (0.78-1.50) 0.65
All except Mills 195 196 60 451 250 270 81 601 0.91 (0.70-1.18) 0.93 (0.63-1.37) 0.57 0.98 (0.68-1.40) 0.91
* CC and CT combined
133
134
Table 6. Mother MTHFR 677C→T SNP and risk of Down syndrome
James 1999 15 34 8 57 24 22 4 50 2.45 (1.06-5.65) 3.16 (0.82-12.3) 0.03 USA 42
Hobbs 2000 51 84 22 157 67 59 14 140 1.87 (1.14-3.05) 2.06 (0.96-4.41) 0.01 USA 43
O’Leary 2002 18 21 2 41 90 84 18 192 1.25 (0.62-2.50) 0.56 (0.12-2.61) 0.89 Ireland 45
Chadefaux- 2002 36 42 7 85 45 46 16 107 1.14 (0.62-2.09) 0.55 (0.20-1.47) 0.47 France 44

Vekemans
Stuppia 2002 20 32 12 64 27 62 23 112 0.70 (0.34-1.43) 0.71 (0.29-1.74) 0.40 Italy 46
All 140 213 51 404 253 273 75 601 1.39 (1.04-1.85) 1.17 (0.76-1.79) 0.15
Figure 3. Findings from published studies of MTHFR 677C→T SNP and risk for Down syndrome (ma-
ternal genotype).
The 677C→T Polymorphism and Cardiac Malformations

The results from two case-control studies suggest a positive association between the 677C→T
polymorphism and cardiac malformations (Table 7).48,49 One of the studies did not provide
the full genotype distribution, allowing calculation only of a summary odds ratio comparing
the combined group of CT and TT individuals with CC individuals. This combined odds ratio
was 1.83 (95% CI: 1.22-2.78; P=0.004).
The 1298A→C Polymorphism and Neural Tube Defects

Six case-control studies provide data on the 1298A→C polymorphism and risk of NTDs
(Table 8).12,14,28,30,32,50 Using the AA genotype as the reference category, we computed sum-
mary odds ratios for the heterozygote variant AC genotype (OR 1.29, 95% CI 1.04-1.61) and
for the homozygote variant CC genotype (OR 1.20, 95% CI 0.81-1.78). The test for linear
trend was of borderline significance (P = 0.053).
Data on maternal genotype for the 1298A→C polymorphism was reported by five
case-control studies.12,28,30,32,50 At variance with what has been seen for infant genotype, the
AC maternal genotype was not associated with an increased risk for an NTD-affected preg-
nancy. The CC genotype was associated with a slightly increased risk that was not statistically
significant.
One should note that the association between the 1298A→C genotype and NTD risk,
suggested by the overall risk estimate, was essentially driven by a single study.50 That study
reported a highly significant dose-response relationship between the number of C alleles and
NTD risk, both for maternal and infant genotype.50 All other studies, however, show no such
relation, in patients and mothers.
Some authors have suggested that individuals who are heterozygotes for both variant alleles
(677C→T and 1298A→C) may be at increased risk of NTD. Thus far, only two studies pro-
vide the complete combined distribution of the two SNPs in cases and controls.12,14 We com-
bined these two studies to estimate such risk, and found that the highest risk was associated
with combined heterozygosity (Fig. 4).
136
Table 7. Fetus or child MTHFR 677C→T SNP and cardiac malformations
Number of Cases Number of Controls CT vs CC TT versus CC P-Trend TT vs CC/CT Ref.
01 Wenstrom 2001 17 9* 26 81 12* 93 3.53 49

(U.S.A.) (1.29-9.62)
P=0.0139
02 Junker 2001 51 42 21 114 129 78 21 228 1.36 (0.83-2.23) 2.52 (1.27-5.00) 0.0092 1.61 48
(Germany) (1.02-2.53)
P=0.0396
All 68 72* 140 210 111* 321 1.83

(1.22- 2.78)
P=0.0044
* CT and TT combined
Table 8. Fetus or patient MTHFR 1298A→C SNP and NTD
Barber 2000 15 4 0 19 56 25 4 85 0.60 (0.18-1.98) 0.00 (0.00-0.00) 0.22 U.S.A. 28
Volcik 2000 168 61 4 233 54 25 3 82 0.79 (0.45-1.37) 0.43 (0.09-1.97) 0.21 U.S.A. 30
DeMarco 2001 75 99 29 203 114 76 12 202 1.98 (1.30-3.00) 3.66 (1.76-7.61) <.0001 Italy 50
Richter 2001 85 87 12 184 123 85 25 233 1.48 (0.99-2.22) 0.70 (0.33-1.46) 0.71 Germany 14
Cunha 2002 14 10 1 25 42 28 5 75 1.07 (0.42-2.73) 0.60 (0.07-5.56) 0.85 Brazil 32
All 394 302 54 750 568 425 87 1080 1.29 (1.04-1.61) 1.20 (0.81-1.78) 0.05
137
Figure 4. Findings from published studies of genotype that include the MTHFR 677C→T and 1298A→C
SNPs with respect to risk for neural tube defects. The figure is based on data from van der Put12 and
Richter.14
Discussion
We have reviewed the association between two SNPs of the MTHFR gene and risk for birth
defects. Overall, the data published thus far support an association between NTD risk and the
677C→T genotype of affected children or fetuses or their mothers. Such a relation appears to
be dose dependent, that is, the relative risk is highest for the TT genotype and intermediate for
the CT genotype. Maternal 677C→T genotype showed a slightly stronger relation with NTDs
than did infant or fetal genotype. Paternal genotype showed only a weak relation with NTD
risk in the child (Table 1). The apparently stronger effect of maternal genotype could indicate
that delivery of adequate folate species to the fetus is compromised in TT mothers. It is unclear
whether such relations vary by type of NTD, though it should be noted that the great majority
of NTDs in the reported studies were cases of spina bifida.
With respect to the1298A→C SNP, the second common MTHFR variant, the data are
much less consistent, and slightly increased risk suggested by the overall estimate is driven by a
single study.50 Two studies suggest an increased risk associated with combined heterozygosity
of the two MTHFR SNPs.12,14 For other malformations such as orofacial clefts or Down syn-
drome, the available data do not support the presence of an association with either of the two
SNPs.
Finally, some studies have suggested that preconceptional intake of multivitamins with folic
acid is associated with a reduced risk for congenital anomalies that have not yet been included
in studies of MTHFR SNPs. Such malformations (including urinary tract defects51,52 and limb
reduction defects53,54) should be addressed in further studies, as should other conditions (such
as heart anomalies and cleft palate) for which current evidence is limited.
Challenges in Evaluation and Interpretation

One of the challenges in drawing overall qualitative and quantitative conclusions relates to
the methodological limitations of several published studies. Among such limitations, small
sample size, inadequate control selection, and lack of information on vitamin status and other
potential effect modifiers, can create considerable bias, imprecision, or both. For example,
most studies reviewed here did not randomly select population controls, but rather used
hospital-based or convenience samples. In some instances, the controls were or could have
been of different ethnic origin compared to the cases, possibly introducing considerable bias
because of the marked geographic and ethnic variation of the 677C→T SNP.10
Because periconceptional use of folic acid prevents NTDs,55 and because MTHFR is in-
volved in folate metabolism, folate status and intake could modify the effect of the 677C→T
SNP on NTD risk, as could other related factors that interfere with homocysteine or folate
metabolism such as smoking, alcohol intake, dietary methionine and folate intake, and use of
vitamin supplements. It will be critical to stratify MTHFR genotype by these factors to display
the full range of associations. For example, studies of subgroups with low or compromised
folate status might better clarify associations between MTHFR SNPs and, for example, risk for
orofacial clefts, heart defects, or Down syndrome. Such data, however, are largely missing,
though some studies are beginning to explore these aspects. For example, studies from Califor-
nia evaluated interactions between the 677C→T SNP and vitamin intake with respect to some
orofacial clefts and spina bifida. In the study of cleft lip (with or without cleft palate), research-
ers from California noted a slightly increased risk for having an affected child in mothers with
the TT genotype who reported not using vitamin supplements, but a lower risk for mothers
with the same genotype who reported using such supplements.38 For spina bifida, the results
for the TT genotype were in the similar direction, but not for the CT genotype.23
Other factors, including alcohol consumption, might modify the effect of variant MTHFR
genotypes. For example, evaluating interactions with vitamin status and alcohol consumption
has proven useful in studies of colorectal cancer,56 and may also add important insight with
respect to birth defects.
For most studies published to date, sample size severely limited the ability to assess interac-
tions. Reported associations between MTHFR SNPs and disease are weak, with odds ratios
typically ranging from 0.5 to 2, and often from 0.8 to 1.25. Confirming odds ratios of this
magnitude requires very large samples. The recent confirmation that the TT genotype is asso-
ciated with an increased risk for adult cardiovascular disease, with estimated odds ratios of
1.16-1.21, was based on data from about 12,000 cases.57,58 Sample size calculations show that
for a case-control study to have 80% power to detect an odds ratio of 1.26 for NTDs (the
estimated effect of the CT genotype compared to CC) using the observed overall genotype
prevalence and two-sided significance level of 5%, that study would have to include approxi-
mately 1,350 cases of NTDs and an equal number of controls. Under similar assumptions,
about 450 cases and 450 controls would be needed to confirm the estimated overall effect of
the TT genotype compared to CC (OR=1.76).
In retrospect, these considerations underscore that even the largest single study could not be
expected to establish a statistically significant relation between MTHFR genotypes and birth
defects risk, assuming that the current overall point estimates are valid. Thus, further advances
in the study of these SNPs require larger samples and careful efforts to combine evidence from
many research groups. Such joint efforts would be made easier if all published studies reported
the full distribution of genotypes. Some published studies present their data by combining CT
and TT or CC and CT genotypes. Whereas investigators may have good reasons to analyze and
present their findings in this manner, for the benefit of future assessments of the cumulated
evidence they should be encouraged to report the full genotype distribution.
Other aspects relating the quality of published studies are more difficult to assess. A detailed
description of study methods, including case-definitions, sample selection for both cases and
controls (individuals and families), and laboratory and quality control procedures can provide
some grounds for an assessment. However, often this information is lacking or severely limited,
and a systematic effort in improving such reporting by authors and journal editors could sig-
nificantly benefit future evaluations.
Though not directly a measure of quality, sample size can provide some indication on the
robustness of the findings and agreement among larger studies can be informative.59 Five stud-
ies of NTD patients and controls had more than 150 cases.14,23,24,26,30 Except for one study
(OR 2.9),26 odds ratios for TT vs. CC genotype were in the range of 1.46 to 1.60 in these large
studies. The remarkable agreement among these larger studies, which include two thirds of all
reported cases, strengthens the plausibility of the association.
Triad-Based Studies
With some added effort it might be possible to obtain samples from the child, mother, and
father (triad). Within such triads, researchers can then study how alleles are transmitted from
heterozygous parents to the affected child. These patterns of transmission can indicate an asso-
ciation between certain alleles or haplotypes and disease, and can be analyzed with an increas-
ing variety of analytic methods.60,61 The stated advantage of such triad studies over classic
case-control studies is that comparisons are done within families, and in certain circumstances
this might improve power and reduce bias due to unrecognized population stratification.60
Triad studies, however, have their own set of limitations, both with respect to their practical
implementation (for example, missing or homozygous parents) and to their interpretation,60
so that discordant findings from case-control and triad-based studies can be difficult to inter-
pret and resolve.
Triad-based studies of MTHFR polymorphisms are still few but, given the popularity of
such a methodologic approach, are likely to increase. One such large study, from a leading
research group in Ireland,26 evaluated the risk for NTD associated with the 677C→T variant,
using several analytic methods on data from over 200 triads. In their triad analysis, they re-
ported an increased risk for NTDs associated with the TT genotype in the child or fetus (OR
1.6). This risk estimate is of the same magnitude as that observed in case-control studies. The
researchers also evaluated the role of parental genotypes and their interaction with the child’s
genotype, and concluded that the genotype that drove NTD risk in their data was that of the
child.
MTHFR Variation as Basis for International Variation in NTD Rates

MTHFR genotype might explain some of the geographic variation in the occurrence of
NTDs. For example, according to the recent data from the International Clearinghouse for
Birth Defect Monitoring Systems (ICBDMS), Finland and Mexico represent extremes both in
NTD prevalence and allele frequencies of the 677 T-allele. Among the 1999 birth cohort, the
rate of spina bifida was 16.5 per 10,000 in Mexico and 3.5 per 10,000 births in Finland.62
Estimated prevalence of the TT genotype in samples of newborns were 32.2% in Mexico and
4.0% in Finland.63 If one standardizes the population with respect to genotype prevalence to a
standard population with TT prevalence of 10 % in Hardy-Weinberg equilibrium, one can
derive predicted rates of spina bifida using the relative risk estimates of 1.76 and 1.26 (TT vs.
CC and CT vs. CC, respectively). Such standardized rates for Mexico and Finland are 13.7 and
3.7 per 10,000 births, respectively, suggesting that variations in MTHFR prevalence would
explain about 18% of the variation in spina bifida rates between Finland and Mexico.
Concluding Comments
The 677C→T SNP in the fetus or mother appears to be consistently associated with an
increased risk for NTDs in a dose-dependent manner. No association with congenital malfor-
mations can be established for the 1298A→C SNP. For other congenital malformations, no
association can be clearly established with the data published to date. However, the quantity
and quality of data for anomalies other than NTDs are limited and the findings frequently
discordant. Replication with large, well-designed, and well-described studies that include as-
sessment of effect modifiers will be necessary to establish whether the lack of association is real
or not.
References
1. Guttmacher AE, Collins FS. Genomic medicine—a primer. N Engl J Med 2002; 347:1512-1520.
2. Lander ES, Linton LM, Birren B et al. Initial sequencing and analysis of the human genome.
Nature 2001; 409:860-921.
3. Venter JC, Adams MD, Myers EW et al. The sequence of the human genome. Science 2001;
291:1304-1351.
4. Sachidanandam R, Weissman D, Schmidt SC et al. A map of human genome sequence variation
containing 1.42 million single nucleotide polymorphisms. Nature 2001; 409:928-933.
cDNA, mapping and mutation identification. Nat Genet 1994; 7:195-200.
6. Rosenblatt DS. Inborn errors of folate and cobalamin metabolism. In: Carmel R, Jacobsen DW,
eds. Homocysteine in health and disease. New York: Cambridge University Press, 2001:244-258.
polymorphism. Trends Pharmacol Sci 2001; 22:195-201.
8. Rozen R. Polymorphisms of folate and cobalamin metabolism. In: Carmel R, Jacobsen DW, eds.
Homocysteine in health and disease. New York: Cambridge University Press, 2001.
9. Clayton D, McKeigue PM. Epidemiological methods for studying genes and environmental factors
in complex diseases. Lancet 2001; 358:1356-1360.
lies: A HuGE review. Am J Epidemiol 2000;151:862-877.
11. van der Put NM, Steegers-Theunissen RP, Frosst P et al. Mutated methylenetetrahydrofolate re-
ductase as a risk factor for spina bifida. Lancet 1995; 346:1070-1071.
Hum Genet 1998; 62:1044-1051.
13. Koch MC, Stegmann K, Ziegler A et al. Evaluation of the MTHFR C677T allele and the MTHFR
gene locus in a German spina bifida population. Eur J Pediatr 1998; 157:487-492.
14. Richter B, Stegmann K, Roper B et al. Interaction of folate and homocysteine pathway genotypes
evaluated in susceptibility to neural tube defects (NTD) in a German population. J Hum Genet
2001; 46:105-109.
15. Stegmann K, Ziegler A, Ngo ET et al. Linkage disequilibrium of MTHFR genotypes 677C/T-1298A/
C in the German population and association studies in probands with neural tube defects(NTD).
Am J Med Genet 1999; 87:23-29.
16. Wenstrom KD, Johanning GL, Owen J et al. Role of amniotic fluid homocysteine level and of
fetal 5, 10-methylenetetrahydrafolate reductase genotype in the etiology of neural tube defects. Am
J Med Genet 2000; 90:12-16.
17. Johanning GL, Tamura T, Johnston KE et al. Comorbidity of 5,10-methylenetetrahydrofolate re-
ductase and methionine synthase gene polymorphisms and risk for neural tube defects. J Med
Genet 2000; 37:949-951.
18. Whitehead AS, Gallagher P, Mills JL et al. A genetic defect in 5,10 methylenetetrahydrofolate
reductase in neural tube defects. Qjm 1995; 88:763-766.
19. Ou CY, Stevenson RE, Brown VK et al. 5,10 Methylenetetrahydrofolate reductase genetic poly-
morphism as a risk factor for neural tube defects. Am J Med Genet 1996; 63:610-614.
20. Papapetrou C, Lynch SA, Burn J et al. Methylenetetrahydrofolate reductase and neural tube de-
fects. Lancet 1996; 348:58.
21. Bjorke-Monsen AL, Ueland PM, Schneede J et al. Elevated plasma total homocysteine and
C677T mutation of the methylenetetrahydrofolate reductase gene in patients with spina bifida.
Qjm 1997; 90:593-596.
22. Mornet E, Muller F, Lenvoise-Furet A et al. Screening of the C677T mutation on the
methylenetetrahydrofolate reductase gene in French patients with neural tube defects. Hum
Genet 1997; 100:512-514.
23. Shaw GM, Rozen R, Finnell RH et al. Maternal vitamin use, genetic variation of infant
methylenetetrahydrofolate reductase, and risk for spina bifida. Am J Epidemiol 1998;
148:30-37.
24. de Franchis R, Buoninconti A, Mandato C et al. The C677T mutation of the
5,10-methylenetetrahydrofolate reductase gene is a moderate risk factor for spina bifida in Italy. J
Med Genet 1998; 35:1009-1013.
25. Boduroglu K, Alikasifoglu M, Anar B et al. Association of the 677C-->T mutation on the
methylenetetrahydrofolate reductase gene in Turkish patients with neural tube defects. J Child
Neurol 1999; 14:159-161.
26. Shields DC, Kirke PN, Mills JL et al. The “thermolabile” variant of methylenetetrahydrofolate
reductase and neural tube defects: An evaluation of genetic risk and the relative importance of the
genotypes of the embryo and the mother. Am J Hum Genet 1999; 64:1045-1055.
27. Christensen B, Arbour L, Tran P et al. Genetic polymorphisms in methylenetetrahydrofolate re-
ductase and methionine synthase, folate levels in red blood cells, and risk of neural tube defects.
Am J Med Genet 1999; 84:151-157.
28. Barber R, Shalat S, Hendricks K et al. Investigation of folate pathway gene polymorphisms and the
incidence of neural tube defects in a Texas hispanic population. Mol Genet Metab 2000; 70:45-52.
29. Yu J, Chen B, Zhang G et al. The 677 C-->T mutation in the methylenetetrahydrofolate reductase
(MTHFR) gene in five Chinese ethnic groups. Hum Hered 2000; 50:268-270.
30. Volcik KA, Blanton SH, Tyerman GH et al. Methylenetetrahydrofolate reductase and spina bifida:
Evaluation of level of defect and maternal genotypic risk in Hispanics. Am J Med Genet 2000;
95:21-27.
31. Garcia-Fragoso L, Garcia-Garcia I, de L et al. Presence of the 5,10-methylenetetrahydrofolate re-
ductase C677T mutation in Puerto Rican patients with neural tube defects. J Child Neurol 2002;
17:30-32.
32. Cunha AL, Hirata MH, Kim CA et al. Metabolic effects of C677T and A1298C mutations at the
MTHFR gene in Brazilian children with neural tube defects. Clin Chim Acta 2002; 318:139-143.
33. Speer MC, Worley G, Mackey JF et al. The thermolabile variant of methylenetetrahydrofolate
reductase (MTHFR) is not a major risk factor for neural tube defect in American Caucasians. The
NTD Collaborative Group. Neurogenetics 1997; 1:149-150.
34. Molloy AM, Mills JL, Kirke PN et al. Low blood folates in NTD pregnancies are only partly
explained by thermolabile 5,10-methylenetetrahydrofolate reductase: Low folate status alone may
be the critical factor. Am J Med Genet 1998; 78:155-159.
35. Davalos IP, Olivares N, Castillo MT et al. The C677T polymorphism of the
methylenetetrahydrofolate reductase gene in Mexican mestizo neural-tube defect parents, control
mestizo and native populations. Ann Genet 2000; 43:89-92.
36. Lucock M, Daskalakis I, Briggs D et al. Altered folate metabolism and disposition in mothers
affected by a spina bifida pregnancy: Influence of 677c --> t methylenetetrahydrofolate reductase
and 2756a --> g methionine synthase genotypes. Mol Genet Metab 2000; 70:27-44.
37. Martinez de Villarreal LE, DelgadoEnciso I, Valdez-Leal R et al. Folate levels and
N(5),N(10)-methylenetetrahydrofolate reductase genotype (MTHFR) in mothers of offspring with
neural tube defects: A case-control study. Arch Med Res 2001; 32:277-282.
38. Shaw GM, Rozen R, Finnell RH et al. Infant C677T mutation in MTHFR, maternal
periconceptional vitamin use, and cleft lip. Am J Med Genet 1998; 80:196-198.
39. Mills JL, Kirke PN, Molloy AM et al. Methylenetetrahydrofolate reductase thermolabile variant
and oral clefts. Am J Med Genet 1999; 86:71-74.
40. Gaspar DA, Pavanello RC, Zatz M et al. Role of the C677T polymorphism at the MTHFR gene
on risk to nonsyndromic cleft lip with/without cleft palate: Results from a case-control study in
Brazil. Am J Med Genet 1999; 87:197-199.
41. Martinelli M, Scapoli L, Pezzetti F et al. C677T variant form at the MTHFR gene and CL/P: A
risk factor for mothers? Am J Med Genet 2001; 98:357-360.
methylenetetrahydrofolate reductase gene may be maternal risk factors for Down syndrome. Am J
Clin Nutr 1999; 70:495-501.
47. Hobbs CA, Cleves MA, Lauer RM et al. Preferential transmission of the MTHFR 677 T allele to
infants with down syndrome: Implications for a survival advantage. Am J Med Genet 2002; 113:9-14.
48. Junker R, Kotthoff S, Vielhaber H et al. Infant methylenetetrahydrofolate reductase 677TT genotype
is a risk factor for congenital heart disease. Cardiovasc Res 2001; 51:251-254.
49. Wenstrom KD, Johanning GL, Johnston KE et al. Association of the C677T methylenetetrahydrofolate
reductase mutation and elevated homocysteine levels with congenital cardiac malformations. Am J
Obstet Gynecol 2001; 184:806-812; discussion 812-807.
50. De Marco P, Calevo MG, Moroni A et al. Polymorphisms in genes involved in folate metabolism as
risk factors for NTDs. Eur J Pediatr Surg 2001; 11(Suppl 1):S14-17.
51. Li DK, Daling JR, Mueller BA et al. Periconceptional multivitamin use in relation to the risk of
congenital urinary tract anomalies. Epidemiology 1995; 6:212-218.
52. Czeizel AE. Reduction of urinary tract and cardiovascular defects by periconceptional multivitamin
supplementation. Am J Med Genet 1996; 62:179-183.
53. Shaw GM, O’Malley CD, Wasserman CR et al. Maternal periconceptional use of multivitamins
and reduced risk for conotruncal heart defects and limb deficiencies among offspring. Am J Med
Genet 1995; 59:536-545.
54. Yang Q, Khoury MJ, Olney RS et al. Does periconceptional multivitamin use reduce the risk for
limb deficiency in offspring? Epidemiology 1997; 8:157-161.
55. Botto LD, Moore CA, Khoury MJ et al. Neural-tube defects. N Engl J Med 1999; 341:1509-1519.
56. Giovannucci E. Epidemiologic studies of folate and colorectal neoplasia: A review. J Nutr 2002;
132:2350S-2355S.
57. Klerk M, Verhoef P, Clarke R et al. MTHFR 677C→T polymorphism and risk of coronary heart
disease: A meta-analysis. JAMA 2002; 288:2023-2031.
from a meta-analysis. BMJ 2002; 325:1202-1208.
59. Ioannidis JPA, Trikalinos TA, Ntzani EE et al. Genetic associations in large versus small studies:
An empirical assessment. Lancet 2003; 361:567-571.
60. Schulze TG, McMahon FJ. Genetic association mapping at the crossroads: Which test and why?
Overview and practical guidelines. Am J Med Genet 2002; 114:1-11.
61. Weinberg CR, Wilcox AJ, Lie RT. A log-linear approach to case-parent-triad data: Assessing effects
of disease genes that act either directly or through maternal effects and that may be subject to
parental imprinting. Am J Hum Genet 1998; 62:969-978.
62. International clearinghouse for birth defects monitoring systems. Annual report 2001 with data for
1999. Roma: International Centre for Birth Defects, 2001.
63. Wilcken B, Bamforth F, Li Z et al. Geographic and ethnic variation of the 677C→T allele of 5,10
methylenetetrahydrofolate reductase (MTHFR): Findings from over 7,000 newborns from 16 areas
world wide. J Med Genet 2003; 40:619-625.
CHAPTER 11
Pregnancy Complications
Willianne L.D.M. Nelen and Henk J. Blom
Abstract
E
levated homocysteine concentrations and the methylenetetrahydrofolate reductase
(MTHFR) 677C→T polymorphism have been identified as risk factors for arterial and
venous thrombosis. More recently, elevated homocysteine concentrations and the
MTHFR 677 TT genotype have also been associated with pregnancy complications, including
chromosomal abnormalities, congenital malformations, recurrent pregnancy loss, placental
disease and preeclampsia. However, in addition to embryonic or foetal consequences early in
pregnancy, hyperhomocysteinemia and homozygosity for the MTHFR 677C→T polymor-
phism (TT genotype) have also been described as a cause of thrombo-embolic processes later in
pregnancy and even during the postpartum period.
This chapter concerns the association of the MTHFR 677C→T polymorphism and
hyperhomocysteinemia with pregnancy complications and the possibilities for intervention
and prevention.
Introduction
Different inborn errors of metabolism lead to extremely high concentrations of homocys-
teine in blood and urine. Classic homocystinuria caused by deficiency of cystathionine β-synthase
is characterized, among other clinical features, by arterial and venous thrombosis.1 There are
only a few case reports concerning the reproductive performances of these women with severe
elevation of homocysteine concentrations. These reports suggest an association between
homocystinuria and Down syndrome,2 congenital anomalies,3 hypertensive disorders,3,4 pre-
mature delivery,5 intra-uterine growth retardation,4 foetal loss3,4,6,7 and maternal venous throm-
bosis.3,4,8
In the general population, homocysteine concentrations are influenced by environmental
factors (e.g., nutrition, medication) as well as by genetic factors. The common 677C→T vari-
ant in the MTHFR gene leads to increased thermolability and decreased activity of the MTHFR
enzyme.9 Homozygosity (TT genotype) for this mutation is associated with elevated homocys-
teine and decreased serum folate concentrations,10,11 and recently it has been demonstrated to
be a genetic risk factor for cardiovascular diseases.12
Risk factors for thrombosis are associated with late pregnancy complications, and because
mild elevations in total homocysteine (tHcy) concentrations have been described as a risk
factor for venous thrombosis, 13 a link between the MTHFR 677C→T polymorphism,
hyperhomocysteinemia and placental or foetal thrombosis has been suggested. Disturbances of
homocysteine metabolism have frequently been studied in relationship to pregnancy compli-
cations.14-18
This chapter reviews the literature on the relationship between disturbances in homocysteine
metabolism and pregnancy-related complications around conception or implantation (e.g., foe-
tal chromosomal abnormalities), disturbances in the first trimester resulting in development of
©2005 Eurekah.com.
Pregnancy Complications 145
congenital anomalies or early loss of pregnancy or placental disease, or maternal vascular com-
plications later in pregnancy.
As part of the literature review, we performed new meta-analyses. A Medline search was
performed with the following search terms: (MTHFR OR methylenetetrahydrofolate) AND
(pregnancy OR reproduct* OR trisomy OR Down syndrome OR chromosomal abnormalit*
OR foetal loss OR fetal loss OR foetal death OR fetal death OR abortion* OR miscarriag* OR
stillbirth OR placental vasculopathy OR placental ischaemia OR placental ischemia OR pla-
cental infarction OR fetal growth retardation OR fetal growth restriction OR foetal growth
retardation OR foetal growth restriction OR abruptio placentae OR placental abruption OR
preterm birth OR prematurity OR preeclampsia OR eclampsia OR pregnancy induced hyper-
tension). For data extraction, we included case-control studies. We calculated odds ratios for
each of the studies with corresponding 95% confidence intervals, using Woolf ’s method,19 and
tested for homogeneity as reported by Greenland.20 A pooled estimate was calculated by the
Mantel-Haenszel method and the 95% confidence intervals by the method of Robins.21
MTHFR 677C→T Polymorphism and Pregnancy-Related

Complications
Chromosomal Abnormalities
The incidence of Down syndrome is 1 per 150 conceptions or 1 in 600 live births. In 95%
of the cases, the cause is a failure in maternal chromosomal segregation (meiosis I) in the
ovum.22 Because abnormal chromosomal segregation is associated with DNA hypomethylation,23,24
James and coworkers investigated the relationship between the 677C→T mutation in the MTHFR
gene and Down syndrome by comparing the prevalence of this mutation in 57 mothers with a
child with Down syndrome (Down syndrome mothers) and 50 control mothers (Table 1).25 Ho-
mozygosity for the 677C→T MTHFR polymorphism (TT genotype) was found in 8% of the
control mothers and in 14% of the Down syndrome mothers. This difference was not statistically
significant. However, comparing the T allele frequency in both groups, 30% and 44% respectively,
results in a significant difference [OR (95% CI): 2.6(1.2-5.8)]. Enlargement of the study groups to
157 women with a child with Down syndrome and 144 control women gave comparable re-
sults.26 The authors concluded that Down syndrome may be related to the 677C→T mutation
in the MTHFR gene, but it could clearly not explain all cases of Down syndrome and therefore
a gene-environment interaction was suggested.
European studies, however, could not confirm these results.27-29 In this review, our overall
OR is estimated to be 0.9 with a 95% CI between 0.6 and 1.4 (Table 1). Although more
research is required, the current data suggest that the maternal MTHFR 677C→T polymor-
phism is not a risk factor for Down syndrome.
In addition, a study investigating the relationship between hyperhomocysteinemia and Down
syndrome did not find an association.29
Recently, both O’Leary and Hobbs and coworkers26,27 suggested a gene-gene interaction
with a three- to four-fold increased risk for Down syndrome if the maternal C/T or TT geno-
types (for the MTHFR 677C→T mutation) were combined with homozygosity for the 66A→G
mutation in the methionine synthase reductase (MTRR) gene. An investigation into the asso-
ciation of the 677C→T mutation in the MTHFR gene with human trisomies other than tri-
somy 21 reported a significantly increased frequency (45%) of the T allele in mothers of a
trisomy 18 conceptus, but not in mothers carrying a foetus with sex-chromosome or other
autosomal trisomies.30
Congenital Anomalies
Homozygosity for the MTHFR 677C→T polymorphism (TT genotype) is an established
risk factor for neural tube defects.10,31 Moreover, congenital oral clefts and heart defects have
also been associated with the 677C→T mutation in the MTHFR gene.31-36 Studies investigating
146
Table 1. Characteristics of studies on the relationship between the MTHFR 677C→T polymorphism and chromosomal abnormalities
Cases Controls
N (%) N (%)
TT Genotype/ TT Genotype/
Reference Country* Study Method Conducted In Total Group Total Group OR (95% CI)†‡
James et al 199925 US & CA maternal case- ? 8/57 (14) 4/50 (8) 3.2 (0.8-12.5)
control trisomy 21
O’Leary et al 200227§ IRL maternal case- 1986-1990 2/41 (5) 18/192 (9) 0.5 (0.1-2.2)
control trisomy 21
Stuppia et al 200228§ IT maternal case- ? 12/64 (19) 23/112 (21) 0.5 (0.2-1.3)
control trisomy 21
Hobbs et al 200026§ US & CA case-control study 1989-1998 22/157 (14) 14/140 (10) 2.1 (1.0-4.4)
trisomy 21 different
previous studies,
including James et al25
Chadefaux-Vekemans F case-control study ? 7/85 (8) 16/107 (15) 0.5 (0.2-1.3)

et al 200229§ trisomy 21
Hassold et al 200130 UK & US case-control study ? 9/44 (20) 5/56 (9) 2.6 (0.8-8.5)
trisomy 18
Overall odds ratio trisomy 21 test for heterogeneity: χ2 = 3.8; p>0.05 0.9 (0.6-1.4)
* Countries given as ISO-codes according to http://europa.eu.int/
† Odds ratio and 95% confidence intervals
‡ OR calculated by comparing the TT genotype of the 677 C→T MTHFR polymorphism with the combined C/C and C/T genotypes
§ Study used for the estimation of an overall OR
the relationship between these congenital anomalies and hyperhomocysteinemia support this
association.17,34,37-41 For details, the reader is referred to chapter 10, “Neural tube defects and
other congenital malformations”.
Recurrent Pregnancy Loss

About 15% of all clinically confirmed pregnancies end in a spontaneous loss before 16
weeks menstrual age. In 2% of the couples that want to become parents, this happens more
than once. This high percentage has motivated many investigators to look for risk factors for
this clinical entity. In 1992, Steegers-Theunissen et al42 suggested a relationship with distur-
bances in homocysteine metabolism. In 2000, in a meta-analysis of case-control studies, we
summarised the relationship between homozygosity for the MTHFR 677C→T polymorphism
and recurrent early pregnancy loss43-48 with an overall OR (95% CI) of 1.4 (1.0-2.0) (Table
2).49 After this meta-analysis, several new studies investigated the same relationship, and most
of them concluded that homozygosity for the MTHFR 677C→T polymorphism (TT geno-
type) was of no significance in the development of recurrent miscarriages (Table 2).50-57 In
contrast, Unfried and colleagues reported significantly different prevalences of the TT geno-
type in women with a history of recurrent pregnancy loss (17%) and in women with uncom-
plicated pregnancies (5%).58
If we consider all these reports, an overall odds ratio of 1.3 with a 95% CI from 1.0 to 1.6
is obtained (Table 2). In other words, a woman with the TT genotype has a 30% greater chance
of suffering from recurrent pregnancy loss than a woman with a CC or CT genotype. Most
investigations lacked the power to detect this small increase in risk. The current meta-analysis
again supports the hypothesis that the MTHFR 677TT genotype is a risk factor for recurrent
pregnancy loss.
In addition, some studies have suggested a contribution to pregnancy losses for the TT
genotype in the presence of other thrombophilia factors such as factor V Leiden or the factor II
G20210A mutation.55,59
Early and Late Pregnancy Loss

Most of the studies on pregnancy loss have included early (first trimester) as well as late
(second and third trimester) pregnancy losses.46,47,50-52,54-58,60 Because of the possibility of
different pathologic mechanisms between early and late pregnancy losses, we tried to subdivide
the data into these two groups and were able to estimate an overall odds ratio with a confidence
interval of 1.4(1.0-2.0) for recurrent early pregnancy losses (Table 2). It was not possible to
identify from all of the above studies women with only late recurrent pregnancy losses to esti-
mate an overall OR. However, some studies investigating the association between the MTHFR
polymorphism and late pregnancy losses did not identify the T/T genotype as a risk factor.61-63
Reports on recurrent pregnancy loss and hyperhomocysteinemia support the hypothesis of
disturbed homocysteine metabolism as a risk factor for recurrent pregnancy loss.42,44,64-66 A
meta-analysis on this subject has reported fasting hyperhomocysteinemia as well as
hyperhomocysteinemia after methionine loading, as risk factors for recurrent early pregnancy
loss, with an overall OR (95% CI) of 2.7 (1.4-5.2) and 4.2 (2.0-8.8), respectively.49
In addition, several studies have investigated the foetal survival advantage or disadvantage
of the MTHFR 677C→T polymorphism other than by case-control studies; they concluded
this MTHFR polymorphism to be a risk factor. Zetterberg and colleagues reported decreased
viability among foetuses with combined methylenetetrahydrofolate reductase 677C→T and
1298A→C mutated alleles.67 Isotalo et al observed the combination of mutant alleles of the
MTHFR 677C→T and 1298A→C variants in foetal tissue of spontaneously terminated preg-
nancies, but did not find three or four mutant alleles in liveborn neonates.68 They suggested
that with foetuses carrying three or four mutant alleles of the MTHFR 677C→T and 1298A→C
variants, a spontaneous pregnancy loss was very likely.68 Moreover, Hasbergen et al69 reported
an influence of the MTHFR 677C→T polymorphism on the number of offspring, after
observing a prevalence of the T allele of 0.3 and 0.16 for mothers with a singleton and twin
148
Table 2. Characteristics of studies on the relationship between the MTHFR 677C→T polymorphism and recurrent pregnancy loss
Cases Controls
N (%) N (%)
49
Nelen et al 2000 §| NL case-control 1994-1996
REPL¶ 29/185 (16) 6/113 (5) 3.3 (1.3-8.3)
Quere et al 199844§| F case-control ?
REPL¶ 20/100 (20) 14/100 (14) 1.5 (0.7-3.2)
Grandone et al 199845§| IT case-control ?
REPL¶ 17/94 (18) 28/150 (19) 1.0 (0.5-1.9)
Holmes et al 199946§| UK case-control ?
REPL¶ 11/129 (9) 6/67 (9) 0.9 (0.3-2.7)
RPL# 14/174 (8) 6/67 (9) 0.9 (0.3-2.4)
Kutteh et al 199947§ US case-control ?
RPL# 4/50 (8) 2/50 (4) 2.1 (0.4-11.9)
Lissak et al 199948§| IL case-control ?
REPL¶ 4/41 (10) 4/18 (22) 0.4 (0.1-1.7)
Brenner et al 199960§ IL case-control 1997-1998
RPL# 14/76 (18) 11/106 (10) 2.0 (0.8-4.6)
Wramsby et al 200050§ S case-control 1995-1997
RPL# 3/62 (5) 7/69 (10) 0.5 (0.1-1.8)
Foka et al 200051§ GR case-control ?
RPL# 6/80 (8) 15/100 (15) 0.5 (0.2-1.2)
Murphy et al 200052§ IRL case-control ?
RPL# 3/40 (8) 56/540 (10) 0.7 (0.2-2.3)
Nowak-Göttl et al 200053§| D case-control ?
REPL¶ 9/40 (23) 4/46 (9) 3.0 (0.9-10.8)
Pihusch et al 200154§ D case-control ?
RPL# 14/102 (14) 12/128 (9) 1.5 (0.7-3.5)

Table 2. Continued
Cases Controls
N (%) N (%)
55
Sarig et al 2002 § IL case-control 1995-1998
RPL# 17/145 (12) 19/145 (13) 0.9 (0.4-1.8)
Carp et al 200256§| IL case-control 1997-2000
REPL¶ 7/70 (10) 7/82 (9) 1.2 (0.4-3.6)
RPL# 14/108 (13) 7/82 (9) 1.6 (0.6-4.2)
Dilley et al 200257§ US case-control ?
RPL# 7/60 (12) 9/92 (10) 1.2 (0.4-3.5)
Unfried et al 200258§ AT case-control 1996-1999
RPL# 23/133 (17) 4/74 (5) 3.7 (1.2-11.0)
2
Overall odds ratio RPL# test for heterogeneity: χ = 23.3 ; p>0.05 1.3 (1.0-1.6)
2
Overall odds ratio REPL¶ test for heterogeneity: χ = 9.6 ; p>0.05 1.4 (1.0-2.0)
§ Study used for the estimation of an overall OR for recurrent pregnancy loss
| Study used for the estimation of an overall OR for recurrent early pregnancy loss
¶ Recurrent early pregnancy loss
# Recurrent pregnancy loss, early as well as late
149
pregnancy, respectively. Because the geographical differences in the incidence of dichorionic

twin pregnancies correspond to the geographical pattern of low T allele frequencies, they specu-
lated that a twin pregnancy in pregnant mothers with low folate concentrations and impaired
MTHFR activity might result in an undetected early loss of one of the cotwins. In accordance
with this hypothesis is the higher multiple births rate after periconceptional multivitamin (in-
cluding 0.8 mg folic acid daily) supplementation in Hungary.70 On the other hand, folic acid
supplementation or food fortification in China and California did not seem to be associated
with an increased occurrence of multiple births.71,72
Since the effect of the TT genotype on plasma homocysteine is related to folate intake73 and
folic acid supplementation is particularly effective in homocysteine lowering in TT individu-
als,74,75 one possible explanation for the differences in prevalence of the TT genotype in women
with recurrent pregnancy loss is the time period during which the studies were conducted. It is
plausible that the TT genotype is a risk factor for recurrent pregnancy loss primarily when it is
combined with reduced folate and/or increased homocysteine concentrations, as proposed for
congenital cleft palate risk.36 However, after several reports, in particular that of Czeizel and
Dudas,76 demonstrated the preventive role of periconceptional folic acid supplementation on
the occurrence of neural tube defects, periconceptional vitamin use increased substantially in
most developed countries. Moreover, due to the introduction of food fortification in some
countries, folate deficiency has nearly disappeared.77 Screening women with a history of recur-
rent pregnancy loss for the 677C→T MTHFR polymorphism at a time when many women
were using folic acid supplementation periconceptionally, and comparing them with women
who were pregnant at an earlier time, when there was no folic acid supplementation or food
fortification, may give an underestimation of TT prevalence in women with recurrent preg-
nancy loss.
In conclusion, the 677C→T mutation in the MTHFR gene as well as hyperhomocysteinemia
are risk factors for recurrent (early) pregnancy losses. The effect of the T/T genotype appears to
be similar for recurrent early and late pregnancy losses. Of particular note, the data on the
677C→T MTHFR polymorphism support a causal relationship between disturbed homocys-
teine metabolism and recurrent (early) pregnancy loss.
Placental Vasculopathy
Complications later in pregnancy are often summarised as placental vasculopathy, which
consists of abruption of the placenta, placental infarction or ischaemia and its consequence of
foetal growth retardation.
Abruptio placentae is a total or partial early separation of the placenta from the uterine wall
caused by retroplacental bleeding. Its incidence is between 0.5 to 1.5% of all pregnancies, with
a recurrence rate of about 14%. The main cause of this severe obstetrical complication is thought
to be damage of the maternal vessels, the spiral arteries in the placental bed. This clinical
picture can result in substantial risks for the mother due to the consumptive coagulopathy
induced by the retroplacental bleeding, but the greatest risk is for the unborn foetus, namely
foetal hypoxaemia and foetal death.
The other form of placental vasculopathy is placental ischaemia or infarction leading to
diminished placental function and, consequently, foetal growth restriction. The incidence is
5-10% of all pregnancies, with a high recurrence rate of 50%.
Combining these two clinical pictures as placental vasculopathy, Ray and Laskin summarised
the relationship with the 677C→T mutation in the MTHFR gene in an overall OR (95% CI)
of 2.3 (1.1-4.9).78 Their meta-analysis contained the studies of Van der Molen et al79 and
Kupferminc et al (Table 3).61 More recent studies investigating the relationship between the
677C→T mutation in the MTHFR gene and placental vasculopathy, as well as abruptio pla-
centae or intra-uterine growth retardation as distinct diseases, are summarised in Table 3.80-85
Our calculated overall OR and 95% CI are 1.4 (1.1-1.9), 1.2 (0.3-4.4) and 1.4 (1.0-1.9) for
placental vasculopathy, abruptio placentae and placental ischaemia, respectively (Table 3).
Table 3. Characteristics of studies on the relationship between the MTHFR 677C→T polymorphism and placental vasculopathy
Cases Controls
N (%) N (%)
Kupferminc et al 199961§| ¶ IL case-control 1996-1997

PV# 15/64 (23) 9/110 (8) 3.4 (1.4-8.4)
AP** 3/20 (15) 9/110 (8) 2.2 (0.5-8.9)
IUGR†† 12/44 (27) 9/110 (8) 4.2 (1.6-10.9)
Van der Molen et al 200079§| NL case-control 1992-1997
PV# 19/165 (12) 7/139 (5) 2.5 (1.0-6.0)
Martinelli et al 200180§¶ IT case-control 1997-1998
IUGR†† 17/61 (28) 19/93 (20) 1.5 (0.7-3.2)
Gebhardt et al 200181§| ¶ ZA case-control 1996-1999
PV# 0/54 (0) 2/114 (2) NC‡‡
AP** 0/18 (0) 2/114 (2) NC‡‡
IUGR†† 0/36 (0) 2/114 (2) NC‡‡
Verspyck et al 200282§¶ F case-control 1998-1999
IUGR†† 11/97 (11) 15/97 (15) 0.7 (0.3-1.6)
Infante-Rivard et al 200283§¶ CA case-control 1998-2000
IUGR†† 45/490 (9) 35/467 (7) 1.2 (0.8-2.0)
Kupferminc et al 200284§¶| IL case-control 1999-2001
IUGR†† 5/26 (19) 5/52 (10) 2.2 (0.6-8.6)
Hira et al 200285§| ZA case-control ?
AP** 0/100 (0) 2/217 (1) NC‡‡
Overall odds ratio placental vasculopathy test for heterogeneity: χ2 = 8.6; p>0.05 1.4 (1.1-1.9)
Overall odds ratio abruptio placentae test for heterogeneity: χ2 = 0; p>0.05 1.2 (0.3-4.4)
Overall odds ratio placental ischaemia or foetal growth retardation test for heterogeneity: χ2 = 8.5; p>0.05 1.4 (1.0-1.9)
* Countries given as ISO-codes according to http://europa.eu.int/ ¶ Study used for the estimation of an overall OR for intra-uterine growth retardation
† Odds ratio and 95% confidence intervals # Placental vasculopathy
‡ OR calculated by comparing the TT genotype of the 677 C→T MTHFR ** Abruptio placentae
polymorphism with the combined C/C and C/T genotypes †† Intra-uterine growth retardation
§ Study used for the estimation of an overall OR for placental vasculopathy ‡‡ Not calculated
| Study used for the estimation of an overall OR for abruptio placentae
151
Infante-Rivard and coworkers suggested an increased risk for intra-uterine growth retarda-
tion in the absence of multivitamin supplement use.83 They also investigated the role of the
neonate’s homozygosity for the MTHFR 677C→T variant in the development of intra-uterine
growth retardation and found it to be protective with an OR (95% CI) of 0.5 (0.3-0.9).83
However, maternal homozygosity for the MTHFR 1298A→C variant was found to be pro-
tective for intra-uterine growth retardation.83 This was in contrast with Gebhardt et al81 who
found an OR and 95% CI of 2.0 (0.1-4.6) and 3.2 (1.0-10.4) for intra-uterine growth retarda-
tion and abruptio placentae, respectively.
Steegers-Theunissen42 suggested an association of intra-uterine growth retardation and abrup-
tio placentae with hyperhomocysteinemia, which was confirmed by Goddijn-Wessel et al86
and Owen et al.87 Ray and Laskin78 summarised these studies with an overall OR (95% CI) for
hyperhomocysteinemia, fasting and after methionine loading, of 5.3 (1.8-15.9) and 4.2
(1.2-15.0), respectively.
In addition to these case-control studies combining abruptio placentae and placental is-
chaemia as placental vasculopathy, Vollset and coworkers17 analysed these groups separately in
the Hordaland Homocysteine Study. They calculated an OR (95% CI) of 2.0 (1.2-3.3) for
very low birth weight and 3.1 (1.6-6.0) for placental abruption, in cases with elevated ho-
mocysteine concentrations.
One might expect to find thrombotic lesions in the placenta of a woman with a genetically
thrombophilic state. Some studies have found such lesions in the form of infarction, fibrin
deposition, fibrinoid necrosis or blood vessel obliteration, while others have not.88-90 None of
these studies described the relationship between placental histology and the MTHFR 677C→T
polymorphism. Two studies investigating placental pathology in cases with hyper-
homocysteinemia91,92 reported more placental calcifications and abnormal placentation.
Therefore, based on these data, we conclude that hyperhomocysteinemia and homozygos-
ity for the 677C→T mutation in the MTHFR gene are risk factors for placental vasculopathy,
despite its clinical heterogeneity.
Preterm Birth
Preterm birth, delivery at < 37 weeks of pregnancy, is a key outcome measure in developed
countries because of its strong relationship to infant mortality, long-term morbidity and high
health-care costs. Its incidence is between 4 and 10% of all pregnancies.
Several authors have hypothesized disturbances in homocysteine metabolism as one of the
causal pathways of preterm delivery.93-95 However, Lauszus et al found preterm delivery in
diabetic women not to be associated with a higher frequency of the 677C→T MTHFR poly-
morphism.96
Preeclampsia
The clinical picture of preeclampsia consists of the triad of hypertension, albuminuria and
maternal oedema. The incidence is about 5% of all pregnancies, but it occurs mostly in a
woman’s first pregnancy with a specific partner. A syndrome associated with preeclampsia is the
combination of haemolysis, elevated liver enzymes and low platelets (HELLP). A severe com-
plication of preeclampsia is eclampsia, synonymous with seizures. Eclampsia is the leading
cause of maternal death in developed countries.
For the relationship between (pre)eclampsia and the maternal MTHFR 677C→T polymor-
phism, we refer to the meta-analysis of Morrison et al.97 They summarised 14 studies on the
relationship between the MTHFR 677C→T polymorphism and gestational hypertension re-
lated diseases, such as isolated pregnancy induced hypertension, (pre)eclampsia and HELLP
syndrome.61,97-109 The authors reported an OR (95% CI) of 1.0 (0.8-1.3) and 1.5 (1.0-2.2) for
preeclampsia and severe preeclampsia, respectively.
Extension of this meta-analysis with five additional studies110-114 leads to our pooled risk
estimation of 1.2 with a 95% CI from 1.0 to 1.4 (Table 4).
Table 4. Characteristics of studies on the relationship between the MTHFR 677C→T polymorphism and (pre)eclampsia
Cases Controls
N (%) N (%)
Sohda et al 1997109†† JP case-control ? 16/67 (24) 11/98 (11) 2.5 (1.1-5.8)

PE§
Chikosi et al 1999110†† ZA case-control ? 1/105 (1) 0/110 (0) NC**
PE§
O’Shaughnessy et al 199999†† UK case-control 1996-1998 31/283 (11) 23/200 (12) 0.9 (0.5-1.7)
PE§
Grandone et al 199998†† IT case-control ? 34/139 (24) 36/216 (17) 1.6 (1.0-2.7)
PE§ + PIH|
Powers et al 1999115†† US case-control ? 15/99 (15) 14/114 (12) 1.3 (0.6-2.8)
PE§
Kupferminc et al 199961†† IL case-control 1996-1997 7/34 (21) 9/110 (8) 2.9 (1.0-8.5)
severe PE§
Kobashi et al 2000108†† JP case-control ? 12/101 (12) 33/215 (15) 0.7 (0.4-1.5)
PE§ + PIH|
Rigó et al 2000101†† HU case-control ? 8/120 (7) 6/101 (6) 1.1 (0.4-3.4)
PE§
Kupferminc et al 2000100†† IL case-control 1997-1999 15/63 (24) 12/126 (10) 3.0 (1.3-6.8)
severe PE§
Laivuori et al 2000106†† FI case-control 1988-1998 4/113 (4) 7/103 (7) 0.5 (0.1-1.8)
PE§
Rajkovic et al 2000112†† ZW case-control 1995-1996 1/171 (0.5) 0/183 (0) NC**
PE§
Zusterzeel et al 2000104†† NL case-control 1985-1996 21/176 (12) 36/403 (9) 1.4 (0.8-2.4)
PE§ + HELLP¶
153
154
Table 4. Characteristics of studies on the relationship between the MTHFR 677C→T polymorphism and (pre)eclampsia
Cases Controls
N (%) N (%)
102
Kim et al 2001 †† US case-control 1996-1999 33/281 (12) 41/360 (11) 1.0 (0.6-1.7)
PE§ + HELLP¶
Lachmeijer et al 2001105†† NL case-control 1997 14/174 (8) 11/120 (9) 0.9 (0.4-2.0)
PE§ + HELLP¶ + E#
Kaiser et al 2001111†† AU case-control 1984-1999 19/156 (12) 11/79 (14) 0.9 (0.4-1.9)
PE§ + E#
Livingston et al 2001103†† US case-control severe 1998-1999 10/110 (9) 7/97 (7) 1.3 (0.5-3.5)
PE§
Prasmusinto et al 2002114†† case-control ?
D 1/15 (7) 7/34 (21) 0.3 (0.0-2.5)
HR PE§ 2/25 (8) 5/38 (13) 0.6 (0.1-3.2)
ID 1/41 (2) 0/27 (0) NC**
D’Elia et al 2002113†† IT case-control ? 6/58 (10) 9/74 (12) 0.8 (0.3-2.5)
PE§
Morrison et al 200297†† UK case-control 1951-1970 42/404 (10) 17/164 (10) 1.0 (0.6-1.8)
PE§ + PIH|?
Overall odds ratio (pre)eclampsia χ2 = 20.6 ; p>0.05 1.2 (1.0-1.4)
§ Preeclampsia
| Isolated pregnancy induced hypertension
¶ Syndrome of haemolysis, elevated liver enzymes and low platelets
# Eclampsia
** Not calculated
†† Study used for the estimation of an overall OR
Foetal homozygosity for the 677C→T MTHFR polymorphism may not be associated with
preeclampsia.103,114,115 Maternal 1298A→C MTHFR polymorphism was found not to be a
risk factor for gestational hypertension associated diseases by Zusterzeel et al,104 Kaiser et al116
and Lachmeijer et al.105
Dekker and coworkers117 reported hyperhomocysteinemia as a risk factor for preeclampsia
in 1995, a finding confirmed by others.107,118-125 In agreement with these results, the Hordaland
Homocysteine Study showed a tendency for increased risk for preeclampsia with elevating total
homocysteine concentrations.17 In contrast, other studies could not confirm an association
between mild hyperhomocysteinemia and preeclampsia.112,126-129
In conclusion, both hyperhomocysteinemia and homozygosity for the 677C→T mutation
in the MTHFR gene may well lead to an elevated risk for developing preeclampsia.
Treatment
Many patients with classic or severe homocystinuria respond to treatment with pyridoxine
(vitamin B6) by reducing homocysteine concentration, which results in a reduction of the
number of vascular accidents.130 Additional supplementation with folic acid, hydroxycobalamin
(vitamin B12) and betaine results in a decrease in homocysteine concentration in the case of
pyridoxine-resistance.
Reduction of homocysteine concentration in mild hyperhomocysteinemia has also been
reported to occur with B vitamin supplementation and, in particular, with the use of folic
acid.131,132 This effect of folic acid supplementation seems to be most pronounced in individu-
als homozygous for the 677C→T mutation.74,75,133 However, the beneficial effect of lowering
homocysteine concentrations for the prevention of cardiovascular events is still unclear. Fur-
thermore, except for the MRC Vitamin Study134 and the study by Czeizel and Dudas76 which
investigated the preventive effect of periconceptional vitamin supplementation on the recur-
rence and occurrence of neural tube defects, the prevention of obstetrical complications by
folic acid or B-vitamin therapy has not been investigated in a randomised controlled trial. An
important consequence of the prevention of neural tube defects by folic acid is the recommen-
dation in many countries for women to use folic acid periconceptionally. This has made it
ethically impossible to investigate the effect of folic acid supplementation, by a placebo-controlled
randomised trial, on the prevalence of obstetrical complications in the first weeks of pregnancy.
In addition, folic acid food fortification programs in some countries have also made it difficult
to conduct these types of investigations.
Nonetheless, observational supplementation studies suggest that some populations of women
—particularly poor women—may benefit from folic acid supplementation during, as well as
before, pregnancy.135-137
Another continuing debate is the optimum dose of folic acid. Some suggest folic acid forti-
fication levels or an increase in the dose of folic acid supplementation,138,139 while others have
reported lower doses to be effective.140
Pathophysiology
Many pathogenic mechanisms have been postulated to explain the clinical consequences of
hyperhomocysteinemia, but, at present, it is still unclear as to which of these mechanisms is the
most relevant. Risk factors for venous thrombosis, such as the factor V Leiden mutation, are
also risk factors for late pregnancy complications,141,142 but possibly also for early pregnancy
loss.143 Because an elevated homocysteine concentration is also a risk factor for venous throm-
bosis,13 the relationship between disturbed homocysteine metabolism and late pregnancy com-
plications is likely due to thrombotic events. For the obstetrical complications, other mecha-
nisms such as oxidative stress,144 disturbances in the differentiation and outgrowth of neural
crest and neuroepithelial cells,145 inhibition of retinoic acid synthesis,146 increased apoptosis,147
toxicity148,149 and reduced chorionic vascularization150 have been described. In addition, the
preventive effects of folic acid supplementation and food fortification have been suggested to
Figure 1. Pooled odds ratios and 95% confidence intervals of the 677C→T mutation in the MTHFR gene
in case-control studies of trisomy 21 (n=4), recurrent pregnancy loss (n=16), placental vasculopathy (n=8)
and preeclampsia (n=19).
be attributed to terathanasia,151 increased human cell stability, reduced hypomethylation of

DNA152 and improved vascular endothelial function.153 However, extensive research is re-
quired to elucidate the exact pathogenic role of disturbances in homocysteine metabolism in
pregnancy complications.
Conclusions
Based on the data presented in this chapter, we conclude that hyperhomocysteinemia and
homozygosity for the 677C→T mutation in the MTHFR gene are associated with many preg-
nancy complications (Fig. 1). Nevertheless, there is a need for randomised controlled trials to
investigate whether supplementation with B vitamins, in particular with folic acid, will prevent
these complications.
References
1. Kluijtmans LA, Boers GH, Kraus JP et al. The molecular basis of cystathionine beta-synthase defi-
ciency in Dutch patients with homocystinuria: Effect of CBS genotype on biochemical and clinical
phenotype and on response to treatment. Am J Hum Genet 1999; 65:59-67.
2. Brenton DP, Cusworth DC, Biddle SA et al. Pregnancy and homocystinuria. Ann Clin Biochem
1977; 14:161-2.
3. Levy HL, Vargas JE, Waisbren SE et al. Reproductive fitness in maternal homocystinuria due to
cystathionine beta-synthase deficiency. J Inherit Metab Dis 2002; 25:299-314.
4. Constantine G, Green A. Untreated homocystinuria: A maternal death in a woman with four
pregnancies. Case report. Br J Obstet Gynaecol 1987; 94:803-6.
5. Ritchie JW, Carson NA. Pregnancy and homocystinuria. J Obstet Gynaecol Br Commonw 1973;
80:664-9.
6. Mudd HS, Skovby F, Levy HL et al. The natural history of homocystinuria due to cystathionine
b- synthase deficiency. Am J Hum Genet 1985; 37:1-31.
7. Burke G, Robinson K, Refsum H et al. Intrauterine growth retardation, perinatal death, and ma-
ternal homocysteine levels. N Engl J Med 1992; 326:69-70.
8. Minkhorst AG, van Dongen PW, Boers GH et al. Cerebral infarction after caesarean section due
to heterozygosity for homocystinuria; a case report. Eur J Obstet Gynecol Reprod Biol 1991;
40:241-3.
10. Van der Put NM, Steegers-Theunissen RPM, Frosst P et al. Mutated methylenetetrahydrofolate
reductase as a risk factor for spina bifida. Lancet 1995; 346:1070-1.
93:7-9.
13. Den Heijer M, Rosendaal FR, Blom HJ et al. Hyperhomocysteinemia and venous thrombosis: A
meta-analysis. Thromb Haemost 1998; 80:874-7.
14. Aubard Y, Darodes N, Cantaloube M. Hyperhomocysteinemia and pregnancy—review of our present
understanding and therapeutic implications. Eur J Obstet Gynecol Reprod Biol 2000; 93:157-65.
15. Eskes TK. Homocysteine and human reproduction. Clin Exp Obstet Gynecol 2000; 27:157-67.
16. Obwegeser R, Hohlagschwandtner M, Sinzinger H. Homocysteine—a pathophysiological corner-
stone in obstetrical and gynaecological disorders? Hum Reprod Update 1999; 5:64-72.
17. Vollset SE, Refsum H, Irgens LM et al. Plasma total homocysteine, pregnancy complications, and
adverse pregnancy outcomes: The Hordaland Homocysteine study. Am J Clin Nutr 2000; 71:962-8.
18. Nelen WLDM. Hyperhomocysteinaemia and human reproduction. Clin Chem Lab Med 2001;
39:758-63.
19. Schlesselman JJ. Basic methods of analysis. Case-control studies. Design, conduct, analysis. New
York: Oxford University Press, 1982: 176-177.
20. Greenland S. Quantitative methods in the review of epidemiologic literature. Epidemiol Rev 1987;
9:1-30.
21. Rothman KJ. Modern epidemiology. Boston: Little, Brown & Co, 1986.
22. Antonarakis SE, Petersen MB, McInnis MG et al. The meiotic stage of nondisjunction in trisomy
21: Determination by using DNA polymorphisms. Am J Hum Genet 1992; 50:544-50.
23. Leyton C, Mergudich D, de la Torre C et al. Impaired chromosome segregation in plant anaphase
after moderate hypomethylation of DNA. Cell Prolif 1995; 28:481-96.
24. Lengauer C, Kinzler KW, Vogelstein B. DNA methylation and genetic instability in colorectal
cancer cells. Proc Natl Acad Sci USA 1997; 94:2545-50.
Clin Nutr 1999; 70:495-501.
30. Hassold TJ, Burrage LC, Chan ER et al. Maternal folate polymorphisms and the etiology of hu-
man nondisjunction. Am J Hum Genet 2001; 69:434-9.
32. Martinelli M, Scapoli L, Pezzetti F et al. C677T variant form at the MTHFR gene and CL/P: A
risk factor for mothers? Am J Med Genet 2001; 98:357-60.
33. Mills JL, Kirke PN, Molloy AM et al. Methylenetetrahydrofolate reductase thermolabile variant
and oral clefts. Am J Med Genet 1999; 86:71-4.
34. Wenstrom KD, Johanning GL, Johnston KE et al. Association of the C677T methyl-
enetetrahydrofolate reductase mutation and elevated homocysteine levels with congenital cardiac
malformations. Am J Obstet Gynecol 2001; 184:806-12.
35. Junker R, Kotthoff S, Vielhaber H et al. Infant methylenetetrahydrofolate reductase 677TT geno-
type is a risk factor for congenital heart disease. Cardiovasc Res 2001; 51:251-4.
36. van Rooij IALM, Vermeij-Keers C, Kluijtmans LAJ et al. Does the interaction between maternal
folate intake and the methylenetetrahydrofolate reductase polymorphisms affect cleft lip with or
without cleft palate risk? Am J Epidemiol 2003; 157:583-91.
37. Wong WY, Eskes TK, Kuijpers-Jagtman AM et al. Nonsyndromic orofacial clefts: Association with
maternal hyperhomocysteinemia. Teratology 1999; 60:253-7.
38. Steegers-Theunissen RPM, Boers GHJ, Trijbels FJM et al. Maternal hyperhomocysteinemia: A risk
factor for neural-tube defects? Metabolism 1994; 43:1475-80.
39. Mills JL, McPartlin JM, Kirke PN et al. Homocysteine metabolism in pregnancies complicated by
neural- tube defects. Lancet 1995; 345:149-51.
40. Van der Put NM, Thomas CM, Eskes TK et al. Altered folate and vitamin B12 metabolism in
families with spina bifida offspring. QJM 1997; 90:505-10.
41. Kapusta L, Haagmans ML, Steegers EA et al. Congenital heart defects and maternal derangement
of homocysteine metabolism. J Pediatr 1999; 135:773-4.
42. Steegers-Theunissen RPM, Boers GHJ, Blom HJ et al. Hyperhomocysteinaemia and recurrent spon-
taneous abortion or abruptio placentae. Lancet 1992; 339:1122-3.
43. Nelen WLDM, Steegers EAP, Eskes TKAB et al. Genetic risk factor for unexplained recurrent
early pregnancy loss. Lancet 1997; 350:861.
44. Quere I, Bellet H, Hoffet M et al. A woman with five consecutive fetal deaths: Case report and
retrospective analysis of hyperhomocysteinemia prevalence in 100 consecutive women with recur-
rent miscarriages. Fertil Steril 1998; 69:152-4.
45. Grandone E, Margaglione M, Colaizzo D et al. Methylene tetrahydrofolate reductase (MTHFR)
677T-->C mutation and unexplained early pregnancy loss. Thromb Haemost 1998; 79:1056-7.
46. Holmes ZR, Regan L, Chilcott I et al. The C677T MTHFR gene mutation is not predictive of
risk for recurrent fetal loss. Br J Haematol 1999; 105:98-101.
47. Kutteh WH, Park VM, Deitcher SR. Hypercoagulable state mutation analysis in white patients
with early first-trimester recurrent pregnancy loss. Fertil Steril 1998; 71:1048-53.
48. Lissak A, Sharon A, Fruchter O et al. Polymorphism for mutation of cytosine to thymine at loca-
tion 677 in the methylenetetrahydrofolate reductase gene is associated with recurrent early fetal
loss. Am J Obstet Gynecol 1999; 181:126-30.
49. Nelen WLDM, Blom HJ, Steegers EA et al. Hyperhomocysteinemia and recurrent early pregnancy
loss: A meta-analysis. Fertil Steril 2000; 74:1196-9.
50. Wramsby ML, Sten-Linder M, Bremme K. Primary habitual abortions are associated with high
frequency of factor V Leiden mutation. Fertil Steril 2000; 74:987-91.
51. Foka ZJ, Lambropoulos AF, Saravelos H et al. Factor V leiden and prothrombin G20210A muta-
tions, but not methylenetetrahydrofolate reductase C677T, are associated with recurrent miscar-
riages. Hum Reprod 2000; 15:458-62.
52. Murphy RP, Donoghue C, Nallen RJ et al. Prospective evaluation of the risk conferred by factor
V Leiden and thermolabile methylenetetrahydrofolate reductase polymorphisms in pregnancy.
Arterioscler Thromb Vasc Biol 2000; 20:266-70.
53. Nowak-Gottl U, Sonntag B, Junker R et al. Evaluation of lipoprotein(a) and genetic prothrombotic
risk factors in patients with recurrent foetal loss. Thromb Haemost 2000; 83:350-1.
54. Pihusch R, Buchholz T, Lohse P et al. Thrombophilic gene mutations and recurrent spontaneous
abortion: Prothrombin mutation increases the risk in the first trimester. Am J Reprod Immunol
2001; 46:124-31.
55. Sarig G, Younis JS, Hoffman R et al. Thrombophilia is common in women with idiopathic preg-
nancy loss and is associated with late pregnancy wastage. Fertil Steril 2002; 77:342-7.
56. Carp H, Salomon O, Seidman D et al. Prevalence of genetic markers for thrombophilia in recur-
rent pregnancy loss. Hum Reprod 2002; 17:1633-7.
57. Dilley A, Benito C, Hooper WC et al. Mutations in the factor V, prothrombin and MTHFR
genes are not risk factors for recurrent fetal loss. J Matern Fetal Neonatal Med 2002; 11:176-82.
58. Unfried G, Griesmacher A, Weismuller W et al. The C677T polymorphism of the methyl-
enetetrahydrofolate reductase gene and idiopathic recurrent miscarriage. Obstet Gynecol 2002;
99:614-9.
59. Grandone E, Margaglione M, Colaizzo D et al. Presence of FV Leiden and MTHFR mutation in
a patient with complicated pregnancies. Thromb Haemost 1997; 77:1036-7.
60. Brenner B, Sarig G, Weiner Z et al. Thrombophilic polymorphisms are common in women with
fetal loss without apparent cause. Thromb Haemost 1999; 82:6-9.
61. Kupferminc MJ, Eldor A, Steinman N et al. Increased frequency of genetic thrombophilia in women
with complications of pregnancy. N Engl J Med 1999; 340:9-13.
62. Martinelli I, Taioli E, Cetin I et al. Mutations in coagulation factors in women with unexplained
late fetal loss. N Engl J Med 2000; 343:1015-8.
63. Gris JC, Quere I, Monpeyroux F et al. Case-control study of the frequency of thrombophilic
disorders in couples with late foetal loss and no thrombotic antecedent - The nimes obstetricians
and haematologists study (NOHA). Thromb Haemost 1999; 81:891-9.
64. Wouters MGAJ, Boers GHJ, Blom HJ et al. Hyperhomocysteinemia: A risk factor in women with
unexplained recurrent early pregnancy loss. Fertil Steril 1993; 60:820-5.
65. Nelen WLDM, Blom HJ, Steegers EAP et al. Homocysteine and folate levels as risk factors for
recurrent early pregnancy loss. Obstet Gynecol 2000; 95:519-24.
66. Coumans ABC, Huijgens PC, Jakobs C et al. Haemostatic and metabolic abnormalities in women
with unexplained recurrent abortion. Hum Reprod 1999; 14:211-4.
67. Zetterberg H, Regland B, Palmer M et al. Increased frequency of combined methylenetetrahydrofolate
reductase C677T and A1298C mutated alleles in spontaneously aborted embryos. Eur J Hum Genet
2002; 10:113-8.
68. Isotalo PA, Wells GA, Donnelly JG. Neonatal and fetal methylenetetrahydrofolate reductase ge-
netic polymorphisms: An examination of C677T and A1298C mutations. Am J Hum Genet 2000;
67:986-90.
69. Hasbargen U, Lohse P, Thaler CJ. The number of dichorionic twin pregnancies is reduced by the
common MTHFR 677C-->T mutation. Hum Reprod 2000; 15:2659-62.
70. Czeizel AE, Metneki J, Dudas I. Higher rate of multiple births after periconceptional vitamin
supplementation. N Engl J Med 1994; 330:1687-8.
71. Li Z, Gindler J, Wang H et al. Folic acid supplements during early pregnancy and likelihood of
multiple births: A population-based cohort study. Lancet 2003; 361:380-4.
72. Shaw GM, Carmichael SL, Nelson V et al. Food fortification with folic acid and twinning among
California infants. Am J Med Genet 2003; 119A:137-40.
73. de Bree A, Verschuren WM, Kromhout D et al. Homocysteine determinants and the evidence to
what extent homocysteine determines the risk of coronary heart disease. Pharmacol Rev 2002;
54:599-618.
75. Nelen WLDM, Blom HJ, Thomas CMG et al. Methylenetetrahydrofolate reductase polymorphism
affects the change in homocysteine and folate concentrations resulting from low dose folic acid
supplementation in women with unexplained recurrent miscarriages. J Nutr 1998; 128:1336-41.
76. Czeizel AE, Dudas I. Prevention of the first occurrence of neural-tube defects by periconceptional
vitamin supplementation. N Engl J Med 1992; 327:1832-5.
77. Jacques PF, Selhub J, Bostom AG et al. The effect of folic acid fortification on plasma folate and
total homocysteine concentrations. N Engl J Med 1999; 340:1449-54.
78. Ray JG, Laskin CA. Folic acid and homocyst(e)ine metabolic defects and the risk of placental
abruption, preeclampsia and spontaneous pregnancy loss: A systematic review. Placenta 1999;
20:519-29.
79. Van der Molen EF, Arends GE, Nelen WL et al. A common mutation in the 5,10-methylenetetra-
hydrofolate reductase gene as a new risk factor for placental vasculopathy. Am J Obstet Gynecol
2000; 182:1258-63.
80. Martinelli P, Grandone E, Colaizzo D et al. Familial thrombophilia and the occurrence of fetal
growth restriction. Haematologica 2001; 86:428-31.
81. Gebhardt GS, Scholtz CL, Hillermann R et al. Combined heterozygosity for methyl-
enetetrahydrofolate reductase (MTHFR) mutations C677T and A1298C is associated with abrup-
tio placentae but not with intrauterine growth restriction. Eur J Obstet Gynecol Reprod Biol 2001;
97:174-7.
82. Verspyck E, Le CD, Goffinet F et al. Thrombophilia and immunological disorders in pregnancies
as risk factors for small for gestational age infants. BJOG 2002; 109:28-33.
83. Infante-Rivard C, Rivard GE, Yotov WV et al. Absence of association of thrombophilia polymor-
phisms with intrauterine growth restriction. N Engl J Med 2002; 347:19-25.
84. Kupferminc MJ, Many A, Bar-Am A et al. Mid-trimester severe intrauterine growth restriction is
associated with a high prevalence of thrombophilia. BJOG 2002; 109:1373-6.
85. Hira B, Pegoraro RJ, Rom L et al. Polymorphisms in various coagulation genes in black South
African women with placental abruption. BJOG 2002; 109:574-5.
86. Goddijn Wessel TA, Wouters MGAJ, Van der Molen EF et al. Hyperhomocysteinemia: A risk
factor for placental abruption or infarction. Eur J Obstet Gynecol Reprod Biol 1996; 66:23-9.
87. Owen EP, Human L, Carolissen AA et al. Hyperhomocysteinemia—a risk factor for abruptio pla-
centae. J Inherit Metab Dis 1997; 20:359-62.
88. Mousa HA, Alfirevic. Do placental lesions reflect thrombophilia state in women with adverse preg-
nancy outcome? Hum Reprod 2000; 15:1830-3.
89. Many A, Schreiber L, Rosner S et al. Pathologic features of the placenta in women with severe
pregnancy complications and thrombophilia. Obstet Gynecol 2001; 98:1041-4.
90. Sikkema JM, Franx A, Bruinse HW et al. Placental pathology in early onset preeclampsia and
intra-uterine growth restriction in women with and without thrombophilia. Placenta 2002;
23:337-42.
91. Khong TY, Hague WM. The placenta in maternal hyperhomocysteinaemia. Br J Obstet Gynaecol
1999; 106:273-8.
92. Bohles H, Arndt S, Ohlenschlager U et al. Maternal plasma homocysteine, placenta status and
docosahexaenoic acid concentration in erythrocyte phospholipids of the newborn. Eur J Pediatr
1999; 158:243-6.
93. Kramer MS, Goulet L, Lydon J et al. Socio-economic disparities in preterm birth: Causal pathways
and mechanisms. Paediatr Perinat Epidemiol 2001; 15 Suppl 2:104-23.
94. Wang X, Zuckerman B, Kaufman G et al. Molecular epidemiology of preterm delivery: Methodol-
ogy and challenges. Paediatr Perinat Epidemiol 2001; 15(Suppl 2):63-77.
95. Ferguson SE, Smith GN, Walker MC. Maternal plasma homocysteine levels in women with preterm
premature rupture of membranes. Med Hypotheses 2001; 56:85-90.
96. Lauszus FF, Gron PL, Klebe JG. Association of polymorphism of methylene-tetrahydro-folate-re-
ductase with urinary albumin excretion rate in type 1 diabetes mellitus but not with preeclampsia,
retinopathy, and preterm delivery. Acta Obstet Gynecol Scand 2001; 80:803-6.
97. Morrison ER, Miedzybrodzka ZH, Campbell DM et al. Prothrombotic genotypes are not associ-
ated with preeclampsia and gestational hypertension: Results from a large population-based study
and systematic review. Thromb Haemost 2002; 87:779-85.
98. Grandone E, Margaglione M, Colaizzo D et al. Prothrombotic genetic risk factors and the occur-
rence of gestational hypertension with or without proteinuria. Thromb Haemost 1999; 81:349-52.
99. O’Shaughnessy KM, Fu B, Ferraro F et al. Factor V Leiden and thermolabile methyl-
enetetrahydrofolate reductase gene variants in an East Anglian preeclampsia cohort. Hypertension
1999; 33:1338-41.
100. Kupferminc MJ, Fait G, Many A et al. Severe preeclampsia and high frequency of genetic
thrombophilic mutations. Obstet Gynecol 2000; 96:45-9.
101. Rigo Jr J, Nagy B, Fintor L et al. Maternal and neonatal outcome of preeclamptic pregnancies:
The potential roles of factor V Leiden mutation and 5,10 methylenetetrahydrofolate reductase.
Hypertens Pregnancy 2000; 19:163-72.
102. Kim YJ, Williamson RA, Murray JC et al. Genetic susceptibility to preeclampsia: Roles of
cytosineto-thymine substitution at nucleotide 677 of the gene for methylenetetrahydrofolate reduc-
tase, 68-base pair insertion at nucleotide 844 of the gene for cystathionine beta-synthase, and fac-
tor V Leiden mutation. Am J Obstet Gynecol 2001; 184:1211-7.
103. Livingston JC, Barton JR, Park V et al. Maternal and fetal inherited thrombophilias are not related
to the development of severe preeclampsia. Am J Obstet Gynecol 2001; 185:153-7.
104. Zusterzeel PL, Visser W, Blom HJ et al. Methylenetetrahydrofolate reductase polymorphisms in
preeclampsia and the HELLP syndrome. Hypertens Pregnancy 2000; 19:299-307.
105. Lachmeijer AM, Arngrimsson R, Bastiaans EJ et al. Mutations in the gene for methyl-
enetetrahydrofolate reductase, homocysteine levels, and vitamin status in women with a history of
preeclampsia. Am J Obstet Gynecol 2001; 184:394-402.
106. Laivuori H, Kaaja R, Ylikorkala O et al. 677 C-->T polymorphism of the methylenetetrahydrofolate
reductase gene and preeclampsia. Obstet Gynecol 2000; 96:277-80.
107. Powers RW, Evans RW, Majors AK et al. Plasma homocysteine concentration is increased in preec-
lampsia and is associated with evidence of endothelial activation. Am J Obstet Gynecol 1998;
179:1605-11.
108. Kobashi G, Yamada H, Asano T et al. Absence of association between a common mutation in the
methylenetetrahydrofolate reductase gene and preeclampsia in Japanese women. Am J Med Genet
2000; 93:122-5.
109. Sohda S, Arinami T, Hamada H et al. Methylenetetrahydrofolate reductase polymorphism and pre
eclampsia. J Med Genet 1997; 34:525-6.
110. Chikosi AB, Moodley J, Pegoraro RJ et al. 5,10 methylenetetrahydrofolate reductase polymorphism
in black South African women with preeclampsia. Br J Obstet Gynaecol 1999; 106:1219-20.
111. Kaiser T, Brennecke SP, Moses EK. C677T methylenetetrahydrofolate reductase polymorphism is
not a risk factor for preeclampsia/eclampsia among Australian women. Hum Hered 2001; 51:20-2.
112. Rajkovic A, Mahomed K, Rozen R et al. Methylenetetrahydrofolate reductase 677 C --> T poly-
morphism, plasma folate, vitamin B(12) concentrations, and risk of preeclampsia among black Af-
rican women from Zimbabwe. Mol Genet Metab 2000; 69:33-9.
113. D’Elia AV, Driul L, Giacomello R et al. Frequency of factor V, prothrombin and methyl-
enetetrahydrofolate reductase gene variants in preeclampsia. Gynecol Obstet Invest 2002; 53:84-7.
114. Prasmusinto D, Skrablin S, Hofstaetter C et al. The methylenetetrahydrofolate reductase 677 C-->T
polymorphism and preeclampsia in two populations. Obstet Gynecol 2002; 99:1085-92.
115. Powers RW, Minich LA, Lykins DL et al. Methylenetetrahydrofolate reductase polymorphism, folate,
and susceptibility to preeclampsia. J Soc Gynecol Investig 1999; 6:74-9.
116. Kaiser T, Brennecke SP, Moses EK. Methylenetetrahydrofolate reductase polymorphisms are not a
risk factor for preeclampsia/eclampsia in Australian women. Gynecol Obstet Invest 2000; 50:100-2.
117. Dekker GA, De Vries JI, Doelitzsch PM et al. Underlying disorders associated with severe early-onset
preeclampsia. Am J Obstet Gynecol 1995; 173:1042-8.
118. Rajkovic A, Catalano PM, Malinow MR. Elevated homocyst(e)ine levels with preeclampsia. Obstet
Gynecol 1997; 90:168-71.
119. Wang J, Trudinger BJ, Duarte N et al. Elevated circulating homocyst(e)ine levels in placental
vascular disease and associated preeclampsia. BJOG 2000; 107:935-8.
120. De Falco M, Pollio F, Scaramellino M et al. Homocysteinaemia during pregnancy and placental
disease. Clin Exp Obstet Gynecol 2000; 27:188-90.
121. van Pampus MG, Dekker GA, Wolf H et al. High prevalence of hemostatic abnormalities in women
with a history of severe preeclampsia. Am J Obstet Gynecol 1999; 180:1146-50.
122. Sorensen TK, Malinow MR, Williams MA et al. Elevated second-trimester serum homocyst(e)ine
levels and subsequent risk of preeclampsia. Gynecol Obstet Invest 1999; 48:98-103.
123. Rajkovic A, Mahomed K, Malinow MR et al. Plasma homocyst(e)ine concentrations in eclamptic
and preeclamptic African women postpartum. Obstet Gynecol 1999; 94:355-60.
124. Laivuori H, Kaaja R, Turpeinen U et al. Plasma homocysteine levels elevated and inversely related
to insulin sensitivity in preeclampsia. Obstet Gynecol 1999; 93:489-93.
125. Sanchez SE, Zhang C, Rene MM et al. Plasma folate, vitamin B(12), and homocyst(e)ine concen-
trations in preeclamptic and normotensive Peruvian women. Am J Epidemiol 2001; 153:474-80.
126. Hogg BB, Tamura T, Johnston KE et al. Second-trimester plasma homocysteine levels and preg-
nancy-induced hypertension, preeclampsia, and intrauterine growth restriction. Am J Obstet Gynecol
2000; 183:805-9.
127. Mayerhofer K, Hefler L, Zeisler H et al. Serum homocyst(e)ine levels in women with preeclamp-
sia. Wien Klin Wochenschr 2000; 112:271-5.
128. Raijmakers MT, Zusterzeel PL, Steegers EA et al. Hyperhomocysteinaemia: A risk factor for preec-
lampsia? Eur J Obstet Gynecol Reprod Biol 2001; 95:226-8.
129. Hietala R, Turpeinen U, Laatikainen T. Serum homocysteine at 16 weeks and subsequent preec-
lampsia. Obstet Gynecol 2001; 97:527-9.
130. Mudd SH, Levy HL, Skovby F. The metabolic basis of inherited disease. In: Scriver CR, Beaudet
AL, Sly WS, Vall D, eds. Disorders of transsulfuration. New York: McGraw-Hill, 1989:693-734.
131. Ubbink JB, Vermaak WJH, Van der Merwe A et al. Vitamin requirements for the treatment of
hyperhomocysteinemia in humans. J Nutr 1994; 124:1927-33.
mediate hyperhomocysteinemia (> or = 40 micromol/liter). The Hordaland Homocysteine Study. J
Clin Invest 1996; 98:2174-83.
133. Fohr IP, Prinz-Langenohl R, Bronstrup A et al. 5,10-Methylenetetrahydrofolate reductase genotype
determines the plasma homocysteine-lowering effect of supplementation with 5-methyltetrahydrofolate
or folic acid in healthy young women. Am J Clin Nutr 2002; 75:275-82.
134. Prevention of neural tube defects: Results of the Medical Research Council Vitamin Study. MRC
Vitamin Study Research Group. Lancet 1991; 338:131-7.
135. Scholl TO, Johnson WG. Folic acid: Influence on the outcome of pregnancy. Am J Clin Nutr
2000; 71:1295S-303S.
136. Leeda M, Riyazi N, De Vries JI et al. Effects of folic acid and vitamin B6 supplementation on
women with hyperhomocysteinemia and a history of preeclampsia or fetal growth restriction. Am J
Obstet Gynecol 1998; 179:135-9.
137. Quere I, Mercier E, Bellet H et al. Vitamin supplementation and pregnancy outcome in women
with recurrent early pregnancy loss and hyperhomocysteinemia. Fertil Steril 2001; 75:823-5.
138. Wald NJ, Law MR, Morris JK et al. Quantifying the effect of folic acid. Lancet 2001; 358:2069-73.
139. Davis RE. Effects of folic acid. Lancet 2002; 359:2038-9.
140. Daly S, Mills JL, Molloy AM et al. Minimum effective dose of folic acid for food fortification to
prevent neural-tube defects. Lancet 1997; 350:1666-9.
141. Hague WM, Dekker GA. Risk factors for thrombosis in pregnancy. Best Pract Res Clin Haematol
2003; 16:197-210.
142. Bloomenthal D, von Dadelszen P, Liston R et al. The effect of factor V Leiden carriage on mater-
nal and fetal health. CMAJ 2002; 167:48-54.
143. Rey E, Kahn SR, David M et al. Thrombophilic disorders and fetal loss: A meta-analysis. Lancet
2003; 361:901-8.
144. Raijmakers MT, Zusterzeel PL, Roes EM et al. Oxidized and free whole blood thiols in preeclamp-
sia. Obstet Gynecol 2001; 97:272-6.
145. Boot MJ, Steegers-Theunissen RP, Poelmann RE et al. Folic acid and homocysteine affect neural
crest and neuroepithelial cell outgrowth and differentiation in vitro. Dev Dyn 2003; 227:301-8.
146. Limpach A, Dalton M, Miles R et al. Homocysteine inhibits retinoic acid synthesis: A mechanism
for homocysteine-induced congenital defects. Exp Cell Res 2000; 260:166-74.
147. Steegers-Theunissen RP, Smith SC, Steegers EA et al. Folate affects apoptosis in human tropho-
blastic cells. Br J Obstet Gynaecol 2000; 107:1513-5.
148. Vanaerts LA, Blom HJ, Deabreu RA et al. Prevention of neural tube defects by and toxicity of
L-homocysteine in cultured postimplantation rat embryos. Teratology 1994; 50:348-60.
149. Greene NE, Dunlevy LE, Copp AJ. Homocysteine is embryotoxic but does not cause neural tube
defects in mouse embryos. Anat Embryol (Berl) 2003; 206:185-91.
150. Nelen WLDM, Bulten J, Steegers EAP et al. Maternal homocysteine and chorionic vascularization
in recurrent early pregnancy loss. Hum Reprod 2000; 15:954-60.
151. Hook EB, Czeizel AE. Can terathanasia explain the protective effect of folic-acid supplementation
on birth defects? Lancet 1997; 350:513-5.
152. Czeizel AE, Dudas I, Metneki J. Pregnancy outcomes in a randomised controlled trial of
periconceptional multivitamin supplementation. Final report. Arch Gynecol Obstet 1994; 255:131-9.
153. Chambers JC, Ueland PM, Obeid OA et al. Improved vascular endothelial function after oral B
vitamins: An effect mediated through reduced concentrations of free plasma homocysteine. Circu-
lation 2000; 102:2479-83.
Neuropsychiatric Disease and Methylenetetrahydrofolate Reductase 163
CHAPTER 12
Neuropsychiatric Disease
and Methylenetetrahydrofolate Reductase
Björn Regland
Abstract
D
isturbances in single-carbon metabolism appear to be related to all sorts of neuro-
psychiatric disorders, which reflect the central importance of single-carbon units in brain
cellular metabolism. Such a disturbance does not have to be specifically related to the
pathogenesis, but may be a factor that renders a person vulnerable and thus triggers the disease
or precipitates its course. In this way, MTHFR is a factor of great interest, because it has a key
function in single-carbon metabolism. The 677C→T mutation increases the risk for late-onset
depression in conjunction with vascular disease. In schizophrenia, a relation to the 677C→T
mutation is supported by some studies but not confirmed by others, which may reflect the
considerable heterogeneity of the schizophrenic population and the need to look more care-
fully on subtypes. The risk and course of dementia disorders are somehow importantly related
to increased homocysteine. In dementia, however, increased homocysteine is not commonly
explained by a MTHFR mutation. The metabolism of L-dopa is dependent on methylation. In
Parkinson’s disease, the 677C→T mutation is a significant factor for the hyperhomocysteinemia
found in a majority of L-dopa-treated individuals.
Introduction
Neuropsychiatry can be regarded as the aspect of psychiatry which, like neurology, seeks to
advance understanding of clinical problems through increased knowledge of brain structure
and molecular function, as is evident for neurodegenerative disorders such as dementia and
Parkinson’s disease. In this sense, however, depression and schizophrenia are also neuropsychi-
atric disorders although their diagnoses are based on behavioural criteria. All the aforemen-
tioned neuropsychiatric disorders are considered as complex disorders to which polygenic as
well as environmental factors are contributing in a complex interaction. From this aspect, the
MTHFR gene may serve a special interest, since the enzyme has a key function in single-carbon
metabolism and interacts with environmental factors such as nutrition and toxic exposure.
Depression
Folate deficiency has long been associated with a wide range of abnormal mental states, and
particularly with affective disorders.1 One possible explanation is that folate probably has a
modulatory function in neurotransmission. It has an important role in the production of
tetrahydrobiopterin, an essential cofactor for tryptophan hydroxylase in serotonin synthesis,
and for tyrosine hydroxylase in dopamine synthesis.2
©2005 Eurekah.com.
Subjects with low folate levels are more likely to suffer from the most severe forms of depres-
sion and are significantly less likely to respond to treatment with antidepressants.1,3,4 In a
recent study, 52% of inpatients with severe depression had raised plasma homocysteine levels.5
The few controlled clinical trials of vitamin therapy in addition to standard psychotropic
medication have all reported positive effects on patient’s mental state. In a double-blind
placebo-controlled clinical trial, Godfrey et al added 15 mg of methylfolate to standard psy-
chotropic medication and reported significant and increasing clinical and social recovery of
folate-deficient depressed groups over six months.6 In depressive patients treated with lithium,
the addition of 0.2 mg of folic acid for one year significantly improved their status.7 Similarly,
the addition of 0.5 mg of the vitamin to fluoxetine for 10 weeks significantly improved antide-
pressant response.4
In one study on the MTHFR mutant 677C→T, there was an increased odds ratio for major
depression,8 which was not confirmed in another study.9 In a third study, it was found that
patients with late-onset depression, i.e., with an onset >50 years of age, had a higher prevalence
of the homozygous or heterozygous forms of the 677C→T mutation than controls or patients
with early-onset disorders.10 Moreover, the patients with late-onset depression had significantly
more vascular risk factors. This does not necessarily mean that vascular risk factors have to be
directly related to the 677C→T mutation. However, the findings are supported by a previous
study on depressed elderly people, in which the homocysteine levels were highest in the pa-
tients who had concomitant vascular disease.11 This may also relate to the fact that late life
depression is often associated with cognitive impairment.12
In summary, the available studies support the concept that folate deficiency is a risk factor
for developing a severe depressive state of mind, and that the MTHFR 677 C→T mutation
increases the risk for late-onset depression in conjunction with vascular disease.
Schizophrenia
Schizophrenia is a psychotic disorder that usually has its onset at a fairly young age, i.e.,
between 18 and 30 years of age. The schizophrenia phenotype is heterogeneous with respect to
clinical presentation, long-term response to medication, and outcome, possibly reflecting ge-
netic heterogeneity and/or the presence of modifier genes. By tradition, the schizophrenic syn-
drome is divided into two main groups: the paranoid and the hebephrenic groups. The para-
noid patients exhibit florid (positive) psychotic symptoms of delusions and hallucinations, and
they tend to respond well to anti-dopaminergic drugs (neuroleptics). The hebephrenics, on the
other hand, follow a more quiet course of disease, respond less well to neuroleptics, and even-
tually tend to exhibit a degenerative (negative) character. The distinction of the two subgroups
was supported by a post-mortem study of brains comparing monoaminergic indices of schizo-
phrenic patients and age-matched controls.13 The results clearly suggested the existence of two
different forms of the disease, both of which could be distinguished from the controls as well as
from each other. One of the schizophrenic groups consisted of paranoid cases and had a rela-
tively mild family history, whereas the other group, mainly consisting of hebephrenic cases,
had a severe family history. The two patterns thus tend to be mutually opposite in several
aspects and it seems unlikely that the two should have similar pathogenesis or aetiology.
Fifty years ago, a ‘transmethylation’ hypothesis on schizophrenia was proposed, based on
the idea of excessive and misdirected transmethylation, resulting in the production of psy-
chotoxic methylated metabolites.14 The hypothesis was later modified to suggest insufficient,
rather than excessive, transmethylation in the patients, and was referred to as the ‘one carbon
cycle’ hypothesis.15 In an attempt to test the original transmethylation hypothesis, a large number
of studies were performed in the period from 1961 to 1971 to evaluate the effect of methionine
in chronic schizophrenia. The administration of methionine yielded varying results, but lead to
acute psychotic reactions in some 40% of patients. According to Antun et al16 chronic
schizophrenics fall into two clear-cut populations: methionine reactors and nonreactors. This
intriguing result has unfortunately not (yet) found its explanation. Schizophrenics are evi-
dently in sharp contrast to normal humans in whom methionine, even at 10g doses, has no
behavioural effects.17
Sargent et al18 tested the general concept of the hypothesis by administering methionine
that was radioactively labelled in the S-methyl carbon. They measured the expiration of such
labelled CO2, and found that the rate and total expiration of labelled CO2 were three times
lower in schizophrenic patients than in controls, with no overlap of data points in the groups.
It is important to note that the patients were relatively young (mean age 37 years), were not
medicated and had active positive symptoms of schizophrenia.
Yet additional evidence for a link between schizophrenia and methylation was recently pre-
sented by the observation that phospholipid methylation in lymphocytes had four-fold lower
activity in samples from schizophrenics in comparison with the controls.19
A few years ago, we drew attention to the ‘one carbon cycle’ hypothesis. Firstly, we were
intrigued by the finding of a case with hyperhomocysteinemia and paranoid schizophrenia
who repeatedly responded to vitamin B12 treatment in a positive although transient way.20
Secondly, this prompted us to investigate the homocysteine concentrations of 20 consecutive
patients with a schizophrenia-like psychotic disorder and of 20 controls, who were matched for
age and gender. Nine of the patients had increased homocysteine levels in serum, which were
evidently not attributable to psychopharmacological medication. In the control group, only a
single individual had an increased homocysteine level, which was explained as the first sign of
pernicious anaemia.21 We thus concluded that schizophrenic patients are more likely than
controls to have increased homocysteine levels. Thirdly, we studied the gene for MTHFR in 11
schizophrenic patients with hyperhomocysteinemia.22 Seven were homozygous and one was
heterozygous for the 677C→T mutation. We also reported that the schizophrenic patients
with hyperhomocysteinemia appeared to be of a paranoid rather than a hebephrenic type.
Shortly thereafter, a study from Japan confirmed the existence of an association between the
MTHFR 677T genotype and schizophrenia.8 The homozygous 677T genotype was signifi-
cantly over-represented in schizophrenics. However, this study was followed by two negative
studies, one from Japan9 and one from Spain,23 which could not confirm such an association in
schizophrenic in-patients of a relatively high age (mean 58 years). A possible reason for the lack
of consensus might be that the investigations were not identical with respect to the distribution
of schizophrenic subgroups. Such a notion is supported by Joober et al24 in Canada who re-
ported a significant association between schizophrenia and the 677T allele of MTHFR. How-
ever, this association was entirely due to an over-representation of the 677T allele in patients
responding to neuroleptic medication; the nonresponder patients did not differ from controls.
It was concluded that their results ‘strongly suggested that the MTHFR gene is involved in the
pathogenesis of schizophrenia when characterized by rapid and sustained therapeutic response
to typical neuroleptics’.24 The importance of subgrouping, according to age and gender for
example, was emphasized in a recent study from Israel. The authors reported that the mean
homocysteine level was markedly increased in patients with schizophrenia in comparison with
controls. However, the difference was almost entirely attributable to the homocysteine levels of
young male patients.25
In Wales, 56 ‘nuclear families’, consisting of mothers, fathers and ‘affected off-spring’ with
schizophrenia, were investigated. The subtyping was not reported and there was no control
group. From statistical tests, it was concluded that the MTHFR gene may be associated with
schizophrenia.26
In summary, the schizophrenic population is considerably heterogeneous and is comprised
of distinct populations with respect to symptomatology, course of disease, familial history and
in the response to methionine treatment. There are findings compatible with the ‘one carbon
cycle’ hypothesis that a folate-sensitive defect contributes to some cases of schizophrenia. A
relation between the 677C→T mutation and schizophrenia is supported by some studies,
although not confirmed by others. The possible relation may result from the selection of pa-
tients who are younger and tend to have a paranoid rather than a hebephrenic character.
Dementia
Dementia is the advanced stage of a neurodegenerative process that starts as mild cognitive
impairment. Even in the early stage of dementia, plasma homocysteine is increased in a signifi-
cant number of patients.27
Two longitudinal prospective studies have confirmed that elevated serum levels of homocys-
teine predict cognitive decline in healthy elderly and are a potential risk factor for the develop-
ment of dementia.28-29 In the Framingham Study of 720 subjects, the adjusted odds ratio for
dementia was 2.63 per SD increase in log-homocysteine measured 3 years earlier, and 2.67 per
SD increase in log-homocysteine measured 11 years earlier.29
Alzheimer’s disease (AD) and vascular dementia (VAD) are the most common types of
dementia disorders. AD is characterized by a temporo-parietal symptomatology corresponding
to a cortical distribution of the senile plaque neuropathy. VAD, on the other hand, is predomi-
nantly characterized by a fronto-subcortical symptom profile corresponding to the subcortical
white matter lesions seen on brain imaging, which represent small vessel disease.
In clinical practice, it is often difficult to clearly separate the two entities because of the high
frequency of mixed cases. The pure forms of AD and VAD should be seen as two different
pathogenic principles that cause dementia. The frequent intermixture and interplay of cortical
Alzheimer and subcortical vascular pathologies is well illustrated by the unique Nun Study.30
Its findings suggested that a few small infarcts in strategic regions may be sufficient to produce
dementia in those made vulnerable by abundant AD lesions in the neocortex. Moreover, low
serum folate was strongly associated with post-mortem atrophy in the cerebral cortex in the
Nun Study.30
A British case-control trial, within the OPTIMA project in Oxford, involved 164 patients
with dementia. Blood homocysteine was higher, and both serum folate and vitamin B12 lower,
both in the patients with clinical AD and in the patients with histopathologically-confirmed
AD than in controls. The association with homocysteine was independent of age, sex, social
class, smoking, and the apoE4 genotype. A follow-up reported that patients with higher ho-
mocysteine at presentation showed a more rapid progression of their disease over a 3-year
period. Moreover, brain imaging revealed a decreased temporal lobe thickness.31 Similarly,in
the large Rotterdam Study on apparently nondemented people, individuals with more cortical
and hippocampal atrophy, i.e., the individuals at risk for AD, had higher plasma homocysteine
levels.32
Cerebral small vessel disease is the hallmark of VAD, and homocysteine is also associated
with an increased risk for such subcortical brain lesions.33-34 In recent years, a number of
studies have been performed on MTHFR genotyping in demented patients. A study of a Japa-
nese population (n=100) suggested that the 677C→T mutation is associated with the clinical
phenotype and clinical onset of dementia.35 A large Chinese study reported that homozygosity
was significantly more frequent in VAD patients (n=143) than in a contrast group of cere-
brovascular nondemented patients (n=122) or in healthy controls (n=217).36 Similar results
were found in a study from Northern Ireland on VAD, AD, nondemented stroke patients and
healthy controls; the 677T allele was found to increase the risk for VAD.37
All studies on AD patients have been negative,37-41 as were three studies which also in-
cluded VAD patients.42-44 None of the studies included the MTHFR 1298 A→C mutant. A
possible role for MTHFR in combination with apoE4, which is a documented genetic risk
factor for AD, was described in a study of 140 AD patients.45
In summary, there are now abundant data to show that increased homocysteine is somehow
importantly related to the risk and the course of dementia disorders. Homocysteine is a risk
factor for vascular disease, which may explain why the 677T allele appears to increase the risk
for VAD in some studies. On the other hand, pure AD is not caused by vasculopathy and the
increased homocysteine levels in AD are not commonly explained by an over-representation of
the specific 677C→T mutation. Thus, there must be mechanisms other than MTHFR mu-
tations to explain the increased homocysteine in AD patients. There is abundant evidence
implicating oxidative stress in AD,46 and it has been suggested that oxidative stress might also
be an important cause of hyperhomocysteinemia in these patients.47
Parkinson
A recent Japanese study showed that homocysteine levels were increased in 60% of 90
L-dopa-treated patients with Parkinson’s disease, and that patients homozygous for the 677C→T
mutation had the most marked elevation. In this group, homocysteine and folate levels were
inversely correlated.48 Additional recent data49-50 lend firm support to the 677C→T mutation
as being a significant factor for hyperhomocysteinemia in L-dopa-treated patients.
The plausible explanation is that the metabolism of L-dopa requires S-adenosylmethionine
(AdoMet) and that L-dopa treatment consumes AdoMet to such an extent that it might be
depleted; this risk is especially high in an individual homozygous for the 677C→T mutation.
L-dopa treatment causes significant reductions in the concentrations of AdoMet in rat brain,51
and L-dopa lowers AdoMet in cerebrospinal fluid of humans.52
In relation to Parkinson’s disease, the 677C→T mutation exemplifies how an enzyme vari-
ant may increase the impact of a drug interaction and thereby have an indirect influence on the
drug treatment outcome.
Conclusions
Disturbances in single-carbon metabolism appear to be related to all sorts of neuropsychiat-
ric disorders, which reflect the central importance of single-carbon units in brain cellular me-
tabolism. Such a disturbance does not have to be specifically related to the pathogenesis, but
may be a factor that renders a person vulnerable and thus triggers the disease or precipitates its
course. In this way, MTHFR is a factor of great interest, firstly because it has a key function in
single-carbon metabolism, and secondly because its function can be improved by supplemen-
tation of folate and other B vitamin coenzymes. Such supplementation has the important
potential to alleviate the outcome of a number of neuropsychiatric disorders.
References
1. Reynolds EH, Preece JM, Bailey J et al. Folate deficiency in depressive illness. Br J Psychiatry
1970; 117:287-92.
2. Bottiglieri T, Hyland K, Laundy M et al. Folate deficiency, biopterin and monoamine metabolism
in depression. Psychol Med 1992; 22:871-6.
3. Fava M, Borus JS, Alpert JE et al. Folate, vitamin B12, and homocysteine in major depressive
disorder. Am J Psychiatry 1997; 154:426-8.
4. Coppen A, Bailey J. Enhancement of the antidepressant action of fluoxetine by folic acid; a
randomised, placebo controlled trial. J Affect Disord 2000; 60:121-30.
5. Bottiglieri T, Laundy M, Crellin R et al. Homocysteine, folate, methylation, and monoamine me-
tabolism in depression. J Neurol Neurosurg Psychiatry 2000; 69:228-32.
6. Godfrey PSA, Toone BK, Carney MWP et al. Enhancement of recovery from psychiatric illness by
methylfolate. Lancet 1990; 336:392-5.
7. Coppen A, Chaudry S, Swade C. Folic acid enhances lithium prophylaxis. J Affect Disord 1986;
10:9-13.
8. Arinami T, Yamada N, Yamakawa-Kobayashi K et al. Methylenetetrahydrofolate reductase variant
and schizophrenia/depression. Am J Med Genet 1997; 74:526-8.
9. Kunugi H, Fukuda R, Hattori M et al. C677T polymorphism in methylenetetrahydrofolate reduc-
tase gene and psychoses. Mol Psychiatry 1998; 3:435-7.
10. Hickie I, Scott E, Naismith S et al. Late-onset depression: Genetic, vascular and clinical contribu-
tions. Psychol Med 2001; 31:1403-12.
11. Bell IR, Edman JS, Selhub J et al. Plasma homocysteine in vascular disease and in nonvascular
dementia of depressed elderly people. Acta Psychiatr Scand 1992; 86:386-90.
12. Gottfries CG. Late life depression. Eur Arch Psychiatry Clin Neurosci 2001; 251(Suppl 2):II57-61.
13. Hansson LO, Waters N, Winblad B et al. Evidence for biochemical heterogeneity in schizophre-
nia: A multivariate study of monoaminergic indices in human post-mortal brain tissue. J Neural
Transm [GenSect] 1994; 98:217-35.
14. Osmond H, Smythies J, Harley-Mason J et al. Schizophrenia: A new approach. J Mental Science
1952; 98:309-14.
15. Smythies JR. The transmethylation hypotheses of schizophrenia reevaluated. Trends in Neurosciences
1984; 7:45-7.
16. Antun FT, Burnett GB, Cooper AJ et al. The effects of L-methionine (without MAOI) in schizo-
phrenia. J Psychiatr Res 1971; 8:63-71.
17. Baldessarini RJ, Stramentinoli G, Lipinski JF. Methylation hypothesis. Arch Gen Psychiatry 1979;
36:303-7.
18. Sargent T, Kusubov N, Taylor SE et al. Tracer kinetic evidence for abnormal methyl metabolism
in schizophrenia. Biol Psychiatry 1992; 32:1078-90.
19. Sharma A, Kramer M, Wick PF et al. D4 dopamine receptor-mediated phospholipid methylation
and its implications for mental illness such as schizophrenia. Mol Psychiatry 1999; 4:235-46.
20. Regland B, Johansson BV, Gottfries CG. Homocysteinemia and schizophrenia as a case of methy-
lation deficiency. Case report. J Neural Transm [GenSect] 1994; 98:143-52.
21. Regland B, Johansson BV, Grenfeldt B et al. Homocysteinemia is a common feature of schizophre-
nia. J Neural Transm [GenSect] 1995; 100:165-9.
22. Regland B, Germgård T, Gottfries CG et al. Homozygous thermolabile methylenetetrahydrofolate
reductase in schizophrenia-like psychosis. J Neural Transm 1997; 104:931-41.
23. Virgos C, Martorell L, Simo JM et al. Plasma homocysteine and the methylenetetrahydrofolate
reductase C677T gene variant: Lack of association with schizophrenia. Neuroreport 1999; 10:2035-8.
24. Joober R, Benkelfat C, Lal S et al. Association between the methylenetetrahydrofolate reductase
677C→T missense mutation and schizophrenia. Mol Psychiatry 2000; 5:323-6.
25. Levine J, Stahl Z, Sela BA et al. Elevated homocysteine levels in young male patients with schizo-
phrenia. Am J Psychiatry 2002; 159:1790-2.
26. Wei J, Hemmings GP. Allelic association of the MTHFR gene with schizophrenia. Mol Psychiatry
1999; 4:115-6.
27. Lehmann M, Gottfries CG, Regland B. Identification of cognitive impairment in the elderly: Ho-
mocysteine is an early marker. Dement Geriatr Cogn Disord 1999; 10:12-20.
28. McCaddon A, Hudson P, Davies G et al. Homocysteine and cognitive decline in healthy elderly.
Dement Geriatr Cogn Disord 2001; 12:309-13.
29. Seshadri S, Beiser A, Selhub J et al. Plasma homocysteine as a risk factor for dementia and
Alzheimer’s Disease. N Engl J Med 2002; 346:476-83.
30. Snowdon DA, Greiner LH, Mortimer JA et al. Brain infarction and the clinical expression of
Alzheimer disease. The Nun Study. JAMA 1997; 277:813-7.
31. Clarke R, Smith AD, Jobst K et al. Folate, vitamin B12, and serum total homocysteine levels in
confirmed Alzheimer disease. Arch Neurol 1998; 55:1449-55.
32. Den Heijer T, Vermeer SE, Clarke R et al. Homocysteine and brain atrophy on MRI of
nondemented elderly. Brain 2002; 126:170-5.
33. Bertsch T, Mielke O, Holy S et al. Homocysteine in cerebrovascular disease: An independent risk
factor for subcortical vascular encephalopathy. Clin Chem Lab Med 2001; 39:721-4.
34. Vermeer SE, vanDijk EJ, Koudstaal PJ et al. Homocysteine, silent brain infarcts, and white matter
lesions: The Rotterdam scan study. Ann Neurol 2002; 51:285-9.
35. Nishiyama M, Kato Y, Hashimoto M et al. Apolipoprotein E, methylenetetrahydrofolate reductase
(MTHFR) mutation and the risk of senile dementia - an epidemiological study using the poly-
merase chain reaction (PCR) method. J Epidemiol 2000; 10:163-172.
36. Yoo JH, Choi GD, Kang SS. Pathogenicity of thermolabile methylenetetrahydrofolate reductase
for vascular dementia. Arterioscler Thromb Vasc Biol 2000; 20:1921-5.
37. McIlroy SP, Dynan KB, Lawson JT et al. Moderately elevated plasma homocysteine,
methylenetetrahydrofolate reductase genotype, and risk for stroke, vascular dementia, and Alzheimer
disease in Northern Ireland. Stroke 2002; 33:2351-6.
38. Bottiglieri T, Parnetti L, Arning E et al. Plasma total homocysteine levels and the C677T muta-
tion in the methylenetetrahydrofolate reductase (MTHFR) gene: A study in an Italian population
with dementia. Mech Ageing Dev 2001; 122:2013-23.
39. Brunelli T, Bagnoli S, Giusti B et al. The C677T methylenetetrahydrofolate reductase mutation is
not associated with Alzheimer’s disease. Neurosci Lett 2001; 315:103-5.
40. Postiglione A, Milan G, Ruocco A et al. Plasma folate, vitamin B-12, and total homocysteine and
homozygosity for the C677T mutation of the 5,10-methylene tetrahydrofolate reductase gene in
patients with Alzheimer’s dementia. A case-control study. Gerontology 2001; 47:324-9.
41. Prince JA, Feuk L, Sawyer SL et al. Lack of replication of association findings in complex disease:
An analysis of 15 polymorphisms in prior candidate genes for sporadic Alzheimer’s disease. Eur J
Hum Genet 2001; 9:437-44.
42. Chapman J, Wang N, Treves TA et al. ACE, MTHFR, factor V Leiden, and APOE polymor-
phisms in patients with vascular and Alzheimer’s dementia. Stroke 1998; 29:1401-4.
43. Pollak RD, Pollak A, Idelson M et al. The C677T mutation in the methylenetetrahydrofolate
reductase (MTHFR) gene and vascular dementia. J Am Geriatr Soc 2000; 48:664-8.
44. Zuliani G, Ble’ A, Zanca R et al. Genetic polymorphisms in older subjects with vascular or
Alzheimer’s dementia. Acta Neurol Scand 2001; 103:304-8.
45. Regland B, Blennow K, Germgård T et al. The role of the polymorphic genes apolipoprotein E
and methylenetetrahydrofolate reductase in the development of dementia of the Alzheimer type.
Dement Geriatr Cogn Disord 1999; 10:245-51.
46. Smith MA, Rottkamp CA, Nunomura A et al. Oxidative stress in Alzheimer’s disease. Biochim
Biophys Acta 2000; 1502:139-44.
47. McCaddon A, Regland B, Hudson P et al. Functional vitamin B12 deficiency and Alzheimer dis-
ease. Neurology 2002; 58:1395-9.
48. Yasui K, Kowa H, Nakaso K et al. Plasma homocysteine and MTHFR C677T genotype in
levodopa-treated patients with PD. Neurology 2000; 55:437-40.
49. Kuhn W, Hummel T, Woitalla D et al. Plasma homocysteine and MTHFR C677T genotype in
levodopa-treated patients with PD (Letter). Neurology 2001; 56:281-2.
50. Yasui K, Kowa H, Nakaso K et al. Plasma homocysteine and MTHFR C677T genotype in
levodopa-treated patients with PD (Reply to Letter). Neurology 2001; 56:281-2.
51 Ordonez LA, Wurtman RJ. Folic acid deficiency and methyl group metabolism in rat brain: Ef-
fects of L-dopa. Arch Biochem Biophys 1974; 160:372-6.
52. Surtees R, Hyland K. L-3,4-dihydroxyphenylalanine (levodopa) lowers central nervous system
S-adenosylmethionine concentrations in humans. J Neurol Neurosurg Psychiatry 1990; 53:569-72.
CHAPTER 13
Polymorphisms and Renal Failure
Manuela Födinger and Gere Sunder-Plassmann
Abstract
H
yperhomocysteinemia is present in more than 90% of dialysis patients and in
approximately 60% to 80% of kidney transplant patients. It is well established that
the MTHFR 677TT genotype aggravates hyperhomocysteinemia of renal failure pa-
tients independent of the type of renal replacement therapy. Moreover, MTHFR 677TT influ-
ences folate status particularly in patients without folic acid supplementation. MTHFR
1298A→C alone neither influences total homocysteine nor plasma folate. On the other hand,
compound heterozygosity for MTHFR 677C→T and MTHFR 1298A→C may influence folate
status. Recently, the MTHFR 677TT genotype has been shown to be a vascular disease risk
factor in the particular setting of low folate status. Therefore, the increased vascular disease risk
of renal failure patients is possibly also attributable to alterations of the folate cycle. Response
to treatment with folate derivatives is influenced by the MTHFR 677TT genotype and com-
pound heterozygosity for MTHFR 677C→T and MTHFR 1298A→C. Unfortunately, many
patients with renal failure show resistance to homocysteine lowering therapies with different
doses of folic or folinic acid, as well as a combination of folic acid with other B-vitamins. In this
article, we summarize the impact of genetic polymorphisms in the MTHFR gene in renal
failure patients. Particular emphasis is given to folate status, cardiovascular disease risk and
pharmacogenetic aspects, as well as the interaction of MTHFR mutations with other common
polymorphisms in genes of the homocysteine/folate cycle.
Introduction
Renal failure represents the most severe stage of chronic kidney disease (stage 5) and is
defined as either (i) a level of the glomerular filtration rate (GFR) < 15 mL/min/1.73 m2
(normal GFR: ≥ 90 mL/min/1.73 m2) which is accompanied in most cases by signs and symp-
toms of uremia or (ii) a need for initiation of kidney replacement therapy (dialysis or transplan-
tation; National Kidney Foundation (NKF) Kidney Disease Outcome Quality Initiative (K/
DOQI) guideline 2002: www.kidney.org/professionals/doqi/kdoqi/p4_class_g1.htm). In the
Unites States, the prevalence of stage 5 chronic kidney disease is 0.2% (95% confidence inter-
val: 0.1%-0.3%; United States Renal Data System (USRDS) 1998 including approximately
230,000 patients treated by dialysis and assuming 70,000 additional patients not on dialysis
treatment).
End-stage renal disease (ESRD) is a term that includes renal failure patients treated by
dialysis or transplantation, irrespective of the level of GFR (K/DOQI guideline 2002). In the
past 10 years, the incidence and the prevalence of reported ESRD have doubled in the United
States. Data from the 2000 Annual Data Report of the USRDS documents the incidence of
©2005 Eurekah.com.
Methylenetetrahydrofolate Reductase Polymorphisms and Renal Failure 171
ESRD in 1998 of more than 85,000 per year. The point prevalence of ESRD on December 31,
1998 was more than 320,000 or 1,160 per million population, of whom 72% were treated by
dialysis and 28% had functioning kidney transplants. The number of individuals with a GFR
< 15 mL/min/1.73 m2 not on dialysis has not been estimated reliably.
A decrease of the GFR is a predictor of cardiovascular disease independent of blood pressure
and serum cholesterol levels.1 Patients with a GFR of < 70 mL/min/1.73 m2 have an increased
risk of all-cause-related, cardiovascular disease-related, and coronary artery disease-related death
as compared to individuals with a rate of > 90 mL/min/1. 73 m2.1 One factor associated with an increased
cardiovascular disease risk in renal failure patients, is hyperhomocysteinemia, defined as plasma total
homocysteine concentrations > 15 µmol/L.2 It is well established that renal function is an
important determinant of plasma total homocysteine. Each decline of renal function is associ-
ated with a parallel increase of plasma total homocysteine levels. This inverse relationship seems
to be independent of the primary kidney disease and is present throughout the whole range of
renal function.3 However, the mechanisms resulting in hyperhomocysteinemia in renal failure
are only poorly understood. Animal models suggested substantial homocysteine uptake and
metabolism in the kidney, supporting the concept that hyperhomocysteinemia in renal failure
is due to the loss of the metabolizing capacity of the kidney.4 However, in humans with normal
renal function, no significant net renal uptake of homocysteine occurs,5 raising again the ques-
tion of the mechanisms ultimately leading to hyperhomocysteinemia.
Moderate (15-30 µmol/L) and intermediate hyperhomocysteinemia (31-100 µmol/L) is
present in more than 90% of patients undergoing maintenance dialysis and in approximately
60% to 80% of chronic renal transplant patients.6 In all these patients, hyperhomocysteinemia
shows a clear association with cardiovascular complications and cardiovascular mortality.7-9
Determinants that influence homocysteine concentration of renal failure patients are folate
status, vitamin B12 supply/intake, vitamin B6 and vitamin B2 availability, as well as albumin
level and age.10,11 Furthermore, total homocysteine concentration is influenced by genetic fac-
tors. Among the genetic polymorphisms that have been investigated for a putative relation to
plasma homocysteine concentration, two polymorphisms in the MTHFR gene encoding
5,10-methylenetetrahydrofolate reductase (MTHFR 677C→T, MTHFR 1298A→C) can in-
fluence homocysteine and folate metabolism.12 In this context, it is worth mentioning that
renal failure patients, as shown in hemodialysis patients, exhibit significantly decreased
remethylation of homocysteine to methionine.13 This observation could explain, at least in
part, the high prevalence of hyperhomocysteinemia in renal failure and suggests that abnormal
folate metabolism is a major cause of hyperhomocysteinemia in these patients. However, it has
been shown that absorption of folic and folinic acid as well as their conversion into
5-methyltetrahydrofolate is not defective in hemodialysis patients,14 excluding these parts of
folic and folinic acid metabolism as a cause of hyperhomocysteinemia in hemodialysis patients.
5,10-Methylenetetrahydrofolate reductase (MTHFR; EC 1.5.1.20) is the key enzyme of
the folate cycle required for remethylation of homocysteine to methionine.15 In the MTHFR
gene (OMIM 607093), there are several polymorphisms including those that are located at
nucleotide position 677 (MTHFR 677C→T),16 position 1298 (MTHFR 1298A→C),17-19 po-
sition 1317 (MTHFR 1317T→C),18 and position 1793 (MTHFR 1793G→A).20 MTHFR
677C→T is located in the catalytic domain, changing an alanine into a valine residue (A222V).
MTHFR 1298A→C is located in the presumptive regulatory domain changing glutamic acid
into an alanine residue (E429A). MTHFR 1317T→C is a silent mutation. MTHFR 1793G→A
is located in exon 11 and leads to an amino acid substitution of arginine to glutamine (R594Q).
MTHFR 677C→T, MTHFR 1298A→C, as well as compound heterozygosity for 1298A→C
and 677C→T are associated with reduced enzyme activity of 45%, 68% and 42%, respec-
tively.21 Whether MTHFR 1793G→A affects the functional activity of the enzyme is currently
unknown.
Many polymorphisms in candidate genes involved in the metabolism of homocysteine have
been investigated in renal failure patients to gain insights into the pathogenesis of
hyperhomocysteinemia. These include MTR 2756A→G in the gene encoding methionine

synthase,22 MTRR 66A→G in the gene coding for methionine synthase reductase,23 RFC1
80G→A of the gene coding for the reduced folate carrier 1,24,25 GCP2 1561C→T encoding
glutamate carboxypeptidase II,24,25 TCN2 776G→C coding for transcobalamin II,26 and the
MTHFR 677C→T and MTHFR 1298A→C mutations. Among these, only the 677C→T
mutation of MTHFR, compound heterozygosity for MTHFR 677C→T and MTHFR
1298A→C, as well as GCP2 1561C→T showed a significant influence on homocysteine and/
or folate status.
MTHFR 677C→T and 1298A→C and Hyperhomocysteinemia

in Renal Failure Patients
The MTHFR 677TT genotype predisposes individuals with and without renal failure to
increased plasma total homocysteine levels.15,27 Among renal failure patients, the homocys-
teine elevating effect of the MTHFR 677TT genotype is present in patients maintained on
chronic hemodialysis treatment, in peritoneal dialysis patients, as well as in renal transplant
recipients. In the majority of studies, this effect is even observed in patients receiving folic acid
in doses up to 10 mg per day (Table 1). In contrast, neither the MTHFR 1298A→C polymor-
phism alone nor compound heterozygosity for the MTHFR 677T and the 1298C allele influ-
enced plasma homocysteine concentration in hemodialysis and peritoneal dialysis patients,28
or kidney graft recipients29 (Table 1).
Table 1. Plasma homocysteine and folate levels of renal failure patients are modulated
by MTHFR 677C→T and MTHFR 1298A→C polymorphisms
Influence on Daily Folate

Patients No. Influence on Plasma tHcy Plasma/Serum Folate Therapy Reference
677C→T 1298A→C 677C→T 1298A→C
HD 69 yes n.d. no n.d. 0.16 mg 54

HD 106 no n.d. no n.d. 2 mg 55
HD 464 yes n.d. yes n.d. no 56
HD 415 yes no no no 0.16 mg# 28
HD 151 yes n.d. no n.d. 10 mg 40
HD 120 yes yes n.d. n.d. 5 mg 57
HD 450 yes n.d. yes n.d. no 58
HD 168 yes n.d. n.d. n.d. no 59
HD 114 yes no no no 0.7 mg 60
HD, PD 82 yes n.d. yes n.d. no 61
HD, PD 459 yes n.d. no n.d. 0.67 mg§ 38
PD 179 yes no no no no’’ 28
PD 154 yes n.d. yes n.d. yes$ 62
TX 189 yes n.d. yes n.d. no 49
TX 733 yes no yes yes** no 29
TX 70 yes n.d. n.d. n.d. no 32
TX 108 yes n.d. yes*** n.d. no 63
Abbreviations: tHcy= total homocysteine; HD= hemodialysis; PD= peritoneal dialysis; TX= renal
transplantation; n.d= not done. # Most patients; §80 % of patients; ’’ 31 patients received multivitamin;
$ 26 patients received low dose folic acid; ** combined MTHFR 677CT/1298AC genotype; *** trend
toward significance
MTHFR 677C→T and 1298A→C Polymorphisms and Folate Status

in Renal Failure Patients
Folate and B-vitamins enhance homocysteine metabolism and thereby facilitate its meta-
bolic clearance. Patients with renal failure are prone to development of deficiencies of water
soluble vitamins, as a result of restricted intake, impaired absorption, uremia-induced alter-
ations in metabolism and enzyme activities, as well as losses during dialysis and erythropoietin
treatment.30 In renal graft recipients, the MTHFR 677TT genotype and compound heterozy-
gosity for the 677T and the 1298C alleles were associated with lower plasma folate levels.29 In
contrast, in hemodialysis and peritoneal dialysis patients, only MTHFR 677TT but not com-
pound heterozygosity for MTHFR 677CT / 1298AC influenced plasma folate levels in pa-
tients without folate supplementation.28 This influence of the MTHFR 677TT genotype on
plasma folate concentrations was not present in dialysis patients routinely receiving folic acid
and vitamin supplementation (Table 1).
Interaction of MTHFR 677C→T with Other Genetic Polymorphisms

and Homocysteine / Folate Status in Renal Failure
As mentioned above, the MTHFR 677TT/1298AA genotype but not the compound het-
erozygous genotype (MTHFR 677CT/1298AC) was associated with increased plasma total
homocysteine concentrations in renal failure patients maintained under different forms of re-
nal replacement therapy.28,29 A combination of MTHFR 677TT/1298AA or MTHFR 677CT/
1298AC together with MTR 2756AG or 2756GG has been more frequently identified among
a population of hemodialysis patients, peritoneal dialysis patients and renal graft recipients
showing plasma total homocysteine concentrations within the highest decile.22 The number of
patients with wild-type MTHFR and MTR alleles was three times higher in the lowest as com-
pared with the highest decile, suggesting that MTR 2756A→G in combination with MTHFR
677TT/1298AA and 677CT/1298AC can predispose to extremely high plasma total homocys-
teine concentrations in renal failure patients.22
In contrast, a combination of MTHFR 677TT/1298AA with RFC1 80G→A, GCP2
1561C→T or MTRR 66A→G was not associated with a further increase of plasma total ho-
mocysteine levels in kidney transplant patients.23,25 This was also the case for TCN2 776G→C,
RFC1 80G→A, and GCP2 1561C→T which, together with MTHFR 677C→T and MTHFR
1298A→C, showed no effect on total homocysteine levels of dialysis patients.26,24
MTHFR 677C→T and Cysteine Levels in Renal Failure Patients

Sulfur amino acids are potential cardiovascular disease risk factors. Cysteine is a
sulfur-containing trans-sulfuration product of homocysteine. High plasma cysteine concentra-
tions have been suggested to represent a risk factor for deep vein thrombosis in patients with-
out renal disease independent of the presence of hyperhomocysteinemia.31 Renal failure pa-
tients frequently present with increased plasma total cysteine levels. The prevalence of
hypercysteinemia in a population of kidney graft recipients was 35.8%32 and 91% in dialysis
patients.33 In patients maintained on hemodialysis treatment, serum cysteine concentration
was positively associated with total homocysteine, folate and vitamin B12 concentration and
negatively associated with the MTHFR 677T allele. MTHFR 677C→T could therefore repre-
sent an independent predictor of total cysteine concentrations.34 Moreover, in patients on
hemodialysis, the ratio of total cysteine to total homocysteine was related to cardiovascular
disease risk, raising the possibility that cysteine plays a role in the development of cardiovascu-
lar disease of renal failure patients.33
MTHFR 677C→T and Cardiovascular Disease Risk in Renal Disease

In many studies of patients with35 and without renal failure,36 increased plasma total ho-
mocysteine levels were associated with increased cardiovascular disease risk. This risk seems to
be graded in that a 25% lower homocysteine level (about 3 µmol / L) was associated with an 11
% lower risk of ischemic heart disease and 19% lower risk of stroke.36 Whether this association
is causal is still uncertain. In this context, it must be mentioned that recent studies revealed
even lower total homocysteine concentrations in renal failure patients with cardiovascular dis-
ease than in individuals without a cardiovascular disease history.37,38 This observation could be
explained, at least in part, by the increased prevalence of malnutrition in the study population
with cardiovascular disease.37 Interestingly, in one of these studies of dialysis patients, the MTHFR
677C→T polymorphism was independently associated with cardiovascular disease.38 This ob-
servation of an independent role of MTHFR 677C→T in cardiovascular disease risk is sup-
ported by a recent meta-analysis of 11,162 cases with coronary artery disease and 12,758 con-
trol individuals without renal failure, clearly demonstrating that subjects with the MTHFR
677TT genotype have a significantly higher risk of coronary heart disease, particularly in the
setting of low folate status.39 The idea of an independent association of MTHFR 677TT with
coronary artery disease risk is further supported by the observation of an association of the
MTHFR 677T allele with increased intima-media thickness of the common carotid arteries.40
Based on these observations, it appears likely that the MTHFR 677TT genotype in individuals
with and without renal failure is an independent cardiovascular disease risk factor and supports
the concept that impaired folate metabolism is causally related to increased coronary heart
disease risk.
Effect of MTHFR 677C→T and 1298A→C Polymorphisms on Total

Homocysteine Lowering Therapy of Renal Failure Patients
Folic acid and B-vitamin supplementation allows for reduction of increased plasma total
homocysteine concentration in individuals without kidney diseases. In renal failure patients,
folic acid combined with vitamin B12 and vitamin B6 is more effective in lowering total ho-
mocysteine levels than either cofactor alone.41 In dialysis patients, the optimal dose of folic
acid seems to be 1 mg orally per day.41 However, in comparison to the general population,
hyperhomocysteinemia of renal failure patients is partially refractory to treatment with folic
acid.30 A daily oral dose of 15 mg of folic acid (together with vitamins B6 and B12) allows for
normalization of homocysteine concentrations in only 33 % of hemodialysis patients.42 Even
oral folic acid in doses up to 60 mg per day does not increase the rate of normalization of total
homocysteine.43 The reasons for this partial folate resistance are unknown.
Response to homocysteine lowering therapy of renal failure patients is influenced by the
MTHFR polymorphisms. Hemodialysis patients with the MTHFR 677TT genotype receiving
high dose folic acid orally more frequently attain normal total homocysteine levels than pa-
tients with the CC / CT genotypes.43 The better response to therapy has also been observed in
hemodialysis patients with the MTHFR 677TT/1298AA and the MTHFR 677CT/1298AC
genotype intravenously treated with 15 mg of folic acid and an equimolar amount of folinic
acid44 as well as in MTHFR 677TT genotype patients receiving 15 mg of folic acid orally.45
This better response to folate therapy of patients with the MTHFR 677TT genotype may be
due to the correction of the intracellular deficiency of 5-methyltetrahydrofolate in TT geno-
type patients.46
MTHFR 677C→T and Kidney Transplant Survival

Chronic allograft nephropathy is the major limitation to long-term kidney transplant sur-
vival. It manifests as obliterative arteriopathy or graft vascular disease, interstitial fibrosis and
atrophy of parenchyma that finally result in allograft failure. The pathogenesis is multifactorial
and can include inflammatory or immunologic reactions, as well as atherosclerosis. Because
hyperhomocysteinemia of kidney transplant patients could contribute to the pathogenesis of
chronic allograft nephropathy, the MTHFR 677C→T and MTHFR 1298A→C polymorphisms
were studied for a possible relation with kidney transplant survival. The studies available thus
far revealed (i) that neither MTHFR 677C→T47,48 nor MTHFR 1298A→C48 are important
determinants of renal-transplant survival and (ii) that the MTHFR genotype of the donor
kidney does not influence plasma total homocysteine levels.49
MTHFR 677C→T and 1298A→C Polymorphisms and

Hyperhomocysteinemia in Children with Renal Failure
Children with chronic renal failure frequently present with hyperhomocysteinemia. In pa-
tients maintained under dialysis treatment, the prevalence is higher (87%) as compared to
those without dialysis treatment (35%).50 Among kidney-transplanted children, the preva-
lence is 73%.51 Comparable with the observation in adult hemodialysis patients, the MTHFR
677C→T polymorphism is associated with significantly higher total homocysteine concentra-
tions as compared to CC genotype patients.50 In children with renal transplants and the MTHFR
677TT/1298AA genotype, plasma total homocysteine is significantly higher as compared to
677CC/1298AA patients. In contrast, MTHFR 677CT/1298AC is not a predictor of homocys-
teine levels in these patients, in agreement with the findings in adults.51
The effect of the MTHFR genotype on homocysteine lowering vitamin therapy has been
investigated in kidney transplanted children. Folate treatment allowed for a significant reduc-
tion of increased total homocysteine concentrations in patients with the MTHFR 677CC and
CT genotype, whereas this was not the case for TT genotype patients.52 This observation is in
clear contrast to adult renal failure patients and healthy individuals,53 who show a better re-
sponse to folic acid treatment in the presence of the 677TT genotype. Therefore, this observa-
tion in children has to be further investigated in larger study populations.
Conclusions
It is well established that the MTHFR 677TT genotype aggravates hyperhomocysteinemia
of renal failure patients independent of the type of renal replacement therapy. Furthermore, the
MTHFR 677TT genotype particularly influences folate status of patients not receiving folic
acid supplementation. In contrast, the influence of MTHFR 1298A→C alone on plasma total
homocysteine and folate concentrations is negligible. Compound heterozygosity for MTHFR
677C→T and MTHFR 1298A→C may influence folate status but not total homocysteine
concentrations. Among the genetic polymorphisms that have thus far been evaluated for a
putative influence on total homocysteine and folate plasma concentrations of renal failure
patients (MTR 2756A→G, MTRR 66A→G, RFC1 80G→A, GCP2 1561C→T, TCN2
776G→C), the MTHFR 677TT genotype is the strongest predictor. Recent data suggest that
the MTHFR 677TT genotype is a risk factor for vascular disease in individuals with and with-
out renal failure particularly when folate status is low. In dialysis patients, response to treat-
ment with folic and folinic acid is influenced by the MTHFR 677TT genotype and compound
heterozygosity for MTHFR 677C→T and MTHFR 1298A→C in that these patients more
frequently attain normal total homocysteine levels. Unfortunately, many patients with renal
failure show resistance to homocysteine lowering therapies with different doses of folic acid
alone, as well as in combination with other B-vitamins. In this context, it should be mentioned
that ongoing studies are aimed at identifying putative genetic defects that could contribute to
our understanding of folate resistance in renal failure. An interesting goal in the near future is
to find novel therapeutic strategies for normalization of increased total homocysteine levels and
to determine whether normalization indeed has a beneficial effect on vascular disease risk.
References
1. Muntner P, He J, Hamm L et al. Renal insufficiency and subsequent death resulting from cardio-
vascular disease in the United States. J Am Soc Nephrol 2002; 13:745-753.
2. American Society of Human Genetics / American College of Medical Genetics Test and Technol-
ogy Transfer Committee Working Group. ASHG / ACMG statement. Measurement and use of
total plasma homocysteine. Am J Hum Genet 1998; 63:1541-1543.
3. van Guldener C, Stam F, Stehouwer CDA. Homocysteine metabolism in renal failure. Kidney Int
2001; 59(Suppl 78):S234-S237.
4. Bostom A, Brosnan JT, Hall B et al. Net uptake of plasma homocysteine by the rat kidney in
vivo. Atherosclerosis 1995; 116:59-62.
5. van Guldener C, Donker AJM, Jakobs C et al. No net renal extraction of homocysteine in fasting
humans. Kidney Int 1998; 54:166-169.
6. Shemin D, Bostom AG, Selhub J. Treatment of hyperhomocysteinemia in end-stage renal disease.
Am J Kidney Dis 2001; 38(Suppl 1):S91-S94.
7. Bostom AG, Shemin D, Verhoef P et al. Elevated fasting total plasma homocysteine levels and
cardiovascular disease outcomes in maintenance dialysis patients. A prospective study. Arterioscler
Thromb Vasc Biol 1997; 17:2554-2558.
8. Ducloux D, Motte G, Challier B et al. Serum total homocysteine and cardiovascular disease occur-
rence in chronic, stable renal transplant recipients: A prospective study. J Am Soc Nephrol 2000;
11:134-137.
9. Mallamaci F, Zoccali C, Tripepi G et al. Hyperhomocysteinemia predicts cardiovascular outcomes
in hemodialysis patients. Kidney Int 2002; 61:609-614.
10. van Guldener C, Robinson K. Homocysteine and renal disease. Semin Thromb Hemost 2000;
26:313-324.
11. Skoupy S, Födinger M, Veitl M et al. Riboflavin is a determinant of total homocysteine plasma
concentrations in end-stage renal disease patients. J Am Soc Nephrol 2002; 13:1331-1337.
12. Födinger M, Buchmayer H, Sunder-Plassmann G. Molecular genetics of homocysteine metabolism.
Miner Electrol Metab 1999; 25:269-278.
13. van Guldener C, Kulik W, Berger R et al. Homocysteine and methionine metabolism in ESRD: A
stable isotope study. Kidney Int 1999; 56:1064-1071.
14. Ghandour H, Bagley PJ, Shemin D et al. Distribution of plasma folate forms in hemodialysis
patients receiving high daily doses of L-folinic or folic acid. Kidney Int 2002; 62:2246-2249.
15. Födinger M, Hörl WH, Sunder-Plassmann G. Molecular biology of 5,10-methylenetetrahydrofolate
reductase. J Nephrol 2000; 13:20-33.
mutation in methylenetetrahydrofolate reductase [letter]. Nat Genet 1995; 10:111-113.
17. Viel A, Dall’Agnese L, Simone F et al. Loss of heterozygosity at the 5,10-methylenetetrahydrofolate
reductase locus in human ovarian carcinomas. Br J Cancer 1997; 75:1105-1110.
reductase (MTHFR) associated with decreased enzyme activity. Mol Genet Metab 1998; 64:169-172.
19. van der Put NMJ, Gabreëls F, Stevens EMB et al. A second common mutation in the
Hum Genet 1998; 62:1044-1051.
(MTHFR) and methionine synthase reductase (MTRR) in ethnic populations in Texas; a report of
a novel MTHFR polymorphic site, G1793A. Am J Med Genet 2002; 107:162-168.
156:409-415.
22. Feix A, Fritsche-Polanz R, Kletzmayr J et al. Increased prevalence of combined MTR and MTHFR
genotypes among individuals with severely elevated total homocysteine plasma levels. Am J Kidney
Dis 2001; 38:956-964.
23. Feix A, Winkelmayer WC, Eberle C et al. Effect of methionine synthase reductase MTRR 66A>G
on total homocysteine plasma level of renal transplant patients. Athersclerosis 2004; 174(1):43-8.
24. Födinger M, Dierkes J, Skoupy S et al. Effect of glutamate carboxypeptidase II and reduced folate
carrier polymorphisms on folate and total homocysteine concentrations in dialysis patients. J Am
Soc Nephrol 2003; 14:1314-1319.
25. Winkelmayer WC, Eberle C, Sunder-Plassmann G et al. Effects of the glutamate carboxypepidase
II (GCP2 1561C>T) and reduced folate carrier (RFC1 80G>A) allelic variants on folate and total
homocysteine levels in kidney transplant patients. Kidney Int 2003; 63:2280-2285.
26. Födinger M, Veitl M, Skoupy S et al. Effect of TCN2 776C>G on vitamin B12 cellular availability
in end-stage renal disease patients. Kidney Int 2003; 64:1095-1100.
93:7-9.
28. Födinger M, Buchmayer H, Heinz G et al. Association of two MTHFR polymorphisms with total
homocysteine plasma levels in dialysis patients. Am J Kidney Dis 2001; 38:77-84.
29. Födinger M, Buchmayer H, Heinz G et al. Effect of MTHFR 1298A-->C and MTHFR 677C-->T
genotypes on total homocysteine, folate, and vitamin B12 plasma concentrations in kidney graft
recipients. J Am Soc Nephrol 2000; 11:1918-1925.
30. De Vriese AS, Verbeke F, Schrijvers BF et al. Is folate a promising agent in the prevention and
treatment of cardiovascular disease in patients with renal failure? Kidney Int 2002; 61:1199-1209.
31. den Heijer M, Willems HPJ, Bos GMJ et al. Plasma cysteine as a risk factor for venous thrombo-
sis. Thromb Haemost 1999; 99(suppl):506 (abstract).
32. Marcucci R, Fedi S, Brunelli T et al. High cysteine levels in renal transplant recipients: Relation-
ship with hyperhomocysteinemia and 5,10-MTHFR polymorphism. Transplantation 2001;
71:746-751.
33. Suliman ME, Stenvinkel P, Heimbürger O et al. Plasma sulfur amino acids in relation to cardio-
vascular disease, nutritional status, and diabetes mellitus in patients with chronic renal failure at
start of dialysis therapy. Am J Kidney Dis 2002; 40:480-488.
34. Kimura H, Gejyo F, Suzuki S et al. A C677T mutation in the methylenetetrahydrofolate reductase
gene modifies serum cysteine in dialysis patients. Am J Kidney Dis 2000; 36:925-933.
35. Eikelboom JW, Hankey GJ. Associations of homocysteine, C-reactive protein and cardiovascular
disease in patients with renal disease. Curr Opin Nephrol Hypertens 2001; 10:377-383.
36. The Homocysteine Studies Collaboration. Homocysteine and risk of ischemic heart disease and
stroke: A meta- analysis. JAMA 2002; 288:2015-2022.
37. Suliman ME, Qureshi AR, Barany P et al. Hyperhomocysteinemia, nutritional status, and cardiovas-
cular disease in hemodialysis patients. Kidney Int 2000; 57:1727-1735.
38. Wrone EM, Zehnder JL, Hornberger JM et al. An MTHFR variant, homocysteine, and cardiovas-
cular comorbidity in renal disease. Kidney Int 2001; 60:1106-1113.
40. Lim PS, Hung WR, Wei YH. Polymorphism in methylenetetrahydrofolate reductase gene: Its im-
pact on plasma homocysteine levels and carotid atherosclerosis in ESRD patients receiving hemodi-
alysis. Nephron 2001; 87:249-256.
41. Hagen W, Födinger M, Hörl WH et al. Therapy of hyperhomocysteinemia in end-stage renal
disease patients. Minerva Urol Nefrol 2001; 53:159-170.
42. Bostom AG, Shemin D, Lapane KL et al. High dose B-vitamin treatment of hyperhomocysteinemia
in dialysis patients. Kidney Int 1996; 49:147-152.
43. Sunder-Plassmann G, Födinger M, Buchmayer H et al. Effect of high dose folic acid therapy on
hyperhomocysteinemia in hemodialysis patients: Results of the Vienna Multicenter Study. J Am
Soc Nephrol 2000; 11:1106-1116.
44. Hauser AC, Hagen W, Rehak PH et al. Efficacy of folinic versus folic acid for the correction of
hyperhomocysteinemia in hemodialysis patients. Am J Kidney Dis 2001; 37:758-765.
45. Billion S, Tribout B, Cadet E et al. Hyperhomocysteinaemia, folate and vitamin B12 in
unsupplemented haemodialysis patients: Effect of oral therapy with folic acid and vitamin B12.
Nephrol Dial Transplant 2002; 17:455-461.
USA 1998; 95:13217-13220.
47. Liangos O, Kreutz R, Beige J et al. Methylenetetrahydrofolate-reductase gene C677T variant and
kidney-transplant survival. Nephrol Dial Transplant 1998; 13:2351-2354.
48. Hagen W, Födinger M, Heinz G et al. Effect of MTHFR genotypes and hyperhomocysteinemia
on patient and graft survival in kidney transplant recipients. Kidney Int 2001; 59(Suppl
78):S253-S257.
49. Födinger M, Wölfl G, Fischer G et al. Effect of MTHFR 677C>T on plasma total homocysteine
levels in renal graft recipients. Kidney Int 1999; 55:1072-1080.
50. Merouani A, Lambert M, Delvin EE et al. Plasma homocysteine concentration in children with
chronic renal failure. Pediatr Nephrol 2001; 16:805-811.
51. Aldámiz-Echevarría L, Sanjurjo P, Vallo A et al. Hyperhomocysteinemia in children with renal
transplants. Pediatr Nephrol 2002; 17:718-723.
52. Szabó AJ, Tulassay T, Melegh B et al. Hyperhomocysteinaemia and MTHFR C677T gene poly-
morphism in renal transplant recipients. Arch Dis Child 2001; 85:47-49.
54. Födinger M, Mannhalter C, Wölfl G et al. Mutation (677 C to T) in the methylenetetrahydrofolate
reductase gene aggravates hyperhomocysteinemia in hemodialysis patients. Kidney Int 1997;
52:517-523.
55. Lee HA, Choi JS, Ha KS et al. Influence of 5,10-methylenetetrahydrofolate reductase gene poly-
morphism on plasma homocysteine concentration in patients with end-stage renal disease. Am J
Kidney Dis 1999; 34:259-263.
56. Kimura H, Gejyo F, Suzuki S et al. The C677T methylenetetrahydrofolate reductase gene muta-
tion in hemodialysis patients. J Am Soc Nephrol 2000; 11:885-893.
57. Haviv YS, Shpichinetsky V, Goldschmidt N et al. The common mutations C677T and A1298C in
the human methylenetetrahydrofolate reductase gene are associated with hyperhomocysteinemia and
cardiovascular disease in hemodialysis patients. Nephron 2002; 92:120-126.
58. Nakamura T, Saionji K, Hiejima Y et al. Methylenetetrahydrofolate reductase genotype, vitamin
B12, and folate influence plasma homocysteine in hemodialysis patients. Am J Kidney Dis 2002;
39:1032-1039.
59. Morimoto K, Haneda T, Okamoto K et al. Methylenetetrahydrofolate reductase gene polymor-
phism, hyperhomocysteinemia, and cardiovascular diseases in chronic hemodialysis patients. Neph-
ron 2002; 90:43-50.
60. Anwar W, Guéant JL, Abdelmouttaleb I et al. Hyperhomocysteinemia is related to residual glom-
erular filtration and folate, but not to methylenetetrahydrofolate-reductase and methionine syn-
thase polymorphisms, in supplemented end-stage renal disease patients undergoing hemodialysis.
Clin Chem Lab Med 2001; 39:747-752.
61. Dierkes J, Domröse U, Ambrosch A et al. Response of hyperhomocysteinemia to folic acid supple-
mentation in patients with end-stage renal disease. Clin Nephrol 1999; 51:108-115.
62. Vychytil A, Födinger M, Wölfl G et al. Major determinants of hyperhomocysteinemia in perito-
neal dialysis patients. Kidney Int 1998; 53:1775-1782.
63. Cossu M, Carru C, Pes GM et al. Plasma homocysteine levels and C677T MTHFR gene polymor-
phism in stable renal graft recipients. Transplant Proc 2001; 33:1156-1158.
MTHFR Polymorphisms and Colorectal Neoplasia 179
CHAPTER 14
MTHFR Polymorphisms
and Colorectal Neoplasia
Jimmy W. Crott and Joel B. Mason
Introduction
F
olate is essential for the synthesis, repair and methylation of DNA, processes that are
central to maintaining the integrity of the genome. It is therefore not surprising that
aberrations in folate metabolism, either due to folate deficiency or altered activity of
folate-dependent enzymes, can modify our risk for cancer. Folate depletion compromises both
genetic and epigenetic features of DNA by altering DNA methylation patterns,1,2 increasing
the content of uracil in DNA3,4 and also increasing the frequency of breaks in DNA5 and
chromosomes.6,7 It is becoming increasingly evident that these types of changes to the struc-
ture and composition of DNA are involved in the etiology of cancer and probably other
age-associated diseases as well.8
As discussed in previous chapters, 5,10- methylenetetrahydrofolate reductase (MTHFR) is
a key folate-metabolizing enzyme, which catalyzes the reduction of 5,10- methylenetetrahydrofolate
(5,10 -methyleneTHF) to 5- methyltetrahydrofolate (5-methylTHF). MTHFR is intricately
linked to the integrity of the genome because its substrate functions as the one-carbon unit
donor for thymidine synthesis (and indirectly for purine synthesis) and the product contrib-
utes to the supply of methyl groups destined for DNA methylation. Perturbations in the nor-
mal functioning of MTHFR might therefore impact on cancer risk by altering the rate at
which DNA accumulates deleterious genetic and epigenetic modifications. It is also important
to realize that the biological methylation of other macromolecules, such as RNA9 and pro-
tein,10 is dependent on the ready supply of methyl groups and may play a role in the etiology of
disease as well.
A thermolabile variant of MTHFR, arising from a C to T nucleotide transition at position
677 of the gene and a subsequent alanine to valine substitution in the protein, has been the
focus of much research over the past decade. This 677C→T polymorphism has been shown to
cause a 70% reduction in the in vitro activity of the enzyme in homozygotes (TTs).11 It is also
clear that the C to T polymorphism reduces in vivo MTHFR activity by virtue of the fact that
in homozygotes plasma homocysteine levels are significantly elevated11 and formylated folates
accumulate at the expense of 5-methylTHF in red blood cells.12 A second polymorphism in
MTHFR involves an A to C nucleotide transition at position 1298 of the gene.13 In people
who are wild-type with respect to the 677 transition, homozygosity for the 1298A→C muta-
tion reduces the in vitro MTHFR (specific) activity by approximately 39% but does not confer
thermolability or seem to affect blood total folate and homocysteine concentrations.13
Since the 677C→T polymorphism is known to impact on cellular folate pools, it is perhaps
not surprising that the presence of the polymorphism affects the ability of cells to sustain
appropriate levels and patterns of DNA methylation. Recent evidence indicates that TTs have
©2005 Eurekah.com.
a compromised ability to sustain normal levels of DNA methylation under low folate condi-
tions and therefore have relatively hypomethylated DNA compared to CCs.14 It remains un-
clear whether the capacity for nucleotide synthesis is altered in TTs, although several groups15,16
have speculated that the shift towards more nonmethylated folates in TT individuals would
improve nucleotide synthesis and therefore DNA synthesis and repair.
This review summarizes the epidemiological evidence for an association of the MTHFR
677C→T and 1298A→C polymorphisms with colorectal neoplasia and discusses the postu-
lated mechanisms behind these relationships. It appears that the critical importance of the
MTHFR reaction is due, in large part, to the fact that the reaction determines the partitioning
of nonmitochondrial folate between two major pathways, biological methylation and nucle-
otide synthesis, aberrations of which are each implicated in carcinogenesis. Currently only two
publications report data on the1298A→C polymorphism and therefore this review focuses
mainly on the common 677C→T polymorphism. However, we begin this review by first sum-
marizing the molecular mechanisms underlying folate deficiency-induced genetic damage be-
cause these are central to understanding how MTHFR polymorphisms might affect an
individual’s risk for cancer.
Importance and Metabolism of Folate

One of the major functions of folate is to provide one-carbon units for the thymidylate
synthase-mediated conversion of uracil (dUMP) to thymidine (dTMP). In this reaction, the
methyl group from 5,10-methyleneTHF is transferred to uracil by thymidylate synthase to
form thymidine and dihydrofolate (Fig. 1). In addition to the synthesis of thymidine, folate, in
the form of 10-formylTHF, provides two carbons for the synthesis of the purine ring, the
precursor for both adenylate and guanylate.17
Besides being involved in the synthesis of DNA bases, folate is important in the synthesis of
methionine, an essential amino acid and precursor to S-adenosylmethionine (AdoMet), the
universal methyl group donor for the methylation of DNA, RNA, proteins and phospholipids.
In this reaction, methionine synthase transfers a methyl group from 5-methylTHF to ho-
mocysteine (via vitamin B12, the cofactor) to form methionine and THF. DNA methylation
occurs at the #5 carbon of certain cytosine residues in CpG dinucleotides. These dinucleotides
are underrepresented in the genome but occur in the expected frequency in clusters (CpG
islands) within the promoter region of most genes.18,19 Mutations occur much more frequently
at 5-methylcytosine than at unmethylated cytosine, a phenomenon thought to be at least par-
tially explained by the fact that spontaneous deamination of these bases results in the forma-
tion of thymidine and uracil, respectively. Because thymidine is a normal constituent of DNA,
as opposed to uracil, the DNA repair machinery more frequently fails to properly correct the
base mismatch, resulting in a mC to a T transition mutation.18
DNA methylation appears to be an important mechanism of gene silencing because the
majority of tissue-specific genes exhibit a strong correlation between hypomethylation of the
promoter region and gene expression.18 The silencing of an allele in the phenomenon of gene
imprinting also seems to be a function of permanent methylation of the promoter.20
Perturbations in gene methylation, particularly of the promoter region, are frequently asso-
ciated with carcinogenesis. For example, hypomethylation of c-fos, c-myc, Ha-ras, Ki-ras, Erb-A1
and bcl-2 proto-oncogenes18 and hypermethylation of p5321 and the mismatch repair gene
hMLH122,23 are common events in various cancers. Thus, in the absence of critical mutations
or deletions, aberrations in the methylation of cancer-related genes may be sufficient to cause
abnormal expression of these genes and thereby promote carcinogenesis.
It is also emerging that the methylation of lysine and arginine residues within histone pro-
teins is another important epigenetic phenomenon that may be involved in the regulation of
gene expression.24 Similarly, aberrations in the methylation of certain subspecies of RNA, such
as snRNA25 and tRNA,26 may interfere with transcription and translation, respectively.
MTHFR Polymorphisms and Colorectal Neoplasia
Figure 1. Metabolism of folate. B6= vitamin B6; B12= vitamin B12; DHF= dihydrofolate; dTMP= deoxythymidylate; dUMP= deoxyuridylate; FAD= flavin adenine
dinucleotide; MS= methionine synthase; MTHFR= methylenetetrahydrofolate reductase; AdoHcy= S-adenosylhomocysteine; AdoMet= S-adenosylmethionine; THF=
181
tetrahydrofolate; TS= thymidylate synthase.

Effects of Folate Depletion

Clinical
Classical folate deficiency causes megaloblastic changes in the bone marrow, peripheral blood
and intestinal epithelium, which include increased mean cell volume and neutrophil
hypersegmentation. Eventually, folate deficiency leads to anemia, and often diarrhea.27 Dimin-
ished folate status of a magnitude insufficiently large to cause megaloblastic anemia causes
hyperhomocysteinemia,28 increases the risk of congenital neural tube defects29 and is associ-
ated with Alzheimer’s disease30 and perhaps with Down syndrome.31
Evidence indicating a relationship between low folate status and an elevated risk for cancer
has rapidly accumulated over the past decade, including evidence from epidemiological,32 ani-
mal33 and cell culture studies.34 Because rapidly dividing tissues have a high demand for thymi-
dine bases and methyl groups for the maintenance of DNA integrity, it follows that these
tissues would be affected the worst, in terms of the accumulation of mutagenic lesions, by low
folate availability. Indeed, it is in the rapidly replicating epithelium of the colorectum where
the strongest evidence for a link between low folate and cancer exists.1 Several large prospective
cohort studies collectively indicate a 30-40% reduction in the risk of colorectal neoplasms
among those consuming high versus low amounts of dietary folate.32,35
Biochemical
During folate deficiency, reduced availability of 5-methylTHF slows the conversion of ho-
mocysteine to methionine, and it is well documented that plasma homocysteine concentra-
tions increase in response to folate deficiency.2,28 In rodent models, folate deficiency is known
to reduce AdoMet concentrations in the liver36 and produce an elevation in colonic37 and
blood AdoHcy38 concentrations. It is beginning to emerge that AdoHcy is more strongly asso-
ciated with DNA hypomethylation than reductions in AdoMet,38 apparently due to its com-
petitive inhibition of DNA methyltransferase enzymes.
Molecular and Cellular Effects

Genetic Effects
Folate deficiency causes an imbalance in nucleotide pools that is characterized by an accu-
mulation of dUMP and a depletion of dTMP. In cell culture studies, it is reported that folate
depletion increases the dUMP:dTMP ratio roughly ten-fold, compared to replete cultures.34,39
A subsequent increase in DNA uracil content during folate depletion has been observed both
in vitro4,5,39-41 and in vivo.2,3,42 To date there are no data on the effect of folate status on uracil
incorporation into cells of the colorectum in vivo. However, it is reported that immortalized
human colonic epithelial cells accumulate uracil in DNA when grown in low concentrations of
folic acid.41
The role of uracil in folate deficiency-induced chromosome damage was realized more than
a decade ago. In his pioneering work, Reidy showed that culturing blood cells in folate-deficient
media caused chromosome breakage, which was enhanced by the addition of deoxyuridine and
abolished by the addition of deoxythymidine.43 Dianov and colleagues44 then showed that the
excision of uracils from DNA could cause double-stranded DNA breakage (DSB). Using a
circular plasmid, they showed that the presence of two uracils, 12 bases apart and on opposing
strands, facilitated DSB formation when the plasmid was exposed to glycosylase-containing
crude cell extracts.
DNA hypomethylation (see section Epigenetic Effects) may also indirectly contribute to
DSB formation because unmethylated cytosines can be deaminated to form uracil, thus pre-
senting a second mechanism for the introduction of uracil into DNA. Shen et al45 showed that
at low concentrations of AdoMet, as seen during folate depletion, DNA methyltransferase
facilitated the deamination of cytosine to uracil in a bacterial cell model.
Folate deficiency-induced chromosome breakage may be an important health issue because

the accumulation of chromosome aberrations, such as double-stranded breaks, is an estab-
lished risk factor for the development of cancer.46-48 For example, Bonassi and colleagues47
report that, compared to people grouped into the lowest tertile of chromosome aberrations,
people in the highest tertile had a 2.35-(95% CI 1.31-4.23) and 2.66-(95% CI 1.26-5.62) fold
increased risk of developing (any) cancer in Nordic and Italian cohorts, respectively. It is pos-
sible that MTHFR activity may influence the risk for colorectal cancer by altering the rate at
which uracil-induced DNA breaks accumulate in dividing cells.
It is thought that the accumulation of DSBs increases the risk for cancer through a number
of mechanisms. Firstly, unrepaired DSBs may directly cause the deletion of tumor suppressor
genes, a common finding in many human solid tumors.49 Furthermore, during the repair of
DSBs by homologous recombination or nonhomologous end-joining, errors may occur that
result in chromosome translocations. Translocations are thought to play an etiologic role in
cancers such as acute and chronic myelogenous leukemia, acute lymphoblastic leukemia, Burkitt’s
lymphoma and desmoplastic round cell tumor.50 Balanced translocations can cause cancer
either by altering the expression of normal genes or by the production of novel hybrid proteins
with oncogenic potential.51
In addition to the potential for the loss or corruption of the genetic code, unrepaired DSBs
may initiate and fuel gene amplification by bridge-fusion-breakage (BFB) cycles.52;a
Dihydrofolate reductase gene amplification by BFB cycles has been observed in Chinese ham-
ster ovary cells in response to methotrexate treatment.53 Furthermore, folate deficiency induces
nuclear budding, a morphological abnormality of the nucleus associated with gene amplifica-
tion, and nuclear bridge formation (an intermediate in the BFB cycle) in a primary human
lymphocyte culture system.6 Gene amplification is thought to be a key event in cellular resis-
tance to drugs like methotrexate,54 and also in tumor progression.54 For example, amplifica-
tion of the N-myc gene occurs in approximately 38% of neuroblastomas and is strongly asso-
ciated with tumor aggressiveness.56
Epigenetic Effects
As mentioned above, folate deficiency induces deleterious changes in DNA methylation
patterns. For example, women who underwent five weeks of folate depletion (56 µg folate/day)
exhibited genomic DNA hypomethylation in lymphocytes characterized by a more than two-fold
increase in DNA methyl acceptance.2 Furthermore, in immortalized normal human colon
cells, folate deficiency significantly increases genomic hypomethylation in vitro.41 In rats, folate
deficiency has been shown to induce both genomic57 and gene-specific (coding region)
hypomethylation.57,58
Interestingly, folate deficiency has also been reported to cause localized DNA
hypermethylation. Pogribny and colleagues42 report that 36 weeks of folate deficiency induced
hypomethylation in exons 6-7 of the p53 gene in rat liver. When the same region was examined
after 54 weeks in tumor DNA, almost all CpG dinucleotides were methylated despite the
continued folate depletion. This reversal of methylation was associated with a four-fold eleva-
tion in de novo methylation capacity compared to control and preneoplastic (36 weeks) tissue.
In this regard, it is important to note that evidence is accumulating to suggest that
a
In a variation of McClintock’s classic model,52 replication of a broken chromatid during S-phase results
in the formation of two homologous chromatids with ‘sticky’ ends. These sticky sister chromatids may then
fuse to form a hairpin-like dicentric chromosome structure. During anaphase, dicentric chromosomes are
drawn towards both poles of the cell and are subsequently broken. Because there are two copies of certain
genes between the centromeres (one from each chromatid), unequal breakage will result in the partitioning
of two copies of certain genes to one daughter chromatid, and a deletion of these genes in the other. In the
next division, the chromatid with duplicated genes may fuse again and propagate the bridge-breakage-
fusion cycle.
hypermethylation of certain genes is also involved in carcinogenesis. For example, promoter

region methylation of p53 and the mismatch repair gene hMLH1 is associated with reduced
expression of these transcripts in hepatic21 and colorectal22 cancer cells, respectively.
Genomic DNA hypomethylation is an early and consistent feature of carcinogenesis al-
though it remains unclear how important a mechanistic role it plays.18,59 In many human
cancers, including those of the stomach, uterine cervix and colorectum, there is a graduated
decrease in genomic DNA methylation in normal tissue versus precancerous dysplastic tissue
versus the cancer tissue itself.60-62 Nevertheless, given the importance of critical loci in cancer,
gene-specific methylation probably plays a more important role than genomic methylation.
Experimentally-induced DNA hypomethylation has been shown to cause the loss of chro-
mosomes 1, 9, 15, 16 and Y during mitosis due to the under-condensation of peri-centromeric
heterochromatin.63 Chromosome loss leads to aneuploidy, a recognized risk factor for cancer.64
Furthermore, because many genes are silenced by methylation of the promoter region,
hypomethylation of specific growth-promoting genes such as proto-oncogenes may confer a
selective advantage to cells, which may lead to transformation.18
DNA microarray technology may prove invaluable in the search for genes affected by folate
depletion. The first study utilizing this technology to examine the effects of folate deficiency
has shown that, of approximately 2000 genes associated with oncogenesis, folate deficiency
caused the up-regulation of 3 genes and down-regulation of 5 genes in human nasopharyngeal
epidermoid carcinoma cells.65 Of interest was a 2.5-fold down- regulation of H-cadherin, a
gene that is commonly down-regulated in cancer cell lines and tumors. This reduced expres-
sion was linked to a 40% increase in the methylation of CpGs within the H-cadherin gene.
It is conceivable that folate depletion may also be involved in carcinogenesis by altering the
expression of foreign genes with oncogenic potential. DNA methylation is involved in a de-
fence system that suppresses parasitic sequences or transposons, which make up at least 35% of
the human genome.66 Transcripts of these transposons are known to be more abundant in
tumor cells compared to normal cells and hypomethylation-induced transcription of transposons
is thought to initiate a series of mutagenic events that can lead to cancer.66
MTHFR 677C→T Polymorphism

Epidemiology
Effect on Risk for Colorectal Cancer
Of all the tissues that are susceptible to cancer, it is the colon and rectum that have received
the most attention from researchers investigating the effect of MTHFR polymorphisms on
cancer risk. Indeed, it was in the colorectum that the first association between MTHFR 677C→T
genotype and neoplasia was observed approximately six years ago.15 Currently, most epidemio-
logical evidence indicates that the MTHFR 677C→T polymorphism has a significant impact
on the risk for colorectal (CR) neoplasia. From a mechanistic perspective, this was an impor-
tant observation since it lent support to the concept of a true casual link between one-carbon
nutrients (folate, B2, B6, B12, choline and methionine) and CR carcinogenesis. The 677C→T
polymorphism is an excellent example of a gene-nutrient interaction and is unique in this
regard since it is possibly the only known gene polymorphism that switches from being a risk
factor to a protective factor depending on nutrient status. The evidence for a role of the 677C→T
polymorphism in CR cancer is summarized in Table 1.
The first association of the 677C→T polymorphism with CR cancer was reported by Chen
et al.15 Data from their case-control study, which was nested within the large US-based Health
Professionals Follow-up Study, showed that TTs had approximately half the risk for CR cancer
compared to CTs and CCs (OR 0.57, 95% CI, 0.30-1.06). After stratification by alcohol
intake, it was found that within the group drinking one glass of alcohol or less per week, TTs
enjoyed a reduction in cancer risk compared to CCs and CTs (OR 0.11, 95% CI 0.01-0.85).
Table 1. Summary of the evidence for a role of the MTHFR 677C→T polymorphism in colorectal cancer
Cases v Plasma/ Overall O.R./R.R

Reference Design Population Controls Endpoint [RBC] Folate for TT v CC Comments
Chen et al, Case-ctrl U.S. M health 144 cases, CR cancer n/a 0.57 (0.3-1.06) Reduced risk abolished by > 5
199615 (nested) professionals 627 ctrl alcoholic drinks/week. No interaction
40-75 yr with folate intake.
Ma et al, Case-ctrl U.S. M 202 cases, CR cancer cases, 5.45; ctrl, 0.45 (0.24-0.86) Reduced risk abolished when folate <
199770 (nested) physicians 326 ctrl 5.84 ng/mlNS. TT, 3.0 ng/ml but risk not greater than CC/
40-80 yr 3.43; CC, CTs.

6.44 ng/ml*
Chen et al, Case-ctrl same as ref. 70 211 cases, CR cancer TT, 3.51: CC, 0.54 (0.29-0.99) Risk for 677 TTs same as before (65).
200271 (nested) 343 ctrl 4.46*. 1298AA, 1298A→C polymorphism had no effect
4.11; 1298CC on CR cancer risk, folate or
4.5 ng/ml NS homocysteine.
Slattery et al, Incident U.S. M + F, 1467 cases, Colon cancer n/a 0.9 (0.7-1.1) In high tertiles of folate & B6 intakes,
199972 case-ctrl 30-79 yr 1821 ctrl TTs may be more protected especially
against proximal tumors and in people
> 60 yrs. OR 0.5 (0.2-0.9) for proximal
tumors for TTs with high folate v CCs
with low folate
185
186
Table 1. Continued
Cases v Plasma/ Overall O.R./R.R

Reference Design Population Controls Endpoint [RBC] Folate for TT v CC Comments
Le Marchand Case-ctrl Hawaiian M + F, 548 cases, CR cancer n/a 0.7 (0.5-1.0) P-trend for reduced risk =0.04. Lowest
et al, 200273 57-74 yr 656 ctrl risk in TT with mid folate intake OR
0.4 (0.2-0.9). TTs with highest B6
protected v low B6 CCs OR 0.4 (0.2-
0.8). Additive effect of 677C→T &
1298A→C.
Shannon et al, Case-ctrl Australian M +F, 501 cases CR cancer n/a n/a For ≥ 70 yrs, TT more frequent in cases
200274 20-92 yr (75 MSI+), (12%) than ctrl (7%)*. For CRC, TT
1207 ctrl more frequent in MSI+ than MSI- *. TTs
diagnosed with CRC-MSI+ at older age
than CC*.
Ryan et al75 Case-ctrl UK M+F. ctrl <1, 136 cases CR cancer n/a n/a T allele frequency higher in cases
case 68a yr 848 ctrlb (0.37) than newborn controls (0.29)*
95% Confidence intervals in parenthesis. Ctrl= control; CC= person nullizygous for 677C→T polymorphism; CR= colorectal; F= female; HPP= hyperplastic polyp;
HRA= high-risk adenoma; M= male; MSI+/-= microsatellite instability present/absent; NS= non-significant difference; OR= odds ratio; TT= person homozygous for
the 677C→T polymorphism; yr= years of age. * P < 0.05. a Median age. b The controls in this study were newborns who were genotyped from Guthrie cards.
However, this protection enjoyed by TT homozygotes was abolished in the group drinking five
or more alcoholic beverages per week. Alcohol is reported to significantly reduce colonic folate
levels in rats, an effect possibly mediated by microbially-produced acetaldehyde.67 Further-
more, chronic ethanol ingestion reduces the AdoMet/AdoHcy ratio, methionine concentra-
tion and methionine synthase activity, increases homocysteine and DNA strand breakages in
the liver of micropigs,68 and induces hypomethylation of genomic DNA in the colon of rats.69
The protective effect of the T allele may be facilitated by an altered distribution of folate
coenzymes, whereby the increased proportion of nonmethylated folate would promote effi-
cient thymidine synthesis.15 More specifically, in the cells of TTs, diminished conversion of
5,10-methyleneTHF to 5-methylTHF by MTHFR is thought to cause accumulation of
5,10-methyleneTHF and therefore increase the availability of one-carbon groups required for
the synthesis of dTMP and purines. This ‘diversion’ of methylfolate towards nucleotide synthe-
sis would presumably enhance DNA synthesis and repair and minimize the intracellular
dUMP:dTMP ratio, the chance of uracil being incorporated into DNA and hence the fre-
quency of uracil-associated double-stranded DNA breaks. However, as a consequence of this
diversion of folate away from 5-methylTHF synthesis and homocysteine remethylation, the
methylation capacity of TT cells appears be compromised under low folate conditions (See
section Methylation).
Data from a second case-control study, nested within the (male) Physicians Health Study,
provided support for the interesting association of the mutant MTHFR gene with reduced
cancer risk.70 In this study, TTs again had half the likelihood of developing colorectal cancer
compared to CCs and CTs (OR 0.45, CI 0.24-0.86). This relationship was driven by the group
who had plasma folate concentrations above 3.0 ng/ml (OR 0.32, CI 0.15-0.68), with the
protection afforded by the polymorphism being absent in the group with low plasma folate
(folate < 3.0 ng/ml). In the first study, Chen et al15 failed to detect such an interaction using
folate intake, as opposed to plasma concentration, possibly due to the imprecision of food
frequency questionnaires. Although the second study detected an effect of folate status on the
relationship between MTHFR and CR cancer, it failed to detect any significant interaction
with alcohol intake, as had been observed by Chen et al.15 The reason for this discrepancy is
unclear.
When the cohort from the Physicians Health Study was reanalyzed with further adjustment
for the MTHFR 1298A→C polymorphism,71 the protection enjoyed by the TTs persisted,
with an OR of 0.54 (0.29-0.99) for CR cancer compared to CCs. This time, however, no
difference was detected between the risk of TTs when divided into groups with folate concen-
trations above and below the median. Analysis of the distribution of the 1298A→C polymor-
phism revealed that neither heterozygous nor homozygous mutants had an altered risk for CR
cancer compared to wild-types and, furthermore, that the polymorphism did not have any
impact on plasma folate or homocysteine concentrations.
Further evidence for a protective effect of the 677C→T polymorphism on CR cancer was
provided by Slattery72 in the largest study on MTHFR and CR cancer to date. While the crude
cancer risks were similar for CCs and TTs (OR 0.9, CI 0.7-1.1), the polymorphism provided a
moderate level of protection against proximal tumors for those in the highest tertiles of folate
(OR 0.5, CI 0.2-0.9) and vitamin B6 intakes (OR 0.5, CI 0.3-0.9). Similar associations were
observed for all tumors (distal and proximal) in the group over the age of 60 yrs.
In the Hawaiian population, the T allele was also found to be weakly associated with de-
creased risk for CR cancer (P-trend= 0.04).73 Curiously, when stratified according to total
folate intake, it was TTs in the middle tertile, rather than the highest, who enjoyed the greatest
protection (OR 0.4, CI 0.2-0.9). This may be related to the fact that the mid-tertile included
those with very high folate intake, i.e., up to 1.6 mg/day. Also of note was the observation that
TTs with the highest intake of vitamin B6 had a reduced risk compared to CCs with the lowest
and were more protected than fellow TTs consuming lesser amounts of B6. When considering
both the 677C→T and 1298A→C polymorphisms, people with one or two copies of both
mutations were significantly protected against CR cancer compared to wild-type individuals

(OR 0.6, CI 0.4-0.9) and also enjoyed greater cancer protection than people with the 1298AA/
677CT or TT genotype. This is the first report of an additive effect (albeit weak) of both
polymorphisms in reducing the risk for CR cancer. No effect of the 1298A→C polymorphism
alone was observed. It would be interesting to investigate the effect of folate status on this
gene-gene interaction.
Recently, Shannon et al74 reported that the TT genotype is associated with a later age of
diagnosis (and presumably onset) for proximal colon cancer than CCs (76 v 67 years, P= 0.03).
In addition to improved survival in TTs, the higher frequency of TTs among cases than con-
trols in the over 70 age group may be explained by the fact that because CCs develop CRC
earlier than TTs, they also die earlier, thereby causing TTs to be over- represented in the older
age groups. Furthermore, of particular interest, the researchers subdivided cases according to
the presence or absence of microsatellite instability (MSI) in their tumors and discovered that
the proportion of TTs was elevated among those with MSI+ tumors. The authors hypothesized
that this may be due to methylation-associated inactivation of mismatch repair genes (e.g.,
hMLH1), a so-called methylator phenotype, in TTs. Although this explanation is plausible,
there is currently only evidence of global hypomethylation in TTs,14 not site-specific or pro-
moter DNA hypermethylation. While there is evidence of folate depletion-induced
hypermethylation of p53 exons 6-7 following a period of hypomethylation in rat liver, there is
no comparable data for promoter methylation status.42
In a comparison of the genotype distribution between newborn controls and CR cancer
cases, Ryan et al75 reported that, among the cases, there was an over-representation of the CT
genotype and a higher T allele frequency compared to controls. Although this suggests that the
T allele is a risk factor for CR cancer, the analysis is complicated by the fact that some of the
newborns would have invariably developed cancer. Of greater interest is their observation that
in CTs, loss of MTHFR heterozygosity occurred in 18% of the tumors tested and that it was
always the T allele that was lost. A similar result has previously been reported,76 and may
indicate that the loss of the C allele in CT cells, and subsequent hemizygosity for the mutant T
allele, may be incompatible with cell survival.
The above studies represent the sum of our knowledge to date on the association of mutant
MTHFR with CR cancer. Taken together, these studies suggest that the 677C→T polymor-
phism is protective against CR cancer when alcohol intake is low and folate intake is high.
Preliminary evidence suggests that the MTHFR 1298A→C polymorphism alone is not associ-
ated with CR cancer risk,71,73 but may provide additional protection to people who also carry
the 677C→T polymorphism.73 More data are required to make definitive conclusions on the
role of the 1298A→C polymorphism, alone and in combination with the 677C→T polymor-
phism, in carcinogenesis.
Effect on Risk for Colorectal Adenoma

The evidence for an association of the 677C→T polymophism with CR adenomas is sum-
marized in Table 2. Following their first study on MTHFR and CR cancer, Chen et al77 unex-
pectedly failed to detect any association of the 677C→T genotype with risk for CR adenoma
in a cohort of female nurses. Again exploiting the availability of a large prospective study (the
Nurses Health Study) in a nested case-control manner, they reported that the frequency of the
mutant T allele was similar between cases and controls (OR 1.35, CI 0.84-2.17). Stratifying
the volunteers according to adenoma size and folate, methionine, and alcohol intake also failed
to reveal any association of the polymorphism with the precancerous lesion. The authors sug-
gested that an effect of the polymorphism might only be manifested when colonic epithelial
cells have a particularly high nucleotide demand, such as when cell division is markedly in-
creased during the transition from adenoma to carcinoma.
Ulrich and colleagues78 provided the first hint of an association between MTHFR 677C→T
genotype and adenoma risk. They showed that while the frequency of T alleles was similar in
Table 2. Summary of the evidence for an association of the MTHFR 677C→T polymorphism with colorectal adenomatous and
hyperplastic polyps
Cases vs Plasma/ Overall O.R./

Reference Design Population Controls Endpoint [RBC] Folate R.R for TT v CC Comments
Chen et al, Case-ctrl U.S. F nurses, 257 cases, CR adenoma n/a 1.35 (0.84-2.17) No effect of T allele. No interaction of
199877 (nested) 30-55 yr 713 ctrl genotype with folate, alcohol or
methionine intake.
Ulrich et al, Incident U.S. M +F 527 cases, CR adenoma n/a 0.8 (0.5-1.3) TTs had trend for decreased risk in
199978 case-ctrl 30-74 yr 645 ctrl highest tertiles of folate, B6, B12 &
methionine intakes but for increased
risk in low tertiles.
Ulrich et al, Incident U.S. M + F, 200 cases, CR HPP n/a 0.9 (0.5-1.6) No effect of polymorphism. No real
200079 case-ctrl 30-74 yr 645 ctrl interaction of genotype with folate, B6,
B12 alcohol or methionine intake.
Levine et al, Incident U.S. M + F, 471 cases, CR adenoma cases, 11.9; ctrl, 1.11 (0.71-1.71) TTs with RBC folate < 165 ng/ml had
200081 case-ctrl 50-74 yr 510 ctrl 13.5 ng/ml. TT, increased risk: O.R. 2.04 (0.60- 6.96).
11.3 [305.8]; CC, Trend for TT to switch from risk factor
12.7NS at low folate to protective at high
[260.2*] ng/ml folate.
Marugame Incident Japanese military 205 cases, CR adenoma n/a 1.17 (0.61-2.23) No real effect of polymorphism.
et al, 2000 case-ctrl M 47-55 yr 220 ctrl
Ulvik et al, Cross- Norwegian. 327 Polyp, CR adenoma CC, 10.9 [227]; HRA, 2.41 When RBC folate < 263 nM/l, TTs had
200183 sectional M +F, 63-72 yr. 215 HPP, and HPP TT, 8.2* (0.82-7.06) increased risk for HRA (O.R. 6.06 (1.43
177 - 25.7)). In CT/TT, low folate increases
adeboma smoking assoc. HRA risk. In CC, high
folate increases smoking assoc. HRA
risk.
95% Confidence intervals in parenthesis. Ctrl= control; CC= person nullizygous for 677C→T polymorphism; CR= colorectal; F= female; HPP= hyperplastic polyp;
HRA= high-risk adenoma; M= male; MSI+/-= microsatellite instability present/absent; NS= non-significant difference; OR= odds ratio; TT= person homozygous for
189
the 677C→T polymorphism; yr= years of age. * P < 0.05.

CR adenoma cases and controls (OR 0.8, CI 0.5-1.3), among those in the highest tertiles of
folate, methionine, vitamin B6 and vitamin B12 intake, there was a weak inverse relationship
between the number of T alleles and adenoma frequency which did not reach statistical signifi-
cance. Stratifying volunteers according to age, polyp location and number did not increase the
strength of these relationships. When this cohort was reanalyzed with CR hyperplastic polyps
as the outcome of interest, no effect of the polymorphism, regardless of dietary intakes, was
observed.79 Since it is unlikely that hyperplastic lesions have cancerous potential and it remains
unclear whether they are even associated with the presence of adenomatous polyps,80 their
relationship with MTHFR polymorphisms (or lack thereof ) is of less interest.
Levine and colleagues81 also reported that, overall, the 677C→T polymorphism had no
bearing on the risk for distal CR adenomas in a cohort of cases and controls from the general
public. However, dividing the volunteers into quartiles of plasma folate revealed a significant
trend (P-trend= 0.04) for increased adenoma risk in TTs in the lowest quartile and decreased
risk in the highest quartile when the CC/CT low folate group was used as the reference group.
This result is somewhat difficult to interpret because the genotype and folate status parameters
are combined in a single analysis. It would be of interest to know whether a ‘genotype only’
effect exists within each tertile of folate in this data set.
In a study of males in the Japanese military, Marugame and colleagues82 also failed to detect
any differences in the overall incidence of the 677C→T polymorphism between adenoma
cases and controls. However, since there was no measure of folate intake or blood concentra-
tions, this study is difficult to interpret.
Ulvik and colleagues83 published one of the most thorough examinations of MTHFR sta-
tus and CR neoplasia to date. They reported that MTHFR 677C→T genotype, RBC folate
concentration and smoking status strongly interact to affect the risk for high-risk adenomas
(HRA), which are adenomas that carry a high risk of evolving into cancers. In accordance with
previous studies on adenomas,77,78,81,82 there was no significant effect of MTHFR genotype on
overall HRA risk. However, after dividing volunteers into groups with RBC folate above and
below the mean (263 nmol/L), a strong genotype effect emerged. In the low folate group, CTs
and TTs had a 3.53-(95% CI 1.38-9.05) and 6.06-(95% CI 1.43-25.7) fold elevated risk of
HRA, respectively, compared to CCs, while in the high folate group the risk for all three geno-
types was similar. Introducing smoking to this model produced some novel findings. In the low
folate group, smoking increased the risk for CT/TTs from 2.96 (95% CI 0.94-9.28) to 8.7
(95% CI 2.4-28.1). Meanwhile, in the high folate group, this smoking-associated increase in
risk suffered by CT/TTs in the low folate group was abolished whereas, in CCs, high folate
substantially increased HRA risk from 1.54 (0.34-7.03) to 11.85 (2.86-49.1). In other words,
the effect of folate on HRA risk associated with smoking was opposite for CCs and CT/TTs. In
CCs it is the high folate group who have a high risk of HRA due to smoking, whereas in CT/
TTs it is the low folate group who have a high risk of HRA due to smoking. The authors
hypothesized that in both the high and low folate group, smoking acts as an initiator of the
cancer process and in the low folate group, carcinogenesis is driven by hypomethylation in the
CT/TTs, whereas in the high folate group cancer cell growth is fuelled by higher folate concen-
trations in CCs. If confirmed, these results dictate caution in implementing initiatives in folate
supplementation since high folate concentrations seem to increase HRA risk in smokers with
the CC genotype but are protective in nonsmokers and smokers with the T allele.
It is likely that future studies will identify other one-carbon nutrients (e.g., methionine, B6,
B12) as factors that further modify the relationship between folate, the MTHFR 677C→T
polymorphism and CR cancer. For instance, recent evidence suggests that plasma riboflavin
and homocysteine concentrations are inversely associated in TTs but not CCs or CTs84,85 and
it is therefore likely that riboflavin will factor into this complicated gene-nutrient interaction
for CR neoplasia. Riboflavin concentrations may affect MTHFR activity because riboflavin is
the precursor to MTHFR’s cofactor, FAD. The 677C→T polymorphism specifically affects
the FAD-binding domain and reduces the enzyme’s affinity for FAD.86 In TTs, high riboflavin
(and hence FAD) concentrations restore MTHFR activity by counteracting this reduced affinity.84
Presently, the body of evidence linking MTHFR 677C→T with CR neoplasia is stronger
for cancer than adenoma, with only one study providing compelling evidence for an associa-
tion with adenoma.83 The adenoma studies are generally consistent with the cancer studies
since there is a suggestion in both that the 677C→T polymorphism is protective against neo-
plasia when folate intake is high. Remarkably, these studies also suggest that at low folate in-
takes, TTs may actually have a higher risk for developing adenomas than CCs. In other words,
it is beginning to emerge that, depending on folate intake, the 677C→T polymorphism can
change from being a protective factor, to being a risk factor for CR neoplasia. The disparity
among adenoma studies may, inpart, be due to confounding by the existence of two distinct
subpopulations of adenomas. Ulvik et al83 were interested in HRA, which, in terms of cancer
progression, is linked much more closely to carcinoma than low-risk small adenomas. As sug-
gested by Chen et al,77 an effect of genotype on neoplasia risk may only be detectable later on
in the development of a carcinoma when nucleotide demand is highly increased. All of the
other adenoma studies lack stratification into low- versus high-risk adenomas and it is there-
fore possible that an effect of genotype on the incidence of HRA was obscured by the lack of
effect in low-risk lesions. It is also feasible, although less likely, that the molecular mechanisms
underlying the development of low- and high-risk adenomas are different and it is only the
mechanism of HRA formation that interacts with MTHFR and folate status to alter the risk of
neoplasia.
Effect on Biochemical and Molecular Endpoints

Hypomethylation-driven carcinogenesis and thymidine-driven protection are generally given
as explanations of any association of the MTHFR polymorphism with neoplasia, but six years
after the first report,15 the molecular mechanisms underlying these associations with colorectal
neoplasia remain unclear. There is a clear need to answer these questions by comparing DNA
methylation patterns, DNA uracil-content and markers of chromosome damage in colorectal
biopsies between the MTHFR 677C→T genotypes in case-control studies of CR neoplasia.
Blood Folate and Homocysteine

As well as having significantly lower plasma70,71,83 and RBC folate levels14 people with two
copies of the T allele also have an altered distribution of folates. Such effects are likely related to
the impact that the polymorphism has on CR neoplasia risk. To date, altered folate distribution
in TTs has been documented in RBCs12 and, in a preliminary fashion, in the colonic mucosa.87
The change in profile of folate coenzymes is characterized by a depletion of methylfolate (mainly
5-methylTHF), presumably due to reduced conversion of 5,10-methyleneTHF to 5-methylTHF
by the less active mutant MTHFR.12,14 Furthermore, lower MTHFR activity results in the
accumulation of formylated folates in RBCs when none are normally present in these cells.12
As mentioned earlier, there is a wealth of data showing that plasma homocysteine concen-
trations are elevated in TTs (reviewed elsewhere in this book). This effect appears to be the
result of a three-way interaction between the polymorphism, folate status and homocysteine,
since it is only those TTs with low folate status who express higher homocysteine levels than the
corresponding CCs.85 The limited availability of methylfolate in TTs, particularly in the set-
ting of low folate status, significantly impairs a cell’s ability to remethylate homocysteine and
therefore homocysteine accumulates. Recently, Hustad et al84 showed that vitamin B2 concen-
trations affect this relationship as well, with low vitamin B2 concentrations accentuating the
homocysteine elevation in TTs but not CCs.84
To our knowledge, there have been no comparisons of AdoMet and AdoHcy concentra-
tions between CC and TTs. One could reasonably speculate, however, that excessive elevations
in AdoHcy among TTs would follow the same pattern as elevations in homocysteine, since the
AdoHcy hydrolase reaction favors the reverse reaction,88 and elevations in homocysteine al-
most invariably raise AdoHcy levels as well.38
Methylation
Because the phenotypic expression of the MTHFR 677C→T polymorphism retards the
conversion of 5,10-methyleneTHF to 5-methylTHF, there is a reduced potential for cells to
synthesize AdoMet from recycled methionine. In addition, the elevated homocysteine levels in
these individuals would almost certainly raise intracellular AdoHcy levels, a potent inhibitor of
AdoMet-dependent methylation reactions.89 Recent evidence indicates that TTs have relatively
hypomethylated DNA compared to CCs. In a preliminary study of 9 CCs and 10 TTs, blood
mononuclear cell DNA from TTs accepted 1.6-fold more radiolabelled methyl groups than
DNA from CCs, which is indicative of genomic hypomethylation.90
These results were recently confirmed in a larger cohort of 105 TTs and 187 CCs from
northern Italy.14 Overall, blood mononuclear cell DNA from TTs contained approximately
half as much 5-methylcytosine as DNA from CCs. When volunteers were divided into groups
with plasma folate concentrations above and below the median value, CCs from both groups
and TTs from the high folate group showed similar abundances of 5-methylcytosine in DNA.
However, TTs in the low folate group had 60% lower 5-methylcytosine than TTs in the high
folate group. This phenomenon may be the basis for the abolished protection or even increased
risk for CR neoplasia in TTs when dietary folate and methionine intake are low and alcohol
consumption is high. Under conditions of ample folate availability, the increased size of the
nonmethylfolate pool may benefit TTs since nucleotide synthesis would presumably be facili-
tated. However, under low folate conditions, the development of hypomethylated DNA (and
RNA/protein/phospholipids) would outweigh the benefits on nucleotide synthesis and drive
the cell towards neoplasia.
Uracil
Unlike altered DNA methylation status, there is currently little evidence on the effect of the
677C→T polymorphism on the rate of uracil incorporation into DNA in any cell type. Al-
though many researchers have hypothesized that a lower rate of uracil incorporation into DNA
in TTs constitutes the underlying mechanism for reduced cancer risk in TTs, only one study
has addressed this issue experimentally.4 In a comparison of the amount of uracil incorporated
into DNA by primary lymphocytes from CCs and TTs, no difference was observed when cells
from each genotype were grown in media containing between 12 and 120 nM folic acid for
nine days,4 even though the lowest folic acid concentration induced a 2.7- and 3-fold increase
in DNA uracil content and micronucleus frequency,b respectively, compared to the highest
concentration. However, since it is now clear that riboflavin and FAD concentrations are im-
portant determinants of MTHFR stability,84,85 it is feasible that high riboflavin concentrations
in the culture media may have obscured any differences between genotypes.
More research is needed to elucidate how MTHFR polymorphisms affect thymidine syn-
thesis and methylation reactions over a range of folate, methionine and riboflavin concentra-
tions. Coming to a better appreciation of these gene-nutrient interactions is critical for our
understanding of how folate metabolism affects DNA integrity.
Conclusions
To date, the majority of the epidemiological evidence supports a moderate protective role
for the 677TT genotype against CR neoplasia, but only when folate and methionine (and
possibly vitamins B6 and B12) availabilities are adequate and alcohol intake is low. Given that
riboflavin interacts with MTHFR genotype to influence remethylation of homocysteine to
methionine, it is of interest to determine whether this vitamin might also play a role in deter-
mining CR cancer risk.
b
Micronuclei are small nuclear bodies that contain chromosome fragments or whole chromosome that lag
behind at anaphase of mitosis. Therefore the scoring of micronuclei is commonly used as a measure of
chromosome damage.
The most recent epidemiologic data suggest that the homozygous state for the MTHFR
677C→T polymorphism not only loses its association with protection against CR neoplasia as
folate status diminishes, but may actually become a significant risk factor. Interestingly, data
from Ulvik et al83 suggest that this effect is stronger in smokers than nonsmokers. Studies
examining the interrelationships between folate status, genotype, and DNA methylation pro-
vide a biological rationale for this switch in the ‘polarity of risk’. A better understanding of the
interdependence of MTHFR polymorphism and folate status in determining cancer risk will
enable us to successfully target those segments of the population who may benefit from addi-
tional folate intake. Just as important, however, is that the 677C→T mutation provides an
invaluable probe to unravel the mechanisms by which folate status modulates colorectal car-
cinogenesis.
Folate status is not the only factor which modulates the impact that the MTHFR polymor-
phism has on cancer risk: accumulating evidence indicates that vitamins B2, B6, B12 as well as
other polymorphisms in the one-carbon scheme may further modify the relationship. Further-
more, although this review has focused on colorectal cancer, it is likely that folate status and the
MTHFR polymorphism impact on cancer risk in other tissues, and there is much to be learned
as to why some tissues are much more sensitive to these factors than others.
Finally, a definitive understanding of how the 677C→T MTHFR polymorphism modu-
lates cancer risk not only requires a complete understanding of the molecular consequences of
the pertinent gene-nutrient interactions, but a better understanding of the pathway that leads
from anomalies in DNA methylation, uracil incorporation, and such, to the actual evolution
of cancer.
References
1. Mason J. Folate status: Effect on carcinogenesis. In: Bailey L, ed. Folate in health and disease.
Marcel Dekker, New York: 1995:361- 378.
128:1204-1212.
3. Blount BC, Mack MM, Wehr CM et al. Folate deficiency causes uracil misincorporation into
human DNA and chromosome breakage: Implications for cancer and neuronal damage. Proc Natl
Acad Sci 1997; 94:3290-3295.
4. Crott JW, Mashiyama ST, Ames BN et al. MTHFR C677T polymorphism does not alter folic
acid deficiency-induced uracil incorporation into primary human lymphocyte DNA in vitro. Car-
cinogenesis 2001; 22(7):1019-1025.
5. Duthie SJ, Hawdon A. DNA instability (strand breakage, uracil misincorporation, and defective
repair) is increased by folic acid depletion in human lymphocytes in vitro. FASEB J 1998;
12:1491-1497.
6. Crott JW, Mashiyama ST, Ames BN et al. The effect of folic acid deficiency and MTHFR C677T
polymorphism on chromosome damage in human lymphocytes in vitro. Cancer Epidemiol Biomark
Prev 2001; 10:1089-1096.
7. Fenech M, Aitken C, Rinaldi J. Folate, vitamin B12, homocysteine status and DNA damage in
young Australian adults. Carcinogenesis 1998; 19:1163-1171.
8. Holliday R. Causes of Aging. Annal N Y Acad.Sci 1998; 854:61-71.
9. Luongo CL, Contreras CM, Farsetta DL et al. Binding site for S-adenosyl-L-methionine in a cen-
tral region of mammalian reovirus lambda2 protein. Evidence for activities in mRNA cap methyla-
tion. J Biol Chem 1998; 273(37):23773-80.
10. Clarke S. Protein methylation. Curr Opin Cell Biol 1993; 5(6):977-83.
USA 1998; 95:13217-13220.
Hum Genet 1998; 62(5):1044-51.
Natl Acad Sci 2002; 99(8):5606-11.
15. Chen J, Giovannucci E, Kelsey K et al. A methylenetetrahydrofolate reductase polymorphism and
the risk of colorectal cancer. Cancer Res 1996; 56:4862-4864.
16. Mason JB, Choi SW. Folate and Carcinogenesis: Developing a unified hypothesis. Advan Enzyme
Regul 2000; 40:127-141.
17. Wagner C. Biochemical role of folate in cellular metabolism. In: Bailey L, ed. Folate in health and
disease. Marcel Deker, New York: 1995:23-42.
18. Laird PW, Jaenisch R. DNA methylation and cancer. Hum Mol Genet 1994; 3:1487-1495.
19. Robertson KD, Jones PA. DNA methylation: past present and future directions. Carcinogenesis
2000; 21(3):461-467.
20. Li E, Beard C, Jaenisch R. Role for DNA methylation in genomic imprinting. Nature 1993;
366(6453):362-5.
21. Pogribny I, James SJ. Reduction of p53 gene expression in human primary hepatocellular carci-
noma is associated with promoter region methylation without coding region mutation. Cancer
Letters 2002; 176:169-174.
22. Deng G, Peng E, Terdiman J et al. Methylation of hMLH1 promoter correlates with the gene
silencing with a region-specific manner in colorectal cancer. Br J Cancer 2002; 86:574-579.
23. Kang YH, Bae SI, Kim WH. Comprehensive analysis of promoter methylation and altered expres-
sion of hMLH1 in gastric cancer cell lines with microsatellite instability. J Cancer Res Clin Oncol
2002; 128(3):119-24.
24. Kouzarides T. Histone methylation in transcriptional control. Curr Opin Genet Dev 2002;
12(2):198-209.
25. Esfandiari F, Moor MJ, Paulson E et al. Folate deficiency induces hypomethylation of the
trimethylguanosine cap of small nuclear RNA in Chinese hamster ovary cells: Implications for
altered gene expression. Proc Am Assoc Cancer Res 1999; 40:548(A3616).
26. Wainfan E, Moller ML, Maschio FA et al. Ethionine-induced changes in rat liver transfer RNA
methylation. Cancer Res 1975; 35(10):2830-5.
27. Lindenbaum J, Allen RH. Clinical spectrum and diagnosis of folate deficiency. In: Bailey L, ed.
Folate in health and disease. Marcel Deker, New York: 1995:43-74.
28. Kang SS, Wong PW, Norusis M. Homocysteinemia due to folate deficiency. Metabolism 1987;
36:458-462.
29. EskesTKAB. From anaemia to spina bifida- the story of folic acid. A tribute to Professor Richard
Smithells. Euro J Obstet Gynecol Reprod Biol 2000; 90:119-123.
30. Clarke R, Smith AD, Jobst KA et al. Folate, vitamin B12, and serum total homocysteine levels in
confirmed Alzheimer’s disease. Arch Neurol 1998; 55:1449-1455.
Clin Nutr 1999; 70:495-501.
32. Giovannucci E, Rimm EB, Ascherio A et al. Alcohol, low-methionine—low-folate diets, and risk
of colon cancer in men. J Natl Cancer Inst 1995; 87(4):265-73.
33. KimY-I, Salomon R, Graeme-Cook F et al. Dietary folate protects against the development of
macroscopic colonic neoplasms in a dose responsive manner in the dimethylhydrazine rat model.
Gut 1996; 39:732-740.
34. James SJ, Basnakian AG, Miller BJ. In vitro folate deficiency induces deoxynucleotide pool imbal-
ance, apoptosis, and mutagenesis in Chinese hamster ovary cells. Cancer Res 1994; 54:5075-5080.
35. Baron JA, Sandler RS, Haile RW et al. Folate intake, alcohol consumption, cigarette smoking, and
risk of colorectal adenomas. J Natl Cancer Inst 1998; 90(1):57-62.
36. Miller JW, Nadeau MR, Smith J et al. Folate-deficiency-induced homocysteinaemia in rats: Dis-
ruption of S-adenosylmethionine’s coordinate regulation of homocysteine metabolism. Biochem J
1994; 298(2):415-419.
37. Choi S-W, Friso S, Dolnikowski GG et al. Biochemical and molecular aberrations in the colon
due to folate depletion are age specific. Submited to J Nutrition.
38. James SJ, Melnyk S, Pogribna M et al. Elevation in S-adenosylhomocysteine and DNA
hypomethylation: Potential epigenetic mechanism for homocysteine-related pathology. J Nutr 132(8
Suppl) 2002; 2361S-2366S.
40. Koury MJ, Horne DW, Brown ZA et al. Apoptosis of late-stage erythroblasts in megaloblastic
anemia: association with DNA damage and macrocyte production. Blood 1997; 89(12):4617-23.
41. Duthie SJ, Narayanan S, Blum S et al. Folate deficiency in vitro induces uracil misincorporation
and DNA hypomethylation and inhibits DNA excision repair in immortalized normal human co-
lon epithelial cells. Nutr Cancer 2000; 37(2):245-251.
42. Pogribny IP, Miller BJ, James SJ et al. Alterations in hepatic p53 gene methylation patterns during
tumor progression with folate/methyl deficiency in the rat. Cancer Lett 1997; 115:31-38.
43. Reidy JA. Folate- and deoxyuridine-sensitive chromatid breakage may result from DNA repair dur-
ing G2. Mutat Res 1987; 192:217-219.
44. Dianov GL, Timchenko TV, Sinitsina OI et al. Repair of uracil residues closely spaced on the
opposite strands of plasmid DNA results in double-strand break and deletion formation. Mol Gen
Genet 1991; 225(3):448-452.
45. Shen JC, Rideout 3rd WM, Jones PA. High frequency mutagenesis by a DNA methyltransferase.
Cell 1992; 71(7):1073-80.
46. Hagmar L, Brogger A, Hansteen IL et al. Cancer risk in humans predicted by increased levels of
chromosomal aberrations in lymphocytes: Nordic study group on the health risk of chromosome
damage. Cancer Res 1994; 54:2919-2922.
47. Hagmar L, Bonassi S, Stromberg U et al. Chromosomal aberrations in lymphocytes predict human
cancer: A report from the European Study Group on Cytogenetic Biomarkers and Health (ESCH).
Cancer Res 1998; 58:4117-4121.
48. Bonassi S, Hagmar L, Stromberg U et al. Chromosomal aberrations in lymphocytes predict human
cancer independently of exposure to carcinogens. European Study Group on Cytogenetic Biomarkers
and Health. Cancer Res 2000; 60(6):1619-1625.
49. Moynahan ME, Jasin M. Loss of heterozygosity induced by a double-strand break. Proc Natl Acad
Sci 1997; 94:8988-8993.
50. Elliot B, Jasin M. Human Genome and Diseases: Review. Double-strand beaks and translocations
in cancer. Cell Mol Life Sci 2002; 59:373-385.
51. Mitelman F. Recurrent chromosome aberrations in cancer. Mutat Res 2000; 462:247-253.
52. McClintock B. The fusion of broken ends of chromosomes following nuclear fusion. Proc Natl
Acad Sci 1942; 28:458-463.
53. Ma C, Martin S, Trask B et al. Sister chromatid fusion initiates amplification of the dihydrofolate
reductase gene in Chinese hamster cells. Genes Dev 1993; 7:605-620.
54. Biedler JL, Spengler BA. Metaphase chromosome anomaly: Association with drug resistance and
cell-specific products. Science 1976; 191:185-7.
55. Brison O. Gene amplification and tumor progression. Biochim Biophys Acta 1993; 1155:25-41.
56. Seeger RC, Brodeur GM, Sather H et al. Association of multiple copies of N-myc oncogene with
rapid progression of neuroblastomas. N Eng J Med 1985; 313(18):1111-1116.
57. Pogribny IP, Basnakian AG, Miller BJ et al. Breaks in genomic DNA and within the p53 gene are
associated with hypomethylation in livers of folate/methyl-deficient rats. Cancer Res 1995;
55:1894-1901.
58. KimYI, Pogribny IP, Basnakian AG et al. Folate deficiency in rats induces DNA strand breaks and
hypomethylation within the p53 tumor suppressor gene. Am J Clin Nutr 1997; 65:46-52.
59. Jones PA. DNA methylation errors and cancer. Cancer Res 1996; 56:2463-2467.
60. Cravo M, Fidalgo P, Pereira AD et al. DNA methylation as an intermediate biomarker in colorectal
cancer: Modulation by folic acid supplementation. Eur J Cancer Prev 1994; 3:473-479.
61. Fang JY, Xiao SD, Zhu SS et al. Relationship of plasma folic acid and status of DNA methylation
in human gastric cancer. J Gastroenterol 1997; 32:171-175.
62. Kim YI, Giuliano A, Hatch KD et al. Global DNA hypomethylation increases progressively in
cervical dysplasia and carcinoma. Cancer 1994; 74:893-899.
63. Guttenbach M, Schmid M. Exclusion of specific human chromosomes into micronuclei by
5-azacytidine treatment of lymphocyte cultures. Exp Cell Res 1994; 211:127-132.
64. Rasnick D, Duesberg PH. How aneuploidy affects metabolic control and causes cancer. Biochem J
1999; 340(3):621-630.
65. Jhaveri MS, Wagner C, Trepel JB. Impact of extracellular folate levels on global gene expression.
Molec Pharmacol 2001; 60(6):1288-1295.
66. Yoder JA, Walsh CP, Bestor TH. Cytosine methylation and the ecology of intragenomic parasites.
Trends Genet 1997; 13(8):335-340.
67. Homann N, Tillonen J, Salaspuro M. Microbially produced acetaldehyde from ethanol may in-
crease the risk of colon cancer via folate deficiency. Int J Cancer 2000; 86(2):169-73.
68. Halsted CH, Villanueva JA, Devlin AM et al. Folate deficiency disturbs hepatic methionine me-
tabolism and promotes liver injury in the ethanol-fed micropig. Proc Natl Acad Sci 2002;
99(15):10072-10077.
69. Choi SW, Stickel F, Baik HW et al. Chronic alcohol consumption induces genomic but not
p53-specific DNA hypomethylation in rat colon. J Nutr 1999; 129(11):1945-50.
71. Chen J, Ma J, Stampfer MJ et al. Linkage diequilibrium between the 677C>T and 1298A>C
polymorphisms in human methylenetetrahydrofolate reductase gene and their contributions to risk
of colorectal cancer. Pharmacogenetics 2002; 12:339-342.
72. Slattery ML, Potter JD, Samowitz W et al. Methylenetetrahydrofolate reductase, diet, and risk of
colon cancer. Cancer Epidemiol Biomarkers Prev 1999; 8(6):513-518.
73. Le Marchand L, Donlon T, Hankin JH et al. B-vitamin intake, metabolic genes, and colorectal
cancer risk (United States). Cancer Causes Control 2002; 3(3):239-48.
74. Shannon B, GnanaSampanthan S, Beilby J et al. A polymorphism in the methylenetetrahydrofolate
reductase gene predisposes to colorectal cancers with microsatellite instability. Gut 2002; 50:520-524.
75. Ryan BM, Molloy AM, McManus R et al. The methylenetetrahydrofolate reductase (MTHFR)
gene in colorectal cancer: Role in tumor development and significance of allelic loss in tumor
progression. Int J Gastro Cancer. In press.
76. Pereira P, StantonV, Jothy S et al. Loss of heterozygosity of methylenetetrahydrofolate reductase in
colon carcinomas. Oncol Rep 1999; 6(3):597-9.
77. Chen J, Giovannucci E, Hankinson SE et al. A prospective study of methylenetetrahydrofolate
reductase and methionine synthase gene polymorphisms, and risk of colorectal adenoma. Carcino-
genesis 1998; 19:2129-2132.
78. Ulrich CM, Kampman E, Bigler J et al. Colorectal adenomas and the C677T MTHFR polymor-
phism: evidence for gene-environment interaction? Cancer Epidemiol Biomarkers Prev 1999;
8(8):659-668.
79. Ulrich CM, Kampman E, Bigler J et al. Lack of association between the C677T MTHFR poly-
morphism and colorectal hyperplastic polyps. Cancer Epidemiol Biomarkers Prev 2000; 9(4):427-33.
80. Provenzale D, Garrett JW, Condon SE et al. Risk for colon adenomas in patients with rectosig-
moid hyperplastic polyps. Ann Intern Med 1990; 113(10):760-3.
81. Levine AJ, Siegmund KD, Ervin CM et al. The methylenetetrahydrofolate reductase 677C-->T
polymorphism and distal colorectal adenoma risk. Cancer Epidemiol Biomark Prev 2000; 9:657-663.
82. Marugame T, Tsuji E, Inoue H et al. Methylenetetrahydrofolate reductase polymorphism and risk
of colorectal adenomas. Cancer Lett 2000; 151(2):181-186.
83. Ulvik A, Evensen ET, Lien EA et al. Smoking, folate and methylenetetrahydrofolate reductase as
interactive determinants of adenamatous and hyperplastic polyps of colorectum. Am J Med Genet
2001; 101:246-254.
2000; 46(8.Pt.1):1065-1071.
85. Jacques PF, Kalmbach R, Bagley PM et al. The relationship between riboflavin and plasma total
Struct Biol 1999; 6(4):359-65.
87. Choi S-W, Bagley P, Mason JB. Individuals possessing the MTHFR mutant allele have a greater
percentage of formylated folates in their colonic mucosa than wild-type individuals. Colon Cancer:
Genetics to Prevention. Proceedings Am Assoc Cancer Res 2002; [abstract] B13.
88. Briske-Anderson M, Duerre JA. S-adenosylhomocysteine hydrolase from rat liver. Can J Biochem
1982; 60(2):118-23.
89. Hoffman DR, Marion DW, Cornatzer WE et al. S-Adenosylmethionine and S-adenosylhomocystein
metabolism in isolated rat liver. Effects of L-methionine, L-homocysteine, and adenosine. J Biol
Chem 1980; 255(22):10822-7.
90. Stern LL, Mason JB, Selhub J et al. Genomic DNA hypomethylation, a characteristic of most
cancers, is present in peripheral leukocytes of individuals who are homozygous for the C677T
polymorphism in the methylenetetrahydrofolate reductase gene. Cancer Epidemiol Biomarkers Prev
2000; 9(8):849-53.
Methylenetetrahydrofolate Reductase Polymorphisms: Pharmacogenetic Effects 197
CHAPTER 15
Polymorphisms:
Pharmacogenetic Effects
Bernd Christian Schwahn and Rima Rozen
Abstract
T
he MTHFR enzyme is not a primary target of drug therapy. However, the investigation
of possible pharmacogenetic effects of MTHFR polymorphisms is an emerging
field that is being explored for an increasing number of pharmaceutical compounds
and dietary supplements. This chapter reviews the current literature on the pharmacogenetic
effects of the MTHFR 677C→T polymorphism in response to various nutrients in homocys-
teine metabolism and in response to such medications as antifolates, anticonvulsants and some
neuropsychotropic drugs.
Unequivocal pharmacogenetic effects of the MTHFR 677 TT genotype have been demon-
strated for the response of plasma homocysteine to folate supplementation in humans. How-
ever, other effects might be too weak to be reliably detected, particularly in the absence of
large-scale studies. This situation is well reflected in the available data that often present con-
flicting results and that, in many cases, must be regarded as preliminary until additional studies
with larger sample sizes are performed. Nonetheless, this effort is warranted, particularly for
therapies with such common medications as the antifolates and the neuropsychotropic drugs.
Introduction
The human phenotype is determined by the genome of the individual, but is heavily influ-
enced by nongenetic interacting factors such as nutrition, behaviour, exposure to microorgan-
isms, and physicochemical environmental conditions. Phenotypic variability in the pathogen-
esis of disease or in the response to exogenous chemical compounds results, in part, from
interindividual variation in gene structure or gene expression. Treating patients of similar dis-
ease phenotype with a chemically defined substance often reveals wide interindividual variabil-
ity in efficacy and toxicity, i.e., in the pharmacokinetics of a compound and in the pharmaco-
dynamic response of the individual to medication. Drug concentrations and kinetics in various
body compartments are influenced by genetic variation in the gene products involved in drug
transport and metabolism. Furthermore, drug action may be altered by polymorphisms that
affect drug targets, either directly (e.g., by variant structure of a drug receptor) or indirectly
(e.g., by changing the biochemical conditions required for certain drug effects).
Though it is somewhat arbitrary to distinguish between pharmacogenetics and
pharmacogenomics, we follow a recent definition1 that uses the term pharmacogenetics to
describe “the differential effects of a drug in vivo in different patients, depending on the pres-
ence of inherited gene variants”, whereas pharmacogenomics applies to “differential effects of a
©2005 Eurekah.com.
number of compounds in vivo or in vitro on gene expression, among the entirety of expressed
genes”.1 Pharmacogenetics hence focuses on patient variability and pharmacogenomics on com-
pound variability.
The MTHFR enzyme is not a primary target of drug therapy. However, the exploration of
possible pharmacogenetic effects of MTHFR polymorphisms is a new field that is challenging
many investigators. MTHFR activity controls the distribution of one-carbon units between
nucleotide synthesis and the labile methyl pool. Polymorphisms in the MTHFR gene that lead
to altered MTHFR activity will change the individual metabolome and could therefore exert a
pharmacogenetic effect whenever supplements or drugs (e.g., folic acid, folinic acid,
methyltetrahydrofolate, or antimetabolites such as methotrexate or fluoropyrimidines) directly
interact with compounds in one-carbon metabolism. Indirect effects of MTHFR polymor-
phisms may be expected for other drugs whose metabolism is methylation-dependent, e.g.,
niacin, mercaptopurine or arsenic trioxide. One can imagine an even more indirect effect of
MTHFR polymorphisms, through alteration of gene-specific or general DNA methylation,
leading to differential gene expression and thereby to a different transcriptome and proteome,
which can influence drug response.
A number of drugs known to influence plasma homocysteine concentrations have also been
evaluated for a possible pharmacogenetic effect of MTHFR polymorphisms, although there are
no clear mechanistic hypotheses as to how they interact with one-carbon metabolism.
The functional consequences of the two main MTHFR genetic variants on the human
enzyme have been examined in a recent paper2 and the evidence for clinical and epidemiologi-
cal consequences is the subject of other chapters in this book. Two recent reviews have specifi-
cally addressed the pharmacogenetic aspects.3,4
This chapter reviews the increasing number of reports of pharmacogenetic effects of the
MTHFR 677C→T polymorphism and the few available reports on the functionally less sig-
nificant MTHFR 1298A→C polymorphism.
Pharmacogenetic Effects of MTHFR Variants

The pharmacogenetic effects of the 677C→T polymorphism in the MTHFR gene have
been investigated for direct effects on the response of plasma homocysteine levels to supple-
mentation with folate, cobalamin, riboflavin, and betaine. Other direct effects have been evalu-
ated for therapy with methotrexate, either low-dose for rheumatic disease or high-dose for
leukemia, for treatment of solid cancers with fluoropyrimidines and for the therapeutic use of
anticonvulsants in epileptic patients. Indirect effects have been investigated for anti-Parkinsonian
drugs, neuroleptic drugs, hormone replacement therapy, and lipid-lowering drugs. Table 1
provides an overview of compounds that have been explored for possible pharmacogenetic
effects of MTHFR polymorphisms. All the data except for those on folate supplementation
must still be regarded as preliminary since the number of studies for each agent is limited.
Direct Effects
MTHFR Polymorphism and Response to Folate Supplementation
The lower circulating total folate concentration in blood of homozygotes for the MTHFR
677T allele, due to decreased methyltetrahydrofolate synthesis, provides a rationale for the
evaluation of a pharmacogenetic effect of the MTHFR 677 variant on supplementation with
folate. Plasma total homocysteine (tHcy) and folate concentrations serve as endpoints.
Several studies have investigated the responsiveness of tHcy to high intake of dietary folate
or to supplementation with folic acid. Some recent studies elegantly summarize and extend the
findings of earlier investigators.
A recent cross-sectional population-based study of 2051 Dutch5 found significantly lower
serum folate associated with the 677CT and 677TT genotypes among individuals with
self-reported low folate intake. The 677TT genotype even conferred lower serum folate on
Table 1. Overview of compounds that have been explored for possible

pharmacogenetic effects of MTHFR polymorphisms
Folate5-8
Cobalamin 11,12
Riboflavin14-18
Betaine22,23
Methotrexate24-30
Fluoropyrimidines33,34
Raltitrexed35
Carbamazepine36,37
Phenytoin36
Valproic acid36,37
Levodopa39
Typical neuroleptics40
Estrogen44,45
Cholestyramine46
Fenofibrate47
Bezafibrate47
individuals with high folate intake compared to other genotype groups. As expected, individu-
als with the 677TT genotype had elevated tHcy only when folate status was low.
Three dietary intervention studies with randomized controlled design studied the influence
of the MTHFR polymorphism on folate effects measured as changes in tHcy.6-8 More than 40
healthy individuals of each genotype completed 3 dietary interventions, each of 4 months
duration, with a low-folate diet, a folate-rich diet, and a low-folate diet supplemented with
0.4mg folic acid. Individuals with the 677TT genotype had consistently higher tHcy and
lower folate than those with the 677CC genotype throughout the study.6 Women carrying the
677TT genotype had a more pronounced drop in tHcy and a bigger increase in serum folate
concentrations than those with the 677CC or 677CT genotype after switching from a low-folate
to a high-folate diet8 or after supplementation with either folic acid or methyltetrahydrofolate.7
Both latter studies showed a higher requirement for folate in subjects with the 677TT geno-
type, presumably due to decreased formation of methyltetrahydrofolate, and provided a clear
pharmacogenetic effect of the MTHFR 677 polymorphism on the response to folate supple-
mentation.
MTHFR Polymorphism and Response to Cobalamin Supplementation

Folate and cobalamin metabolism are linked through the vitamin B 12-dependent
remethylation of homocysteine to methionine by methionine synthase; 5-methyltetrahydrofolate,
generated by MTHFR, is the methyl donor for this reaction.
Two relatively small studies have reported the association of low cobalamin with
hyperhomocysteinemia in homozygotes for the MTHFR 677T allele9,10 and one study in ho-
mozygotes for the MTHFR 1298C allele,10 compared to the MTHFR wild type genotype. The
pathophysiological basis of this interaction, however, is not clear. In a small cohort of
hyperhomocysteinemic patients undergoing hemodialysis, no significant effect of cobalamin
supplementation on tHcy was observed in any of the MTHFR 677 genotype groups.11 In
contrast, in a similar study, patients responded to cobalamin supplementation and the decrease
of tHcy was most pronounced in those with the MTHFR 677TT genotype.12
MTHFR Polymorphism and Response to Riboflavin Supplementation

Riboflavin is the precursor of flavin adenine dinucleotide (FAD), the cofactor of MTHFR
which binds and stabilizes the enzyme. The enzyme product of the MTHFR 677T allele loses
its FAD cofactor faster than the wild type enzyme,2 and FAD has been shown to stabilize both
the wild type and mutant enzymes.13 Low cytosolic FAD may therefore contribute to the
functional impairment of the thermolabile MTHFR 677T variant and increase the tHcy eleva-
tion, particularly in individuals with low-folate status. An effect of riboflavin supplementation
would be expected in the latter group. Four studies have reported an association of low ribofla-
vin status with elevated tHcy,14-17 and three of them have confirmed the interaction between
the MTHFR 677TT genotype and riboflavin status.15-17 One recent preliminary report de-
scribes a statistically significant homocysteine-lowering effect of riboflavin supplementation in
individuals with the MTHFR 677TT genotype,18 suggesting higher riboflavin requirements in
this group.
MTHFR Polymorphism and Response to Betaine Supplementation

Folate-dependent remethylation of homocysteine to methionine is observed in all tissues
but an alternate remethylation pathway in liver and kidney is carried out by betaine homocys-
teine methyltransferase (BHMT), which utilizes betaine as the methyl donor. When
folate-dependent remethylation is disrupted, the alternate pathway may be particularly impor-
tant. Large doses of oral betaine are the mainstay of therapy for patients with homocystinuria
due to severely compromised MTHFR activity, although betaine can also lower homocysteine
in healthy subjects.19 The betaine effect can be monitored by measurement of plasma ho-
mocysteine concentrations. Using serial measurements of plasma homocysteine, betaine, and
dimethylglycine (the product of the BHMT reaction), it is possible to describe the pharmaco-
kinetics of betaine20 and to model its pharmacodynamic effects.21,22
The elimination rate of betaine and the homocysteine-lowering effect are increased in pa-
tients with severe MTHFR deficiency,22 but there are no studies thus far on the impact of
MTHFR polymorphisms on betaine administration. However, corresponding data have been
generated in a mouse model for mild and severe MTHFR deficiency. The relative decrease of
homocysteine (approximately 50%) with a betaine supplement was not influenced by MTHFR
activity, but the absolute decrease in homocysteine was higher in mice with mild MTHFR
deficiency than that in wild-type mice.23 Humans with the MTHFR 677TT genotype might
therefore be expected to show a greater homocysteine-lowering response to betaine than those
with the other MTHFR genotypes.
MTHFR Polymorphism and Antifolates

The antifolate drugs methotrexate (MTX), pernetrexed, edatrexate, and lometrexol are struc-
tural folate analogues that target different enzymes of the folate cycle and/or purine or pyrimi-
dine synthesis. Pharmacogenetic effects of the MTHFR polymorphisms in antitumor therapy
with these substances are of utmost importance, given their widespread use, efficacy and unex-
pected deleterious toxicity after standardized dosing. Altered antiproliferative efficacy and tox-
icity might be observed in those with the 677TT genotype due to a relative preponderance of
oxidized folates that are available for nucleotide synthesis, while other toxic effects, such as
neurotoxicity, could occur due to the hyperhomocysteinemia and disruption of transmethyla-
tion reactions. In contrast to the potential clinical significance of pharmacogenetic effects, only
a few reports have been published thus far.
The first case report observed a high prevalence of the MTHFR 677TT genotype (5 out of
6) in patients who were treated for breast cancer with a combination of cyclophosphamide,
methotrexate and 5-fluorouracil and who experienced major antiproliferative toxicity.24 One
retrospective study addressed the pharmacogenetic effect of the MTHFR 677 polymorphism
on acute MTX toxicity in 220 patients with chronic myeloid leukemia undergoing post bone
marrow transplant methotrexate prophylaxis for graft-versus-host disease.25 A borderline sig-
nificant higher degree of mucositis and a trend towards slower recovery from thrombocytope-
nia was found in those with the 677TT genotype. However, the MTHFR genotype of the bone
marrow graft was not available.25 Two groups reported increased antiproliferative26,27 and hepa-
totoxic27 effects associated with the 677TT genotype in retrospective studies on 43 patients
with ovarian cancer and 61 patients with acute leukemia undergoing low-dose maintenance
therapy with MTX.26,27
Low-dose MTX is effective in immunomodulation of rheumatic disease, probably through
release of adenosine, in combination with other anti-inflammatory drugs such as sulfasalazine.
Long-term MTX therapy in rheumatoid arthritis resulted in elevation of tHcy in 93 patients
with the MTHFR 677CC and CT genotypes to a final level similar to that of 10 patients with
the 677TT genotype, who remained at their elevated tHcy level throughout the study. No
difference in plasma folate or toxicity was observed between genotype groups.28 Another study
on 106 patients with rheumatoid arthritis and MTX therapy reported an association with
slightly increased efficacy in patients with the 1298C allele and slightly higher toxicity in those
with the 677T allele29 while a third similar study showed no pharmacogenetic association of
the MTHFR genotype with MTX in 93 patients.30 The latter two studies, however, did not
take into account the folate status of their study subjects.
The widely-used thymidylate synthase (TS) inhibitors such as 5-fluorouracil (FU),
capecitabine, raltitrexed, nolatrexed, and tegafur interact with folate metabolism in so far as the
substrate of MTHFR, 5,10-methylenetetrahydrofolate, is also a substrate for TS and binds
together with dUMP, or with the competitive inhibitor 5-fluorouracil, to the enzyme.31 The
MTHFR 677TT variant confers a higher availability of 5,10-methylenetetrahydrofolate32 and
should therefore increase the efficacy of FU, similarly to the therapeutic coadministration of
leucovorin.
The first study did not observe increased efficacy or toxicity in 4 patients with the MTHFR
677TT genotype among 47 patients with colon cancer treated with FU and leucovorin.33 A
second study evaluated 43 patients with advanced colorectal cancer and different TS-inhibitor
treatment regimens (38 with FU and leucovorin). The MTHFR 677T allele conferred a signifi-
cantly higher response rate to therapy without evidence of higher antiproliferative or hepatic
toxicity.34 One other study investigated the combined use of irinotecan, an inhibitor of
topoisomerase 1, and raltitrexed in 39 patients with different solid cancers and found signifi-
cantly reduced raltitrexed-associated toxicity in homozygotes for the MTHFR 677T allele.35
All three association studies were limited due to their small sample size and, although some
results suggest a pharmacogenetic influence, additional data are required to interpret the effect
of MTHFR polymorphisms on the response to TS inhibitors.
In summary, the data are inconclusive thus far and do not definitively answer the question
as to whether patients with the MTHFR 677TT genotype have an altered response when ex-
posed to antifolate therapy.
MTHFR Polymorphism and Anticonvulsant Drugs

A number of commonly used anticonvulsant drugs have been shown to interfere with folate
metabolism, possibly due to some interaction with the MTHFR enzyme. For a review see ref.
3. In a study of 103 patients with epilepsy, only those who were homozygous for the MTHFR
677T allele and who were receiving phenytoin or carbamazepine, but not those receiving valproic
acid, had significantly higher tHcy and lower plasma folate, compared with matched healthy
individuals with the same genotype; the few patients with the homozygous 677TT genotype
receiving concomitant folate supplements had normal plasma homocysteine.36 In another study
of 136 patients with childhood epilepsy, the MTHFR genotype influenced plasma folate and
tHcy in patients receiving carbamazepine, but not in those with valproic acid therapy.37 A third
study in 81 epileptic patients showed that hyperhomocysteinemia and folate deficiency oc-
curred more often in individuals with the 677TT genotype than in those with other genotypes,
but this was only true in the subgroup receiving multidrug therapy and not in patients with
monotherapy.38 These few studies hint to a possible pharmacogenetic effect of MTHFR in
patients treated with carbamazepine and phenytoin.
Indirect Effects
MTHFR Polymorphism and Neuropsychotropic Drugs
Levodopa, the most common drug in the treatment of Parkinson’s disease, is a potent me-
thyl acceptor and has been shown to be associated with depletion of labile methyl groups, a
condition that might be aggravated in individuals with the MTHFR 677TT genotype. In a
study of patients with Parkinson’s disease receiving combined therapy with levodopa, carbidopa,
and other anti-Parkinsonian drugs, all patients had elevated tHcy compared with healthy con-
trols, but this elevation was only significant for those with the 677 TT genotype.39
Patients with severe MTHFR deficiency can show psychotic symptoms and the MTHFR
677TT genotype might be associated with risk for schizophrenia (for review see ref. 3). Based
on these findings and on the observation of interindividual heterogeneity in the response of
patients to neuroleptic treatment, Joober et al40 studied the association of the MTHFR poly-
morphism in patients stratified by treatment outcome with conventional neuroleptics. They
identified a clear over-representation of the MTHFR 677T allele in patients with schizophrenia
that responded to therapy with conventional neuroleptics compared to the frequency in con-
trols. Nonresponders did not differ from controls in their allele frequencies. The authors sug-
gested that the MTHFR 677TT genotype might confer a more rapid and sustained response to
neuroleptics. However, these findings remain to be confirmed by other groups.
Gene expression studies of multiple genes including the MTHFR 677 locus did not reveal a
significant effect of MTHFR on the response to treatment for Alzheimer’s disease41 or demen-
tia.42 Considering the importance of the folate cycle and methylation reactions for the metabo-
lism of biogenic amines that function as neurotransmitters, for cell proliferation, for DNA
methylation, and for phospholipid/membrane biosynthesis, it is likely that MTHFR variation
may modulate the response to many other pharmaceutical compounds that affect brain func-
tion.
MTHFR Polymorphisms and Hormone Replacement Therapy

Premenopausal women generally have lower fasting tHcy than men. These values rise after
menopause to the level of those of men of similar age, but can be reduced by estrogen replace-
ment therapy.43 One report of 90 women with adequate folate status suggested that the
homocysteine-lowering effect of hormone replacement therapy was less pronounced in women
who were homozygous for the 677T allele.44 This result could not be reproduced in a similar
study of 217 Japanese women, in whom the MTHFR 677 genotype did not impair the re-
sponse of tHcy to hormone replacement therapy.45
MTHFR Polymorphisms and Lipid-Lowering Drugs

Two classes of lipid-lowering drugs have been shown to interfere with homocysteine me-
tabolism by unknown mechanisms. In one study, children with hypercholesterolemia who
were heterozygous or homozygous for the MTHFR 677T allele had an increase in plasma
homocysteine levels upon treatment with cholestyramine.46 Treatment with fenofibrate or
bezafibrate, drugs that lower triglyceride levels and increase HDL levels in certain dyslipidemias,
may increase fasting and postprandial tHcy, but MTHFR genotypes did not modify the
fenofibrate-induced homocysteine changes in one study of men with cardiovascular disease
and hypertriglyceridemia.47 The sample size in that study, however, was quite small.
Conclusions
Given the complex interaction of any drug with many body systems and a presumed mean
prevalence of more than five amino-acid-changing polymorphisms per coding sequence in the
human genome, a pharmacogenetic effect of a single genetic polymorphism must either be
very strong or the polymorphism must have a high prevalence to generate reproducible results
in association studies (see review in ref. 48). The MTHFR 677C→T polymorphism appears to
fulfill both criteria when applied to direct pharmacogenetic effects. However, indirect effects
might be too weak to be reliably detected without large-scale studies.
Although the pharmacogenetic data on the MTHFR 677 variant are somewhat preliminary
for many of the compounds examined thus far, additional efforts in this field are clearly war-
ranted considering the widespread use of such common medications as antifolates and neuro-
psychotropic drugs. Furthermore, given the fact that folate-dependent one-carbon transfer re-
actions and many pharmaceutical reagents can influence such fundamental cellular processes as
DNA replication, neurotransmitter function and membrane biology, it is likely that a moder-
ate disruption in folate metabolism could influence response to a wide variety of medications.
References
1. Lindpaintner K. Pharmacogenetics and the future of medical practice. J Mol Med 2003; 81:141-153.
binant human methylenetetrahydrofolate reductase. Proc Natl Acad Sci 2001; 98:14853-14858.
3. Schwahn B, Rozen R. Polymorphisms in the Methylenetetrahydrofolate reductase gene – Clinical
consequences. Am J Pharmacogenomics 2001; 1:189-201.
4. Ulrich CM, Robien K, Sparks R. Pharmacogenetics and folate metabolism – a promising direction.
Pharmacogenomics 2002; 3:299-313.
5. De Bree A, Verschuren WMM, Bjorke-Monsen A-L et al. Effect of the methylenetetrahydrofolate
reductase 677C-->T mutation on the relations among folate intake and plasma folate and ho-
mocysteine concentrations in a general population sample. Am J Clin Nutr 2003; 77:687-693.
6. Ashfield-Watt PA, Pullin CH, Whiting JM et al. Methylenetetrahydrofolate reductase 677 C-->T
genotype modulates homocysteine responses to a folate-rich diet or a low-dose folic acid supple-
ment: A randomized controlled trial. Am J Clin Nutr 2002; 76:180-186.
7. Fohr IP, Prinz-Langenohl R, Brönstrup A et al. 5,10-Methylenetetrahydrofolate reductase genotype
determines the plasma homocysteine-lowering effect of supplementation with 5-methyltetrahydrofolate
or folic acid in healthy young women. Am J Clin Nutr 2002; 75:275-282.
8. Silaste M-L, Rantala M, Sämpi M et al. Polymorphisms of key enzymes in homocysteine metabo-
lism affect diet responsiveness of plasma homocysteine in healthy women. J Nutr 2001;
131:2643-2647.
9. D’Angelo A, Coppola A, Madonna P et al. The role of vitamin B-12 in fasting hyperhomocysteinemia
Haemost 2000; 83:563-570.
10. Bailey LB, Duhaney RL, Maneval DR et al. Vitamin B-12 status is inversely associated with plasma
homocysteine in young women with C677T and/or A1298C methylenetetrahydrofolate reductase
polymorphisms. J Nutr 2002; 132:1872-1878.
11. Billion S, Tribout B, Cadet E et al. Hyperhomocysteinemia, folate and vitamin B12 in
unsupplemented haemodialysis patients: Effect of oral therapy with folic acid and vitamin B12.
Nephrol Dial Transplant 2002; 17:455-461.
12. Hyndman ME, Manns BJ, Snyder FF et al. Vitamin B12 decreases, but does not normalize, ho-
mocysteine and methylmalonic acid in end-stage renal disease: A link with glycine metabolism and
possible explanation of hyperhomocysteinemia in end-stage renal disease. Metabolism 2003;
52:168-172.
14. Moat SJ, Ashfield-Watt PA, Powers HJ et al. Effect of riboflavin status on the homocysteine-lowering
effect of folate in relation to the MTHFR (C677T) genotype. Clin Chem 2003; 49:295-302.
16. McNulty H, McKinley MC, Wilson B et al. Impaired function of thermolabile methylenetetrahydrofolate
Nutr 2002; 76:436-441.
2000; 46:1065-1071.
18. McNulty H, Dowey LC, Scott JM et al. Riboflavin supplementation lowers plasma homocysteine
in individuals homozygous for the MTHFR C677T polymorphism. J Inherit Metab Dis 2003;
26(Suppl 1):12.
19. Brouwer IA, Verhoef P, Urgert R. Betaine supplementation and plasma homocysteine in healthy
volunteers. Arch Intern Med 2000; 160:2546-2547.
20. Schwahn BC, Hafner D, Hohlfeld T et al. Pharmacokinetics of oral betaine in healthy subjects
and patients with homocystinuria. Br J Clin Pharmacol 2003; 55:6-13.
21. Matthews A, Johnson TN, Rostami-Hodjegan A et al. An indirect response model of homocysteine
suppression by betaine: Optimising the dosage regimen of betaine in homocystinuria. Br J Clin
Pharmacol 2002; 54:140-6.
22. Balkenhol ND, Laryea MD, Hafner D et al. Pharmacokinetic-pharmacodynamic modeling of oral
betaine in patients with homocystinuria. J Inherit Metab Dis 2003; 26(Suppl 1):19.
23. Schwahn BC, Chen Z, Laryea MD et al. Homocysteine - Betaine Interactions in a murine model
of 5,10-Methylenetetrahydrofolate reductase deficiency. FASEB J 2003; 17:512-514.
24. Toffoli G, Vernosi A, Boiocchi M et al. MTHFR gene polymorphism and severe toxicity during
adjuvant treatment of early breast cancer with cyclophosphamide, methotrexate, and fluorouracil
(CMF). Ann Oncol 2000; 11:373-374.
25. Ulrich CM, Yasui Y, Storb R et al. Pharmacogenetics of methotrexate: Toxicity among marrow
transplantation patients varies with the methylenetetrahydrofolate reductase polymorphism. Blood
2001; 98:231-234.
26. Toffoli G, Russo A, Innocenti F et al. Effect of methylenetetrahydrofolate reductase 677C→T
polymorphism on toxicity and homocysteine plasma level after chronic methotrexate treatment of
ovarian cancer patients. Int J Cancer 2003; 103:294-299.
27. Chiusolo P, Reddiconto G, Casorelli I et al. Preponderance of methylenetetrahydrofolate reductase
C677T homozygosity among leukemia patients intolerant to methotrexate. Ann Oncol 2002;
13:1915-1918.
28. Haagsma CJ, Blom HJ, van Riel PLCM et al. Influence of sulphasalazine, methotrexate, and the
combination of both on plasma homocysteine concentrations in patients with rheumatoid arthritis.
Ann Rheum Dis 1999; 58:79-84.
29. Urano W, Taniguchi A, Yamanaka H et al. Polymorphisms in the methylenetetrahydrofolate re-
ductase gene were associated with both the efficacy and toxicity of methotrexate used for the treat-
ment of rheumatoid arthritis, as evidenced by single locus and haplotype analyses. Pharmacogenet-
ics 2002; 12:183-190.
30. Kumagai K, Hiyama K, Oyama T et al. Polymorphisms in the thymidylate synthase and
methylenetetrahydrofolate reductase genes and sensitivity to the low-dose methotrexate therapy in
patients with rheumatoid arthritis. Int J Mol Med 2003; 11:593-600.
31. Zhang ZG, Rustum YM. Pharmacologic rationale for fluoropyrimidine-leucovorin combination:
Biochemical mechanisms. Semin Oncol 1992; 19:46-50.
32. Bagley PJ, Selhub JA. Common mutation in the methylenetetrahydrofolate reductase gene is asso-
USA 1998; 95:13217-13220.
33. Wisotzkey JD, Toman J, Bell T et al. MTHFR (C677T) polymorphism and stage III colon cancer:
Response to therapy. Mol Diagn 1999; 4:95-99.
34. Cohen V, Panet-Raymond V, Sabbaghian N et al. Methylenetetrahydrofolate reductase polymor-
phism in advanced colorectal cancer: A novel genomic predictor of clinical response to
fluoropyrimidine-based chemotherapy. Clin Cancer Res 2003; 9:1611-1615.
35. Stevenson JP, Redlinger M, Kluijtmans LA et al. Phase I clinical and pharmacogenetic trial of
irinotecan and raltitrexed administered every 21 days to patients with cancer. J Clin Oncol 2001;
19:4081-4087.
36. Yoo J-H, Hong SB. A common mutation in the methylenetetrahydrofolate reductase gene is a
determinant of hyperhomocysteinemia in epileptic patients receiving anticonvulsants. Metabolism
1999; 8:1047-1051.
37. Vilaseca MA, Monros E, Artuch R et al. Anti-epileptic drug treatment in children: Hyperhomocysteinemia,
B-vitamins and the 677C-->T mutation of the methylenetetrahydrofolate reductase gene. Europ J
Paediatr Neurol 2000; 4:269-277.
38. Ono H, Sakamoto A, Mizoguchi N et al. The C677T mutation in the methylenetetrahydrofolate
reductase gene contributes to hyperhomocysteinemia in patients taking anticonvulsants. Brain Dev
2002; 24:223-226.
39. Yasui K, Kowa H, Nakaso K. Plasma homocysteine and MTHFR C677T genotype in
levodopa-treated patients with PD. Neurology 2000; 55:437-440.
40. Joober R, Benkelfat C, Lal S et al. Association between the methylenetetrahydrofolate reductase
677C-->T missense mutation and schizophrenia. Mol Psychiatry 2000; 5:323-326.
41. Cacabelos R. Pharmacogenomics in Alzheimer’s disease. Mini Rev Med Chem 2002; 2:59-84
42. Cacabelos R. Pharmacogenomics for the treatment of dementia. Ann Med 2002; 43:357-379.
43. Mijatovic V, Kenemans P, Jakobs C et al. A randomized controlled study of the effects of
17beta-estradiol-dydrogesterone on plasma homocysteine in postmenopausal women. Obstet Gynecol
1998; 91:432-436.
44. Brown CA, McKinney KQ, Young KB et al. The C677T methylenetetrahydrofolate reductase poly-
morphism influences the homocysteine-lowering effect of hormone replacement therapy. Mol Genet
Metab 1999; 67:43-48.
45. Somekawa Y, Kobayashi K, Tomura S et al. Effects of hormone replacement therapy and
methylenetetrahydrofolate reductase polymorphism on plasma folate and homocysteine levels in
postmenopausal Japanese women. Fertil Steril 2002; 77:481-486.
46. Tonstad S, Refsum H, Ose L et al. The C677T mutation in the methylenetetrahydrofolate reduc-
tase gene predisposes to hyperhomocysteinemia in children with familial hypercholesterolemia treated
with cholestyramine. J Pediatr 1998; 132:365-368.
47. Bissonnette R, Treacy E, Rozen R et al. Fenofibrate raises plasma homocysteine levels in the fasted
and fed states. Atherosclerosis 2001; 155:455-462.
48. Ryan SG. Regression to the truth: Replication of association in pharmacogenetic studies.
Pharmacogenomics 2003; 4:201-207.
Index
Symbols B
1298A→C 14, 15, 21, 23-25, 31, 38, 41, 43, bcl-2 180
44, 54, 55, 66, 119-121, 125-127, 129, Betaine 17, 41-43, 45, 46, 48-50, 55, 81, 88,
135, 136, 138-140, 147, 152, 155, 155, 198, 200
170-175, 179, 180, 184, 187, 188, 198 Betaine-homocysteine methyltransferase
677C>T 103, 104 (BHMT) 55, 81, 200
677C→T 1, 2, 12-16, 21, 23-26, 31, 33, 35, Bezafibrate 198, 202
38, 39, 41, 43, 44, 49, 54, 55-57, 59-62, Birth defect 86, 92, 126, 129, 138-140
65, 66, 71, 73-76, 80, 89, 100, 101, Brain imaging 42, 166
103-106, 108, 113, 116-120, 125-135,
137-140, 144, 145, 147, 149, 150, 152, C
155-157, 163-167, 170-175, 179, 180,
184, 187, 188, 190-193, 197, 198, 203 c-fos 180
c-myc 180
A CAMP-responsive element (CRE) 80
Cancer 76, 83, 88-90, 92, 126, 139, 179,
Abruptio placentae 145, 150, 152 180, 182-184, 187, 188, 190-193, 198,
Adenosylhomocysteine (AdoHcy) 13, 17, 32, 200, 201
38, 39, 78-82, 85-87, 91, 94, 97, 99, Cardiac malformation 125, 135, 137
180, 182, 187, 191, 192 Cardiovascular disease (CVD) 1, 12, 13, 17,
Adenosylmethionine (AdoMet) 13, 17, 31-34, 41, 56-58, 66, 76, 100, 101, 105, 126,
38, 39, 41-43, 48, 50, 66, 71-73, 78-83, 139, 144, 170, 171, 173, 174, 202
86, 91, 167, 180, 182, 187, 191, 192 Carnitine 48
Alcohol 106, 107, 126, 139, 184, 187, 188, CC genotype 35, 55, 56, 60, 65, 73, 80, 100,
192 101, 103-106, 108, 113, 120, 125-127,
Allele frequency 15, 21, 145, 150, 187, 188, 135, 140, 175, 190, 199
202 CH3-H4folate 31, 32, 36, 38
Allele-specific PCR assay 21, 24, 27 CH3-H4folate-menadione oxidoreductase assay
Alzheimer’s disease (AD) 86, 87, 166, 167, 38
182, 202 Choline 88, 91, 184
Anticonvulsant 197, 198, 201 Chromatin 78, 82-87, 89
Anticonvulsant drug 201 Chromosomal abnormalities 144, 145, 147
Antidepressant 164 Cleft lip 127, 132, 139
Antifolate 90, 197, 200, 201, 203 Cleft palate (CLP) 127, 129, 132, 138, 139,
Apoptosis 83, 86, 90, 91, 155 150
Arabidopsis 85 Cobalamin 32, 43, 44, 48, 71-74, 198, 199
Arterial occlusive disease (AOD) 100, 101, Colorectal adenoma 188
103, 108 Colorectal cancer 139, 183, 184, 187, 193,
Arterial thrombosis 42 201
Ataxia 42, 43, 47, 87 Congenital anomalies/malformation 125, 138,
Australian National Heart Foundation Perth 140, 144, 145, 147
Risk Factor 56 Coronary artery/heart disease (CAD/CHD)
12, 33, 57, 58, 62-64, 101, 103-108,
119, 120, 171, 174
CT genotype 33, 54, 56, 61, 72, 101, 103, Folate 1, 7, 12, 13, 17, 32, 33, 35, 36, 39, 41,
117, 120, 125, 126, 138, 139, 147, 174, 43-46, 48, 50, 54-66, 71-76, 78-82, 85,
175, 188, 199, 201 86, 88-92, 100, 101, 103-105, 108, 113,
Cystathionine β-synthase (CBS) 55, 72, 81, 119, 121, 126, 138, 139, 144, 150,
82, 144 163-167, 170-175, 179, 180, 182-184,
187, 188, 190-193, 197-203
D Folate receptor 86
Folate/methyl deficiency 86, 89, 90, 92
Dementia 47, 87, 163, 166, 202 Folbp1 86
Demyelination 42, 43, 45 Fragile X 87, 88
Depression 163, 164
Developmental delay 7, 33, 41, 42, 45, 46, G
47, 54
DNA hypomethylation 17, 79, 80, 82, 86-92, GenBank 2-5, 15, 16
145, 182-184 Gene structure 3, 16, 17, 197
DNA methylation 13, 17, 78-80, 82-92, 179, Genotyping 21, 24-28, 166
180, 183, 184, 191-193, 198, 202 Global hypomethylation 86, 90, 92, 188
DNA methyltransferase 79-83, 85, 87, 90-92, Glomerular filtration rate (GFR) 170, 171
182 Glutathione reductase 73
DNA polymerase 25, 27, 80, 88
DNA repair 90-92, 180 H
DNA strand break 80, 89, 91, 92, 187
DNA synthesis 13, 14, 17, 78-80, 82, 92, Ha-ras 180
180, 187 Haemolysis, elevated liver enzymes and low
Dnmt protein 84 platele (HELLP) 152, 155
Down syndrome 88, 89, 125, 127, 134, 135, Heteroduplex 21, 24, 25, 27
138, 139, 144, 145, 182 Heteroduplex assay 21, 24, 25, 27
Heteroduplex generator (HDG) 24, 25
E Heterogeneity 49, 103-106, 108, 147, 148,
150, 152, 163, 164, 202
5´-exonuclease 25 HinfI 13, 21, 23
Edatrexate 200 Histone 82, 84-87, 180
Electroencephalogram (EEG) 42 Histone deacetylase (HDAC) 84, 87
Embryonic development 83, 85, 86 Histone methylation 84-86
End-stage renal disease (ESRD) 75, 170, 171 Homocysteine 1, 2, 13, 15, 17, 31-33, 35, 36,
Epigenetic effect 182, 183 39, 41, 42, 44, 48-50, 54-60, 66, 71-75,
Epilepsy 46, 201 79-82, 86, 87, 89, 100, 101, 103-105,
Erb-A1 180 108, 113, 119, 120, 126, 139, 144, 147,
Escherichia coli 13, 34-38, 59, 72, 80 150, 152, 155, 156, 163-167, 170-175,
179, 180, 182, 184, 187, 190-192,
F 197-202
Homocysteine studies collaboration 105
Factor V Leiden 43, 117-120, 147, 155 Homocystinuria 1, 2, 7, 12, 17, 33, 44, 54,
Fenofibrate 198, 202 144, 155, 200
Fetal growth retardation 145, 150 Homogenous assay 25, 27, 28
Fibroblast 7, 8, 44, 48, 49, 65, 83 Hormone replacement therapy 198, 202
Flavin adenine dinucleotide (FAD) 10, 13, Hybridization 2, 4, 25-27
31-33, 35-38, 59, 71-73, 80, 108, 180, Hybridization probe assay 26, 27
190, 192, 199, 200 Hyperhomocysteinemia 1, 2, 12, 13, 17, 33,
Flavin mononucleotide 71 39, 44, 54, 55, 66, 75, 76, 82, 113, 119,
Fluorescence in situ hybridization (FISH) 4 144, 145, 147, 150, 152, 155, 156, 163,
FMN 71-73 165, 167, 170-175, 182, 199-201
Fnu4HI 15, 23
Index 209
Hypermethylation 83, 86, 90-92, 180, 183, Methyltetrahydrofolate 1, 13, 31, 32, 36, 38,
184, 188 41, 43, 44, 48, 54, 55, 71, 72, 100, 113,
Hypertriglyceridemia 202 171, 174, 179, 198, 199
Hypomethioninemia 33 Microarray analysis 17
Hypothyroidism 73 Missense mutation 10, 21, 54
Molecular beacon assay 25
I Mouse model 1, 79, 200
MRI 42
ICF syndrome 87 MTHFR 1-7, 10-14, 16-18, 21, 22, 24, 25,
Ischemic heart disease (IHD) 103, 105, 174 31-39, 41-50, 54-57, 59-66, 71-76,
78-80, 82, 86, 88, 89, 92, 100, 101,
K 103-106, 108, 113, 115-121, 125-140,
144, 145, 147-150, 152, 155-157,
Ki-ras 180 163-167, 170-175, 179, 180, 183, 184,
Kidney transplant 170, 171, 173-175 187, 188, 190-193, 197-203
Mthfr 2, 3, 4, 5, 17
L MTHFR 677C→T polymorphism 57, 62,
74-76, 89, 100, 101, 103-106, 113,
L-dopa 163, 167 116-120, 144, 145, 147, 149, 150, 152,
L-homocysteine 32 155, 174, 175, 184, 188, 190, 192, 193,
Lactobacillus casei 48 197, 198, 203
Limb reduction defect 125, 138 MTHFR mRNA 4
Lipid 17, 198, 202 Mthfr mRNA 3, 4
Lometrexol 200 Mthfr pseudogene 5
Mthfr-deficient mice 2
M MTX 200, 201
Mutation 1-4, 7-16, 21, 24, 31, 33, 35-39,
5,10-methylenetetrahydrofolate reductase 15, 41, 43, 45-47, 49, 50, 54-56, 59-61, 65,
41, 126, 171, 179 66, 83, 85, 87, 88, 90, 113, 116-120,
MboII 15, 23 126, 144, 145, 147, 150, 152, 155-157,
MeCP2 84, 87 163-167, 170-172, 179, 180, 188, 193
Mental retardation 42, 45-47, 87, 88
metF 2, 34, 35 N
Methionine 1, 13, 16, 31, 32, 39, 41-44,
48-50, 54, 55, 66, 71-73, 78-81, 86, 88, Neural tube defect (NTD) 13, 41, 62-64, 76,
89, 91, 100, 113, 126, 139, 145, 147, 86, 89, 120, 125, 127-131, 133, 135,
152, 164, 165, 171, 172, 180, 182, 184, 136, 138-140, 145, 147, 150, 155, 182
187, 188, 190, 192, 199, 200 Neuropsychiatry 163
Methionine synthase reductase (MTRR) gene Neuropsychotropic drug 197, 202
89, 145 Neurospora crassa 85
Methotrexate 183, 198, 200 NHLBI Family Heart Study 13, 33, 56
Methylation 13, 14, 17, 32, 78-92, 163, 165, Nonsense mutation 7, 10
179, 180, 183, 184, 187, 188, 191-193, Nonsense-mediated mRNA degradation
198, 202 (NMD) 7
Methylenetetrahydrofolate 1, 13, 15, 21, 31,
32, 34, 41, 44, 48, 54, 55, 59, 71, 72, O
100, 113, 125, 126, 144, 145, 147, 171,
179, 180, 201 Orofacial cleft 125, 127, 138, 139
Methylenetetrahydrofolate reductase 1, 15,
21, 31, 32, 34, 41, 48, 54, 71, 72, 100,
113, 126, 144, 147, 171, 179, 180
P Seizure 7, 41-43, 45-47, 54, 152

Single nucleotide polymorphism (SNP)
p53 83, 91, 92, 180, 183, 184, 188 21-24, 26-28, 44, 125-128, 130-140
Parkinson 163, 167, 198, 202 Site-directed mutagenesis 12, 14, 24, 25
Pathophysiology 41, 42, 155 Smoking 106, 107, 126, 139, 166, 188, 190
PCR 2, 4-7, 15, 17, 21-27 Spina bifida 56, 127, 138-140
PCR/restriction fragment length Stillbirth 145
polymorphism based (PCR/RFLP) 21, Stroke 47, 57, 63, 101, 103, 105, 166, 174
24, 26, 27 Subacute combined degeneration 42, 43, 48
Pediococcus acidilactici 48
Pediococcus cerevisiae 48 T
Peripheral neuropathy 42, 46
Pernetrexed 200 TaqMan 25, 26
Pharmacogenetic 170, 197-203 Thermolabile MTHFR 1, 55, 71, 200
Placental vasculopathy 145, 150, 152, 157 Trisomy 89, 145, 147, 157
Plasma tHcy 55, 56, 59-61, 66, 73-75, 173 TT genotype 13, 21, 22, 33, 35, 36, 39, 54,
Polymorphism 1, 7, 10-15, 17, 21-26, 28, 31, 56, 61, 65, 66, 71, 73-76, 79, 80, 92,
33, 34, 38, 39, 41, 43, 44, 49, 50, 54, 55, 100, 101, 103-106, 108, 113, 115-120,
57, 62, 71, 74-76, 78, 79, 89, 100, 101, 125-127, 138-140, 144, 145, 147-150,
103-106, 113, 116-121, 125-127, 129, 152, 155, 170, 172-175, 188, 192,
135, 140, 144, 145, 147-150, 152, 155, 197-202
170-175, 179, 180, 184, 187, 188,
190-193, 197-203 U
Preeclampsia 144, 145, 152, 155, 157
Pregnancy 42, 43, 76, 85, 89, 113, 127, 135, Uracil 80, 88, 89, 91, 92, 179, 180, 182, 183,
144, 145, 147-150, 152, 155-157 187, 191-193
Prenatal diagnosis 41, 43, 44 Urinary tract defect 138
Preterm birth 145, 152
Promoter hypermethylation 90
V
Prothrombin 24, 118-120
Pyridoxine 48, 49, 61, 72, 155 Vascular dementia (VAD) 166
Venous thrombosis 66, 76, 101, 113, 115,
R 116-120, 144, 155
Vitamin B2 108, 171, 191
RBC folate 56, 58, 61, 65, 66, 188, 190, 191 Vitamin B6 55, 71-73, 75, 155, 171, 174,
Regional hypermethylation 90-92 180, 184, 187, 188, 190, 192, 193
Renal failure 82, 170-175 Vitamin B12 32, 44, 55, 56, 79, 80, 88, 89,
Restriction enzyme based assay 21, 23 91, 119, 155, 165, 166, 171, 173, 174,
Retinoic acid 155 180, 184, 188, 190, 192, 193, 199
Rett syndrome 87, 88
Riboflavin 10, 13, 39, 48, 59, 60, 71-76, 108,
X
190, 192, 198-200
Riboflavin intervention study 75 X-ray crystallography 36, 81
RNA polymerase 80
S
S-adenosylhomocysteine 13, 17, 32, 78, 79,
81, 180
S-adenosylmethionine 13, 17, 32, 41-43, 48,
50, 66, 71, 78, 79, 81, 167, 180
Saccharomyces cereviseae 33
Schizophrenia 163-165, 202
MEDICAL INTELLIGENCE UNIT
Landes Bioscience, a bioscience publisher,

is making a transition to the internet as
UELAND • ROZEN
Eurekah.com.
INTELLIGENCE UNITS
Biotechnology Intelligence Unit
Medical Intelligence Unit
Molecular Biology Intelligence Unit
Neuroscience Intelligence Unit MIU
Tissue Engineering Intelligence Unit
The chapters in this book, as well as the chapters

of all of the five Intelligence Unit series,
are available at our website.
ISBN 1-58706-217-8
9 781587 062179

MTHFR Polymorphisms and Disease

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

MTHFR Polymorphisms and Disease

Uploaded by

Copyright:

Available Formats

MEDICAL INTELLIGENCE UNIT

Per Magne Ueland and Rima Rozen

Per Magne Ueland, M.D.

Rima Rozen, Ph.D., FCCMG

LANDES BIOSCIENCE EUREKAH.COM

Copyright ©2005 Eurekah.com

Library of Congress Cataloging-in-Publication Data

1. Molecular Biology of Methylenetetrahydrofolate Reductase

2. Assays for Methylenetetrahydrofolate Reductase Polymorphisms ........... 21

3. Biochemical Characterization of Human Methylenetetrahydrofolate

4. Severe Methylenetetrahydrofolate Reductase Deficiency ...................... 41

7. The Molecular Dynamics of Abnormal Folate Metabolism

8. Methylenetetrahydrofolate Reductase 677C→T Polymorphism

9. Methylenetetrahydrofolate Reductase and Venous Thrombosis ......... 113

13. Methylenetetrahydrofolate Reductase Polymorphisms

15. Methylenetetrahydrofolate Reductase Polymorphisms:

Lorenzo D. Botto Manuela Födinger

Methylenetetrahydrofolate reductase (MTHFR) is a critical enzyme in both

Rima Rozen and Per M. Ueland

cDNA and Genomic Structure of MTHFR

Table 1. Description of MTHFR gene-related entries in current databases

Tool or Database URL

Exploring the Introns

Table 2. Primers used to amplify individual MTHFR exons

Exon Direction Annealing Length of PCR

1a Sense Intron 1 AACCTGCCACTCAGGTGTCTTG 66 389

Mutations in Severe MTHFR Deficiency

Table continued on next page

K Turkish 1 mo 0 1084C→T R358X 6 22 1084C→T R358X 6 22 VV NA NA

Table 4. In vitro assessment of MTHFR activity for different mutations

Mutationa Constructb % Controlc Activity Study

None AE 100 ref. 27d

None AA 100 ref. 25d

None AE 100 ref. 20d

None AE 100 ref. 26e

Severe Mutations and Mild Polymorphisms

Biochemical and Molecular Studies

As discussed in subsequent chapters, the 677C→T polymorphism may only be a modest

Other Polymorphisms in MTHFR

Table 5. Polymorphic mutations in 5,10-methylenetetrahydrofolate reductase

Mutationa Affected Codonb Exon or Intronc Reference

129C→T P39P Exon 1 46

Recommendations for Nomenclature of MTHFR Mutations

Animal Model of MTHFR Deficiency

48. Weisberg I, Tran P, Christensen B et al. A second genetic polymorphism in methylenetrahydrofolate

Assays for Methylenetetrahydrofolate

Methods for Genotyping MTHFR SNPs

Restriction Enzyme Based Assays, the 677C→T Variant

Allele-Specific PCR Assays, General Description

Allele-Specific PCR Assays for MTHFR Genotyping

Figure 5. Principle of a heteroduplex assay. A heteroduplex generator (HDG) is constructed by site-directed

Real-Time PCR with Fluorogenic Probes—Homogenous Assays

5´Exonuclease and Molecular Beacon Assays

Hybridization Probe Assays

Comparison of the Different Homogenous Formats

Preparation of DNA Material

New Technologies and Future Developments

Biochemical Characterization of Human

NAD(P)H + H+ + E-FAD → NAD(P)+ + E-FADH2 (1)

E-FADH2 + CH2-H4folate → E-FAD + CH3-H4folate (2)

E-FADH2 + menadione → E-FAD + menadiol (3)

Severe Methylenetetrahydrofolate Reductase

Studies on Cultured Cells