<%BANNER%>

Effect of the Methylenetetrahydrofolate Reductase 677C to T Polymorphism on Folate status and DNA methylation response i...

University of Florida Institutional Repository

PAGE 1

EFFECT OF THE METHYLENETETRAHYDROFOLATE REDUCTASE 677C T POLYMORPHISM ON FOLATE STATUS AND DNA METHYLATION RESPONSE IN YOUNG WOMEN By KARLA PAGN SHELNUTT A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY UNIVERSITY OF FLORIDA 2003

PAGE 2

This dissertation is dedicated to my aunt Nellie J. Mendz whose successful career as a dietitian inspired me to study nutrition. Although her career ended early as a result of health complications, she continues to incorporate good nutrition into her daily life. I hope that one day I will leave my mark in nutrition research as she has done in community nutrition.

PAGE 3

ACKNOWLEDGMENTS I would like to thank the members of my supervisory committee (Lynn B. Bailey, Ph.D.; Gail P. A. Kauwell, Ph.D.; Jesse F. Gregory III, Ph.D.; and Michael S. Kilberg, Ph.D.). I would like to extend special thanks to my major professor Dr. Bailey for giving me the opportunity to work with her on such an important and exciting project. Her guidance over the years has been invaluable to me. She has always assisted me in any capacity that I needed and treated me as a colleague instead of a student. She is an amazing scientist who is successful in balancing her family life with her career. I also would like to acknowledge Dr. Gail Kauwell for all of her scientific knowledge and moral support. Without her expertise, this project would not have been successful. I would also like to thank all of the members of our research group (Carrie Chapman, Aisha Cuadras, David Maneval, Karen Novak, and Angeleah Browdy, Ph.D.) who were essential to the completion of this project. Further thanks are extended to George Henderson, Ph.D.; Eoin Quinlivan, Ph.D.; and Steve Davis, Ph.D. for their technical assistance and advice. I would like to thank the staff at the General Clinical Research Center for their assistance with the feeding study, and especially Doug Theriaque for his essential assistance with the statistical analyses. Special thanks are extended to the many friends I have made over the years who have constantly encouraged me throughout this process, especially Carrie Chapman, Suzie Cole, Britton McPherson, and Sara Rathman. The prayers of these amazing women have been crucial during this experience. iii

PAGE 4

I wish to thank my family for all of their love and support over the years. I wish to especially thank my parents, who continue to pray for me on a daily basis. Their constant words of encouragement have provided guidance to me when I felt I had lost my way. Last but not least, I would like to thank my husband and best friend, David Shelnutt. His unconditional love has sustained me through the toughest parts of graduate school. His help over the years has given me the opportunity to be successful. I look forward to being able to spend the rest of our lives continuing to support each other in all of our endeavors. iv

PAGE 5

TABLE OF CONTENTS page ACKNOWLEDGMENTS.................................................................................................iii LIST OF TABLES...........................................................................................................viii LIST OF FIGURES...........................................................................................................ix LIST OF ABBREVIATIONS..............................................................................................x ABSTRACT.....................................................................................................................xiii CHAPTER 1 INTRODUCTION........................................................................................................1 Hypotheses....................................................................................................................3 Specific Aim.................................................................................................................3 2 BACKGROUND AND LITERATURE REVIEW......................................................4 Folate............................................................................................................................4 Chemistry..............................................................................................................4 Dietary Sources.....................................................................................................4 Bioavailability.......................................................................................................5 Absorption.............................................................................................................7 Transport................................................................................................................8 Biochemical Functions........................................................................................10 FolateRelated Polymorphisms..................................................................................17 Methylenetetrahydrofolate Reductase (MTHFR)................................................17 Methionine Synthase (MS)..................................................................................41 Methionine Synthase Reductase (MSR)..............................................................44 Other Polymorphisms..........................................................................................45 DNA Stability.............................................................................................................47 DNA Methylation................................................................................................47 DNA Strand Breaks.............................................................................................61 Micronuclei Formation........................................................................................64 Dietary Reference Intakes (DRIs)..............................................................................65 Folate Status in Women of Reproductive Age...........................................................67 Effect of Fortification..........................................................................................67 v

PAGE 6

Effect of Ethnicity...............................................................................................68 Folate Deficiency........................................................................................................69 Folate Status Assessment............................................................................................70 Analytical Methodology.............................................................................................72 Blood Folate Analysis.........................................................................................72 Plasma Homocysteine and SAM/SAH Ratio Analysis.......................................73 DNA Stability......................................................................................................74 Genotype Determination.....................................................................................77 Research Significance.................................................................................................77 3 STUDY DESIGN AND METHODS.........................................................................80 Subject Screening and Description.............................................................................80 Study Design...............................................................................................................81 General Clinical Research Center (GCRC) Protocol..................................................82 Dietary Treatment and Supplementation Description................................................83 Sample Collection and Processing..............................................................................86 Analytical Methods.....................................................................................................87 Food Folate Extraction........................................................................................87 Food Choline Analysis........................................................................................88 Supplemental Folic Acid.....................................................................................88 Microbiological Assay.........................................................................................89 Plasma Homocysteine Concentration..................................................................90 DNA Extraction...................................................................................................91 DNA Quantitation...............................................................................................91 MTHFR Genotype Determination.......................................................................92 Methyl Acceptance Assay...................................................................................93 Liquid Chromatography/Mass Spectrometry/Mass Spectrometry Assay...........94 Statistical Analysis......................................................................................................97 4 RESULTS.................................................................................................................101 Folate Content of Menus..........................................................................................101 Serum Folate Concentration.....................................................................................101 Red Blood Cell Folate Concentration.......................................................................105 Homocysteine Concentration....................................................................................108 DNA Methylation.....................................................................................................112 DISCUSSION AND CONCLUSIONS...........................................................................118 SUMMARY.....................................................................................................................135 APPENDIX A VITAMIN-MINERAL SUPPLEMENT COMPOSITION......................................136 B CHOLINE CONTENT OF DIET.............................................................................137 vi

PAGE 7

C POLYMERASE CHAIN REACTION GEL............................................................138 LIST OF REFERENCES.................................................................................................139 BIOGRAPHICAL SKETCH...........................................................................................170 vii

PAGE 8

LIST OF TABLES Table page 2-1 Compartmentalization of enzymes and reactions.....................................................16 3-1 Five-day cycle menu................................................................................................84 3-2 Selective reaction monitoring fragmentation table..................................................96 4-1 Folate content of meals and total daily intake (g DFE).......................................101 4-2 Serum folate concentration.....................................................................................103 4-3 Red blood cell folate concentration........................................................................106 4-4 Homocysteine concentration..................................................................................110 4-5 DNA [ 3 H]methyl group acceptance.......................................................................113 4-6 Percentage (%) of methylated cytosine..................................................................113 4-7 Percent (%) change in DNA [ 3 H]methyl group acceptance...................................113 4-8 Raw change in mCyt/tCyt ratio..............................................................................114 A-1 Vitamin/mineral RDA, five-day diet average, and supplement calculations.........136 B-1 First analysis of dietary choline content.................................................................137 B-2 Second analysis of dietary choline content............................................................137 viii

PAGE 9

LIST OF FIGURES Figure page 2-1 Structure of folate.......................................................................................................4 2-2 Folate metabolism....................................................................................................11 2-3 Histidine metabolism................................................................................................13 2-4 Compartmentalization of folate pools......................................................................15 3-1 Study design.............................................................................................................81 3-2 LC-MS/MS analysis of a standard...........................................................................96 3-3 LC-MS/MS analysis of a hydrolyzed sample..........................................................97 4-1 Weekly unadjusted mean serum folate concentrations..........................................102 4-2 Mean serum folate concentration by genotype......................................................104 4-3 Weekly unadjusted mean red blood cell folate concentrations..............................105 4-4 Mean red blood cell folate concentration by genotype..........................................107 4-5 Weekly unadjusted mean plasma homocysteine concentrations............................108 4-6 Mean plasma homocysteine concentration by genotype........................................111 4-7 Weekly unadjusted mean serum folate and plasma homocysteine concentrations by genotype....................................................................................112 4-8 Percent change in [ 3 H]methyl group acceptance....................................................116 4-9 Percent change in mCyt/tCyt ratio.........................................................................117 C-1 Sample gel with amplified DNA fragments...........................................................138 ix

PAGE 10

LIST OF ABBREVIATIONS Abbreviation Meaning AI adequate intake ALL acute lymphoid leukemia ANCOVA analysis of covariance ANOVA analysis of variance AP apyrimidinic BHMT betaine:homocysteine methyltransferase CBC-D complete blood counts with differentials CDC Centers for Disease Control and Prevention cDNA complementary DNA CHES 2-[N-cyclohexylamino]ethanesulfonic acid CpG cytosine-guanine dinucleotide cpm counts per minute CV coefficient of variation d day dCTP deoxycytosine triphosphate DFE dietary folate equivalents DHF dihydrofolate DNA deoxyribonucleic acid DNMT DNA methyltransferase dNTP deoxynucleotide triphosphate dpm disintegrations per minute DRI dietary reference intakes dsDNA double-stranded DNA dTMP deoxythymidylate monophosphate dTTP deoxythymidylate triphosphate dUMP deoxyuridylate monophosphate dUTP deoxyuridylate triphosphate EAR estimated average requirement EDTA ethylenediaminetetraacetic acid FAD flavin adenine dinucleotide FBPs folate binding proteins FDA Food and Drug Administration FIGLU formiminoglutamate FMN flavin mononucleotide g gram GCPII glutamate carboxypeptidase II GCRC General Clinical Research Center x

PAGE 11

h hour HCG human chorionic gonadotropin HDAC histone deacetylase HEPES hydroxyethylpiperazine-N-[2-ethanesulfonic acid] HPLC high performance liquid chromatography IOM Institute of Medicine kDa kilodalton kg kilograms L liter LC/ESI-IDMS liquid chromatography/electrospray ionization-isotope dilution mass spectrometry LC/MS liquid chromatography/mass spectrometry LC-MS/MS liquid chromatography-mass spectrometry/mass spectrometry LS least squares LSES low socioeconomic status mCyt methylcytosine mg milligrams min minute ml milliliter mmol millimole mo month MRC Medical Research Council MS methionine synthase MS/MS mass spectrometry/mass spectrometry MSR methionine synthase reductase MTHFR methylenetetrahydrofolate reductase NaOH sodium hydroxide NEB New England Biolabs ng nanogram NHANES National Health and Nutrition Examination Survey nmol nanomole NTDs neural tube defects o old PCR polymerase chain reaction RA radiobinding assay Rb retinoblastoma protein RDA recommended dietary allowance RFC reduced folate carrier RNA ribonucleic acid ROPS random oligonucleotide primed synthesis s seconds SAH S-adenosylhomocysteine SAM S-adenosylmethionine SD standard deviation SEA socioeconomically advantaged SHMT serine hydroxymethyltransferase xi

PAGE 12

SNP single nucleotide polymorphism SST serum separator tubes tCyt total cytosine TE tris EDTA THF tetrahydrofolate UL upper limit VTE venous thromboembolism wk week y year g microgram l microliter mol micromole xii

PAGE 13

Abstract of Dissertation Presented to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Doctor of Philosophy EFFECT OF THE METHYLENETETRAHYDROFOLATE REDUCTASE 677CT POLYMORPHISM ON FOLATE STATUS AND DNA METHYLATION RESPONSE IN YOUNG WOMEN By Karla Pagn Shelnutt August 2003 Chair: Lynn B. Bailey Major Department: Food Science and Human Nutrition Methylenetetrahydrofolate reductase (MTHFR) is a key enzyme in folate metabolism. A CT substitution at base pair 677 of MTHFR may impair folate status when folate intake is marginal, a particular concern for women of reproductive age. Folate status response to a controlled folate depletion-repletion feeding protocol (14 wk) was investigated in young women (20 to 30 y) with (TT) (n = 19) and without (CC) (n = 22) the MTHFR 677CT variant. Subjects consumed a moderately folate-deplete diet (115 g DFE/d) for 7 wk, followed by folate repletion for 7 wk with the current RDA for this population (400 g DFE/d). Overall serum folate decreased (P < 0.0001) during depletion and increased (P < 0.0001) during repletion with lower (P = 0.03) post-depletion serum folate in women with the TT versus CC genotype (LS mean 15.3 1.2 vs 19.5 1.3 nmol/L, respectively). Folate status was low (serum folate < 13.6 nmol/L) in more women with xiii

PAGE 14

the TT (59%) versus the CC (15%) genotype post-depletion. Red blood cell folate for all subjects decreased during depletion (P < 0.0001) and repletion (P = 0.02) with lower (P = 0.04) red blood cell folate in women with the TT versus the CC genotype post-repletion (LS mean 1036 53 vs 1203 53 nmol/L, respectively). Homocysteine increased (P < 0.0001) during depletion and decreased for subjects with the CC (P = 0.02) but not the TT (P = 0.47) genotype during repletion. Homocysteine tended to differ by genotype post-depletion [10.5 3.3 (TT) vs 8.9 1.9 (CC) (P = 0.09)] and post-repletion [8.9 1.6 (TT) vs 7.2 1.9 (CC) (P = 0.08)]. Overall percent change in [ 3 H]methyl group acceptance tended to increase during depletion (P = 0.08). Women with the TT genotype had an increase in raw and percent change in methylated cytosine during repletion (P < 0.05). Except for global DNA methylation, these data suggest that young women with the MTHFR 677 TT genotype respond more negatively to folate depletion and less positively to repletion with the current RDA compared to women with the CC genotype. xiv

PAGE 15

CHAPTER 1 INTRODUCTION Folate is an essential vitamin that functions to accept and transfer one-carbon units in nucleotide synthesis, amino acid interconversions including methionine regeneration, and the formation of S-adenosylmethionine (SAM)the primary methylating agent in the body. An important reaction in folate metabolism is the reduction of 5,10-methylenetetrahydrofolate (5,10-methyleneTHF) to 5-methyltetrahydrofolate (5-methylTHF) by the 5,10-methylenetetrahydrofolate reductase (MTHFR) enzyme. This reaction generates folate in the form of 5-methylTHF, essential for the remethylation of homocysteine to methionine. The conversion of homocysteine to methionine is essential for the synthesis of SAM, which is the methyl donor in > 100 physiological reactions, including methylation of DNA, RNA, and membrane phospholipids as previously reviewed (1). Emerging science involving single-nucleotide polymorphisms offers new insight and a more precise understanding of how individual genetic variations influence folate-dependent metabolic pathways (2). Single-nucleotide polymorphisms, genetic variations present in > 1% of the population, can act alone or synergistically with nutritional deficiencies to accelerate and accentuate metabolic pathologies. A very common polymorphism and one that has been studied extensively is a C-to-T substitution at base pair 677 in the gene that codes for MTHFR (3). The MTHFR 677CT variant is prevalent in the overall population: ~12% homozygous (TT); and ~50% heterozygous (CT) (4). This variant results in an alanine-to-valine substitution in the enzyme resulting 1

PAGE 16

2 in a marked reduction in enzyme stability that can be enhanced by the addition of 5-methylTHF (5). In individuals who are homozygous for the MTHFR 677CT variant (TT) the impaired enzyme stability is associated with reduced methylated blood folate (6) and elevated plasma homocysteine concentration that can be significantly improved in response to improved folate status (7). Different studies have provided evidence that methylation of DNA plays a role in genome stability and gene expression (8,9). The potential for impaired folate status coupled with the MTHFR 677 TT genotype to negatively influence genome stability (10,11), including DNA methylation (12,13), provided one incentive for the present study. The effect of folate inadequacy in individuals with the TT genotype on indicators of global DNA methylation has not been previously evaluated in a metabolic study in which nutrient intake was carefully controlled. The present study was designed to monitor changes in DNA methylation by MTHFR genotype in conjunction with other folate status indicators in response to a folate depletion-repletion protocol. An estimated 120,000-150,000 infants are born with a major birth defect in the US each year, representing more than 3% of all live births (14). Emerging scientific evidence involving single nucleotide polymorphisms such as the MTHFR 677CT variant, including data from the present study, may lead to a precise understanding of how individual genetic variations influence folate-dependent metabolic pathways and potentially increase birth defect risk. Various studies have shown that the MTHFR 677 TT genotype is a significant risk factor for neural tube defects when folate intake is not sufficient to maintain metabolic homeostasis (15,16). Since the combined presence of the MTHFR 677CT variant and low folate status has been associated with increased risk

PAGE 17

3 for birth defects, the present study was designed to address the specific aim in females of reproductive age. Hypotheses 1. Consumption of a low-folate diet will result in more significant impairment in indicators of folate status and global DNA methylation in young women with the homozygous MTHFR 677CT TT genotype compared to those with the CC genotype. 2. Consumption of the current Recommended Dietary Allowance for folate (400 g dietary folate equivalents/d) following a low-folate diet will not be sufficient to significantly improve folate status or global DNA methylation in females with the TT compared to the CC genotype. Specific Aim The specific aim of the present study is to determine the combined effects of the MTHFR 677CT variant and dietary folate depletion-repletion on indicators of folate status response (serum and red blood cell folate and plasma homocysteine concentrations) and global DNA methylation ([ 3 H]methyl group acceptance and methylcytosine/total cytosine ratio) in women of reproductive age.

PAGE 18

CHAPTER 2 BACKGROUND AND LITERATURE REVIEW Folate Chemistry Folate is a general term that refers to naturally occurring food folate and synthetic folic acid found in supplements and fortified foods. Synthetic folic acid is yellow with a molecular weight of 441.4. Folate is made up of three parts that are all necessary for vitamin activity. A pteridine bicyclic ring is linked by a methylene group to para-aminobenzoic acid whose carboxy group is peptide bound to the -amino group of glutamate to form folate (17) (Figure 2-1). COOH COOH Figure 2-1. Structure of folate. Modified from Bailey, L. B., Moyers, S., and Gregory, J.F. (2001) Folate. In: Present Knowledge in Nutrition (Bowman, B. A. and Russell, R.M., ed.), p. 214. ILSI Press, Washington, D.C. Dietary Sources Mammals are unable to synthesize folate de novo and must therefore obtain it from the diet. Folate is not considered to be prevalent in the food supply. Two different types of folate are available for consumption: folate found naturally in the food supply and synthetic folic acid added to enriched food products and supplements. 4

PAGE 19

5 Naturally occurring food folate. Some good sources of naturally occurring food folates are orange juice, dark green leafy vegetables, asparagus, strawberries, peanuts, and legumes such as kidney and lima beans (18). Folate concentration in raw foods is higher than in cooked foods due to the folate losses incurred by cooking (19). Variability of losses associated with cooking is attributable to differences in oxygen exposure, ascorbic acid content, and the amount of water present (19). The folates that occur naturally in food are the fully reduced tetrahydrofolates and usually have 5 to 7 glutamates in the polyglutamate side chain (19). Folic acid in fortified foods. Another major source of folate is synthetic folic acid used in enriched foods and fortified, ready-to-eat breakfast cereals. Effective January 1, 1998, the U.S. Food and Drug Administration (FDA) mandated the fortification of all cereal grain products labeled as enriched (e.g., bread, pasta, flour, rice) with 140 g folic acid per 100 g of product (20). Thousands of food items have been affected under this mandate (21). A large number of ready-to-eat breakfast cereals contain folic acid in varying amounts. Most ready-to-eat cereals are fortified to provide 25% the Daily Value of folate, with some brands providing four times this amount (18). Bioavailability Folate bioavailability refers to the overall efficiency of utilization of the vitamin, including absorption, transport, metabolism, catabolism, and excretion (1,22). The bioavailability of food folate differs greatly from the bioavailability of synthetic folic acid. Food folate bioavailability. The bioavailability of folate in different foods varies considerably because of differences in digestion, absorption, and metabolism. Some possible digestion/absorption issues that can affect folate bioavailability include altered

PAGE 20

6 pH, conjugase activity, and transit time (22). The effect of different food components on folate utilization further complicates the issue by trapping the folate in the food matrix or inhibiting conjugase activity (22). Food processing can cause oxidative damage of folates (22) and can account for 50 to 95% of any folate lost (23). Alcohol and pharmaceuticals also can inhibit the absorption of folate (22). All of these factors attribute to the high variability of folate bioavailability, which can be as high as 96% and as low as 25% (24). Synthetic folic acid bioavailability. Synthetic folic acid is much more bioavailable than food folate. When folic acid is administered without food it is 100% bioavailable (25). Sauberlich et al. (23) estimated that when compared to synthetic folic acid, the bioavailability of food folate is no more than 50% available. Pfeiffer et al. (26) examined the bioavailability of 13 C 5 -labeled folic acid administered in apple juice when given with or without food. Consumption of the folic acid supplement with food slightly decreases its bioavailability by approximately 15% (26). The Dietary Reference Intake Committee of the Institute of Medicine (IOM) used these data to derive an estimated bioavailability of folic acid when consumed with food (85%) (27). This estimate of folic acid bioavailability was used by the IOM as a basis for the new term, dietary folate equivalents (DFEs), which was used to express the 1998 Dietary Reference Intakes (DRIs). Dietary folate equivalents (DFEs). The DFEs convert all forms of dietary folate, including synthetic folic acid in fortified products, to an amount that is equivalent to food folate (28). The rationale for the DFEs is that folic acid consumed with food is only 85% bioavailable and naturally occurring food folate is only 50% bioavailable. Using these

PAGE 21

7 estimates of folate bioavailability, folic acid consumed with food is estimated to be 1.7 times (i.e., 85/50) more available than natural food folate (27). The total folate content can be calculated by multiplying the micrograms of synthetic folic acid by 1.7 and then adding this number to the number of micrograms of food folate present in the meal (18). To compare folic acid with food folate the following conversions are used: 1 g DFE = 1 g food folate = 0.5 g folic acid taken on an empty stomach = 0.6 g folic acid taken with meals. When estimating food folate intake only, no adjustment is needed (27). Absorption Before absorption of ingested food folate can occur, it must be hydrolyzed to the monoglutamate form by glutamate carboxypeptidase II (EC 3.4.17.21) (29) also called pteroylpolyglutamate hydrolase or folate conjugase. This enzyme is located in the jejunal brush border membrane and has an optimum pH of 6.5 to 7.0 (30). Under normal conditions, monoglutamyl folate is transported across the intestinal membrane by a saturable, pH-dependent carrier-mediated intestinal folate carrier (31). The expression of this intestinal folate carrier may be upregulated in response to folate deficiency (32). When folate concentrations are high, a nonsaturable mechanism involving passive diffusion transports folate across the intestinal membrane. This mechanism may be more important in the absorption of supplemental folic acid as opposed to food folate (22). Most folate is converted to the reduced forms, dihydrofolate (DHF) or tetrahydrofolate (THF), by dihydrofolate reductase before portal blood entry. This mechanism is saturable and large amounts of oxidized folic acid have been found in the plasma and urine of individuals ingesting 400-800 g/d of folic acid (22). Further metabolism to a methylated or formylated form of folate also may occur in the mucosal cells before portal blood entry (22).

PAGE 22

8 Transport Once the folate is absorbed in the monoglutamate form, it travels in the portal circulation, mostly as 5-methylTHF, to the liver where it is reduced and conjugated for retention. The major forms of folate in the liver are 5-methylTHF and 10-formylTHF, which can be secreted into the bile and reabsorbed via enterohepatic circulation (33). Approximately two-thirds of folates in the plasma are protein bound, 50% of which are bound to albumin. The remaining one-third is tightly bound to folate binding protein (34). Folate binding proteins (FBPs). Membrane associated FBPs, also known as folate receptors, transport folate across membranes from circulating blood into cells, are highly localized and expressed in specific tissues and cells (35), and have a higher affinity for oxidized folic acid than for reduced folates (36). The FBPs are essential for normal embryonic development and can cause embryonic lethality if the gene that codes for FBP is knocked out in mice (37). The FBP knockout mice have been brought to term with folinic acid supplementation and were born with normal phenotypes (35). Although the FBP has been characterized, little is known about how cells obtain their folate. Antony (38) reviewed two possible mechanisms by which cells take up folate. The first is the well-known process of endocytosis, and the second is referred to as potocytosis. In potocytosis the FBPs are gathered in clusters at the plasma membrane, which move into caveolae and concentrate circulating folates. Following sudden acidification of the caveloae, the folates dissociate from the folate receptors and are transported into the cytoplasm by anion channels. This is not the only mechanism for folate transport (38). This membrane-associated FBP has extensive homology with the plasma FBP (17).

PAGE 23

9 Reduced folate carrier (RFC). A RFC also exists that transports reduced folates into the cell (36). The RFC protein is the same folate carrier protein that is expressed in the intestine. The RFCs are located in all tissues and cells (35) and have a higher affinity for reduced folates than oxidized folic acid. They function to form channels in the plasma membrane through which reduced folates, mostly 5-methylTHF, can cross into the cytoplasm of the cell (39). Zhao et al. (40) determined the consequence of inactivation of the RFC in embryonic RFC knockout mice who died in utero. They also observed that near normal development of RFC knockout mice could be attained with daily supplementation of the heterozygous pregnant dams with folic acid, but these pups died within 12 d of birth. Zhao et al. (40) concluded that the folic acid supplementation was sufficient in utero, but that there was probably insufficient folate in the mothers milk to sustain life. A polymorphism of the RFC gene that results in a GA substitution at base pair 80 replaces arginine with histidine (36) in the protein. This polymorphism has not been found to affect plasma folate or homocysteine concentrations in adults but may be more important to the developing embryo (36). De Marco et al. (41) observed that neural tube defect (NTD)-affected Italian children with the homozygous variant genotype (GG) had a significantly higher risk for NTDs than control children with the normal genotype (AA). In addition, mothers with the GG genotype had a higher risk of giving birth to an NTD-affected child compared to mothers with the AA genotype (40). Folates enter cells in the monoglutamate form and if in the oxidized form must be reduced by dihydrofolate reductase to THF and polyglutamated by folylpoly--glutamate synthetase. Conjugation retains folates in the cells and is required for participation in

PAGE 24

10 one-carbon metabolism. The different coenzymatic forms of folate are all derivatives of THF and can contain a methyl (CH 3 ), methylene (-CH 3 -), methenyl (-CH=), formyl (-CH=O), or formimino (-CH=NH) group (1). Tetrahydrofolate accepts the one-carbon moieties, which become bonded to the N 5 or N 10 atoms or both. The polyglutamate folates must be hydrolyzed before release from the tissues into circulation with -glutamylhydrolase converting the folates back to the monoglutamate form (17). The total body content of folate is approximately 15 to 30 mg with the liver containing 50% of the bodys folate stores (42). Biochemical Functions Folates main biochemical functions are to accept and transfer one-carbon units involved in amino acid metabolism, purine and pyrimidine synthesis, and the formation of SAM, the main methyl donor in > 100 reactions. Tetrahydrofolate is the main acceptor molecule and ultimately ends up in pathways required for nucleotide biosynthesis or methylation reactions. An overview of the different folate pathways is presented in Figure 2-2. Amino acid metabolism. Folate functions as an intermediate in the metabolism of the amino acids serine, glycine, methionine, homocysteine, and histidine. The metabolism of serine, glycine, methionine, and homocysteine are closely related. Tetrahydrofolate is converted to 5,10-methyleneTHF via serine hydroxymethyltransferase (SHMT) (Figure 2-2, reaction 3). Serine is converted to glycine in the process, and both reactions are reversible. Pyridoxal phosphate is required as a cofactor for this reaction. The degradation of glycine requires THF and NAD+. The -carbon of serine provides the most one-carbon units in all mammalian systems. These

PAGE 25

DNA Methylation 1. dihydrofolate reductase 2. thymidylate synthase 3. serine hydroxymethyltransferase 4. cyclohydrolase 5. formate activa ting enzyme 6. transformylases 7. methylenetetrahydrofolate reductase 8. methionine synthase 9. S-adenosylmethionine synthase 10. cellular methyltransferases 11. S-adenosylhomocysteine hydrolase 12. cystahionine -synthase 13. cystathionase 14. betaine:homocysteine methyltransferase Methionine Homocysteine SAM SAH Purine Synthesis + THF Cobalamin 9 10 11 Cystathionine Cysteine12 13 Betaine 14 DNA Synthesis dTMP dUMP DHF THF 5-methylTHF 1 2 Serine Glycine 3 5,10 methyleneTHF 10 formylTHF 4 5 6 7 8 11 Figure 2-2. Folate metabolism. Adapted from Bailey, L. B., Moyers, S., and Gregory, J.F. (2001) Folate. In: Present Knowledge in Nutrition (Bowman, B.A. and Russell, R.M., ed.), p. 217. ILSI Press, Washington, D.C.

PAGE 26

12 one-carbon units get transferred to THF (43). After 5,10-methyleneTHF is formed, it is reduced to 5-methylTHF by the irreversible reaction requiring the enzyme MTHFR (Figure 2-2, reaction 7). The substrate and cofactor for methionine synthase, the enzyme that remethylates homocysteine to methionine, is 5-methylTHF (Figure 2-2, reaction 8) and requires vitamin B12, or cobalamin, as a coenzyme. A vitamin B12 deficiency can lead to a secondary folate deficiency that investigators have termed the methyl trap. Methionine synthase depends on vitamin B12 as a coenzyme and its activity is reduced during a vitamin B12 deficiency. This results in the accumulation of 5-methylTHF, which gets trapped in this form and results in the decreased availability of THF and all of the other forms of folate. This can lead to megaloblastic anemia due to insufficient folate coenzymes available for DNA synthesis (17). Homocysteine can also be remethylated to methionine by betaine:homocysteine methyltransferase (BHMT) (Figure 2-2, reaction 14), which is a betaine-dependent reaction found in the pathway of choline oxidation. Choline is a methyl rich compound present in food that is first oxidized to betaine and then demethylated by BHMT (44). Choline also can be synthesized by the body in the SAM dependent methylation of phosphatidylethanolamine to form phosphatidylcholine, which can be further catabolized to choline. Choline oxidation primarily takes place in the liver and kidney. Choline is first oxidized to betaine aldehyde via choline oxidase. Betaine aldehyde is then oxidized to betaine via betaine aldehyde dehydrogenase (44). Betaine can then be used in the BHMT reaction. The interdependence of choline and folate has been demonstrated in a variety of studies with rats. Rats fed a choline or choline and methionine deficient diet

PAGE 27

13 for 2 wk to 12 mo had decreased hepatic folate stores (45-47). Adverse effects from a choline deficiency were reversed within 2 wk with adequate dietary choline (47). Histidine is the other amino acid dependent on folate metabolism. The deamination of histidine produces urocanic acid, which becomes formiminoglutamate (FIGLU) after further metabolism. Formiminoglutamate loses the formimino group via formimino transferase to THF to produce N 5 -formiminoTHF. Glutamic acid is the final product of this reaction. The pathway of histidine metabolism is shown in Figure 2-3. THF FIGLUHistidine Glutamic Aci d 5-formiminoTHF Formimino transferase Figure 2-3. Histidine metabolism. Purine and pyrimidine synthesis. The nucleotide biosynthesis pathway utilizes the 5,10-methyleneTHF coenzyme. Pyrimidine synthesis involves the formation of deoxythymidylate monophosphate (dTMP) from deoxyuridylate monophosphate (dUMP) via thymidylate synthase (Figure 2-2, reaction 2), which requires 5,10-methyleneTHF as a coenzyme. This is the rate-limiting step in the cell cycle and allows DNA replication to continue. Once thymidylate synthase utilizes 5,10-methyleneTHF as a coenzyme, it is oxidized to DHF and then regenerated to THF via dihydrofolate reductase (Figure 2-2, reaction 1). Purine synthesis depends on the conversion of 5,10-methyleneTHF to

PAGE 28

14 10-formylTHF via 10-formylTHF synthetase (Figure 2-2, reaction 4). The folate coenzyme is used to donate formyl groups to different positions of the purine ring and depends on a formate activating enzyme (Figure 2-2, reaction 5). S-Adenosyl methionine (SAM) synthesis. The formation of SAM is dependent on the metabolism of homocysteine and methionine. Under normal conditions, methionine synthase transfers the methyl group from 5-methylTHF to vitamin B12 and then to homocysteine. In this reaction, 5-methylTHF is reconverted to THF. The newly synthesized methionine can then be converted to SAM by SAM synthase (Figure 2-2, reaction 9) when combined with adenosine. S-Adenosylmethionine can then act to donate methyl groups in over 100 different methylation reactions, including DNA and RNA methylation. Once a methyltransferase uses the methyl group provided by SAM, it is converted to S-adenosylhomocysteine (SAH) (Figure 2-2, reaction 10). S-Adenosylhomocysteine is hydrolyzed to homocysteine and adenosine via the reversible SAH hydrolase (Figure 2-2, reaction 11). Homocysteine can be remethylated to methionine or converted to cysteine through the transulfuration pathway where cystathionine -synthase (CBS) (Figure 2-2, reaction 12) converts homocysteine to cystathionine, a reaction that is dependent on pyridoxal phosphate and serine. Cystathionase (Figure 2-2, reaction 13) converts cystathionine to cysteine. Cytosolic folate metabolism differs from mitochondrial folate metabolism. The composition of folate pools within the cytosol and mitochondria varies among different tissues (1). In the cytosol, one source of one-carbon groups is provided by formate from the mitochondria and cytosol that is converted to 10-formylTHF by 10-formyl synthetase (Figure 2-4, enzyme C3). It serves as a one-carbon source in the cytosol for purine

PAGE 29

Cytosol Mitochondrial Matrix CO 2 + THF Purines + THF 10 formylTHF 5,10 methenylTHF 5,10 methyleneTHF Formate + THF DNA dTMP + THF 5 methylTHF SAM Inhibited Homocysteine Methionine Sarcosine Glycine 5-methylTHF inhibited SAH SAM Biosynthetic Pathways Serine THF Glycine Serine 5,10-methyleneTHF THF 5,10-methenylTHF 10-formylTHF THF Formate C14 C3 C2 C1 C9 C6 C7 C8 C4 C5 M5 M13 Choline N H 4 +CO2 1 C units M12 M11 M1 M2 M3 M14 15 Figure 2-4. Compartmentalization of folate pools between the cytosol and mitochondria. Reproduced from Bailey, L. B., Moyers, S., and Gregory, J.F. (2001) Folate. In: Present Knowledge in Nutrition (Bowman, B.A. and Russell, R.M., ed.), p. 218 Figure 3. ILSI Press, Washington, D.C.

PAGE 30

16 Table 2-1. Compartmentalization of enzymes and reactions. Reaction Enzyme C1:M1 5,10-MethyleneTHF dehydrogenase C2:M2 5,10-MethyleneTHF cyclohydrogenase C3:M3 10-FormylTHF synthetase C4 Glycine N-methyltransferase C5:M5 Serine hydroxymethyltransferase C6 Methionine synthase C7 Glycinamide ribonucleotide transformylase C8 Phosphoribosylamino-imidazole carboxamide transformylase C9 5,10-MethyleneTHF reductase C10 Thymidylate synthase M11 Dimethylglycine dehydrogenase M12 Sarcosine dehydrogenase M13 Glycine cleavage system C14:M14 10-FormylTHF dehydrogenase Reproduced from Bailey, L. B., Moyers, S., and Gregory, J.F. (2001) Folate. In: Present Knowledge in Nutrition (Bowman, B.A. and Russell, R.M., ed.), p. 218 Table 2. ILSI Press, Washington, D.C. synthesis. 10-FormylTHF can be further reduced to 5,10-methenylTHF by 5,10-methenylTHF cyclohydrolase (Figure 2-4, enzyme C2) and reduced once again by 5,10-methyleneTHF dehydrogenase to 5,10-methyleneTHF for pyrimidine synthesis. The enzyme MTHFR (Figure 2-4, enzyme C9) reduces 5,10-methyleneTHF to 5-methylTHF for homocysteine remethylation to methionine in the cytosol. The interconversion of serine and glycine not only provides one-carbon units to the cytosol, but also to the mitochondria. The difference between cytosolic and mitochondrial folate metabolism can be explained by the form of SHMT (Figure 2-4, enzymes C5:M5), the enzyme used to catalyze this reaction, since cytosolic SHMT is different than mitochondrial SHMT. Glycine cleavage occurs only in the mitochondria. The final folate-dependent steps of choline catabolism occur in the mitochondria.

PAGE 31

17 The biochemical functions of folate are collectively referred to as one-carbon metabolism (1). Although each function is independently important, they are all dependent on the cellular concentration of THF and each other to function. FolateRelated Polymorphisms Single nucleotide variations in the genome can result in genes that code for enzymes with different activity. If a genetic variation is present in > 1% of the population, it is considered a genetic polymorphism (48). Polymorphisms affecting one-carbon metabolism are of special interest to investigators due to their frequency in the population and their effects on disease risk and developmental abnormalities. The focus of folate research has shifted from evaluating the effect of severe deficiency on clinical indicators and blood folate concentrations to identifying functional indicators of disease risk and how they are affected by genetic polymorphisms. The next section will include a synopsis of recent research findings and key research questions related to polymorphisms affecting folate metabolic function. Methylenetetrahydrofolate Reductase (MTHFR) Although folate metabolism involves over 30 genes, enzymes, and transporters, the most extensively studied polymorphism affects MTHFR, the enzyme that catalyzes the reduction of 5,10-methyleneTHF to 5-methylTHF (3). In 1988 Kang and colleagues (49) discovered a thermolabile variant of MTHFR that was positively correlated to increased cardiovascular risk and increased homocysteine concentrations. In 1994, Goyette et al. (50) isolated a complementary DNA (cDNA) copy of human MTHFR that was 22 kilobases long consisting of 11 exons. Isolation of this cDNA enabled them to identify nine mutations in the gene in patients with severe MTHFR deficiency (50,51). A more recent report increased this number to 33 severe mutations identified in patients with

PAGE 32

18 severe MTHFR deficiency (52). Frosst et al. (3) identified a homozygous variant of this gene that resulted in decreased enzyme activity and increased thermolability in vitro. This autosomal recessive variation results in a CT substitution at base pair 677 that replaces alanine with valine in the enzyme. Any decrease in MTHFR activity from the 677CT variant will result in lower 5-methylTHF available to donate a methyl group for the remethylation of homocysteine to methionine. This can result in elevated homocysteine concentrations in individuals with the homozygous variant (TT) genotype in contrast to the heterozygous variant (CT) or homozygous normal (CC) MTHFR 677CT genotype. The homozygous TT variant is termed thermolabile because of a significant decrease in residual activity compared to CC controls after heat inactivation at 46C for 5 min (49). Frosst et al. (3) compared specific activity and residual activity after heating for 5 min at 46C. They found that subjects with the TT genotype had a specific activity < 50% of that of the subjects with the CC genotype and a residual activity < 35% after heating. Rozen (53) found a 50-60% reduction in residual activity among individuals with the TT genotype at 37C and a 65% decrease at 46C compared to individuals with the CC genotype. Matthews et al. (54) reported that the MTHFR gene codes for an enzyme with two identical 77 kDa subunits, each consisting of a 40 kDa N-terminal domain and a 37 kDa C-terminal domain. In addition, Frosst et al. (3) found a 70 kDa subunit, which they attributed to the presence of isozymes. Sumner et al. (55) discovered that SAM binds to the C-terminal domain, which led them to implicate this domain as the regulatory region because SAM is an allosteric inhibitor of MTHFR. Matthews et al. (56) reviewed the similarities between the N-terminal domain of the human MTHFR and smaller proteins

PAGE 33

19 of enteric bacteria that catalyze the same reaction. They explained that since SAM does not regulate the bacteria proteins, the N-terminal domain must be the catalytic region. This is important because the CT substitution occurs in exon 4 of the N-terminal domain (57). Matthews et al. (56) purified MTHFR from Escherichia coli. They reported that the enzyme was a flavoprotein with flavin adenine dinucleotide (FAD) as its cofactor, which is reduced by NADH and NADPH and then reduces 5,10-methyleneTHF to 5-methylTHF. Guenther et al. (58) characterized the normal (CC) and variant (TT) MTHFR enzymes from Escherichia coli, which revealed that the variant enzyme loses its flavin cofactor more readily than the normal enzyme. They reported that the catalytic domain shared by all MTHFRs is a barrel that binds FAD, and that the variant does not cause decreased activity of the enzyme directly but instead decreases enzyme stability, which facilitates the dissociation of FAD and decreases enzyme activity. They found that folate supplementation confers protection by increasing the affinity of MTHFR for FAD. Matthews (59) hypothesized that the exact mechanism of this protection lies in the positioning of the folate cofactor with respect to the barrel. Folate may bind with the pteridine ring stacked above the flavin, thus stabilizing the enzyme and preventing FAD dissociation. Yamada et al. (5) were the first to characterize human MTHFR by using a baculovirus expression system. They identified important differences between the mammalian enzyme and the prokaryotic enzyme. For example, the prokaryotic enzyme lacks a regulatory domain and uses NADH as a reductant while the mammalian enzyme has a regulatory domain and uses NADPH as a reductant. Another difference is that prokaryotic MTHFR is a tetramer in contrast to the mammalian MTHFR, which is a dimer. Their results agree with the previous findings of Guenther et al. (58) that the

PAGE 34

20 decreased activity of the enzyme is due to decreased protein stability resulting from FAD dissociation. This loss of FAD leads to dissociation of the dimer into monomers and reduction in activity (59) and has been found to occur 10 times faster in the homozygous variant (TT) compared to the normal (CC) enzyme (58). Yamada et al. (5) concluded that individuals with the TT genotype for the MTHFR 677CT polymorphism are at higher risk for increased homocysteine concentrations if their folate concentrations are low. This increased risk is due to less remethylation of homocysteine to methionine as a result of the decreased MTHFR stability. Decreased MTHFR stability will lead to even more dissociation of FAD and result in less enzyme activity. The in vitro studies performed by this group showed that increasing folate, SAM, and FAD concentrations protected against loss of activity in the affected human MTHFR enzyme. The association between riboflavin status and homocysteine concentration has been evaluated. Hustad et al. (60) reported a significant inverse dose-response association between plasma riboflavin and homocysteine concentrations in their combined MTHFR 677 CT/TT group. When plasma riboflavin was separated into quartiles, they observed a 1.4 mol/L higher homocysteine concentration in the lowest compared to the highest quartile. They concluded that riboflavin is an independent determinant of plasma homocysteine concentration regardless of folate status. McNulty et al. (61) separated the riboflavin status of their population into tertiles and found that among subjects in the lowest tertile, the mean homocysteine concentration of subjects with the TT genotype was approximately twice the concentration in either the CT or CC genotype groups. There were no significant differences in mean homocysteine concentrations between genotype groups in either the medium or high tertiles of riboflavin status. They

PAGE 35

21 concluded that unlike folate status, riboflavin status is only a predictor of homocysteine concentration in subjects with the TT genotype, and that a riboflavin-dependent mechanism may contribute to the interrelationship between folate and plasma homocysteine concentrations in this group (61). Jacques et al. (62) further clarified the relationship between riboflavin and homocysteine concentrations by reporting that the association between these indices in their population was only significant in those with lower folate status. When further stratified by genotype, the interaction between folate and riboflavin was only significant in subjects with the TT genotype. Only those with low folate status and the TT genotype showed a significant association between riboflavin and homocysteine concentrations. In contrast to the previously described studies, Moat et al. (63) found an inverse association between riboflavin and homocysteine concentrations in individuals with both the CC and TT genotypes, but further studies are warranted to clarify this discrepancy. Rozen (64) suggested that additional studies to determine the role of folate in the riboflavin-homocysteine association would be useful before drawing conclusions. An extensive review by Botto and Yang (4) provides a description of the population frequency for the MTHFR 677CT polymorphism. They estimated that approximately 22% of Hispanics, 12% of Caucasians, 11% of Asians, and 1 to 2% of African Americans are homozygous for the variant. The variant has been found in ~35% of alleles (3,65). The frequency of the homozygous variant varies among racial and ethnic groups. The lowest prevalence is among blacks and Africans (~2% and 0%). Stevenson et al. (66) reported the frequency of the TT genotype in 151 whites and 146

PAGE 36

22 blacks from South Carolina and found a frequency of 13% and 0%, respectively, which agrees with other reports (4). The effect of the TT genotype can be quantitated by evaluating the change in cellular composition of the folate derivatives (6). If 5,10-methyleneTHF is not reduced to 5-methylTHF, the methylene group of 5,10-methyleneTHF can be donated to dUMP to form dTMP or oxidized to 5,10-methenylTHF, which ultimately is converted to 10-formylTHF for the synthesis of purines (6). One would expect a decrease in MTHFR activity to result in an increase in formylated folates and a decrease in methylated folates. Bagley and Selhub (6) provided data to support this hypothesis by evaluating the effect of the MTHFR genotype on the form of folate within red blood cells. Of eight subjects with the CC genotype, all had 100% methylated folates. Of the 10 subjects with the TT genotype, eight had formylated folates ranging from 0 to 59% of the total folate content. Genotype did not affect total red blood cell folate concentration. Zittoun et al. (67) also evaluated the effect of the TT genotype on methylfolates and reported that in subjects with the CC genotype, 71% of the red blood cell folate concentration was methylfolate compared to 66% and 27% in individuals with the CT and TT genotypes, respectively. After separating their study population into high and low red blood cell folate status, Friso et al. (13) found significantly less methylated folates in subjects with the TT genotype, regardless of folate status. These data support the findings of Bagley and Selhub (6). Another more recently discovered MTHFR polymorphism, 1298AC, results in an AC base pair substitution at base pair 1298 that replaces glutamic acid with alanine in the C-terminal regulatory domain of the enzyme resulting in a homozygous variant

PAGE 37

23 (CC), heterozygous variant (AC), or normal (AA) genotype (57,68). The MTHFR 1298AC polymorphism was first discovered by Viel et al. (69) in patients with ovarian carcinomas, but was not characterized until later (57,68). Yamada et al. (5) were unable to distinguish between the properties of the variant and normal enzymes after baculovirus expression. They determined that the homozygous variant enzyme did not affect catalytic function alone or in combination with the 677 TT genotype, and that it appeared thermostable and unaffected by changes in in vitro folate concentration. This is in contrast to the 60% decrease in activity observed in lymphocyte extracts with the homozygous variant genotype (57) This MTHFR 1298AC polymorphism was reported to affect 10% of the Canadian population with an allele frequency of 33%. Neither individuals with the CC variant nor individuals with the heterozygous variant genotype have higher homocysteine concentrations or lower folate concentrations (68,70-72) than normal individuals. The risk appears when the 677CT variant is combined with the 1298AC variant. Individuals heterozygous for both polymorphisms have been reported to have reduced MTHFR activity based on observed reductions in folate concentration and increased homocysteine concentration in population groups (68). An association between vitamin B12 status and the 677CT and 1298AC MTHFR genotypes was recently reported by Bailey et al. (73). Individuals with the homozygous MTHFR 677CT variant had higher plasma homocysteine concentrations than any other genotype combination, including those heterozygous for both mutations. They reported a negative association between plasma vitamin B12 concentration and serum folate concentration with plasma homocysteine concentration, which was

PAGE 38

24 dependent on MTHFR genotype. Plasma homocysteine concentration significantly decreased as vitamin B12 concentration increased in individuals heterozygous for both MTHFR polymorphisms in contrast to the individuals with the normal genotypes whose plasma vitamin B12 concentration was not associated with changes in homocysteine concentration. This study was the first to report an inverse correlation between plasma vitamin B12 concentration and plasma homocysteine concentration in individuals heterozygous for the MTHFR 1298AC polymorphism even when vitamin B12 was within the normal range. The population frequency of the MTHFR 677CT and 1298AC polymorphisms in the MTHFR gene far exceeds the percentage required to be defined as a genetic polymorphism (i.e., > 1% of the population) (48) and has been reported to affect individual folate and homocysteine concentrations. Further study of these genetic variations in different populations may determine whether individuals with the homozygous variants require higher intakes of different vitamins. Folate, MTHFR, and homocysteine. When Kang et al. (49) tested the thermolability of MTHFR, they found that two of their six subjects who had the thermolabile enzyme had hyperhomocysteinemia and deficient plasma folate concentrations, while the other four subjects had normal homocysteine and folate concentrations. This is the first study to show a possible relationship between deficient folate status and hyperhomocysteinemia in subjects with the thermolabile enzyme. Several subsequent studies have provided data that individuals with the TT genotype have higher homocysteine concentrations than the individuals with the CT or CC genotypes (65,74-83). Frosst et al. (3) concluded that the MTHFR TT genotype is the

PAGE 39

25 most important genetic determinant for moderate hyperhomocysteinemia. Jacques et al. (7) clarified that hyperhomocysteinemia associated with the MTHFR 677CT variant is dependent on folate status, and results from Kauwell et al. (84) support this conclusion. Folate status also has been shown to be impaired in individuals with the TT genotype compared to controls (67,77,85). Specifically, pregnant and nonpregnant women homozygous for the MTHFR 677CT variant have been found to have lower red cell folate concentrations than women with the CC genotype (86). Kauwell et al. (84) demonstrated using a controlled folate dietary protocol with elderly women that subjects with the TT genotype had significantly lower folate concentrations and higher homocysteine concentrations after 7 wk of depletion than subjects with the CT or CC genotypes. Subjects with the TT genotype had the greatest reduction in homocysteine concentration after 7 wk of folic acid supplementation with 415 g/d than subjects with the CT or CC genotype. Guinotte et al. (87) used a similar dietary protocol in a metabolic study of Mexican American women. Subjects with the TT genotype had lower serum and red blood cell folate concentrations throughout depletion (7 wk) and repletion (7 wk) with 400 g DFE/d. Homocysteine concentrations did not differ between genotypes throughout depletion, but subjects with the TT genotype had higher homocysteine concentrations throughout repletion with 400 g DFE/d. Ashfield-Watt et al. (81) examined the influence of the MTHFR 677CT genotype on folate status response in adults aged 18 to 65 who were given dietary intake advice. Subjects were assigned to one of three dietary intake advice groups for 4 mo each: advice to exclude folate-rich and fortified foods, advice to consume a folate-rich diet (400 g folate/d), and advice to take a daily folic acid supplement (400 g folic acid/d). Subjects

PAGE 40

26 with the TT genotype had the greatest increase in homocysteine and decrease in folate concentrations after 4 mo of folate exclusion. Folate supplementation with fortified foods or supplements were effective in lowering plasma homocysteine concentrations of subjects with the TT genotype to a significantly greater extent than that of the subjects with the CT and CC genotypes. Dietary advice to consume a folate-rich diet and advice to take a daily folic acid supplement effectively increased folate concentrations in subjects with the TT genotype above the baseline concentrations of subjects with the CC genotype, but the response was not significantly greater than that observed in the other genotype groups. The investigators recommended folate intakes between 400 and 600 g folate/d (575 to 830 g DFE/d) for individuals with the TT genotype to maintain normal homocysteine concentrations. Although this was not a controlled metabolic study, it demonstrated the beneficial effects of folate intake on folate status and homocysteine concentrations. In a recently published observational study, de Bree et al. (82) determined plasma folate and homocysteine concentrations by genotype in a Dutch population. They observed that subjects with the TT genotype had lower plasma folate and higher plasma homocysteine concentrations than subjects with the CT and CC genotypes. The difference in folate status indicators between MTHFR genotypes is well established in observational data. MTHFR and chronic disease. Kang et al. (49) were the first to suggest that the homozygous MTHFR 677CT variant may be a risk factor for cardiovascular disease. Since then conflicting results have been reported. Some studies have found a positive association between MTHFR, hyperhomocysteinemia, and vascular disease (74,75,88-92)

PAGE 41

27 while others have not found this association (76,93,94). Brattstrm et al. (65) conducted a meta-analysis to determine the relationship between cardiovascular disease risk and MTHFR genotype. They reported that although individuals with the TT genotype had an average 25% higher homocysteine concentration than individuals with the CC genotype, the variant is not associated with an increase in cardiovascular disease risk. A possible reason for the lack of association between cardiovascular disease risk and individuals with the TT genotype is that there was insufficient power in the individual studies. Chen et al. (95) developed a MTHFR knockout mouse model to evaluate the in vivo pathogenic mechanisms of MTHFR deficiency. Plasma homocysteine concentrations were 1.6-fold and 10-fold higher in heterozygous and homozygous knockout mice compared to controls. Both the heterozygous and homozygous knockout mice had significantly decreased SAM concentrations and/or increased SAH concentrations and abnormal lipid deposition in the aorta. Homozygous knockout mice were developmentally retarded with cerebellar pathology. Although this study evaluated the effect of MTHFR deficiency in mice, it is reflective of possible abnormalities in humans with the TT genotype. Only one study has linked homozygosity for the 1298AC variant with a higher risk for early-onset coronary artery disease independent of homocysteine concentration (94). Rothenbacher et al. (96) did not find an association between the MTHFR 1298 CC genotype and risk for coronary heart disease. Most other studies suggest that this polymorphism is benign unless it is combined with the homozygous MTHFR 677CT variant (97).

PAGE 42

28 MTHFR and cancer. Various research groups have evaluated the effect of genetic interactions and cancer. Chen et al. (98) reviewed their previously published studies with respect to MTHFR and colorectal cancer. They found an inverse association between the homozygous TT variant and risk of colorectal cancer in two different case-control studies conducted in the Health Professionals Follow-up Study (99) and Physicians Health Study (100). This protective effect was diminished by a high alcohol and low methionine intake. Ma et al. (100) found that there was no protective effect if folate status was low, but if folate status was adequate, risk for colorectal cancer was reduced by 50% in individuals with the TT genotype versus the CC genotype. Toffoli et al. (101) reported a significantly reduced risk of developing proximal colon cancer in Italian subjects with the TT genotype compared to the CC or CT genotypes. Other studies only have observed a protective effect in subjects with the TT genotype with high plasma folate concentrations compared to subjects with the CC or CT genotypes with low plasma folate concentrations (102,103). Studies also have shown a protective effect of the homozygous TT mutation related to acute lymphocytic leukemia (ALL) in children and adults (104-106). In contrast, the homozygous TT variant appears to be associated with an increased risk for breast (107-110), endometrial (111), gastric (112,113), and esophageal (114) cancers in certain populations, although folate status of the subjects was not reported in any of the studies. Few studies have investigated the influence of the MTHFR 1298AC polymorphism on cancer risk. Chen et al. (115) reported that subjects with the homozygous variant (CC) had a decreased risk for colorectal cancer compared to the normal genotype (AA), but it was a less substantial independent risk factor than the

PAGE 43

29 MTHFR 677CT polymorphism. Sharp et al. (109) investigated the role of the MTHFR 1298AC polymorphism on breast cancer risk and observed a reduced risk in individuals with the CC genotype compared to the AA genotype. No effect was reported for the CC genotype on gastric cancer risk (113). Matthews et al. (59) hypothesized that the reason a potentially harmful polymorphism is so common in humans is this protective effect against colon cancer and certain ALLs. The reason for this protective effect is hypothesized to be that under low folate conditions, the polymorphism decreases the flux of 5,10-methyleneTHF into 5-methylTHF and instead this coenzyme can be used to convert dUMP to dTMP, which is the rate limiting step in cell synthesis (99). Crott et al. (116,117) tested this hypothesis in vitro and did not find a significant decrease in uracil misincorporation (116) or micronuclei formation (117) in human lymphocytes taken from individuals with the TT genotype compared to those with the CC genotype. The investigators argued that their results are due to differences between in vitro and in vivo conditions. Folate, MTHFR, and neural-tube defects (NTDs). Neural tube defects are congenital abnormalities resulting from the malformation of the brain and/or spinal cord or from a failure of the skeleton to cover them, resulting in a protrusion (118). Development of these malformations occurs prior to day 28 of gestation, before many women know they are pregnant. Most groups studying NTDs generally only include anencephaly (failure of the brain to develop) or spina bifida (exposure of the spinal cord due to defective closure of the neural tube), but NTDs also include encephalocele, craniorachischisis, and iniencephaly. Each year anencephaly or spina bifida occur in 1 in 1,000 pregnancies in the US and roughly 300,000 or more births worldwide (119). This

PAGE 44

30 section will focus on key studies linking folate status, MTHFR, and homocysteine with NTD risk. During the 1960s Hibbard suggested that folate deficiency might be a cause for congenital birth defects (120). This hypothesis was supported by many subsequent studies. Smithells et al. (121) and Laurence et al. (122) conducted studies to evaluate the possible protective effect of folic acid supplementation against NTDs in high-risk mothers who already had an NTD-affected pregnancy. In both studies, the risk for a subsequent NTD-affected pregnancy was significantly decreased in the supplemented compared to that of the nonsupplemented mothers. These preliminary findings prompted a large randomized placebo-controlled intervention trial to determine whether it was folic acid alone that reduced the risk of NTDs. The Medical Research Council (MRC) Vitamin Study launched this trial in 1983 and ended it early because they found that 72% of NTDs were prevented with 4 mg folic acid supplementation daily in women who had a previous NTD-affected pregnancy (123). Czeizel and Dudas (124) found that periconceptional folic acid supplementation (800 g/d) could prevent the first occurrence of NTDs in a Hungarian population group compared to subjects receiving no folic acid. Berry and colleagues (125) investigated the effect of periconceptional folic acid supplementation in Chinese women from areas with high rates of NTDs (northern region of China) and low rates of NTDs (southern region of China). They observed that 400 g/d of folic acid reduced the occurrence of NTDs from 4.8 to 1.0 per 1000 pregnancies in the northern region of China and 1.0 to 0.6 in the southern region of China. Kirke et al. (126) evaluated blood samples from 56,049 pregnant women on their first clinic visit. Of this group 81 had NTD-affected babies and their blood folate concentrations were

PAGE 45

31 compared to 247 control females with normal pregnancy outcomes. The median plasma vitamin B12 (pmol/L), serum folate (ng/ml), and red blood cell folate (ng/ml) concentrations in the mothers with NTD-affected pregnancies compared to the control mothers were 243 and 296, 3.5 and 4.6, and 269 and 338, respectively. They concluded that both vitamin B12 and folate were independent risk factors for NTDs. Using data from the Kirke et al. (126) study, Daly et al. (127) were able to stratify the serum and red blood cell folate concentrations into quintiles. They found a greater than 8-fold increase in risk of NTDs in those women with red blood cell folate concentrations < 150 ng/ml versus those with red blood cell folate concentrations > 906 ng/ml. They concluded that NTD risk is reduced as red blood cell folate concentrations increase even in the normal range. Moore et al. (128) observed that NTD risk declines in a dose-responsive manner according to supplemental folic acid intake, dietary folate intake, and total folate intake. They reported that for every additional 500 g DFE/d consumed, the prevalence of NTDs decreased by 0.78 cases per 1000 pregnancies. In addition, compared to women in the lowest quintile of folate intake (0 to 149 DFE/d), the prevalence of NTDs decreased by 34%, 30%, 56%, and 77% among the offspring of women consuming 150 to 399, 400 to 799, 800 to 1199, and 1200 DFE/d, respectively (128). This strong link between NTDs and folate has prompted investigations to determine whether polymorphisms of genes that code for enzymes involved in folate metabolism have an effect on NTD risk. van der Put et al. (15) were one of the first groups to associate the MTHFR 677CT variant with NTD risk. Botto and Yang (4) provided an extensive review of case-control studies evaluating the association between the homozygous MTHFR 677CT variant and risk for spina bifida. Shields et al. (16)

PAGE 46

32 obtained blood samples from a large number of NTD-affected Irish individuals and their parents to determine if there was any genetic association with NTDs. They concluded that the homozygous TT genotype in embryos is the most important risk factor for NTD pregnancy outcome in this population. They also stated that the homozygous TT genotype might not play such a large role in other populations with more adequate folate status. Despite the many positive associations found between individuals with the TT genotype and NTD risk, not all study results support this correlation (129-133). Lucock et al. (133) hypothesized that another genetic insult of folate metabolism may underlie the condition. Recently, Rampersaud et al. (134) reported a significantly higher frequency of the TT genotype in 175 Caucasian NTD-affected subjects compared to controls, but the investigators suggested that an additional gene may be responsible for an increase in NTD risk. After careful consideration of published data related to the MTHFR 677CT polymorphism and NTD risk, the IOM concluded that only approximately 15% of NTD risk is attributable to the homozygous variant of MTHFR (27). Molloy et al. (85) investigated the role of maternal MTHFR TT genotype on the risk of having an NTD-affected pregnancy. Although they found a greater frequency of mothers of affected children with the TT genotype than control mothers, they did not find an effect of the TT genotype on red blood cell folate concentration. Therefore they concluded that maternal risk factors for NTDs are not explained by the TT genotype for MTHFR. Instead, maternal folate status may be the most important risk factor. Few studies have evaluated the association between the 1298AC polymorphism and risk of NTDs (4). van der Put et al. (68) did not find a significant risk for NTDs in individuals with the 1298AC homozygous or heterozygous variant genotypes;

PAGE 47

33 however, they observed a significantly increased risk for individuals with combined heterozygosity for the 1298AC and 677CT polymorphisms. They concluded that this combined heterozygosity accounts for a proportion of NTDs not explained by the TT genotype and should therefore be considered a risk factor for NTDs. Parle-McDermott et al. (135) did not find an association between NTDs and the MTHFR 1298AC polymorphism. Because most of the MTHFR and NTD studies have focused on living infants and parents, Isotalo et al. (136) collected fetal tissue samples from spontaneously aborted and terminated pregnancies to compare genotypes to neonatal umbilical cord samples for controls. They found all possible genotype combinations including the 677CT/1298CC and 677TT/1298CC genotypes in the fetal tissue, which were absent in the neonatal control group. The presence of NTDs could not be determined in the fetal tissue group, but due to the presence of the 677CT/1298CC and 677TT/1298CC genotypes in fetal tissue, they concluded that the combined MTHFR variants contributed to decreased fetal viability. Volcik et al. (137) found a higher rate of the homozygous 1298AC variant in mothers of an NTD-affected pregnancy in a Hispanic population, although it was not significant. Their results also contrast those of Isotalo et al. (136) in that they identified all possible genotype combinations in living individuals, including the 677CT/1298CC and 677TT/1298CC genotypes. It is possible that ethnic differences between the populations studied resulted in different genotypes in the population and, therefore, differences in results between the two studies. Homocysteine concentrations also have been found to be higher in mothers of NTD-affected infants (138,139), but the mechanism is unknown. Based on epidemiological studies, Eskes et al. (140) and Steegers-Theunissen et al. (141)

PAGE 48

34 hypothesized that homocysteine may have a direct teratogenic effect on the embryo. To test this hypothesis, Rosenquist et al. (142) treated avian embryos with incremental physiologic doses of homocysteine and observed the development of NTDs. They concluded that homocysteine is in fact teratogenic to the embryo. In contrast, homocysteine exposure to mammalian embryos (rat or mouse) did not increase the incidence of NTDs (143-145). Greene et al. (145) concluded that elevated homocysteine concentrations in NTD-affected pregnancies are more likely a marker of abnormal folate or methionine metabolism. More recently, high homocysteine concentrations have been found in NTD-affected individuals with the TT genotype. Bjorke-Monsen et al. (130) found significantly higher homocysteine concentrations in NTD-affected patients versus controls, but this increase was confined to patients with the CT or TT genotype. Wenstrom et al. (146) collected amniotic fluid from NTD-affected pregnancies and controls and found significantly more patients with TT genotypes in the NTD-affected fluid versus controls. They also found significantly higher amniotic fluid homocysteine concentrations in fluid from cases compared to controls. High homocysteine concentrations due to a polymorphism in the gene coding for MTHFR may be a risk factor for NTDs (139,147). The association of folate status, MTHFR polymorphisms, and homocysteine concentrations with NTD risk has been widely studied. Although results have been conflicting, the general opinion is that each factor plays a role in the occurrence of NTDs with folate status being the greatest contributor to risk. MTHFR polymorphisms may affect folate status and increase homocysteine concentrations resulting in an increased risk associated with the MTHFR polymorphism. Future studies need to further evaluate

PAGE 49

35 this possibility in order to achieve the goal of preventing the maximal number of folic acid preventable NTDs. Folate, MTHFR, and fetal malformations. Although there is significant evidence indicating a protective effect of folic acid supplementation in the prevention of NTDs, studies are beginning to link poor folate nutriture with other fetal malformations. Folate is involved in DNA synthesis and methionine production, both of which are crucial for normal embryonic development. Low folate status may contribute to embryonic malformations for the following reasons: decreased cell division, homocysteine-associated vascular events, impaired maternal-to-fetal folate transfer, or homocysteine-associated neurotoxicity, as reviewed by Moyers and Bailey (97). An estimated 120,000 to 150,000 infants with fetal malformations are born in the United States yearly (97). This section will focus on the possible role folate status and MTHFR polymorphisms may have on the risk for cleft lip and palate, congenital heart defects, and Down syndrome. Cleft lip with or without cleft palate is one of the more common malformations seen at birth and occurs in every 1/500 to 1/1000 births (148). Early studies documented the beneficial effect of high dose folic acid supplementation (10 mg) in the reduction of recurrent cleft lip with or without cleft palate (149). Two to 3 mg of folic acid also have been shown to significantly protect against cleft lip with or without cleft palate in a Hungarian population (150). The dose of folic acid used is important because 0.8 mg of folic acid in a multivitamin was not found to be protective against cleft lip with or without cleft palate in a Hungarian population (124,151). Hayes et al. (152) did not find a protective effect of periconceptional folic acid supplementation of mothers against the

PAGE 50

36 risk of oral clefts in infants in a North American population. The conflicting data on whether or not folic acid supplementation confers a protective effect against oral clefts in the North American population are reviewed by Moyers and Bailey (97). Wong et al. (153) determined the prevalence of hyperhomocysteinemia in mothers of infants with orofacial clefts and concluded that maternal hyperhomocysteinemia is a risk factor for having a child with an orofacial cleft. Mills (154) listed some possible limitations to the study by Wong et al. (153), including unexpectedly lower homocysteine concentrations in the controls compared to cases, differences in vitamin status between cases and controls, and the vitamin status of the population during the study compared to pregnancy. Mills (154) also listed some possible explanations including a metabolic defect in vitamin B6 and folate metabolism. Additional studies are needed to further define the relationship between maternal homocysteine concentrations and orofacial clefts. Because MTHFR is directly involved in folate metabolism, there may be an association between the homozygous MTHFR 677CT variant and cleft lip with or without cleft palate. An Irish group determined the prevalence of the TT genotype in their population with cleft lip with or without cleft palate. They found that the homozygous 677CT variant was significantly more prevalent in an Irish population with cleft palate only; however, periconceptional maternal vitamin use was not reported (155). In contrast, Gaspar et al. (156) did not find an increased prevalence of individuals with the TT genotype in their cleft lip patients with or without cleft palate. The presence of the TT genotype in mothers of children affected with cleft lip with or without cleft palate also has been shown to be significantly more prevalent when compared to controls

PAGE 51

37 (157). A subsequent study reported a significantly higher frequency of the TT genotype in mothers with affected children who were also affected themselves compared to healthy mothers with affected children (158). In addition, Shotelersuk et al. (159) did not observe an association between patient genotypes and occurrence of cleft lip or palate, but did detect a significantly higher frequency of compound heterozygotes (MTHFR 677CT and 1298AC) in mothers of patients. In a series of studies, an American group determined that infants with the TT genotype are not at increased risk for cleft lip (160) or cleft palate (161). Neither study indicated an interaction between infant genotype and maternal multivitamin use on the malformation occurrence. Reanalysis of the data by Wyszynski et al. (148) revealed significant differences in the risk of clefting between patients with the TT compared to the CC genotypes in patients whose mothers were not supplemented. The effect of the TT genotype of patients with supplemented mothers decreased the risk for orofacial clefts slightly, but not significantly. They concluded that based on their reanalysis of the data that periconceptional vitamin supplementation may protect against clefting (148). In another series of studies in an American population, no association was observed between familial (162) or isolated (163) nonsyndromic cleft lip and palate and the TT genotype. The relationship between cleft lip with or without cleft palate and MTHFR needs further study. Birth defects involving the heart include transposition of the great arteries, conotruncal heart defects, atrial septal defects, and others. Conotruncal heart defects are one of the more prevalent defects and occur in every 4/10,000 births. They are disorders that involve the neural crest cells that are ultimately incorporated into the aortic arch vessels, truncal outflow tract, and vessel walls, as reviewed by Moyers and Bailey (97).

PAGE 52

38 There are three main observational studies that provide evidence regarding the relationship between folate and heart defects (150,164,165). Briefly, high doses (2-3 mg) and low doses (i.e., multivitamin or fortified cereal) of folic acid supplementation significantly reduced the risk for heart defects in one Hungarian and two American studies. Boot et al. (166) attributed the development of conotruncal heart defects to the abnormal differentiation of neural crest cells in the presence of high homocysteine concentrations in a recently published in vitro study. Although the inverse association between folate supplementation and reduced risk for heart defects has been the focus of investigations, the role of MTHFR on conotruncal heart defect risk has been evaluated in only one study. Storti et al. (167) determined the MTHFR 677 and 1298 genotypes in an Italian population of affected children and their parents. Although increased odds ratios for heart defects were detected for different combinations of MTHFR polymorphisms in mothers and affected children, none of the odds ratios were significant. More studies evaluating the role of MTHFR polymorphisms on the risk of conotruncal heart defects are needed. Down syndrome is a genetic disorder that results from the presence of three copies of chromosome 21 (trisomy 21). This extra chromosome is a result of abnormal chromosome segregation during meiosis, also known as meiotic nondisjunction. Ninety-five percent of Down syndrome cases are maternal, with nondisjunction occurring in the oocyte before conception. Down syndrome occurs in every 1/600 live births and in 1/150 conceptions. It is estimated that approximately 80% of all trisomy 21 conceptions results in spontaneous abortion. It is a major public health concern and is the leading cause of mental retardation (168). James and colleagues (168) hypothesized that altered folate

PAGE 53

39 metabolism in mothers of affected children with the TT genotype may affect their DNA methylation and result in nondisjunction that leads to Down syndrome. They found that the presence of the MTHFR 677CT variant on one or both alleles in these mothers significantly increased the risk of having a child with Down syndrome. Their results were supported by Hobbs et al. (169). Other studies have not reported an association between maternal MTHFR genotype and Down Syndrome (170-172). Differences in the ethnicity among population groups may have contributed to the differences in results. Controlled intervention clinical trials will need to be performed before preconceptional folic acid supplementation is recommended to reduce the risk of Down syndrome. The possibility of preventing major malformations with folic acid supplementation has directed the focus of research to the influence of folate status and MTHFR genotype on fetal malformations. More studies are needed to develop concrete evidence in support of folic acid supplementation as a preventive measure against fetal malformations. Homocysteine, MTHFR, and pregnancy outcome. Although the main focus of folate research involves NTDs and fetal malformations, evidence has been reported for the role of homocysteine and MTHFR polymorphisms in preeclampsia and early pregnancy loss. Preeclampsia and recurrent early pregnancy loss are very serious complications of pregnancy. Some recent findings for preeclampsia and early pregnancy loss will be discussed in this section. Preeclampsia is defined as pregnancy-induced proteinuric hypertension with onset of clinical symptoms beyond 20 wk gestation (173). The cause of preeclampsia is thought to be due to increased resistance to uterine artery blood flow (174) and has been associated with elevated homocysteine concentrations (139). Lpez-Quesada et al. (175)

PAGE 54

40 observed a 7.7-fold increased risk for preeclampsia in pregnant women with hyperhomocysteinemia in the third trimester (> 10.5 mol/L) compared to normal pregnant controls. The role of homocysteine as a risk factor for vascular events has been well established (176). Because the homozygous MTHFR 677CT variant is associated with elevated homocysteine concentrations in individuals with low folate status (3), investigators have implicated this polymorphism in mothers as a risk factor for preeclampsia. There have been positive associations (174) and negative associations (177) found between mothers with the homozygous TT variant and the development of preeclampsia. Prasmusinto et al. (173) did not associate this polymorphism in mothers or infants with an increased risk for preeclampsia. Although reported data are not consistent, possible explanations for the variation include the differences in the population groups and subsequent variation in the frequency of the MTHFR polymorphism. Recurrent early pregnancy loss is another serious problem in pregnancy. Although still unresolved, the mechanism responsible for this pregnancy complication has been a topic of intense study. There is evidence to support the hypothesis that abnormal procoagulant activity may be a causative factor for early pregnancy loss (178). As mentioned previously, high homocysteine concentrations are associated with various cardiovascular diseases, including increased prothrombotic tendency (179). Since high homocysteine concentrations may promote thrombotic events, they also may play a significant role in early pregnancy loss. Nelen and colleagues (180,181) concluded that elevated homocysteine concentrations were a risk factor for recurrent early pregnancy loss in a case-controlled study (180) and a meta-analysis (181). Any factor that may

PAGE 55

41 increase homocysteine concentrations also may increase the risk for early pregnancy loss. Results from different studies have supported an association between the MTHFR 677 TT genotype and risk for recurrent early pregnancy loss (80,182). Zetterberg et al. (183) determined the genotype of spontaneously aborted embryos for the 677CT and 1298AC MTHFR polymorphisms and found a high prevalence of MTHFR polymorphic genotypes. Regardless of the population, homozygosity for the 677CT variant has been found to be a risk factor for early pregnancy loss (80,182,184). Other studies have been unable to confirm this association in different populations (184-186). More studies are needed on the 1298AC polymorphism to draw any conclusions. In summary, risks for preeclampsia and recurrent early pregnancy loss are additional examples of pregnancy complications that may be reduced with folic acid supplementation. The possible increased need for folate in individuals homozygous for the MTHFR polymorphisms needs to be substantiated further prior to recommending doses of folic acid for the prevention of these pregnancy complications. Methionine Synthase (MS) Another widely studied enzyme involved in folate metabolism is methionine synthase (MS), which catalyzes the conversion of homocysteine to methionine. This enzyme requires 5-methylTHF and cobalamin to function properly. Bacterial MS has been characterized and studied extensively (187). It consists of 3372 nucleotides and a molecular weight of 123,640. Escherichia coli MS studies have shown the enzyme to be a modular protein consisting of four different regions. The module residing on the N-terminal is the one responsible for binding homocysteine. The second module binds 5-methylTHF, the third binds cobalamin, and the fourth binds SAM (56). Leclerc et al.

PAGE 56

42 (188) were the first to clone human MS cDNA, which shares approximately 58% identity with Escherichia coli, and to describe specifics about its localization, expression, and partial characterization. They reported on a gene that codes for a protein containing 3795 base pairs encoding a polypeptide of 1265 amino acids located near the telomeric region of the long arm on chromosome band 1q43. They also identified an AG transition at base pair 2756 that converted an aspartic acid into a glycine in patients with MS activity deficiency. They hypothesized that this MS deficiency could be associated with mild hyperhomocysteinemia. MS and chronic disease. The role of the MS 2756AG variant in chronic disease has been investigated. Individuals with the homozygous 2756AG variant have been reported to have lower fasting plasma homocysteine concentrations (189-191). Jacques et al. (192) found no evidence for an association between homocysteine and the MS 2756AG variant. van der Put et al. (193) sequenced the entire coding region of MS in eight individuals with hyperhomocysteinemia (four NTD patients and four vascular disease patients) to determine whether mutations in this gene were involved in homocysteine-related diseases. They reported that there was no association between the MS 2756AG substitution and hyperhomocysteinemia in their population. In fact, they also detected a slightly lower homocysteine concentration in patients homozygous for the variant compared to heterozygotes, as described previously. It was hypothesized that when the strong helix breaker glycine was present in the enzyme instead of the moderate helix breaker aspartic acid at the position near the vitamin B12 binding site, enzyme function may be modified. In contrast, Harmon et al. (194) reported that this polymorphism was associated significantly with homocysteine concentrations in their

PAGE 57

43 population of Irish males, with the AA genotype having higher homocysteine concentrations. Silaste et al. (190) investigated the effect of two 5 wk dietary interventions of low and high folate intake to determine whether genetic variations in enzymes involved in homocysteine metabolism affect the responsiveness of folate status to naturally occurring food folate. They reported that individuals with the G allele for the MS variant had a greater decrease in homocysteine concentrations and lower homocysteine concentrations during the high folate period than individuals with the normal enzyme. Since hyperhomocysteinemia also is a risk factor for venous thromboembolism (VTE) an association with the MS 2756AG variant was investigated. den Heijer et al. (194) and Salomon et al. (195,196) did not find and association between the GG variant and VTE. In contrast, Yates and Lucock (197) reported a protective effect for the presence of the G allele in relationship to risk for VTE. Although the association of the MS 2756AG variant and cancer has been investigated, no significant correlation was found (198). Swanson et al. (199) developed a MS knockout mouse model to study the pathophysiology associated with a severe MS deficiency. Heterozygous knockout mice had a 50-60% decrease in enzyme activity and slightly elevated plasma homocysteine and methionine concentrations while complete omission of MS resulted in embryonic lethality, proving that MS activity is essential for early embryonic development in mice. MS and NTDs. The possible association between a polymorphism affecting MS and risk for NTDs has been the focus of investigations. van der Put et al. (193) did not find an increased prevalence of the MS 2756 GG genotype in NTD cases compared to

PAGE 58

44 controls or in mothers of children with NTDs compared to controls. Unlike the homozygous MTHFR 677CT variant, it is hypothesized that the MS 2756 GG genotype confers some sort of protection against NTDs compared to the normal enzyme. Christensen et al. (200) did not find the MS 2756 GG genotype in any of their cases, but 10% of their controls had the 2756 GG genotype. In contrast, Zhu et al. (201) observed a higher frequency of the G allele in NTD cases compared to controls. The GG genotype was detected in very low frequencies in cases. Doolin et al. (202) found an increased risk for spina bifida-affected pregnancy in mothers with the GG genotype compared to the AA genotype. They emphasized the importance of considering maternal and embryonic genotypes when evaluating risk for spina bifida. The role of the MS 2756AG polymorphism on NTD risk is not well defined and needs further investigation. Methionine Synthase Reductase (MSR) Associated with MS is MSR, a flavoprotein responsible for the reductive activation required for the maintenance of MS once cobalamin becomes oxidized over time. Leclerc et al. (203) isolated the cDNA clone for the human MSR gene and found that it had three binding sites for FMN, FAD, and NADPH, which are required for the reduction of MS. The gene consists of 2094 base pairs encoding a polypeptide of 698 amino acids with a molecular mass of 77.7 kDa. It is localized on human chromosome 5. Wilson et al. (204) discovered a variation in the gene coding for MSR in homocystinuric patients with severe MS deficiency. This polymorphism results in a AG transition at base pair 66, which converts an isoleucine to methionine in the protein. Individuals can have the homozygous variant (GG), heterozygous variant (AG), or normal (AA) genotype for this MSR 66AG variation.

PAGE 59

45 MSR and NTDs. The association between the MSR 66AG variant and risk for NTDs has been evaluated. Wilson et al. (205) found that cases and mothers of cases had almost a 2-fold increase in risk for spina bifida compared to controls, although not significant. These investigators observed that when low cobalamin status was coupled with the MSR GG genotype, it conferred an even greater risk for NTDs in the children and mothers. In addition, a combination of homozygous MTHFR 677CT and MSR 66AG variants in both children and mothers conferred a 4-fold increase in risk, a greater increase than for each individual variant. These results were supported by Zhu et al. (201) who observed that the G allele was associated with increased risk for spina bifida. Doolin et al. (202) reported that the risk of having a child with spina bifida increased with increasing maternal G alleles. These results link the homozygous MSR 66AG variant with an increased risk of spina bifida, but more studies are needed to make any concrete associations. Other Polymorphisms Although much of folate research focuses on the preceding polymorphisms, other polymorphisms have been identified that have not received as much attention. Devlin et al. (206) identified a polymorphism in the gene coding for the conjugase enzyme located in the intestine that results in 53% less activity than the normal enzyme. This enzyme termed GCPII is responsible for cleaving polyglutamates into monoglutamates. Glutamate carboxypeptidase II can have a CT transition at base pair 1561, which replaces histidine with tyrosine at codon 475, codes for a 750 amino acid polypeptide, and is located on chromosome 11. Devlin et al. (206) associated the 1561CT transition in GCPII with low serum folate concentrations and higher homocysteine concentrations

PAGE 60

46 in a healthy English population. More studies are needed to corroborate these findings because dietary intakes were not assessed at the time of sample collection, and recent intakes of folate, vitamin B6, vitamin B12 can affect folate and homocysteine concentrations. Any supplements taken by the subjects containing any of these B vitamins also could have had an effect. Vargas-Martinez et al. (207) concluded that the polymorphism is not associated with plasma folate or homocysteine concentrations after determining the GCPII 1516CT genotype in subjects from the Framingham Offspring Study. They took many factors into account known to affect folate and homocysteine concentrations and, after adjusting for these factors, still did not find an association. A possible explanation for variations between these studies are the factors accounted for by Vargas-Martinez et al. (207). Vargas-Martinez et al. (207) concluded that this polymorphism has no effect on folate and homocysteine concentrations when confounding factors are taken into account. Fdinger et al. (208) observed that GCPII was a predictor of red blood cell folate but not homocysteine concentrations in chronic dialysis patients. Other polymorphisms that may have an impact on folate status are the thymidylate synthase promoter region polymorphisms (209,210) and folate receptor polymorphisms (37,211). Multiple studies have illustrated the complexity of the problem of how a genetic variation associated with different diseases and birth defects can be modified by nutrient intake. A perfect example of this comes from the previously discussed study by Jacques et al. (7), which reported that homocysteine concentrations were higher in individuals with the homozygous MTHFR 677CT variant only when plasma folate concentration was below the median. There was no difference in homocysteine between

PAGE 61

47 individuals with the TT and CC genotype with plasma folate concentrations at or above the median. Adequate folate status also can provide protection against certain cancers in individuals with the TT genotype, while inadequate folate status can add to the risk for these cancers (100). DNA Stability For years scientists have known that exposure to chemical mutagens and carcinogens can cause gene mutations and chromosome damage. Only recently has there been more and more evidence linking dietary factors to similar damage (212). Currently, a large body of evidence supports the conclusion that folate deficiency has a negative impact on DNA stability. The proposed mechanisms by which folate deficiency impairs DNA structure include: DNA methylation, uracil misincorporation, DNA strand breaks, and micronuclei formation (11,213-215). This section will summarize research findings related to each mechanism to provide a basis for understanding the links between folate status and DNA stability. DNA Methylation Until recently, not much was known about gene regulation. It is now recognized that it is not just the primary sequence of the DNA that determines whether or not a gene is expressed. Gene transcription depends on modifications referred to as epigenetics, which is defined as any DNA modification that regulates gene activity without changing the primary DNA sequence and can persist through one or more generations (216). It is these epigenetic modifications that allow cells to adapt to external stimuli from the environment and from regulatory molecules within the body without having to change the primary DNA sequence (216). DNA methylation is one epigenetic modification that is critical for normal development of cells and organs (217). Li et al. (218) reported that a

PAGE 62

48 mutation of a key methylating enzyme, Dnmt1, produced a recessive, lethal phenotype. A few years later, Okano et al. (219) confirmed these findings in a study that targeted two other DNA methyltransferases, Dnmt3a and Dnmt3b. Deletion of Dnmt3a produced mice that appeared normal at birth but failed to grow normally and died after 4 wk. In contrast, deletion of Dnmt3b did not produce any viable mice at birth. Deletion of both DNA methyltransferases produced an embryonic lethal phenotype. These series of studies illustrate the importance of methylation for the normal development of mammals. Before the function of DNA methylation can be discussed, the specifics of methylation will be reviewed briefly. As reviewed by Costello and Plass (217), the majority of 5-methylcytosine is present in cytosine-guanine (CpG) dinucleotides. Clusters of CpGs also exist and are referred to as CpG islands. Gardiner-Garden and Frommer (220) originally defined these CpG islands as regions of DNA from 200 base pairs to several kilobases in length with a CpG frequency approximately five times greater than the whole genome. They also determined that these CpG islands comprise 1 to 2% of the genome. It is estimated that 3 to 6% of total mammalian cytosine bases are methylated and that 70% of mammalian CpG dinucleotides are methylated (8). The exact mechanism that determines the pattern of DNA methylation is unknown. Methylation assays using polymerase chain reaction amplification have shown that mouse sperm DNA is primarily methylated at all non-CpG island sites located throughout the genome in contrast to mouse oocyte DNA, which is primarily unmethylated at these same nonCpG island sites (221). Early in the life of the embryo, the CpG sites of germ and somatic cells are demethylated and then de novo methylation reestablishes a methylation pattern, although differently in germ cells versus somatic cells (221). Different

PAGE 63

49 mechanisms can change DNA methylation patterns during different developmental stages. The spontaneous deamination of a methylated cytosine produces thymine, which can be replicated and result in a TG base pair mismatch and subsequent TA transition. This transition is more difficult for the cell to repair (222) and can result in the loss of heritable methylation patterns (8). There are three DNA methyltransferases that have been discovered which transfer a methyl group from SAM to the cytosine in CpG dinucleotides: Dnmt1, Dnmt3a and Dnmt3b (223). These enzymes are necessary for embryonic development (218,219) and possess de novo methylation capabilities (i.e., methylation of a CpG sequence opposite an unmethylated CpG sequences) (224). Dnmt1 has a higher affinity for hemimethylated DNA and therefore works mostly as a maintenance enzyme (225). The exact mechanism as to how the 5-position of cytosine becomes methylated is still unknown. It is hypothesized that cytosine is everted from the DNA helix and inserted deep within the active site of the DNA methyltransferases where a methyl group is transferred (226). The mechanism of DNA methylation involves the methylated parent strand as a guide. The daughter DNA strand is methylated by one of the maintenance DNA methyltransferases shortly after replication, resulting in the exact methylation pattern present in the parent strand. DNA methylation patterns are preserved throughout many rounds of replication by DNA methyltransferases (8). DNA methylation has many functions. Stabilization of DNA by preventing cleavage by nucleases is one proposed function for DNA (8). The primary function, however, is as a gene silencer. Genes make up a small portion of the genome, while the rest is made up of introns, repetitive elements, and potentially active transposable

PAGE 64

50 elements. In order for genes to be successfully expressed, all of the noncoding DNA needs to be silenced. Mammals appear to have evolved to use methylation as a mechanism to silence this DNA. The post-transcriptional addition of a methyl group to the 5-position of cytosine alters the DNA-protein interactions, which in turn keeps the DNA from being transcribed (9). CpG islands often contain the promoter regions near the 5 end of genes. If these promoter regions are unmethylated, this denotes active transcription (227). If they are methylated, transcription is suppressed. As technology has improved, scientists have determined that CpGs located within these CpG islands are mostly unmethylated while CpGs located outside these islands are generally methylated (217). Costello and Plass (217) suggested that these patterns of methylation may act to separate the genome into transcriptionally active and inactive areas. These specific methylation patterns are upheld through DNA replication in order to promote and maintain the transcription of specific genes (228). Methylation within these promoter regions can stop transcription. Cooney et al. (229) reported that dietary methyl group supplementation of rats has a significant effect on DNA methylation and their subsequent methylation-dependent phenotype by successfully changing the coat color of the yellow agouti mouse. Kass et al. (228) reviewed how methylation in the promoter region represses transcription. The most obvious mechansim would be to prevent the transcription factors and proteins from binding to the DNA. However, this cannot be the only explanation because there is transcription machinery that will bind to DNA despite its methylation. Jones et al. (230) reported that the binding of the methyl CpG binding complex 2 to methylated promoter regions recruits transcriptional repressors. These complexes contain

PAGE 65

51 histone deacetylases (HDAC1, HDAC2), which function to deacetylate lysine residues in histone tails, which are associated with DNA and result in the compaction of chromatin. Once the chromatin becomes compacted, it is transcriptionally inactive. Therefore, DNA methylation also has a repressive effect on chromatin, serves to inactivate one of the two X chromosomes in females during development, and determines the expression of imprinted genes (217). Similar machinery is used to establish methylation patterns. Robertson (231) reviewed two models for how DNA methylation patterns may be established in somatic cells. The first model incorporates the use of HDACs, ATP-dependent remodeling complexes, and DNA methyltransferases. Histones destined for silencing are first deacetylated by HDACs. The chromatin remodeling enzymes can now move the nucleosomes that are wrapped by DNA from side to side in an ATP-dependent manner to allow the DNA methyltransferases access to its target DNA sites. These enzymes also may create a specific area on the DNA that is recognized by the DNA methyltransferases. Once DNA is methylated, the methyl-CpG binding proteins bind the methylated cytosines and further repress transcription as described above. The second model involves the maintenance methylase (DNMT1) and the retinoblastoma protein, Rb, which is a protein involved in transcriptional regulation of chromatin (231). In resting cells, Rb is associated with DNMT1 and inhibits its catalytic activity to prevent any aberrant methylation of the genome. Early in cell division (S phase) both proteins colocalize with the replication foci. In late S phase Rb is no longer colocalized with the replication foci and, instead, HDAC colocalizes with DNMT1, which is now active. It is possible that Rb departs from the replication foci when hypermethylated regions are replicated (231).

PAGE 66

52 The main methyl donor used by the DNA methyltransferases is SAM. DNA methylation is dependent on the availability of SAM; therefore, anything that affects the supply of SAM may have an effect on DNA methylation. In humans, the main dietary sources of methyl groups that are transferred to SAM are folate, choline, and methionine (232). SAM is the main methyl donor in over 100 reactions. The methyl group requirement normally exceeds the supply available in food, but the difference is usually made up by the synthesis of methyl groups utilizing folate coenzymes (213). Most of the 5-methylTHF is regenerated through the one-carbon cycle illustrated in Figure 2-2, but a small amount is lost through urinary excretion, skin, bile, and catabolism. If this folate is not replaced, it could decrease the methylation cycle resulting in lower amounts of SAM available for methylation and increased homocysteine concentrations (233). Early studies investigating the role of folate in methylation used rats as a mammalian model. Balaghi and Wagner (213) found that after 4 wk of feeding a folate deficient diet to rats, methylation of hepatic DNA was significantly reduced in folate depleted rats compared to controls. Alonso-Aperte and Varela-Moreiras (234) observed that administration of the folate antagonist methotrexate in rats produced a folate deficiency compared to controls and was associated with significant hypomethylation of brain DNA. Kim et al. (235) investigated global and protooncogene specific (c-myc) DNA methylation. As a model for conditions preceeding colorectal neoplasia in rats and humans, the effect of moderate folate depletion over a longer period of time was evaluated. After 15 and 24 wk, the folate deplete rats had a significantly lower plasma folate concentration than the control rats. There was no significant difference in global methylation after 15 or 24 wk. They also did not find a significant difference in

PAGE 67

53 methylation of c-myc between the two groups after 15 or 24 wk of folate depletion. It was hypothesized that the SAM/SAH ratios were not low enough, the strain of rat used was resistant to hypomethylation, or the betaine pathway compensated for any folate loss and prevented significant hypomethylation. Kim et al. (214) investigated the effect of a more severe folate deficiency resulting from antibiotic treatment on global DNA methylation and strand breaks and methylation and strand breaks within a specific sequence of the p53 tumor suppressor gene. Folate deficient rats at 4 and 6 wk had significantly lower plasma folate concentrations than the controls and significantly more strand breaks within the p53 gene than control rats. They also had reduced DNA methylation of this gene that obtained significance at 6 wk. The investigators concluded that a dietary folate deficiency could have a negative effect within critical regions of the p53 tumor suppressor gene. Deficient rats fed for 6 wk also had significantly higher global strand break accumulation than controls and rats fed for 4 wk, indicating a time dependence of strand breaks. They did not find a significant difference between global methylation between the two groups at 4 or 6 wk, strengthening their previous observations. Although the preliminary rat studies were important in making a connection between folate status and DNA methylation, the links to clinical implications in humans were unclear. Studies correlating low folate status in humans with increases in micronuclei formation and uracil misincorporation were reported, which supported using them as new folate status indicators in addition to the traditional blood folate values (11,236). Jacob et al. (237) evaluated DNA methylation as a potential new biomarker for folate status. This study was designed to assess the correlation between impaired folate

PAGE 68

54 status and DNA methylation in humans. Ten, healthy postmenopausal women lived in a metabolic unit for the 13 wk of this study and consumed a low folate diet and varying amounts of synthetic folic acid to provide intakes ranging from 56 to 516 g/d during this depletion-repletion study. Days 6 to 41 were designed to provide a moderately-deplete folate diet to evaluate the effect of low folate intakes not associated with overt clinical signs of deficiency. Mean plasma folate concentrations dropped significantly from baseline to post-depletion. Lymphocyte DNA hypomethylation was determined using a methyl acceptance assay, whereby acceptance of [ 3 H]methyl groups is inversely associated with methylation. The investigators observed that the marginal folate deficiency induced in these postmenopausal women was associated with reduced DNA methylation, which was reversed with folate supplementation. This was the first study to show that folate intake affects DNA methylation in humans and reflects the results seen in the aforementioned rat studies (213,214). Rampersaud et al. (238) investigated the effects of controlled folate intake on global genomic DNA methylation in leukocytes of elderly women. Thirty-three healthy, elderly women consumed a depletion-repletion diet consisting of a low folate diet (118 g/d) for 7 wk and repletion with 200 or 415 g/d for 7 wk. Blood samples were taken weekly and leukocyte DNA methylation was determined using the methyl acceptance assay (213). Moderate folate depletion in elderly women was severe enough to be associated with increased [ 3 H]methyl incorporation in vitro, which reflects decreased methylation in vivo. Decreased methylation was evident in these women at wk 7, when they were also found to have the lowest folate status. DNA methylation did not increase after 7 wk of repletion with 200 or 415 g folate/d, which suggests that the repletion period may not have been not long

PAGE 69

55 enough to increase the methylation. The results from this study show that low folate status may significantly reduce DNA methylation and support the use of DNA methylation as a functional folate status indicator. The investigators stressed that results related to DNA methylation status within specific cells may not relate to whole body methylation. Epidemiological data also support a correlation between folate status and DNA methylation. Fowler et al. (239) found a significant, inverse correlation between cervical tissue folate concentration and DNA methylation and serum folate concentration and DNA methylation. An inverse correlation between folate status and DNA methylation has not been found by all investigators. Fenech et al. (240) measured the folate and lymphocyte DNA methylation status of young Australian men and women. Volunteers were required to consume their normal diets and were assigned to eat a bowl of fortified or unfortified bran cereal once/d plus a placebo or a vitamin supplement for 24 wk. The folate group received 2700 g folic acid/d and 27 g vitamin B12/d (cereal + tablet), while the placebo group received no folic acid. No significant differences in DNA methylation were detected after treatment relative to baseline for either group or any correlations between DNA methylation and folate status. This is not surprising because this was not a folate depletion protocol, and all volunteers had normal status throughout the study. It is only when folate status is impaired that the availability of methyl groups decreases and there is a decrease in genomic methylation (233). The effect of folate on methylation and subsequent transcription was evaluated in cell culture experiments. Jhaveri et al. (241) performed experiments with human nasopharyngeal epidermoid carcinoma KB cells to determine whether folate deplete

PAGE 70

56 media affects the transcription of genes. They found that eight genes responded to a variation of folate concentrations. Three genes were upregulated and five genes were downregulated in cells grown in folate deficienct media. H-cadherin, a protein involved with cell adhesion molecules, was one of the genes down regulated in folate deficient media. This down regulation was associated with hypermethylation of the CpG island that contained the promoter region. These data indicated that decreased folate positively and negatively influences the expression of certain genes, so that folate deficiency affects specific genes rather than global methylation. Regardless of whether experiments were in vitro or in vivo, results support a role for folate in DNA methylation and transcription. Methionine is an essential amino acid that is the immediate precursor to SAM in the methionine cycle and is required for protein synthesis (242). It is well known that methyl deficient diets (i.e., choline and methionine deficient) can cause global hypomethylation in rats (243,244) and that it can occur within 1 wk of starting the diet (244,245). Choline deficient diets also have been reported to induce significant hypomethylation of brain DNA compared to controls (234). SAM/SAH regulation of DNA methylation. Once SAM is used as a substrate for methylation, SAH is formed within the active site of the methyltransferase enzyme. Most methyltransferases have a higher binding affinity for SAH than SAM; therefore, excess SAH can result in strong product inhibition of these methyltransferases, which may lead to decreased DNA methylation (246). In a review by James and colleagues (247), three defense mechanisms against toxic SAH concentrations are discussed. They suggest that SAH can be bound to proteins, exported into the plasma, or hydrolyzed by SAH hydrolase, which degrades SAH into homocysteine and adenosine.

PAGE 71

57 In order for methyltransferases to work efficiently, SAH concentrations need to be regulated intracellularly, which is primarily accomplished by SAH hydrolase. Homocysteine can then be remethylated to form methionine or shuttled down the transulfuration pathway (248). Previous studies have implicated decreased availability of SAM as a limiting cofactor for methyltransferases, and therefore, decreased cellular methylation (249). The SAM/SAH ratio has been used to predict reduced cellular methylation, with SAM as the main effector (249). Yi et al. (248) estimated whether SAM or SAH had a greater impact on global DNA methylation in healthy, adult women. They found that an increase in homocysteine concentration in these women correlated with a significant increase in SAH concentration but had no relationship with SAM. A strong, correlation between homocysteine concentration and the SAM/SAH ratio was detected, and this decreased ratio was associated with an increased SAH concentration rather than decreased SAM concentration. Lymphocyte DNA hypomethylation increased significantly with increasing concentrations of SAH, but had no relationship with SAM. This study was the first to show that moderate elevations in homocysteine concentration are in fact associated with increases in SAH and decreases in lymphocyte DNA methylation. These investigators suggested that there might be another mechanism of homocysteine pathogenicity via SAH inhibition of DNA methyltransferases. Instead of limiting DNA methyltransferase activity and subsequently causing reduced DNA methylation, it was hypothesized that low SAM concentrations may instead affect DNA methylation by decreasing thymidine and purine synthesis via increased activity of MTHFR (248).

PAGE 72

58 A combined genetic and dietary approach in a mouse model was used by Caudill et al. (250) to investigate the effect of changes in plasma homocysteine and intracellular SAM, SAH, and the SAM/SAH ratio on global DNA methylation in different tissues. Mice were normal or heterozygous for CBS deficiency and were randomized into a methyl-deficient diet group or the control group for 24 wk. The combined results for different tissues indicated that an increase in SAH alone or in conjunction with a decrease in SAM was associated with a decrease in DNA methylation. A decrease in SAM alone was not sufficient to affect DNA methylation. In addition, a decrease in the SAM/SAH ratio was associated with reduced DNA methylation only when associated with an increase in SAH. The negative effects increased SAH concentrations and decreased DNA methylation may have on altered gene expression and chromatin formation have been reviewed (251,252). James et al. (247) hypothesized that SAH-mediated reduced DNA methylation as a result of increased homocysteine concentrations may increase DNA damage from homocysteine-induced free radicals. Further research is needed to determine whether this is a possibility. Genetic polymorphisms and methylation. Individuals homozygous for the MTHFR 677CT variant have diminished methylation capabilities. It has been shown that the distribution of folate forms was altered in the red blood cells of individuals with the TT genotype (6). These individuals had formylated folates in addition to methylated folates in contrast to individuals with the CC genotype who exclusively had methylated folates. Genotype did not affect total red blood cell folate content. This finding prompted Stern et al. (12) to use a methyl acceptance assay to evaluate whether this decrease in methylated folates affected DNA methylation in these individuals. They

PAGE 73

59 found that individuals with the TT genotype had significantly decreased DNA methylation compared to individuals with the CC genotype, and that this methylation was directly correlated with red blood cell folate concentration. This was the first study to report that the TT genotype may be associated with epigenetic alterations. These findings were supported by a second observational study by the same research group in a larger population. Friso et al. (13) evaluated the effect of folate status and MTHFR 677CT TT genotype on DNA methylation in an Italian population. They directly measured methylated cytosines using a liquid chromatography/mass spectrometry method. DNA methylation in individuals with the TT genotype was approximately 50% that of individuals with the CC genotype when plasma folate concentration was below the median for this population group. There was no significant difference in methylation between the individuals with the TT and CC genotypes when plasma folate concentration was above the median. An inverse relationship between DNA methylation and homocysteine concentration also was observed in this study. An important limitation of this study is that dietary folate intake was not controlled. These results have been corroborated using a MTHFR knockout mouse model. Chen et al. (95) found lower global DNA methylation in heterozygous and homozygous knockout mice compared to control mice. This body of evidence supports the conclusion that the homozygous MTHFR 677CT variant is associated with reduced DNA methylation when folate status is low. Methylation and cancer. Recent evidence has shown that cancer is a process that is modified by DNA mutations and epigenetic mechanisms (253). In normal cells, CpG islands are hypomethylated and located in the promoter regions of 40-50% of genes.

PAGE 74

60 Many cancer genes are being discovered that are hypermethylated in the promoter region (253). The role of hypermethylation of DNA in cancer involves the silencing of genes by hypermethylation of promoter regions. Although the information on hypermethylation is extensive, a review by Ehrlich (254) argues that there has been inadequate attention given to global hypomethylation of DNA in cancer. The evidence for global hypomethylation in carcinogenesis is considerable. Rats fed methyl deficient diets had hypomethylated liver DNA that was associated with increased mRNAs for protooncogenes (255). After 1 mo of consumption of a diet adequate in methyl groups, a reversal in methylation and protooncogene expression was observed (255). Rats fed methyl deficient diets also had hypomethylated p53 tumor suppressor genes (10,214), which could enhance tumor production. Sibani et al. (256) reported that reduced DNA methylation in preneoplastic intestinal cells was directly associated with tumor multiplicity, which was increased under low folate conditions. It was hypothesized that if hypermethylation and hypomethylation of DNA can be altered, there may be an opportunity to influence tumor production. Caution is warranted when interpreting data regarding hypermethylation and hypomethylation of DNA. These terms denote more or less methylation of DNA relative to some standard. When studying cancer, normal tissue is the standard (254). A different standard has to be developed for each tissue studied because methylation is species and tissue-specific. Cell types also have to be considered because tissues are comprised of a mixture of cells (254). As reviewed by Ehrlich (254), DNA hypomethylation has been found in leukemia, liver and prostate, and cervical cancer, and DNA hypermethylation has been found in colon, kidney, esophageal, and pancreatic cancer.

PAGE 75

61 DNA Strand Breaks Another measure of genomic stability is the number of strand breaks in the DNA. Strand breaks were initially found to be associated with reduced DNA methylation resulting from a methyl group deficiency. Pogribny et al. (10) fed rats a diet deficient in the methyl donors methionine, choline, and folate and found that genomic strand breaks increased with increasing DNA hypomethylation. They also showed that increased methylation protected the DNA from enzyme-induced strand breaks. These results were supported by a subsequent study from the same group (214) who found increased DNA strand breaks with prolonged folate deficiency. They hypothesized that DNA hypomethylation secondary to folate deficiency may induce strand breaks by changing the conformation of the chromatin and protein-protein interactions. These changes make the DNA more susceptible to DNA-damaging agents or endonucleases (214). Although methyl group deficiency was the first explanation for strand breaks, recent research also has implicated uracil misincorporation as a source for strand breaks. As seen in Figure 2-2, the enzyme thymidylate synthase requires 5,10-methyleneTHF as a coenzyme for the conversion of dUMP to dTMP for DNA synthesis and repair. A folate deficiency will limit the amount of coenzyme available for this conversion and cause a buildup of dUMP with subsequent dUTP misincorporation into DNA by DNA polymerase resulting in a UA base pair (11). Uracil is excised from DNA by uracil deglycosylase, which can cause breaks in the DNA if there is insufficient dTMP available for repair. If two breaks are located within 12 base pairs from each other on the DNA it could result in a double-stranded DNA break (257). Another mechanism of uracil misincorporation involves spontaneous deamination of nonmethylated cytosine residues to uracil, which can result in a UG base pair and a cytosine to thymine transition mutation

PAGE 76

62 if the uracil is not excised before replication. Uracil misincorporation is not a problem unless the capacity for uracil excision is exceeded (258). Uracil excision is an important DNA repair mechanism involving different enzymes. Uracil DNA glycosylase recognizes a conformational change produced by the misincorporated uracil and binds to the DNA. The uracil is flipped out of the double helix into the active site of the enzyme and cleaved. This leaves an apyrimidinic (AP) site, which is cleaved by an AP endonuclease. Deoxyribophosphodiesterase removes the 5-phosphate group, DNA polymerase inserts the correct nucleotide, and DNA ligase seals the gaps in the DNA (258). Cell culture studies have confirmed that folate deficiency causes increased uracil misincorporation and DNA strand breakage. Duthie and Hawdon (259) used single cell gel electrophoresis with uracil deglycosylase on human lymphocytes to determine DNA damage in cells in a variety of experiments. Stimulation of cells in folate deficient media resulted in no growth compared to cells in folate-rich media that grew 6-fold in 8 d. The effect of folate deficiency was graded for cell growth, with cells growing normally in 100 ng/ml, poorly in 10 ng/ml, and not at all in 1 ng/ml. Uracil misincorporation and DNA strand breakage was significantly greater in cells grown in a folate deficient medium compared to cells grown with folic acid. Finally, cells grown in folate deficient media for 5 d were unable to repair oxidant-induced DNA damage as efficiently as controls. Melnyk et al. (260) used Chinese hampster ovary cells to determine the effect of folate on DNA damage. Cell growth was inhibited in folate deficient media. They expanded the experiments and measured intracellular nucleotide concentrations, finding significantly increased dUTP and decreased dTTP concentrations and an increased dUTP/dTTP ratio

PAGE 77

63 during folate deprivation. Uracil misincorporation, AP sites, and DNA strand breaks all increased with increasing duration of folate depletion. These cell culture studies supported an association between poor folate status and increased DNA damage. Pogribny et al. (258) found significantly increased uracil misincorporation and AP sites after just 3 wk of feeding rats a methyl deficient diet low in methionine, choline and folate. DNA strand breaks also were increased but progressed more slowly with significant results appearing after 9 wk of feeding and continued to increase with prolonged feeding. Duthie et al. (261) separated their rats into a control group, a folate deficient group, a choline/methionine deficient group, and a folate/choline/methionine deficient group and determined DNA integrity at 4, 8, and 10 wk. Lymphocyte DNA strand breaks were higher in all groups compared to controls after just 4 wk. The greatest amount of strand breakage was seen in the folate/choline/methionine deficient group. The choline/methionine deficient group had more DNA strand breaks than the folate deficient group at 4 wk, indicating that methyl group deficiency has a greater effect on strand breaks than folate deficiency alone. Uracil misincorporation was highest in the folate deficient group, with no uracil misincorporation seen in the choline/methionine group, indicating the specificity of this biomarker for folate deficiency. The investigators concluded that DNA strand breaks are more affected by methyl-donor status than folate status, and that uracil misincorporation is more affected by folate status than methyl-donor status. The extent of uracil misincorporation in humans has been evaluated. After separation of subjects in normal and deficient folate groups at baseline, Blount et al. (11) supplemented all subjects with 5 mg of folic acid for 8 wk. DNA uracil concentrations

PAGE 78

64 were reduced in all subjects after supplementation with folic acid. The greatest decrease was seen in subjects with the lowest folate status at baseline. Folate deficient subjects had the greatest chromosome breakage as measured by micronucleated reticulocytes and erythrocytes at baseline compared with controls. Folate supplementation reduced chromosome breakage in folate deficient subjects but had no effect on folate replete controls. Independent of whether strand breaks are formed from DNA hypomethylation, low methyl-donor status, or uracil misincorporation, they lead to chromosome damage and are associated with increased risk for cancer. Adequate folate intake is essential for the prevention of chromosome damage and may reduce cancer risk. Micronuclei Formation DNA strand breaks ultimately lead to micronuclei, which are formed by the loss of whole chromosomes or portions of chromosomes from daughter nuclei at mitosis and form small, independent nuclei within the cytoplasm of a cell (262). Although micronuclei appear in almost every cell type, cells from the hematopoietic system are most widely used because of their ease of collection (262). Micronuclei formation is a marker of genetic damage that is used to assess different risk factors for their genotoxic capabilities. The role of folic acid in initiating the formation of micronucei was first investigated in early in vitro experiments that were able to induce micronuclei formation with folic acid deficient media (263). They also found dose-dependent protection from micronuclei formation with increasing folic acid concentrations in the media of cultured lymphocytes (263). Everson et al. (264) supported these findings by reviewing a case study of a subject with an elevated frequency of micronucleated cells, which returned to normal after folic acid supplementation.

PAGE 79

65 Many subsequent studies have substantiated the role of folic acid in micronucleated cells (215,236,265,266). Differences have been found between men and women, with folate having an effect on micronuclei formation in women, but not in men (267). Fenech et al. (268) evaluated the association between folate status and micronuclei formation in older men and found a significant increase in micronuclei formation in men with a folate deficiency without any clinical manifestations. They also conducted an intervention study in older men to determine whether folic acid supplements could affect the genetic damage rate in lymphocytes. They did not find a decrease in micronuclei formation with folic acid intakes of up to 10 times the DRI for 16 wk (268). Homocysteine also has been reported to be an independent risk factor for micronuclei formation, although more studies are needed to confirm these results (269,270). Quantifying micronuclei is an easy and fast method to measure genetic damage. It is an assay that can be used in vitro and in vivo to evaluate the influence of factors like environmental toxins and radiation, on cell carcinogenicity (262). This method, combined with other quantitative and qualitative techniques, can provide good insight relative to the etiology of cancer in mammals. Dietary Reference Intakes (DRIs) The DRIs are recommendations for intakes of specific nutrients and include the Recommended Dietary Allowances (RDA), Estimated Average Requirements (EAR), Tolerable Upper Limits (UL), and Adequate Intakes (AI). The most recent recommendations were published in 1998 by the National Academy of Science IOM (27). Previous recommendations were established to prevent clinical deficiencies of each nutrient. The basis for the 1998 IOM DRI recommendations changed from preventing clinical deficiencies to ensuring optimum health (27). The EAR for folate for women

PAGE 80

66 between the ages of 19 to 50 is 320 g DFE/d and is calculated as the amount of folate needed to meet the requirements of 50% of this population. The RDA for this group, which is based on the EAR, is 400 g DFE/d and is set to cover the needs of 97 to 98% of these individuals (27). The IOM also recommends that all women of childbearing age consume 400 g/d of folic acid in the form of supplements and/or fortified foods in addition to the daily diet to reduce the risk of NTDs. An UL of 1,000 g/d of synthetic folic acid was set for this age to avoid a delay in the diagnosis of a vitamin B12 deficiency that could otherwise lead to neurological damage from a masked vitamin B12 deficiency (27). Folate plays an important role in genomic stability. Fenech (271) argues that because DNA damage increases the risk for degenerative diseases and aging, the basis of dietary intake recommendations should shift to defining optimal intakes of nutrients to prevent DNA damage. Vitamin B12 and folate are the two nutrients with the greatest effects on DNA stability (239). Fenech et al. (240) performed a dietary intervention study to evaluate DNA damage and concluded that an intake of 700 g folic acid/d and 7 g vitamin B12/d was sufficient to minimize chromosome damage, which are amounts that greatly exceed the current DRIs for these nutrients. The results of different intervention studies in humans that indicate DNA damage is minimized when red blood cell folate concentration is > 700 nmol/L were evaluated in a review by Fenech (272). This red blood cell folate concentration is associated with folate intakes greater than the current DRIs. This review also lists different intakes recommended by various investigators to minimize genomic instability that range from 228 to 10,000 g folic

PAGE 81

67 acid/d. Fenech (271) concluded that there is a need for an international collaboration to establish DRIs for nutrients to enhance genomic stability. Folate Status in Women of Reproductive Age Effect of Fortification Earlier research associating the intake of folic acid with NTD risk reduction prompted the Public Health Service in 1992 to recommend that all women of childbearing age consume 400 g/d of folic acid (273). Since then there have been monumental efforts to help women in this age group meet this recommendation, including the mandate by the FDA in the US in 1996 to fortify all enriched cereal grain products with folic acid. The fortification of the food supply with folic acid has had a positive effect on the folate status of the population as a whole. In order to assess the benefit of fortification, the CDC (274) compared serum and red blood cell folate concentration for women of childbearing age who participated in the 1999 National Health and Nutrition Examination Survey (NHANES 1999) to women of childbearing age who participated in the Third National Health and Nutrition Examination Survey (NHANES III, 1988-1994). The serum and red blood cell folate concentrations from the NHANES III were 6.3 and 181 ng/ml and increased significantly to 16.2 and 663 ng/ml, respectively, in NHANES 1999. Choumenkovitch et al. (275) also evaluated the effect of fortification in a cross-sectional study with participants in the Framingham Offspring Cohort. They compared the mean red blood cell folate concentration of subjects before fortification to red blood cell folate concentration of subjects after fortification and observed a 38% increase post fortification.

PAGE 82

68 Fortification also has had a positive effect clinically. Honein et al. (276) reported a 19% reduction in NTD prevalence in the United States, although they concede that other factors also may have contributed to this decrease. The FDA predicted that fortification would increase the average folate consumption by 100 g/d. Studies by Choumenkovitch et al. (277) and Quinlivan and Gregory (278) determined that intakes are approximately twice the predicted level. Effect of Ethnicity Ethnicity may affect the folate status and homocysteine concentration of certain populations. Specifically, certain Hispanic groups may be the most affected. The Office of Management and Budgets revised minimum standards for reporting race, which have been adopted by the NIH, define two ethnic categories: Hispanic and non-Hispanic (279). Hispanic ethnic groups can be further divided into the subcategories Cuban, Mexican, Puerto Rican, South or Central American, or other Spanish culture or origin, regardless of race (279). Ford and Bowman (280) published data from NHANES III on serum and red blood cell folate concentration in non-Hispanic white, non-Hispanic black, and Mexican American men and women. Both Mexican American men and women had significantly lower serum and red blood cell folate concentrations compared to non-Hispanic white men and women even after controlling for dietary folate intake. In addition, Jacques et al. (281) published data from NHANES III on homocysteine concentrations in non-Hispanic white, non-Hispanic black, and Mexican American men and women. Mexican American women had significantly lower homocysteine concentrations compared to non-Hispanic white and non-Hispanic black women. In response to the publication of these two studies, Baggott (282) questioned how lower serum folate concentrations could lead to lower homocysteine concentrations in Mexican Americans, which is inconsistent with

PAGE 83

69 a large body of literature that reports that lower serum folate concentrations are associated with higher homocysteine concentrations. Jacques et al. (283) responded by suggesting that lifestyle factors and racial and genetic differences may influence the homocysteine concentration of this population, and that the lower homocysteine concentrations are not a result of higher serum folate and vitamin B12 concentrations. The basis for the lower serum and red blood cell folate and homocysteine concentrations in Mexican Americans needs further study. Folate Deficiency There is a progressive sequence of events associated with the development of folate deficiency. The first event is a reduction in serum folate concentration, which can occur within 1 to 3 wk. Serum folate concentration is considered to be a short-term indicator of folate status because it is most indicative of recent intake. After a longer period of folate deficiency, red blood cell folate concentration will begin to drop. Red blood cell folate concentration is considered a long-term indicator of folate status. Homocysteine concentration begins to increase around the same time that red blood cell folate concentration decreases (284). After blood folate status is compromised for a period of time, the clinical manifestations of folate deficiency develop. Neutrophil hypersegmentation causes a state of macrocytosis without anemia. Eventually, macrocytic, megaloblastic anemia will develop with decreases in hemoglobin, hematocrit, and red blood cell folate concentrations (284). A decrease in folate availability will cause a reduction in the DNA cycle, ultimately decreasing cell division, and resulting in the formation of the large, immature red blood cells observed in megaloblastic anemia (233).

PAGE 84

70 Folate Status Assessment Folate status assessment can be separated into quantitative assessment and functional assessment. The two quantitative indicators are serum and red blood cell folate concentrations. The functional indicators include plasma homocysteine, SAM/SAH ratio, DNA methylation, uracil misincorporation, and DNA strand breaks. Quantitative measures of folate status can provide an indication to the clinician of the patients folate status. Because serum folate is a more sensitive indicator of recent intake, an isolated serum folate measurement may not differentiate between a temporary reduction in serum folate and chronic folate deficiency (27). Repeated serum folate measurements over time will provide more information as to the status of the individual. A serum folate concentration less than 3 ng/ml (6.8 nmol/L) is reflective of a negative folate balance in the individual at the time the sample was taken (285). Red blood cell folate concentration is the primary indicator chosen to determine folate status because it reflects tissue stores of the vitamin (27). Red blood cells only take up folate during erythropoiesis and have a lifespan of 120 d; therefore, they are indicative of long-term folate status (27). The IOM recognizes a red blood cell folate concentration of < 140 ng/ml (305 nmol/L) as a deficient red blood cell folate concentration and has based this number on a review of studies in which this value was associated with clinical indicators of deficiency (27). Although red blood cell folate concentration is a better indicator of folate status than serum folate concentration, both serum and red blood cell folate concentrations taken together can provide a good measure of folate status. The functional indicators of folate status become affected as a result of changes in the quantitative indicators. Plasma homocysteine concentration becomes elevated during a folate deficiency because methionine synthase, which requires folate as a cofactor, is

PAGE 85

71 unable to remethylate homocysteine to methionine. As mentioned previously, different studies have provided evidence that homocysteine concentration rises as folate concentration decreases. Specifically, Jacob et al. (286) reported that folate depletion in young men led to a rise in homocysteine concentration. OKeefe et al. (287) found an inverse relationship between serum and red blood cell folate concentrations and plasma homocysteine concentration in women. Although there are differences in the cut off values for plasma homocysteine, any value > 14 mol/L is generally considered high (27). Ubbink et al. (288) suggested a reference range of 4.9 to 11.7 mol/L. Regardless of gender, low folate concentrations result in increased homocysteine concentrations. High homocysteine concentrations lead to an increase in SAH concentration via the reversible enzyme SAH hydrolase (Fig. 2-2, reaction 11). Increases in SAH without a concurrent increase in SAM results in a decreased SAM/SAH ratio. Balaghi and Wagner (213) found significantly decreased SAM/SAH ratios in rats fed a folate deficient diet for 4 wk compared to controls. There are currently no published reports related to SAM/SAH ratios in humans fed folate-deficient diets. DNA methylation, uracil misincorporation, and DNA strand breaks are also functional indicators of folate status but cannot be used in the same way as plasma homocysteine concentration. Each indicator has been discussed in detail under the DNA stability section. Briefly, a folate deficiency can affect DNA methylation by limiting the amount of methionine available for the production of SAM. Low folate status causes a decrease in the formation of dTTP from dUTP and results in uracil misincorporation. Finally, uracil misincorporation ultimately leads to DNA strand breaks as a result of DNA repair. There are no generalized norms for these indicators because of individual

PAGE 86

72 variation in DNA and variation in the assessment of these methods. These indicators are only useful when comparing subjects to a control group or to themselves after some kind of treatment within the same study. Analytical Methodology A variety of methods have been used to assess folate status. Not all research groups use the same methods, and even within a given method, protocols may vary between labs. These differences limit the ability to directly compare results from different studies without an evaluation of interlaboratory differences (289). Although there are many methods used, this section reviews some of the more commonly used methods. Blood Folate Analysis Serum and red blood cell folate concentrations can be measured using a microbiological assay or a radiobinding assay. The microbiological assay is the most widely accepted method for determining folate concentrations in blood, urine, tissue, and food samples. The test organism used is Lactobacillus casei subspecies rhamnosis. This organism metabolizes the greatest number of folate derivatives, including polyglutamted residues with up to three glutamates (290). The growth of the organism is calculated by measuring the turbidity of the medium, which is directly proportional to the amount of folate in the sample. This assay cannot be used on samples containing antibiotics or methotrexate because they will inhibit the growth of the organism. The radiobinding assay (RA) also is used to measure folate concentrations of samples because of its speed and the fact that it is not affected by pharmaceuticals. A folate binding protein attached to microbeads and [ 125 I]-labeled folic acid or methyltetrahydrofolate are used to quantitate serum or red blood cell folate concentration.

PAGE 87

73 Unlabeled folate competes with labeled folate for binding to the folate binding protein. Samples are centrifuged and bound folates and microbeads precipitated. The supernatant is discarded and bound labeled folate is counted in a scintillation counter. The decrease in radioactivity is proportional to the folate concentration in the sample. There have been problems associated with using RA assays. Folate values for the first 4 y of the NHANES III were determined using a RA, and results had to be adjusted to correct values after it was determined that the RA results were 30% too high due to the problems with the standards produced by the manufacturer of the assay kit (27). Although the microbiological assay tends to be time consuming and tedious, it has been the preferred method for quantitative analysis of folate status. The RA is a quicker method and tend to be lower than values obtained using the microbiological method (291). A round robin comparison of different lab techniques for the measurement of serum and red blood cell folate concentration was conducted by the CDC (289). Five different analytical techniques were used by 20 different laboratories worldwide to assess the folate concentrations of six serum and six whole blood pools. They reported overall CVs for the serum folate pools and the whole blood pools of 27.6% and 35.7%, respectively. They also reported a 2to 9-fold difference in concentrations of the pools within the different methods. These results support the fact that folate values cannot be compared between labs unless interlaboratory variations are considered. Plasma Homocysteine and SAM/SAH Ratio Analysis Measurement of plasma homocysteine concentration is generally limited to total homocysteine concentration. Before total homocysteine concentration can be measured, the disulfides must be chemically reduced. The preferred approach to homocysteine quantitation is an HPLC method because of the relatively low cost of chemicals and

PAGE 88

74 solvents, the availability of equipment, and the existence of a fully automated assay (292). Some other automated methods include gas chromatography-mass spectrometry, liquid chromatography electrospray tandem mass spectrometry, and immunoassay (293). HPLC methods for homocysteine utilize reversed-phase columns, which have hydrophobic chains that protrude and retain the hydrophobic molecules of interest. These molecules are eluted in order of hydrophobicity using a mobile phase. Once eluted, homocysteine can be detected fluorescently, electrochemically, or colorimetrically. A similar technique also can be used to measure SAM and SAH concentrations (294). DNA Stability The most common ways to assess genomic stability is by evaluating DNA methylation, uracil misincorporation, and DNA strand breaks. Because analysis of these indicators differs between laboratories, results cannot be compared between different studies. Comparisons can be made only within studies between controls and experimental subjects. DNA methylation. The original assay used to determine DNA methylation involved the use of [ 3 H]SAM and a bacterial DNA methylase that only methylates at the 5-position of cytosines. Incorporation of labeled methyl groups is inversely related to the extent of DNA methylation in the sample (213). Some limitations of this method are that damaged DNA can interfere with the methylase, and that DNA strand breaks or abasic sites can give a false positive (295). Pogribny et al. (295) developed a cytosine-extension method based on the use of a methylation-sensitive restriction enzyme that leaves a 5guanine overhang followed by single nucleotide primer extension with [ 3 H]dCTP. The extent of [ 3 H]dCTP incorporation also is inversely proportional to the DNA methylation in the sample. The cytosine-extension method is less subject to error than the methyl

PAGE 89

75 acceptance assay because the integrity of the DNA does not influence the method, which can be applied to nanogram quantities of DNA (295). Fujiwara and Ito (296) made one modification of the cytosine extension protocol to circumvent the use of radioactivity. Biotinylated dCTP was added to the digested DNA by Taq polymerase, which was visualized with a streptavidin-alkaline phosphatase reaction. Friso et al. (13) developed a liquid chromatography/mass spectrometry method that allowed direct quantitation of methylated cytosine residues. DNA is enzymatically hydrolyzed with sequential digestion to nucleotides and separated into the four bases. Methylated cytosine elutes 2 min after cytosine and can be quantitated. This method is more accurate than the previous two methods in determining DNA methylation because it is quantitative rather than semiquantitative, it measures methylated cytosines directly rather than indirectly, and it has small intraand inter-assay CVs. A similar method by Cooney et al. (297) involves the direct measurement of 5-methyldeoxycytidine after sequential enzymatic digestion and HPLC separation. A less utilized method involves bisulfite-induced modification of genomic DNA to convert cytosine to uracil but methylated cytosines remain nonreactive (298). Uracil misincorporation. A high degree of uracil misincorporation can lead to DNA instability when DNA repair enzymes remove the uracil and leave single-strand breaks that could result in the less repairable double-stranded breaks (11). The original assay used to determine uracil misincorporation employed gas chromatography/mass spectrometry in negative chemical ionization mode after DNA was digested with uracil deglycosylase. This allowed for direct measurement of uracil in the DNA sample.

PAGE 90

76 The comet assay can also be used to determine uracil misincorporation (259). The comet assay involves the use of a microscope slide covered with agarose and cells. These slides are washed with uracil deglycosylase to excise the uracil. Slides are subjected to electrophoresis and stained. The DNA sample resembles a comet, with the intact DNA in the head, and any DNA fragments in the tail. Uracil misincorporation is related to the fluorescence in the tail. More recent assays convert misincorporated uracils to DNA strand breaks with addition of the endonuclease Exo III after uracil excision with uracil deglycosylase. DNA strand breaks can then be quantified using a comet assay (299) or by random oligonucleotide-primed synthesis (ROPS) assay (258,260). DNA strand breaks. There are many different methods available to quantify DNA strand breaks. Some of the older methods include alkali elution, DNA unwinding assays, and unscheduled DNA synthesis. Agarose gel electrophopresis, terminal deoxynucleotide transferase, nick translation, and the ROPS assay are more current methods of detection. Most of these assays do not distinguish 3OH from 5OH ends and require large amounts of DNA (300). Basnakian and James (300) developed a ROPS assay based on random oligo-nucleotide synthesis catalyzed by Klenow fragment polymerase in order to detect low frequency 3OH DNA strand breaks. After denaturation and renaturation of the DNA, the single-stranded DNA serves as its own template for extension using [ 32 P]-labeled dNTPs. Incorporation of labeled dNTPs is proportional to the number of strand breaks in the DNA sample. The strength of this assay is that it only requires nanogram concentrations of DNA and it can detect single-stranded and double-stranded DNA breaks.

PAGE 91

77 Genotype Determination With the discovery of single nucleotide polymorphisms (SNPs), determining subject genotypes for these SNPs has become common practice. The most common method of genotyping involves polymerase chain reaction (PCR) to amplify the desired region of DNA. Once the region is amplified, specific restriction enzymes are added depending on the SNP being studied. Fragments are then separated by electrophoresis on an agarose gel. There are several SNPs associated with folate metabolism. In order to simplify the search for multiple SNPs in one sample, Barbaux et al. (301) developed a method that allows genotyping of four SNPs on one gel. This heteroduplexing method involves the use of a heteroduplexing generator instead of restriction enzymes. The generator is a synthetic DNA molecule identical in sequence to the SNP of choice with a microdeletion adjacent to the polymorphic site (301). The generator combined with PCR technology enables multiple genotyping in a single-tube reaction that can be separated on a single gel. Recently, Ulvik and Ueland (302) developed a method utilizing real-time PCR to genotype for multiple folate related SNPs in whole blood or serum in one tube with the goal of eliminating the DNA purification step. Advances in technology keep improving current methods and will one day enable all SNPs associated with folate metabolism to be identified in the most efficient and cost friendly manner. Research Significance The MTHFR 677CT polymorphism affects a large percentage of the population with an estimated frequency of ~12% for the TT genotype with considerable variation between different ethnic groups (4,303). Blood folate concentrations are reduced (80,86), homocysteine concentrations increased (7,65,67,82), and DNA methylation diminished (13) in individuals with the homozygous TT genotype for the MTHFR 677CT variant.

PAGE 92

78 It is well recognized that impaired folate status is associated with abnormal fetal growth and development (304) and increased risk of pregnancy complications (139), and that periconceptional folic acid supplementation significantly reduces the risk of NTDs (119). The metabolic basis of these observations has not been definitively established but may relate to folates role in nucleotide biosynthesis (1), DNA methylation (218,219), and/or maintenance of normal homocysteine concentrations (142). Since the combined presence of the MTHFR 677CT variant and low folate status has been associated with increased risk for birth defects, the present study was designed to address the specific aim in females of reproductive age. Guinotte et al. (87) recently published a metabolic study in this age group with a similar design as the present study to investigate differences in response to folate depletion and repletion with the RDA by MTHFR genotype in young women of Mexican American descent. An important difference between these studies is the ethnicity of the study groups, a factor that has been shown to significantly affect folate status and homocysteine concentration (280,281). The present study is the first controlled metabolic study performed in predominantly non-Hispanic women of childbearing age to determine whether there are differences in response to folate depletion and repletion between MTHFR genotypes. In addition, the effect of controlled folate intake on DNA methylation in women with the CC or TT MTHFR 677CT genotype has not been reported in this age group. Although DNA methylation was found to be significantly reduced after depletion and reversed with folate supplementation in post-menopausal women fed a folate-controlled diet (237), these researchers did not evaluate the effect of genotype on methylation. Similarly, the effect of MTHFR genotype on DNA methylation was was not considered in a group of elderly women fed a

PAGE 93

79 folate-controlled depletion diet (238). The present study is the first to report the effect of folate depletion-repletion on global DNA methylation in women of childbearing age based on MTHFR 677CT genotype. In addition, there are no genotype-specific DRI recommendations, and data are insufficient to determine whether individuals with the TT genotype require more folate to maintain normal folate status than individuals with the CC genotype. Data from this study can be considered when making future revisions of the RDA for folate.

PAGE 94

CHAPTER 3 STUDY DESIGN AND METHODS Subject Screening and Description After approval of the study protocol by the University of Florida Institutional Review Board, nonpregnant, healthy women 20 to 30 y old were recruited from the Gainesville, FL area by distributing flyers and placing announcements in local papers. Inclusion criteria were normal blood chemistry, blood folate concentrations, body weight (within 120% of ideal body weight), and health status as determined by medical history. Exclusion criteria were chronic alcohol consumption or use of tobacco or any medications. Approximately 3500 women were screened initially over the phone by a research nurse. Women who seemed to meet the initial screening criteria (n = 379) reported to our lab to have their blood drawn to determine genotype status for the MTHFR 677CT polymorphism. Only women with the normal CC or homozygous variant TT genotypes were eligible for the study. Fortysix women were selected to participate in the study. All subjects screened and enrolled in the study provided signed informed consent and agreed to participate for the duration of either a 7 wk or 14 wk study and to comply with the study protocol. This study was performed in conjunction with another study whose subjects did not complete the repletion phase of the study. Forty-one women (22 CC, 19 TT) completed the depletion phase of the study (7 wk), and 20 women (10 CC, 10 TT) completed the entire depletion-repletion protocol (14 wk). Serum and red blood cell folate, and plasma vitamin B12, pyridoxal phosphate, and 80

PAGE 95

81 homocysteine concentrations were normal at baseline for all subjects (i.e., 7 nmol/L, 317 nmol/L, 125 pmol/L, 20 nmol/L, and 14 mol/L, respectively). Study Design Subjects adhered to a depletion-repletion feeding protocol divided into two consecutive periods of 49 d (7 wk) each (Fig. 3-1). Subjects consumed a low-folate diet providing 115 20 g dietary folate equivalents (DFE)/d during the first 7 wk of the study. The repletion diet consisted of a combination of the depletion diet plus folic acid and provided 400 g DFE/d [115 + 285 g DFE (168 g folic acid X 1.7 = 285 g DFE)] (27). This controlled metabolic feeding study was conducted in the General Clinical Research Center (GCRC) at Shands Hospital at the University of Florida in Gainesville, FL. Depletion 115 g DFE/d Repletion 400 g DFE/d 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 CBC-D and HCG Blood collection Blood chemistry and leukocyte collection Figure 3-1. Study design. Fasting blood samples were obtained at weekly intervals to determine changes in serum and red blood cell folate and homocysteine concentrations. A blood chemistry

PAGE 96

82 profile was performed at baseline (wk 0), post-depletion (wk 7) and post-repletion (wk 14) to monitor health status (Quest Diagnostic Laboratories; Gainesville, FL). Leukocytes were collected at baseline (wk 0), post-depletion (wk 7), and post-repletion (wk 14) for DNA extraction. Complete blood counts with differentials (CBC-D) were performed biweekly and these measurements were evaluated throughout the study to monitor hematological indices. In the event of a reduction in hematocrit to < 30%, 50 mg of iron as ferrous fumarate from a time-released Ferro-Sequels caplet (Inverness Medical, Inc; Waltham, MA) was provided with dinner until values reverted to > 30%. Additionally, biweekly samples were obtained for quantitative analysis of serum human chorionic gonadotropin (HCG) (Quest Diagnostic Laboratories; Gainesville, FL) in order to detect a pregnancy very early during the folate depletion phase and throughout the repletion phase. Subjects were instructed to use barrier methods for contraception if they were sexually active, and they were informed of the potential risks to maternal and fetal health posed by consumption of the low-folate diet. General Clinical Research Center (GCRC) Protocol Breakfast was consumed at the GCRC between 6:30 am and 7:30 am for the duration of the study. Subjects were provided with a take-out lunch and snacks each day and returned to the GCRC between 5:00 pm and 6:00 pm for dinner. They were required to eat all foods and only those foods provided to them for the duration of the study. Subjects were permitted to take all meals for 2 d away from the GCRC and were supplied with all of the menu items packed for transport with detailed instructions on reheating food items. Compliance with the protocol was monitored through close daily contact by the research team and weekly evaluation of changes in serum folate concentration.

PAGE 97

83 Dietary Treatment and Supplementation Description An experimental diet that provided a limited quantity of folate using a variety of palatable entrees and accompaniments was developed and analyzed. The diet consisted of a 5-d menu cycle as shown in Table 3-1. Many commercially prepared food items could not be used in this study as a consequence of the 1996 FDA mandate that all enriched foods be fortified with folic acid (20). To keep the folate content to a minimum, customized recipes were developed and prepared with unenriched flour obtained from Kansas State University (Manhattan, KS) and other low-folate ingredients. Recipes were modified in the GCRC kitchen and taste-tested at laboratory meetings to select the recipes that were the most palatable. Foods made with the unenriched flour included waffles, pancakes, blueberry muffins, pita bread, biscuits, brownies, cookies, cakes, toppings, and pizza crust. A limited selection of canned, low-folate vegetables were used in this study, and each of these was boiled 3 times with the cooking liquid being discarded after each boiling to help leach endogenous folate from the food (305). All foods were weighed to 0.05 g to ensure that each subject received equal amounts of food and that portion sizes remained constant throughout the study. The macronutrient and micronutrient contents of the diet, excluding folate content, were estimated using the Minnesota Nutrient Data System (Version 4.03; Nutrition Coordinating Center at the University of Minnesota; Minneapolis, MN). According to this analysis, the diet provided 2358 kilocalories distributed as 11% protein, 62% carbohydrate, and 27% fat. Folate content was determined by laboratory analysis involving a trienzyme extraction procedure followed by a microbiological assay using Lactobacillus casei (L. casei) as described below. Analyses indicated that the diet contained an average of 115 20 g DFE/d.

PAGE 98

84 Table 3-1. Five-day cycle menu Day 1 Day 2 Day 3 Day 4 Day 5 Breakfast Shredded wheat Waffle* Cornflakes Pancake* Blueberry Skim milk Syrup Skim milk Syrup muffin* Raisins Peaches, canned Raisins Hash browns Skim milk Brown sugar Cranberry juice Apple juice Applesauce Pears, canned Grape juice Cranapple juice Apple juice Lunch Sandwich, Tuna salad Sandwich, Sandwich, Baked stuffed Pita pocket* Crackers Biscuit* Pita pocket* potato Honey ham Pears, canned Baked ham Turkey breast Turkey ham Cheese Corn chips Cheese Cheese chunks Mayonnaise Soda Peaches, canned Mayonnaise Fruit cocktail Applesauce Popcorn Fruit cocktail, canned Doritos Soda canned Popcorn Soda Corn chips Soda Soda Dinner Enchiladas, Chicken pot pie, Tacos, Pizza, BBQ chicken, Chicken Chicken Seasoned beef Crust* Chicken Cheese Potatoes Taco sauce Sauce BBQ sauce Enchilada Carrots Sour cream Mozzarella Margarine sauce* Green beans Cheese cheese Mashed Corn tortillas Sauce Corn tortilla Turkey ham potatoes Green beans Margarine shells chunks Green beans Blueberry tart Crust Carrots Marinated green Chocolate Ice cream Green beans Orange sherbet beans pudding Cranberry juice Orange sherbet Cranberry juice Apple crisp* Shortbread Grape juice Cranberry juice bar Cranapple juice Snacks Fruit cocktail, Applesauce Fruit cocktail, Pears, canned Peaches, canned Shortbread canned Chocolate chip canned Ginger cookies rounds Snickerdoodle cookies* Oatmeal Caramel Popcorn cookies* Caramel raisin popcorn Apple juice Brownie* popcorn cookies* Apple juice Skim milk Grape juice Grape juice Pound cake* Skim milk Skim milk Skim milk Skim milk *Menu item prepared with unenriched flour Menu item boiled three times to minimize folate content

PAGE 99

85 Subjects consumed a custom-formulated, folic acid free supplement (Westlab Pharmacy; Gainesville, FL) with breakfast and dinner to provide the 1998 Recommended Dietary Allowance (RDA) for all other nutrients except choline. The vitamin-mineral supplement composition is presented in Appendix A. The composition of the custom-formulated supplement was determined by comparing the computerized micronutrient analysis of the low-folate diet to the 1998 RDA for all nutrients except choline. Additionally, the micronutrient analysis was adjusted for the loss of water-soluble vitamins due to boiling the vegetables 3 times by assuming 100% loss and subtracting those amounts from the total value obtained by computer analysis. Nutrients other than folate and choline present in the diet at less than 100% the 1998 RDA were included in the supplement. The choline content of the diet was analyzed as discussed below and found to provide 283 mg/d, which is 67% of the Adequate Intake (AI) for choline (425 mg/d) (306). In addition, a separate calcium supplement (Citracal Mission Pharmacal; San Antonio, TX) provided 200 mg of calcium as calcium citrate to provide calcium that was not included in the supplement. Subjects body weights were maintained within 5% of baseline. Subjects who lost more than 5% of their initial body weight were provided with foods with relatively little or no nutrient value aside from calories (i.e., margarine, candy, Jello, and sweetened beverages). If weight loss was not sustained by these measures, the carbohydrate-based caloric supplement Moducal (Mead Johnson Nutritionals; Evansville, IN) was added to beverages. For weight gain exceeding 5% of initial body weight, margarine was excluded from food preparation, and only unsweetened beverages were permitted.

PAGE 100

86 Caffeinated beverages were limited to two 12 oz carbonated beverages or one cup of coffee/d plus one 12 oz beverage. Sample Collection and Processing Weekly fasting blood samples were collected throughout the 14 wk study by a registered nurse using a # 23 gauge needle and a 1/2 inch butterfly (Vacutainer Blood Collection Sets; Becton Dickinson, Vacutainer Systems; Franklin Lakes, NJ). All blood samples were processed within 1 h of collection, as previously described (238,307). A total of 15 blood collections per subject were obtained during the 14 wk study period. Blood samples were treated with extra precaution during collection, processing, and storage to ensure protection from light by wrapping blood collection tubes in foil, processing under yellow lights (Sylvania Gold; Danvers, MA), and storing in cardboard boxes, respectively. Blood for serum samples was collected in 8.3 ml SST gel and clot activator tubes (Vacutainer ; Becton Dickinson, Rutherford, NJ) and kept at room temperature for 30 to 60 min to allow time for clotting. Serum was obtained by centrifuging the SST gel clot activator tubes at 650 x g for 15 min at 21C (International Equipment Company; Model HN-S II Centrifuge, Needham Heights, MA). Supernatant sera were mixed with sodium ascorbate (1 mg/ml), aliquoted into 200 l samples, and stored at 30C until analysis. Whole blood was collected in 7 ml tubes containing K 3 ethylenediaminetetraacetic acid (EDTA) (Vacutainer ; Becton Dickinson, Rutherford, NJ). Blood for plasma homocysteine was kept on ice until processing. A small aliquot of whole blood held at room temperature was used for hematocrit determination and another portion diluted 20-fold in 1 mg/ml ascorbic acid was aliquotted into 200 l samples and frozen for

PAGE 101

87 measurement of red blood cell folate concentration. The iced blood was centrifuged at 2000 x g at 4C for 30 min (Astel Enterprises, Inc.; Model ALC 4237R Refrigerated Centrifuge, Winchester, VA). The plasma from these samples (500 l/sample) was frozen and used to measure the plasma homocysteine concentration. Following removal of the plasma, approximately 1.0 ml of peripheral leukocytes were carefully removed with a transfer pipette and stored at 30C. These samples were subsequently used to extract DNA. Analytical Methods Food Folate Extraction In order to determine the folate content of the study diet, folate was extracted from the diet using a tri-enzyme extraction method (308). Each meal and its related snacks were prepared in the GCRC kitchen using standardized procedures, homogenized in a blender, and stored frozen for 3 d in sealed plastic bags prior to lyophilization (Vertis Company; Gardner, NY) for 3 d. Weights of the freeze-dried meals were recorded. Duplicate 2 g samples from each meal were mixed with 20 ml boiling extraction buffer (6.0 g HEPES, 5.2 g CHES, 10 g sodium ascorbate, 0.4 ml -mercaptoethanol in 500 ml deionized water; pH 7.85) and boiled for 10 min. After homogenization with a tissue tearer homogenizer (Biospec Products, Inc.; Bartlesville, OK) at a setting of 5 for 1 min, samples were incubated with 33,000 U -amylase (Sigma; St. Louis, MO) and 18 U protease (Sigma; St. Louis, MO) for 1 h at 37C to aid in the release of folate from the carbohydrate and protein matrix, respectively. This was followed by incubation with rat plasma conjugase purified from purchased rat plasma (Pel Freeze; Rogers, AZ) at 37C for 4 h to deconjugate polyglutamyl folates. Samples were placed in a boiling water bath

PAGE 102

88 to inactivate the enzymes and placed on ice for 10 min. After centrifugation at 17000 X g for 20 min at 4C (DuPont Instruments; Sorvall RC-5 Superspeed refrigerated centrifuge, Newton, CT), the supernatant was filtered through Whatman #1 paper in a graduated cylinder. Pellets were resuspended in 5 ml Hepes-Ches extraction buffer and recentrifuged. Additional supernatant was added to the previous filtrate, diluted to the greatest sample volume with Hepes-Ches extraction buffer, and aliquotted into 2-ml microcentrifuge tubes. Extracted samples were sparged with nitrogen for ~20 s to guard against folate degradation and stored at 20C. Each meal was analyzed using the microbiological assay with L. casei subspecies rhamnosis, as described below. The tri-enzyme extraction and folate determination were performed 4 times on different meal samples during the course of the study period to verify uniform folate content over time. Food Choline Analysis Choline content of the 5-d menu cycle was determined twice by Dr. Steve Zeisel, at the University of North Carolina, Chapel Hill. All forms of choline and betaine were quantitated in the diet by liquid chromatography/electrospray ionization-isotope dilution mass spectrometry (LC/ESI-IDMS). Choline was extracted and partitioned from the sample using methanol and chloroform and analyzed directly by LC/ESI-IDMS (309). Choline content of the diet is presented in Appendix B. Supplemental Folic Acid Commercially available folic acid (Sigma; St. Louis, MO) was used to prepare the folic acid supplements containing 84 g folic acid administered in 10 ml apple juice for use in the repletion phase of the study. Each subject consumed two folic acid supplements daily for a total of 168 g folic acid (285 g DFE/d). Stock solutions were

PAGE 103

89 prepared by weighing 10 mg folic acid, dissolving in 200 l 0.1 N NaOH, and filling to 100 ml with deionized water for a final concentration of 100 g/ml. Two hundred microliters of the stock solution were further diluted to 10 g/ml by adding 1.8 ml phosphate buffer (pH 7.0). The exact concentration of the diluted stock solution was determined spectrophotometrically at 282 nm using a molar absorptivity coefficient at pH 7.0 of 27,600 L (mol cm) -1 (310). In order to prepare the supplements, 850 l of the 100 g/ml stock solution was added to 10 ml of apple juice in a 15 ml conical tube, wrapped in foil to protect it from light, and frozen immediately at C. Microbiological Assay Serum Folate. Serum folate concentrations were determined using a modified microtiter plate adaptation of the microbiological assay (311). The microorganism L. casei subspecies rhamnosis (American Type Culture Collection; Manassas, VA), which requires folate for growth, was used as the test organism and was grown in Folic Acid Casei Medium (Difco Laboratories; Detroit, MI). The main folate coenzyme in the serum is 5-methylTHF, which supports the growth of L. casei subspecies rhamnosis (ATCC # 7469) (312). The assay was performed in a 96-well, flat-bottomed, sterile, microtiter MICROTEST tissue culture plate (Becton Dickinson Labware; Franklin Lakes, NJ). For each assay, 20 l aliquots of 10 ng/ml folic acid standard, control serum, and 3 subject samples were added in duplicate to 130 l potassium phosphate buffer. Serial dilutions were made with 150 l potassium phosphate buffer in each well. All wells were inoculated with 150 l of microplate medium containing the L. casei and incubated at 37C for 18 h. Cell growth was determined by absorbance at 650 nm using a microtiter

PAGE 104

90 plate reader interfaced with a computer running SOFTmax for Windows (Version 2.35; Molecular Devices; Sunnyvale, CA). A log-linear plot of absorbance against folate standard concentration was used to interpolate unknown folate concentrations of samples and control. The control serum was used to compare values among assays and determine the interassay coefficient of variation (CV). The interassay CV for this study was less than 15%. Intra-assay variation between duplicate samples on each plate also was less than 15%. Red Blood Cell Folate. Red blood cell folate concentration also was determined using the microbiological assay as described above. In order to allow the conjugase in the blood to cleave the polyglutamate form of the reduced folate to the mono-, di-, and triglutamate forms, blood was held at room temperature for approximately 1 h prior to processing. Red blood cell folate concentrations were determined by multiplying the whole blood folate concentration by the dilution factor of 20 and using these values in the following equation: [whole blood folate] [serum folate x (1 hematocrit / 100)] (hematocrit / 100) Plasma Homocysteine Concentration Plasma homocysteine concentration was analyzed in duplicate using a modified high performance liquid chromatography (HPLC) method of Vester and Rasmussen (313). Standards and samples were prepared identically with the addition of 0.1 M borate buffer. The homocysteine disulfide bonds were reduced by adding 20 l tri-n-butylphosphine in dimethylformamide (Sigma; St. Louis, MO) and incubating at

PAGE 105

91 4C for 30 min. Proteins were removed by precipitation with perchloric acid (Sigma; St. Louis, MO). Homocysteine was derivitized with ammonium-7-fluorobenzo-2-oxa-1,3-diazole-4-sulphonate (Wako Chemicals USA; Richmond, VA) to allow for homocysteine quantitation after separation from other sulfide compounds on a Zorbax 4.6 x 250 mm column (Amersham Pharmacia Biotech; Piscataway, NJ) using an HPLC system (Dionex Corporation; Atlanta, GA) with fluorescence detection. A standard curve was prepared using known L-homocysteine (Sigma; St. Louis, MO) concentrations. DNA Extraction DNA was extracted from the buffy coat layer or whole blood using a genomic DNA extraction kit (Genomic DNA Isolation Kit; Bio-Rad Laboratories; Hercules, CA) and a scaled version of the manufacturers protocol. For buffy coat, a 1:4 dilution in phosphate buffered saline was used. Anucleate red blood cells were lysed to facilitate centrifugal separation from nucleated white blood cells that contain DNA. Cellular and nuclear membranes from pelleted white blood cells were lysed with a DNA lysis solution containing detergent and chaotrope followed by removal of RNA with Rnase for 30 min at 37C. Proteins were precipitated from the solution and removed by centrifugation. Following sequential precipitation and washes with isopropanol and 70% ethanol, the DNA was dried at 65C for 10 min, dissolved in a hydration solution at 65C for 10 min, and stored at C to be used at a later time. DNA Quantitation Comparison of DNA samples between subjects requires precisely equivalent DNA concentration and purity. Therefore, a fluorescent DNA stain was used for these determinations rather than employing the common method of measuring the absorbance at 260 and 280 nm. Picogreen dsDNA Quantitation Kit (Molecular Probes; Eugene, OR)

PAGE 106

92 was used for all DNA quantitation in order to detect quantities as little as 25 pg/ml of double-stranded DNA (dsDNA) for which it is specific. Briefly, DNA samples and a 2 g/ml standard of lambda DNA were serially diluted in TE buffer in a 96-well microtiter plate. Picogreen reagent was added, and the plate was read on an FMAX fluorescent microplate reader (Molecular Devices; Sunnyvale, CA) at excitation ~480 nm and emission ~520 nm. Sample DNA concentrations were determined from a lambda DNA standard curve. In order to check DNA purity defined as a 260/280 nm absorbance ratio of > 1.8, random samples were evaluated spectrophotometrically following Picogreen analysis and found to be pure. MTHFR Genotype Determination The restriction fragment length polymorphism polymerase chain reaction protocol of Frosst et al. (3) was used to detect MTHFR 677CT polymorphisms. Briefly, using appropriate forward and reverse primer pairs bracketing the mutation, DNA fragments were amplified by PCR from template DNA purified as described above. Amplified fragments from the gene that codes for the MTHFR enzyme were separated on an agarose gel. DNA fragments with two C alleles (CC genotype for the MTHFR 677CT polymorphism) form a single band that is 198 base pairs long. DNA fragments containing two T alleles are 175 and 23 base pairs long and run further down the gel due to their lighter molecular weight. When the MTHFR enzyme contains both C and T alleles, as is the case in individuals who are heterozygous for the MTHFR 677CT polymorphism, three bands appear, one 198 base pairs long, one 175 base pairs long, and one 23 base pairs long. A sample gel and more detatiled description are presented in Appendix C.

PAGE 107

93 Methyl Acceptance Assay The extent of in vivo DNA methylation was determined using a modified method of Balaghi and Wagner (213), in which the bacterial DNA methylase SssI is used to incorporate methyl groups into unmethylated CpG residues with tritium labeled ([ 3 H]methyl) SAM. The bacterial methylase SssI is isolated from the wall-less prokaryote Spiroplasma strain MQ1 (314). Although most prokaryotic methylases recognize and methylate sites composed of 4-6 bases, SssI is similar to mammalian methylases that recognize and methylate CpG sequences exclusively (314), making it an ideal enzyme for this assay. The incorporation of [ 3 H]methyl groups in vitro is inversely proportional to DNA methylation in vivo. In order to optimize and determine the reproducibility of the methyl acceptance assay, a series of controls were tested. These controls included a hypermethylated, pooled human DNA control (negative control), an unmethylated, pooled human DNA control, and an unmethylated poly dG-dC (Sigma; St. Louis, MO) control (positive control), the latter of which is a synthetic substrate for CpG methylation. In order to prepare the negative control, ~40 g pooled human DNA was hypermethylated by incubation with 60 U SssI methylase (NEB; Beverly, MA) and 15 l unlabeled SAM at 37C for 1.5 h. To prepare the positive control, 10 g of dG-dC was diluted with 1 ml of deionozed water. Triplicate 250 ng samples were assayed for every subject and control. The assay was performed using the following reaction mixture in a 0.5 ml conical screw cap tube: 250 ng DNA or control, 1 l of [ 3 H]SAM (Perkin Elmer, 0.55 mCi/ml, 15 Ci/mmol; Boston, MA), 1.5 U (0.375 l) SssI methylase (NEB; Beverly, MA), 1.5 l 10X NEB buffer (NEB; Beverly, MA), and sterile-filtered, deionized water to a 15 l total reaction volume and incubated at 30C for 1 h. Samples were placed immediately in

PAGE 108

94 ice to stop the enzymatic reaction. The entire reaction mixture was loaded on a 2.5-cm round DE81 ion exchange paper filter (Whatman; Maidstone, England) and filtered through a 25mm/200 ml filter funnel unit (Gelman Sciences; Ann Arbor, MI) attached to a vacuum source. After successive washes 3 times each with 15 ml of 0.5 M sodium phosphate buffer, twice with 1 ml 70% ethanol, and twice with 1 ml 100% ethanol, filters were allowed to dry for 15 min. Once dried, filters were added to 10 ml liquid scintillation fluid (Fisher Scientific ScintiSafe; Fair Lawn, NJ) and counts per min (cpm) determined using a Beckman LS 2800SC (Beckman Instruments; Fullerton, CA) liquid scintillation counter. Samples were recounted after 24 h at room temperature to allow the counts to stabilize, and these triplicate cpm values for each sample were averaged to determine DNA methyl acceptance. Each subjects samples at baseline, post-depletion, and post-repletion were processed in the same run to ensure consistency. In order to determine residual labeled methyl groups remaining on the filters after washing, a background mixture containing all ingredients except DNA was run with each assay and subtracted from the sample counts. The unmethylated, pooled human DNA control processed with each assay was used to determine an inter-assay CV of 7%. Intra-assay CVs among triplicate samples were less than 15%. Counts were converted to disintegrations per min (dpm) by dividing cpm by 0.6 to account for counting efficiency. Subject counts were multiplied by two and reported as dpm/0.5 g DNA. Liquid Chromatography/Mass Spectrometry/Mass Spectrometry Assay Deoxymethylcytidine (mCyt) and deoxycytidine (Cyt) were quantified to produce mCyt/total Cyt (tCyt) ratios in enzymatically hydrolyzed DNA samples by liquid chromatography/mass spectrometry/mass spectrometry (LC-MS/MS) (Quinlivan et al. unpublished). Before analysis, DNA was hydrolyzed enzymatically using a modified

PAGE 109

95 method of Crain (315). In order to prepare the DNA samples for hydrolysis, 1 g of DNA (50 l) for each subject at baseline, post-depletion, and post-repletion was added to a 0.5 ml conical screw cap tube, placed in a boiling water bath for 4 min to denature the DNA, and then immediately placed in an ice bath. Samples were mixed with 5 l of 0.1 M ammonium acetate (pH 5.3) and incubated with 6 U (3 l) P 1 nuclease (Sigma; St. Louis, MO) at 50C for 2 h to nick various phosphate bonds of the DNA. The second hydrolysis involved the addition of 6 l 1.0 M ammonium bicarbonate (pH 7.75) and incubation with 3.25 mU (2.5 l) phosphodiesterase (Sigma; St. Louis, MO) at 37C for 2 h to complete the digestion of the phosphate backbone. Finally, samples were incubated with 0.5 U (2.5 l) alkaline phosphatase (Sigma; St. Louis, MO) at 37C for 1 h to cleave the sugar-phosphate bonds leaving only the nucleosides (sugar + base). Samples were then chromatographed on a 5-micron Discovery C18 column (100 x 4.6 mm; Supelco; Bellefonte, PA) and eluted with a 50 mM ammonium formate (Solvent A): methanol (Solvent B) gradient. The analysis was started using an eluent of 95% A and 5% B for 4 min, followed by a gradient from 5% B to 65% B over 4 min at a flow rate of 0.6 ml/min, then maintained at 65% solvent B for 3 min, reversed to the original composition (95% solvent A, 5% solvent B) over 1 min, and reequilibrated at that composition for 2 min. Mass spectrometry was performed in the selective reaction monitoring mode using a Finnigan TSQ 7000 (Thermo Finnigan; San Jose, CA) (Table 3-2). The result of the LC-MS/MS analysis of a standard sample is given in Figure 3-2 and that of the DNA hydrolysis products is shown in Figure 3-3. The top panel of each figure (Fig. 3-2 and 3-3) illustrates the sum of all the ions monitored, the middle panel the deoxymethylcytidine peak, and the bottom panel the deoxycytidine peak. In order to quantify the mCyt and

PAGE 110

96 Table 3-2. Selective reaction monitoring fragmentation table. Nucleoside Retention time (min) Parent ion Daughter ion deoxymethylcytidine 7.0 241.6 125.8 deoxycytidine 3.9 227.6 111.8 Collision energy for MS/MS was 20V. D:\Henderson\...\EXPT_GNH372\Gnh-372-11a12/17/02 02:42:33 PMRT:0.00 14.01 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14Time (min) 0 20 40 60 80 100 0 20 40 60 80 100Relative Abundance 0 20 40 60 80 100 RT: 3.96AA: 107082SN: 41985RT: 6.99AA: 6983SN: 1661 RT: 6.99AA: 153068SN: 1549 RT: 3.96AA: 1624158SN: 23470NL: 1.69E5TIC MS GNH-372-11A NL: 1.10E4TIC F: + c APCI SRM ms2 241.60@20.00 [ 125.50-126.10] MS GNH-372-11A NL: 1.69E5TIC F: + c APCI SRM ms2 227.60@20.00 [ 111.50-112.10] MS GNH-372-11A Figure 3-2. LC-MS/MS analysis of a standard. Top panel: sum of all ions monitored; middle panel: deoxymethylcytidine peak; bottom panel: deoxycytidine peak. Cyt in the samples, an external standard consisting of 6% mCyt was prepared by adding 23.5 l deoxymethylcytidine (Sigma; St. Louis, MO) at a concentration of 10 ng/ml and 15 l deoxycytidine (Acros Organics, Fisher Scientific; Pittsburg, PA) at a concentration of 100 ng/mL to 961.5 l sterilefiltered, deionized water. Sample mCyt and Cyt concentrations were quantified by comparing their peak areas to standard curves calculated from the peak areas of deoxymethylcytidine (0.29 to 2.35 ng/ml; r = 0.99) and deoxycytidine (18.75 to 150 ng/ml; r = 0.99). Ratios were calculated by dividing the

PAGE 111

97 D:\Henderson\...\EXPT_GNH372\Gnh372-21a12/17/02 06:53:04 PMRT:0.00 14.02 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14Time (min) 0 20 40 60 80 100 0 20 40 60 80 100Relative Abundance 0 20 40 60 80 100 RT: 3.90AA: 556940SN: 187525 RT: 6.88AA: 504538SN: 3657 RT: 3.90AA: 11182417SN: 34878NL: 8.84E5TIC MS GNH372-21A NL: 2.95E4TIC F: + c APCI SRM ms2 241.60@20.00 [ 125.50-126.10] MS GNH372-21A NL: 8.84E5TIC F: + c APCI SRM ms2 227.60@20.00 [ 111.50-112.10] MS GNH372-21A Figure 3-3. LC-MS/MS analysis of a hydrolyzed sample. Top panel: sum of all ions monitored; middle panel: deoxymethylcytidine peak; bottom panel: deoxycytidine peak. mCyt concentration by the total Cyt concentration (mCyt + Cyt = tCyt). Statistical Analysis At baseline, one-way analysis of variance (ANOVA) was used to test for differences in mean serum and red blood cell folate concentrations, plasma homocysteine concentration, and DNA methylation ([ 3 H]methyl group acceptance and mCyt/tCyt ratio) between genotype groups. Analysis of covariance (ANCOVA) assesses changes between genotype groups by accounting for initial values in order to correct for any variance between values due to differences at baseline. ANCOVA was used as the primary analysis to evaluate potential differences in serum and red blood cell folate concentrations, plasma homocysteine concentration, and DNA methylation between genotype groups at wk 7 (adjusting for wk 0 values) and at wk 14 (adjusting for wk 7 and wk 0 values). The least squares (LS) means, which are the means adjusted for initial

PAGE 112

98 values by using an ANCOVA model, were used to describe the magnitude of the differences between each genotype group. As a secondary analysis, ANOVA was performed on the raw and percent change values from wk 0 to wk 7, wk 7 to wk 14, and wk 0 to wk 14 for serum and red blood cell folate and plasma homocysteine concentrations and DNA methylation. In addition to assessing differences between genotype groups, it also was of interest to evaluate differences by serum folate status. Two methods for categorizing serum folate status were employed. Method one involved classifying subjects based on whether their serum folate values were equal to or above or below the overall median serum folate values at wk 0, 7, and 14. The second method involved categorizing subjects based on whether their serum folate values were equal to or above or below 13.6 nmol/L at wk 0, 7, and 14, a concentration that is considered to represent a moderate folate deficiency (312). For clarity, method one and two will be referred to as "serum folate median status" and "serum folate depletion status", respectively. A 2x2 contingency table was constructed and Fischers Exact test used to evaluate serum folate depletion status by subject genotype (CC vs TT); comparisons of other dependent measure means by folate status and genotype were performed via ANOVA. Pearsons correlation analysis was used to evaluate the strength of the relationships between the dependent variables at each point in time (wk 0, 7, and 14). Specifically, at each time point the following were correlated: serum folate vs red blood cell folate, plasma homocysteine, [ 3 H]methyl group acceptance, and mCyt/tCyt ratio; red blood cell folate vs serum folate, plasma homocysteine, [ 3 H]methyl group acceptance, and mCyt/tCyt ratio; plasma homocysteine vs serum and red blood cell folate, [ 3 H]methyl

PAGE 113

99 group acceptance, and mCyt/tCyt ratio; [ 3 H]methyl group acceptance vs serum and red blood cell folate, plasma homocysteine, and mCyt/tCyt ratio; and mCyt/tCyt ratio vs serum and red blood cell folate, plasma homocysteine, and [ 3 H]methyl group acceptance. Correlation coefficients less than 0.35 were considered a low correlation, while correlation coefficients between 0.35 and 0.70, and greater than 0.70 were used to indicate moderate or high correlation, respectively. A sign test for proportion of trends analysis (316) was used to compare the expected and observed combination of trends for DNA methylation indicators and plasma homocysteine, serum folate, and red blood cell folate concentrations over the depletion and repletion periods. Specifically, regression analysis was used to determine the slope of each individuals response over the specified time period. The signs of the regression slope values (positive or negative) were tallied and the observed proportion tested against the proportion expected by chance alone (i.e., there are 4 possible trends: +/+, +/, /+, and /, such that by chance alone the proportion of any possible combination is 25% or 0.25). To evaluate the strength of the relationship between each status indicator over the depletion and repletion phases, regression and Pearson correlation techniques were used. Specifically, linear regression was used to determine the slope of each individuals response over the specified time period. Correlations of combinations of these coefficients (e.g., plasma homocysteine and serum folate concentrations) were determined to assess the magnitude of the relationship. Exploratory statistical methods employing ANOVA models with main factors for genotype, serum folate status (median status and depletion status separately), as well as the interaction term were used to evaluate potential differences in [ 3 H]methyl group

PAGE 114

100 acceptance, mCyt/tCyt ratio, and plasma homocysteine concentration at each time point (wk 0, wk 7, and wk 14). In addition, one-way ANOVA was used as a secondary exploratory analysis to evaluate potential differences in [ 3 H]methyl group acceptance, mCyt/tCyt ratio, and plasma homocysteine concentration by serum folate status (median status and depletion status separately) within each genotype group (CC and TT separately) at each point in time (wk 0, wk 7, and wk 14). For these analyses, alpha was adjusted as alpha/n where n = the number of comparisons. The adjusted alpha was therefore 0.05/12 = 0.004. For all other comparisons, the alpha level was set apriori to 0.05. All statistics were computed using SAS 8.00 (Cary, NC).

PAGE 115

CHAPTER 4 RESULTS Folate Content of Menus The folate content of each menu in the 5-d cycle is presented in Table 4-1. Table 4-1. Folate content of meals with snacks and total daily intake (g DFE). Day Breakfast & Snack Lunch & Snack Dinner & Snack Average Daily Total (g DFE) A 40.8 6.8 43.4 5.6 55.0 7.1 139.2 B 21.7 3.7 32.8 2.6 35.3 4.3 89.8 C 24.0 3.9 29.9 2.8 40.8 3.1 94.7 D 19.3 3.2 59.4 6.3 41.6 3.8 120.3 E 34.6 5.5 42.8 5.6 40.3 5.1 117.7 The 5-d cycle menu provided an average of 115 20 g DFE/d. All meals and corresponding snacks were prepared, extracted, and measured in duplicate at three different time points during the study to verify that folate content remained constant. Serum Folate Concentration Serum folate concentrations (mean SD) throughout the study are illustrated in Figure 4-1 for subjects by genotype. The mean ( SD) serum folate concentration (nmol/L) at baseline, post-depletion, and post-repletion and changes in serum folate status are presented in Table 4-2. Although no significant (P = 0.12) difference in serum folate concentration existed at baseline between subjects with the TT genotype compared 101

PAGE 116

102 -1 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 0 10 20 30 40 50 60 70 80*CC TT Week of StudySerum Folate(nmol/L SD) Figure 4-1. Weekly unadjusted mean ( SD) serum folate concentrations (nmol/L) by genotype groups throughout the study. *Significant difference between genotype groups at wk 7 (P = 0.03). to the CC genotype, values for subjects with the TT genotype were lower than those with the CC genotype (41.5 19.7 vs 52.2 22.5 nmol/L, respectively) (Fig. 4-2). Overall serum folate concentration decreased (P < 0.0001) during the depletion phase by 59 15%. Serum folate for subjects with the TT genotype decreased (P < 0.0001) during depletion by 61 16% compared to a 57 15% decrease (P < 0.0001) in subjects with the CC genotype. No significant (P = 0.42) difference in serum folate percent change during depletion was detected between genotypes. The raw change (nmol/L) in serum folate concentration throughout depletion was significant overall (P < 0.0001) and by genotype (P < 0.0001) (Table 4-2). Post-depletion, there was a significantly lower (P = 0.03) serum folate concentration in subjects with the TT compared to the CC genotype (LS mean SEM: 19.5 1.2 vs 15.3 1.3 nmol/L, respectively) (Fig. 4-2). Post

PAGE 117

Table 4-2. Serum folate concentration (mean SD) at baseline, post-depletion and post-repletion and mean changes in serum folate concentration during depletion and repletion by genotype 1 MTHFR Genotype Mean serum folate concentration nmol/L SD (range) Baseline Post-depletion 2 Post-repletion 3 Percent change % SD Depletion 5 Repletion 6 Raw change nmol/L SD Depletion 5 Repletion 6 CC 52.2 22.4 (25.9 94.7) n = 22 20.2 6.3 4 (8.9 31.0) n = 20 26.8 4.8 (17.8 32.4) n = 10 57 15 38 39 13.6 8.4 2.7 2.1 TT 41.5 19.7 (12.7 88.3) n = 19 14.5 5.9 4 (7.8 24.2) n = 17 21.9 6.9 (13.7 35.5) n = 10 61 16 57 50 11.7 7.2 3.0 2.2 Overall Mean 47.2 21.7 (12.7 94.8) n = 41 17.6 6.7 (7.8 30.9) n = 37 24.4 6.3 (13.7 35.5) n = 20 59 15 48 45 12.7 7.8 2.9 2.1 103 1 All means listed in table are unadjusted means. 2 Four (2 CC; 2 TT) subjects were only on the depletion diet for 5 wk so they were not included in post-depletion calculations. 3 The current study was performed in conjunction with another study whose subjects did not continue through repletion. 4 Means significantly different between genotypes post-depletion: LS Mean CC: 19.5; TT: 15.3; P=0.0304. 5 Significant change during depletion: CC (P<0.0001); TT (P<0.0001), overall (P<0.0001). 6 Significant change during repletion: CC (P<0.01); TT (P<0.006); overall (P<0.0001).

PAGE 118

104 0714 0 25 50 75CC TT *Week of StudySerum Folate(nmol/L SEM) Figure 4-2. Mean serum folate concentration (nmol/L SEM) by genotype at baseline (wk 0; unadjusted mean), post-depletion (wk 7; adjusted mean), and postrepletion (wk 14; adjusted mean). *Significant difference between genotype groups at wk 7 (P = 0.03). depletion, a greater proportion of subjects with the TT genotype (59%) had low folate status (serum folate < 13.6 nmol/L) compared to the CC genotype (15%) (P = 0.008). Overall serum folate increased over the repletion phase of the study by 48 45%, representing a significant (P < 0.0001) change from wk 7 to wk 14. Mean serum folate concentration for subjects with the TT genotype increased (P = 0.006) by 57 50% during repletion compared to the significant (P = 0.013) increase in serum folate (38 39%) of subjects with the CC genotype. No difference (P = 0.36) in the percent change in serum folate between genotypes was detected during repletion. The differences detected for raw change paralleled the percent change results, with significant increases overall in serum folate and by genotype (Table 4-2). No significant differences were detected (P = 0.60) in the mean serum folate concentration of subjects with the TT vs CC genotype (21.9 6.9 vs 26.8 4.9 nmol/L, respectively) post-repletion (Table. 4-2).

PAGE 119

105 Red Blood Cell Folate Concentration Weekly red blood cell folate concentrations throughout the study for subjects by genotype are illustrated in Figure 4-3. Mean ( SD) red blood cell folate concentration at -1 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 0 500 1000 1500 2000 2500*CC TT Week of StudyRed Blood Cell Folate(nmol/L SD) Figure 4-3. Weekly unadjusted mean ( SD) red blood cell (RBC) folate concentrations by genotype throughout the study. *Significant difference between genotype groups at wk 14 (P < 0.05). baseline, post-depletion, and post-repletion are presented in Table 4-3. Subjects with the TT genotype tended (P = 0.06) to have a lower red blood cell folate concentration at baseline compared to subjects with the CC genotype (1486 564 vs 1806 497 nmol/L, respectively) (Fig. 4-4). Mean red blood cell folate concentration for all subjects decreased (P < 0.0001) by 18 15% during the depletion phase of the study. The red blood cell folate concentration of subjects with the TT and CC genotypes decreased significantly (P = 0.0004 and P < 0.0001, respectively) during depletion by 17 15% vs 19 16%, respectively, but these decreases in percent change were not different between

PAGE 120

Table 4-3. Red blood cell folate concentration (mean SD) at baseline, post-depletion and post-repletion and mean changes in red blood cell folate concentration during depletion and repletion by genotype 1 MTHFR Genotype Mean red blood cell folate concentration nmol/L SD (range) Baseline Post-depletion 2 Post-repletion 3 Percent change % SD Depletion Repletion Raw change nmol/L SD Depletion Repletion CC 1806 497 (1021 2883) n = 22 1456 273 (969 1952) n = 20 1205 150 4 (1028 1428) n = 10 19 16 5 6 18 178 154 5 45 105 TT 1486 564 (637 2505) n = 19 1194 392 (749 1885) n = 17 1033 284 4 (589 1434) n = 10 17 15 5 17 19 6 128 116 5 116 118 6 Overall Mean 1658 547 (637 2883) n = 41 1336 354 (749 1952) n = 37 1120 238 (589 1434) n = 20 18 15 5 11 19 6 155 139 5 81 115 6 106 1 All means listed in table are unadjusted means. 2 Four (2 CC; 2 TT) subjects were only on the depletion diet for 5 wk so they were not included in post-depletion calculations. 3 The current study was performed in conjunction with another study whose subjects did not continue through repletion. 4 Means significantly different between genotypes post-repletion: LS Mean CC: 1203; TT: 1036; P=0.0395. 5 Significant change during depletion: CC (P<0.0001); TT (P0.0004), overall (P<0.0001). 6 Significant change during repletion: TT (P 0.02); overall (P<0.02).

PAGE 121

107 0714 0 500 1000 1500 2000 2500 3000CC TT *Week of StudyRed Blood Cell Folate(nmol/L SEM) Figure 4-4. Mean red blood cell folate concentration (nmol/L SEM) by genotype at baseline (wk 0; unadjusted mean), post-depletion (wk 7; adjusted mean), and post-repletion (wk 14; adjusted mean). *Significant difference between genotype groups at wk 14 (P < 0.05). genotypes (P = 0.64). The raw change analysis for red blood cell folate concentration during depletion produced analogous results, with significant overall (P < 0.0001) and by genotype (TT: P < 0.0003; CC: P < 0.0001) decreases during depletion (Table 4-3). No significant differences were detected (P = 0.33) in red blood cell folate concentration between genotypes post-depletion (TT: 1194 392; CC: 1456 273 nmol/L, respectively) (Table 4-3). The overall red blood cell folate concentration continued to decrease (P = 0.02) during the repletion phase by 11 19%. Only the red blood cell folate concentration of subjects with the TT genotype continued to decrease significantly (17 19%; P = 0.02) during repletion. This is in contrast to the red blood cell folate concentration response of subjects with the CC genotype, which did not decrease significantly (6 18%; P = 0.36). The percent change in red blood cell folate concentration during repletion was not

PAGE 122

108 different between genotypes (P = 0.19). The raw change analysis for red blood cell folate concentration paralleled the percent change analysis, with significant decreases detected for all subjects (P = 0.02) and subjects with the TT genotype (P = 0.02) during repletion (Table 4-3). Post-repletion, the mean red blood cell folate concentration of subjects with the TT genotype was significantly lower (P = 0.04) than that of subjects with the CC genotype (LS mean SEM: 1036 53 vs 1203 53 nmol/L, respectively) (Fig. 4-4). Homocysteine Concentration Weekly mean plasma homocysteine concentrations for subjects by genotype are illustrated in Figure 4-5. Mean ( SD) plasma homocysteine concentrations at baseline, -1 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 0 2 4 6 8 10 12 14 16CC TT Week of StudyPlasma Homocysteine(mol/L SD) Figure 4-5. Weekly unadjusted mean ( SD) plasma homocysteine concentrations by genotype throughout the study. Means did not differ between genotype groups at any time point. post-depletion, and post-repletion are presented in Table 4-4. No significant difference was detected (P = 0.18) in plasma homocysteine concentration at baseline between subjects with the TT compared to the CC genotype (7.1 2.1 vs 6.3 1.4 mol/L,

PAGE 123

109 respectively) (Fig. 4-6). The overall plasma homocysteine concentration increased by 52 42% (P < 0.0001) during the depletion phase. Plasma homocysteine concentration increased (P < 0.0001) by 62 52% for subjects with the TT genotype and 43 32% (P < 0.0001) for subjects with the CC genotype during depletion. The plasma homocysteine percent change during depletion was not different between genotypes (P = 0.19). The plasma homocysteine raw change also was significantly increased overall (P < 0.0001) and for both genotype groups (P < 0.0001) during depletion (Table 4-4). Although the plasma homocysteine percent change during depletion was not significantly (P = 0.19) different between genotype groups, there was a trend (P = 0.09) for subjects with the TT genotype to have a greater raw change in homocysteine concentration during depletion than subjects with the CC genotype (3.9 3.0 vs 2.5 1.7, respectively). These increases paralleled significant decreases in serum folate during depletion (Fig. 4-7). Throughout the study, subjects with the TT genotype had numerically higher plasma homocysteine concentrations and lower serum folate concentrations than subjects with the CC genotype (Fig. 4-7). Women with the TT genotype tended (P = 0.09) to have higher mean homocysteine concentration post-depletion than women with the CC genotype (10.5 3.3 vs 8.9 1.9 mol/L, respectively) post-depletion (Table 4-4). A low but inverse correlation was detected (r = 0.32; P = 0.05) between serum folate and homocysteine concentrations for all subjects post-depletion. In addition, a moderate inverse correlation was detected (r = 0.46; P = 0.004) between the slopes for serum folate and plasma homocysteine concentrations post-depletion. In response to folate repletion, overall plasma homocysteine concentration decreased (P = 0.02) by 10 18%. The mean homocysteine concentration for subjects

PAGE 124

Table 4-4. Homocysteine concentration (mean SD) at baseline, post-depletion and post-repletion and mean changes in plasma homocysteine concentration during depletion and repletion by genotype 1 MTHFR Genotype Mean homocysteine concentration mol/L SD (range) Baseline Post-depletion 2 Post-repletion 3 Percent change % SD Depletion Repletion Raw change mol/L SD Depletion Repletion CC 6.3 1.4 (4.2 9.4) n = 22 8.9 1.9 (5.8 14.1) n = 20 7.2 1.9 (4.4 9.9) n = 10 43 32 4 15 17 5 2.5 1.7 4 1.4 1.6 5 TT 7.1 2.1 (4.3 14.1) n = 19 10.5 3.3 (5.9 17.4) n = 17 8.9 1.6 (6.2 11.1) n = 10 62 51 4 5 19 3.9 3.0 4 0.8 1.9 Overall Mean 6.6 1.8 (4.2 14.1) n = 41 9.6 2.7 (5.8 17.4) n = 37 8.1 1.9 (4.4 11.1) n = 20 52 42 4 10 18 5 3.2 2.5 4 1.1 1.8 5 110 1 All means listed in table are unadjusted means. 2 Four (2 CC; 2 TT) subjects were only on the depletion diet for 5 wk so they were not included in post-depletion calculations. 3 The current study was performed in conjunction with another study whose subjects did not continue through repletion. 4 Significant change during depletion: CC (P<0.0001); TT (P<0.0001), overall (P<0.0001). 5 Significant change during repletion: CC (P<0.03); overall (P<0.02).

PAGE 125

111 0714 0 5 10 15CC TT Week of StudyPlasma Homocysteine(mol/L SEM) Figure 4-6. Mean plasma homocysteine concentration (mol/L SEM) by genotype at baseline (wk 0; unadjusted mean), post-depletion (wk 7; adjusted mean), and post-repletion (wk 14; adjusted mean). Means did not differ (P > 0.05) between genotype groups at any time point. with the CC genotype decreased (P = 0.02) during repletion (15 17%). This decrease in plasma homocysteine concentration paralleled an increase in serum folate concentration during repletion (Fig. 4-7). This is in contrast to the homocysteine concentration in subjects with the TT genotype, which did not change (P = 0.47) during repletion and tended (P = 0.08) to be higher than that of subjects with the CC genotype (8.9 1.6 vs 7.2 1.9 mol/L, respectively) post-repletion (Table 4-4). Although mean homocysteine concentrations were not significantly different between genotypes post-repletion, homocysteine by genotype did not behave the same way over the course of the study. Specifically, based on raw and percent changes from wk 0 to wk 14, the mean homocysteine concentration returned to baseline levels (P = 0.1327 and 0.1088, respectively) for subjects with the CC genotype but not the TT genotype (P = 0.0016 and

PAGE 126

112 -1 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 0 25 50 75 100 0 10 20 CC TT Week of StudySerum Folate(nmol/L SD)Plasma Homocysteine(mol/L SD) ( ) ( ) Figure 4-7. Weekly unadjusted mean ( SD) serum folate and plasma homocysteine concentrations by genotype. 0.0039, respectively), whose levels increased on average 2.3 mol/L (36%) throughout the study. Although plasma homocysteine concentration for subjects with the TT genotype did not decrease during repletion, an increase in serum folate concentration during repletion was detected (Fig. 4-7). DNA Methylation DNA [ 3 H]methyl group acceptance and percentage of methylated cytosine (LC-MS/MS method) are presented in Tables 4-5 and 4-6. No significant difference in DNA [ 3 H]methyl group acceptance (P = 0.20) or mCyt/tCyt ratio (P = 0.53) was detected at baseline between genotype groups. A significant (P = 0.03) interaction was detected for mCyt/tCyt ratio and serum folate status at baseline for all subjects below the median. Specifically, subjects with a serum folate concentration below the median (40 nmol/L) for serum folate tended (P = 0.01; adjusted alpha P = 0.0004) to have a lower mCyt/tCyt

PAGE 127

113 Table 4-5. DNA [ 3 H]methyl group acceptance at baseline, post-depletion, and postrepletion for all subjects and by MTHFR genotype Genotype CC TT Overall [ 3 H]methyl group acceptance (mean dpm X 10 2 /0.5 g DNA SD) Baseline 470 80 (n=22) 500 70 (n=19) 490 70 (n=41) Post-depletion 490 80 (n=22) 520 70 (n=19) 500 80 (n=41) Post-repletion 550 150 (n=10) 470 70 (n=10) 510 120 (n=20) Table 4-6. Percentage (%) of methylated cytosine at baseline post-depletion, and postrepletion for all subjects and by MTHFR genotype Genotype CC TT Overall % of methylated cytosine, mean SD Baseline 4.7 0.5 (n=22) 4.8 1.0 (n=19) 4.7 0.8 (n=41) Post-depletion 4.6 0.7 (n=22) 4.6 0.7 (n=19) 4.6 0.7 (n=41) Post-repletion 4.6 0.4 (n=10) 4.5 0.5 (n=10) 4.6 0.5 (n=20) ratio (i.e., less DNA methylation) than subjects with a serum folate concentration above the median.During depletion, there was a trend (P = 0.08) for all subjects to have an increase in percent change of DNA [ 3 H]methyl group acceptance (Table 4-7), suggestive Table 4-7. Percent (%) change in DNA [ 3 H]methyl group acceptance during folate depletion and repletion for all subjects and by MTHFR genotype Genotype CC TT Overall % change, mean SD Depletion (wk 07) 5 20 (n=22) 6 18 (n=19) 5 19 (n=41) 1 Repletion (wk 714) 5 32 (n=10) 8 21 (n=10) 1 27 (n=20) 1 Trend (P=0.08) for significant increase in % change during depletion for all subjects.

PAGE 128

114 of a decrease in global DNA methylation for all subjects. A nonsignificant (P = 0.12) decrease in mean raw change of mCyt/tCyt ratio during depletion as determined by LC-MS/MS during depletion was observed only in subjects with the TT genotype (raw change = 0.003), suggestive of a slight decrease in methylated cytosine (i.e., less DNA methylation) during folate depletion in these subjects (Table 4-8). In addition, a Table 4-8. Raw change in mCyt/tCyt ratio during folate depletion and repletion for all subjects and by MTHFR genotype Genotype CC TT Overall Raw change, mean SD Depletion (wk 07) 0.0003 0.006 (n=22) 0.003 0.007 (n=19) 0.001 0.006 (n=41) Repletion (wk 714) 0.0002 0.003 (n=10) 0.002 0.002 1 (n=10) 0.001 0.002 (n=20) 1 Significant (P = 0.03) raw change during repletion for subjects with the TT genotype. nonsignificant (P = 0.17) decrease in percent change was observed for subjects with the TT genotype. During depletion and repletion an inverse relationship between serum folate or red blood cell folate concentration and [ 3 H]methyl group acceptance was expected (i.e., it was expected that serum folate concentrations would decrease and [ 3 H]methyl group acceptance would increase with folate depletion). Conversely, a direct relationship between plasma homocysteine concentration and [ 3 H]methyl group acceptance was expected in response to folate depletion. Based on the sign test for trends analysis, an inverse relationship between serum folate and [ 3 H]methyl group acceptance was observed in 61% of all subjects (P < 0.0001), 55% of subjects with the CC genotype (P = 0.0064), and 68% of subjects with the TT genotype (P = 0.0004). An inverse relationship also was observed between red blood cell folate and [ 3 H]methyl group

PAGE 129

115 acceptance in 59%, 50%, and 68% of all (P < 0.0001), CC (P = 0.0162), and TT (P = 0.0005) subjects, respectively. A direct relationship was observed during depletion for plasma homocysteine and [ 3 H]methyl group acceptance in 61% of all subjects (P < 0.0001), 55% of subjects with the CC genotype (P = 0.0064), and 68% of subjects with the TT genotype (P = 0.0005). During depletion and repletion, a direct relationship between serum folate or red blood cell folate and mCyt/tCyt ratio was expected and an inverse relationship between homocysteine and mCyt/tCyt ratio was expected. During depletion a direct relationship between serum folate and mCyt/tCyt ratio was observed in 42% and 50% of all (P = 0.02) and CC (P = 0.0162) subjects, respectively. No significant relationship was observed for subjects with the TT genotype. Only subjects with the CC genotype had a direct relationship between red blood cell folate and mCyt/tCyt ratio (46%; P = 0.0369). There was a trend for all subjects (37%; P = 0.0680) and no significant relationship for subjects with the TT genotype. In addition, an inverse correlation was observed between homocysteine and mCyt/tCyt ratio in 42% and 50% of all (P = 0.02) and CC (P = 0.0162) subjects, respectively. There was no significant relationship for subjects with the TT genotype. No significant difference in [ 3 H]methyl group acceptance (P = 0.30) or mCyt/tCyt ratio (P = 0.39) between genotypes was detected post-depletion. A moderate inverse relationship was found between red blood cell folate concentration and DNA [ 3 H]methyl group acceptance for all subjects (r = 0.37; P = 0.02) and for subjects with the CC genotype (r = 0.48; P = 0.03) post-depletion. In response to repletion, although not significant (P = 0.62) a small positive increase in percent change in [ 3 H]methyl group acceptance was observed in subjects with

PAGE 130

116 the CC genotype (Fig. 4-8), suggestive of a slight decrease in DNA methylation. This is in contrast to the nonsignificant negative trend in percent change in [ 3 H]methyl group acceptance in subjects with the TT genotype (P = 0.24) and for all subjects (P = 0.81) -20 -15 -10 -5 0 5 10 15 20CC TT Overall [3H]CH3Acceptance(% Change SEM)Depletion (wk 0-7)Repletion (wk 7-14) Figure 4-8. Percent (%) change in [ 3 H]methyl group acceptance for all subjects and by MTHFR genotype during depletion and repletion. (Fig. 4-8). These nonsignificant trends suggested by DNA [ 3 H]methyl group acceptance data were more clearly defined by the significant (P = 0.04) increase in raw change (Table 4-8) and percent change (Fig. 4-9) in mCyt/tCyt ratio, which was detected for subjects with the TT genotype during repletion and indicated a positive response to folate repletion in this group. The overall raw change and percent change in mCyt/tCyt ratio tended (P = 0.07 and P = 0.06, respectively) to increase, but no increase was noted for subjects with the CC genotype (P = 0.52) (Table 4-8 and Fig. 4-9). During repletion, an inverse relationship was observed between serum folate and [ 3 H]methyl group acceptance in 60% of all (P = 0.0029), CC (P = 0.0303), and TT (P = 0.0303) subjects (P < 0.03). A direct relationship also was observed between

PAGE 131

117 -10 -5 0 5 10CC TT Overall *mCyt/tCyt Ratio(% Change SEM)Depletion (wk 0-7)Repletion (wk 7-14) Figure 4-9. Percent (%) change in mCyt/tCyt ratio for all subjects and by MTHFR genotype during depletion and repletion. *Significant percent change during repletion for subjects with the TT genotype (P < 0.05). homocysteine and [ 3 H]methyl group acceptance in 50% of all subjects (P = 0.02). A trend towards a significant direct relationship was observed for 50% of CC and TT subjects (P = 0.08). No significant difference in [ 3 H]methyl group acceptance (P = 0.17) or mCyt/tCyt ratio (P = 0.56) between genotypes was detected post-repletion.

PAGE 132

CHAPTER 5 DISCUSSION AND CONCLUSIONS The first objective of this study was to use a low-folate diet as a tool to evaluate the response to suboptimal folate intake (115 g DFE/d) in women of reproductive age by assessing indicators of folate status and global DNA methylation between women with the CC genotype and the TT genotype for the MTHFR 677 CT polymorphism. The second objective was to evaluate differences in folate status and DNA methylation indicators between the CC and TT genotype groups in response to repletion with the current Recommended Dietary Allowance (RDA) for folate (400 g DFE/d). There is a large body of observational data (82,86) indicating that individuals with the TT genotype have lower blood folate and higher plasma homocysteine concentrations (65,67,82) than individuals with the CC genotype. Interpretation of data from these epidemiological cross-sectional studies is complicated by the fact that confounding dietary and environmental variables that may affect folate status and plasma homocysteine concentration are not controlled (e.g., folate intake, caffeine, alcohol and tobacco use). The strength of the current metabolic study is that a controlled folate intake was used to assess folate status and DNA methylation response to moderate folate depletion and repletion with the current RDA in young women by MTHFR genotype while controlling for other dietary and environmental factors. Women with the TT genotype responded more negatively to folate depletion and less positively to folate repletion than women with the CC genotype. This was demonstrated by the significantly lower serum folate concentration and the trend for a 118

PAGE 133

119 higher plasma homocysteine concentration post-depletion. During repletion, subjects with the TT genotype did not have a significant decrease in plasma homocysteine concentration, resulting in a trend for a higher plasma homocysteine concentration compared to the CC genotype post-repletion. In addition, women with the TT genotype had a significantly lower red blood cell folate concentration post-repletion. During repletion, women with the TT genotype also had a significant increase in global DNA methylation based on percent change in mCyt/tCyt ratio. Although there are many observational studies available that provide data on comparisons between the TT and CC genotypes for the MTHFR 677CT polymorphism, this discussion focuses on a very recently published metabolic study (87) in young women and a previously published metabolic study (307) in elderly women with research designs and protocols similar to the present study. A similar controlled folate depletion-repletion feeding study in young women was recently published by Guinotte et al. (87), but an important difference in the current study is the ethnicity of the population group. Guinotte et al. (87) restricted their population to Hispanic women who were all of Mexican American origin, whereas the current study consisted of 90% non-Hispanic white, 7% non-Hispanic Black, and 3% Hispanic (Puerto Rican) women. The Mexican American women in the study by Guinotte et al. (87) had a lower overall baseline mean serum folate concentration (30.8 nmol/L) and a lower overall baseline mean plasma homocysteine concentration (5.5 mol/L) than that observed in the women in the current study (47.2 nmol/L and 6.6 mol/L, respectively). These data agree with previously published data from the third National Health and Nutrition Examination Survey, which indicated that serum and red blood cell folate and plasma

PAGE 134

120 homocysteine concentrations were significantly lower in a Mexican American population compared to non-Hispanic whites and non-Hispanic blacks (281). Serum and red blood cell folate concentrations were still lower in Mexican American women than non-Hispanic white women after controlling for dietary intake assessed by food frequency questionnaires. Additionally, Caudill et al. (317) reported lower serum folate concentrations in a combined group of socioeconomically advantaged (SEA) and disadvantaged (LSES) Hispanic women compared to a combined group of Caucasian women. They also reported a lower plasma homocysteine concentration in the LSES Hispanic women compared to LSES Caucasian women. Although the ethnicity of the Hispanic women was not reported, it is likely that these women were predominantly Mexican American based on other reports published by that group (87,303). The influence of ethnicity on folate status at baseline in the women of the current study and that of Guinotte et al. (87) may explain the differences in results between the two studies, which are discussd below. The magnitude of the effect of the low-folate diet on serum folate as evidenced by the 59% reduction in values is similar to that observed previously by our research group (307) in elderly women following a similar 7 wk folate depletion protocol and that observed by Guinotte et al. (87) in the young Mexican American women (65% and 58%, respectively). After consuming the low-folate diet, approximately one-third of the young women in the current study had a serum folate concentration between 7 and 13.6 nmol/L, suggestive of moderate folate depletion, and none of these subjects was severely deficient (< 7 nmol/L) (312). In contrast, 21% of the elderly women in our previous study were severely deficient by wk 7 (< 7 nmol/L) despite having comparable baseline values and

PAGE 135

121 consuming comparable low folate diets to that of the young women in the current study (46 vs 47 nmol/L, respectively, and 118 vs 115 g DFE/d, respectively) (307). The mean serum folate concentration of the young women at wk 7 was 17.6 nmol/L versus 11.3 nmol/L in the elderly women, suggestive of a possible influence of age on the decline of folate status in response to an inadequate intake (307). A negative effect of age on folate status has been observed in rat studies. Choi et al. (318) reported significantly lower plasma folate concentrations in aged rats compared to young rats fed a folate deficient diet for 8 or 20 wk. The overall post-depletion serum folate concentration in the present study differs from that reported in Mexican American women post-depletion (12.9 nmol/L) (87), which are comparable to the values of the elderly women in our previous study (307). During the 7 wk depletion phase, the young women in the present study decreased 29.6 nmol/L or 4.2 nmol/L/wk. The women in the Guinotte et al. (87) study decreased 19.8 nmol/L or 2.6 nmol/L/wk. The mechanism behind this difference in depletion is not yet understood. These data are consistent with observational data that indicate lower mean blood folate values in Mexican American compared to non-Hispanic women (274,281). Genotypic differences were evident in the present study. The young women with the TT genotype for the MTHFR 677CT polymorphism responded more negatively to the folate deficient diet compared to those with the CC genotype based on the significantly lower serum folate concentration post-depletion. In addition, more women in the current study with the TT genotype had low serum folate concentrations (< 13.6 nmol/L) post-depletion than subjects with the CC genotype (59% vs 15%, respectively). Ethnic differences by genotype also were evident between the populations of the Guinotte

PAGE 136

122 et al. (87) study and the women of the current study. Serum folate concentrations were ~10 and ~20 nmol/L lower in Mexican women with the TT and CC genotypes, respectively, at baseline compared to the predominantly non-Hispanic women in the current study (87). In addition, Guinotte et al. (87) observed that Mexican American women with the TT genotype had significantly lower serum folate concentrations than their subjects with the CC genotype throughout depletion. Increased age coupled with the TT genotype may have a compounding negative effect on folate status. The elderly women in our previous study with the TT genotype had a significantly (P = 0.04) greater decrease in serum folate concentration during depletion (72 21.6%) compared to the decrease in a subset (n = 22) of the young women in the current study with either genotype (CC = 64.6 10.8%; TT = 69.9 12.1%). After repletion with the RDA for folate (400 g DFE/d) for 7 wk, overall mean serum folate increased significantly (P < 0.0001) by 48%. In contrast, Guinotte et al. (87) observed a much smaller yet significant overall increase in serum folate (23%) in response to repletion with 400 g DFE/d in Mexican American women (87). The Mexican American women in the Guinotte et al. (87) study appeared to be less responsive than the women in the present study to folate repletion with 400 g DFE/d based on changes in serum folate concentration. During the 7 wk repletion phase, serum folate concentration increased 6.8 nmol/L or ~1 nmol/L/wk in the present study compared to only 3.2 nmol/L or 0.45 nmol/L/wk in the Guinotte et al. (87). This lower rate of repletion in the Guinotte et al. (87) study is consistent with the lower rate of depletion described earlier. It is likely that ethnicity-based differences, including genetic differences, exist between these groups. In the current study, the serum folate

PAGE 137

123 concentration of subjects with the CC genotype increased 38% in contrast to a 57% increase in subjects with the TT genotype. Although all subjects had normal serum folate concentrations (> 13.6 nmol/L) at wk 14, two subjects with the TT genotype had borderline normal values (15.6 and 13.7 nmol/L), suggesting a more impaired response to folate depletion in subjects with the TT genotype compared to subjects with the CC genotype. This is in contrast to the study by Guinotte et al. (87) in which 50% of the Mexican American women with the TT genotype and 14% of those with the CC genotype were still moderately folate deficient (7 to 13.6 nmol/L) after repletion with the current RDA for folate. It is likely that the lower mean serum folate concentration reported by Guinotte et al. (87) post-depletion explains the greater number of Mexican American women still moderately folate deficient post-repletion compared to the women in the present study. During repletion with the current RDA for folate, red blood cell folate concentrations continued to decrease for all women, culminating with a significantly lower mean red blood cell folate concentration in women with the TT genotype compared to the CC genotype. Due to the long half-life of red blood cells and the fact that folate is only taken up by the developing reticulocyte, it takes longer to detect changes in red blood cell folate concentration as a result of folate depletion or repletion. Only a study with a longer repletion phase would determinehow long it would take to restore normal red blood cell folate concentration. Overall baseline homocysteine concentration was approximately 3.4 mol/L lower in the women in the current study compared to that observed in the elderly women studied previously (307). It is a well-established fact that homocysteine concentration

PAGE 138

124 increases with age (27). Herrmann et al. (319) found significantly higher plasma homocysteine concentrations in elderly individuals aged 65 to 75 compared to a younger control group (19 to 60 y). They attributed these higher concentrations to an age-related decline in cystathionine--synthase or a decrease in intracellular vitamin B12. In addition, a reduced vitamin intake (320), changes in renal function (321) and impaired renal homocysteine metabolism (322) also may account for age-related decreases in homocysteine concentration. The negative influence of the MTHFR 677 TT genotype on plasma homocysteine concentration was evident in the trend for women with the TT genotype to have higher mean homocysteine concentration than the women with the CC genotype post-depletion. It also is reflected in the trend for subjects with the TT genotype to have a greater raw change in plasma homocysteine concentration in response to folate depletion than subjects with the CC genotype. The fact that the mean homocysteine concentration was not significantly different between genotype groups post-depletion differs from our previous study with elderly women in which individuals with the TT genotype had a significantly (P = 0.001) greater mean homocysteine concentration post-depletion than women with the CC genotype (LS Mean 13.7 0.7 vs 10.4 0.5 mol/L, respectively) (84). It is likely that age and the TT genotype may also be having a negative synergistic effect on homocysteine status during depletion. Results from the study in elderly women were compared to the results from the present study. At baseline, elderly women with the TT genotype had a plasma homocysteine concentration that was 1.5 mol/L higher than elderly women with the CC genotype (10.0 vs 8.5 mol/L, respectively). Elderly women with the TT genotype had a 2.9 mol/L higher homocysteine concentration post

PAGE 139

125 depletion than elderly women with the CC genotype. This is almost twice the difference in homocysteine concentration noted for the young women with the TT compared to the CC genotype (1.6 mol/L) post-depletion. Baseline homocysteine concentrations in women with the TT and CC genotype in the present study are higher than values reported by Guinotte et al. (87) in Mexican American women with the TT and CC genotype (7.1 vs 5.4 and 6.3 vs 5.3 mol/L, respectively). Although the mean plasma homocysteine concentration increased significantly in both populations during depletion, the Mexican American women had an overall mean plasma homocysteine concentration of 7.2 mol/L post-depletion compared to 9.6 mol/L in the women of the present study. Thus, the Mexican American women in the Guinotte et al. (87) study had a 1.7 mol/L increase in homocysteine concentration compared to 3 mol/L increase in the women in the present study during depletion. Guinotte et al. (87) attributed this low post-depletion value to the low initial overall mean homocysteine concentration of 5.5 mol/L. Their low baseline values are consistent with pre-fortification reports that Mexican American women have lower plasma homocysteine concentrations than non-Hispanic white and non-Hispanic black women (281). There are currently no plasma homocysteine concentration data in these women post-fortification. Lifestyle factors or genetic differences in Mexican American women may be associated with the lower homocysteine concentration in spite of the lower blood folate concentrations observed in these women compared to predominantly non-Hispanic women. Another key difference between the two studies is that in the study by Guinotte et al. (87) choline was supplemented to provide the Adequate Intake (AI) of 425 mg/d

PAGE 140

126 (306), unlike the current study in which the dietary choline content was 285 mg/d (67% of AI). Choline is a methyl-rich compound that can be oxidized to betaine in order to supply methyl groups to remethylate homocysteine to methionine (323). This reaction takes place via the betaine:homocysteine methyltransferase enzyme, which is only found in the liver and kidneys (306). Choline supplementation in the Guinotte et al. (87) study may have contributed to the smaller increase in plasma homocysteine concentration reported during depletion in their women compared to that observed in the present study. Rat studies have demonstrated that choline is utilized to remethylate homocysteine when folate is not available in the diet (232,324). After performing an experimental choline depletion-repletion study in healthy adult men, Zeisel et al. (325) concluded that choline is essential for humans when methionine and folate are limited in the diet. Steenge et al. (326) supplemented healthy men and women with 6 g of betaine, the oxidized form of choline, daily for 6 wk and observed a significant decrease in plasma homocysteine concentration. In addition, Schwahn et al. (327) supplemented mice with all three genotypes for the MTHFR 677CT polymorphism (CC, CT, TT) with betaine (25 mmol/kg) and detected significant decreases in homocysteine concentration for all three genotype groups. They also reported results from a human observational study in which a weak but highly significant inverse correlation (r = 0.2543; P = 0.0049) between plasma homocysteine concentration and plasma betaine concentration was observed (327). The results of the aformentioned studies support the conclusion that choline supplementation may have affected the plasma homocysteine response in the Guinotte et al. (87) study.

PAGE 141

127 Another explanation is that the lower initial homocysteine values contributed to the smaller overall increase in homocysteine. It is well established that pretreatment homocysteine concentrations affect the extent to which folic acid supplementation decreases homocysteine concentrations. The Homocysteine Lowering Trialists Collaboration (328) reported that individuals with higher pretreatment homocysteine or lower pretreatment folate concentrations benefited the most from folic acid supplementation. Based on these data, it is likely that baseline homocysteine concentration affects homocysteine response in either direction. Jacob et al. (323) retrospectively determined the choline status of men and women participating in previously published folate depletion studies (237,286). The men participated in a 108 d study that included three folate intake periods: a 9-d baseline period consisting of 440 g/d followed by two low folate periods in which subjects consumed 25 /d for 30 d and 99 g/d for 15 d. The three periods were then repeated. The choline content (238 mg) of this low folate diet was comparable to that in the present study (285 mg/d). Jacob et al. (323) observed that the men consuming the low folate, low choline diet for the first trial (48 d), a comparable period of time to the depletion phase in the current study, did not have a decrease in plasma choline concentrations. They concluded that hepatic choline stores must have been adequate to maintain choline status during the study. When the trial was repeated, plasma choline concentrations decreased significantly. Based on the data from this study in men, they concluded that more than 250 mg of choline/d is required to maintain adequate choline status when folate intake is low. Although the choline content of the diet consumed during the 49 d folate depletion phase of the current study was slightly lower (67%) than the current AI for choline, the

PAGE 142

128 choline status indicators in our subjects are not likely to have been affected based on the data from the study by Jacob et al. (323). The women also participated in a feeding study consisting of a low folate diet of 56 g/d for 35 d and 111 g/d for another 30 d. The choline content of this diet was approximately half of the content in the diet consumed by the men (147 mg/d). Plasma choline concentrations decreased significantly after 35 d of the lower folate intake. Weekly plasma choline concentrations would need to be analyzed in order to definitively conclude that choline status remained adequate in the present study. It is likely that all diets low in folate, including those consumed chronically in countries that do not practice fortification, also may be low in choline, which may compound the negative effect on homocysteine concentrations. There also was a significant difference in homocysteine response to folate repletion between genotypes in the current study. Repletion with the current RDA was sufficient to significantly decrease mean plasma homocysteine concentrations for women in the current study with the CC genotype. In contrast, there was no significant change in mean plasma homocysteine concentration during repletion for women with the TT genotype. This differs from the response in the Guinotte et al. (87) study in which 400 g DFE/d was sufficient to significantly decrease plasma homocysteine concentrations in both genotype groups (87). Jacques et al. (7) reported that plasma homocysteine concentrations were elevated only in individuals with the TT genotype who had plasma folate concentrations below the median for the group. In the present study, 59% and 15% of subjects with the TT and CC genotypes, respectively, had low folate status (serum folate < 13.6 nmol/L) post-depletion. The mean plasma homocysteine concentrations of these women with low

PAGE 143

129 folate status were 11.1 mol/L and 8.2 mol/L in subjects with the TT and CC genotypes, respectively. This difference in homocysteine concentration (~3 mol/L), although not significantly different in this relatively small group of subjects, may be physiologically significant. This point is illustrated by a prospective study of plasma homocysteine concentration and risk for myocardial infarction in US physicians (329). These researchers reported a 3-fold increase in risk in physicians in the highest 5% (> 15.8 mol/L) vs the lowest 90% (< 14.1 mol/L) of plasma homocysteine, which is a difference of 1.7 mol/L between the two groups. The 3 mol/L difference between genotype groups in the present study is within a normal range that is not associated with vascular disease (< 14 mol/L) (27). An increase of 3 mol/L in an individual with a baseline homocysteine concentration of 12 mol/L would increase their homocysteine concentration into a range that is associated with cardiovascular disease. This is especially important in European countries where homocysteine concentrations tend to be higher especially in individuals with the TT genotype (82) and an increased risk for cardiovascular disease would be likely. In addition, Boushey et al. (330) conducted a meta-analysis of observational studies investigating the relationship between homocysteine concentrations and vascular disease risk. They reported that throughout a range of homocysteine concentrations from 10 to 15 mol/L, a decrease in homocysteine concentrations by ~1 mol/L was associated with a ~10% reduction in risk. Therefore, a difference of 3 mol/L between genotype groups in the current study may confer a difference in risk for vascular disease. Differences in the means for indicators of global DNA methylation were not detected at any time point during this study in women with either genotype. The first

PAGE 144

130 method used to assess global DNA methylation was the [ 3 H]methyl group acceptance assay. There was a trend for all subjects to have an increase in the percent change in [ 3 H]methyl group acceptance during depletion, which is suggestive of a decrease in global DNA methylation. In our previous study of elderly women (238), a significant increase in labeled methyl group acceptance was observed in response to consumption of a comparable low folate diet. The young women in the current study had been consuming fortified foods for approximately 5 y and had very high mean folate concentration at baseline. The folate depletion protocol used in this study may not have been of sufficient duration to reduce their folate status to detect a change in DNA methylation. In contrast, 21% of elderly women studied previously (307) were severely folate deficient post-depletion despite having baseline values comparable to the young women. Another explanation could be that DNA methylation in young women may not be affected as negatively by folate depletion as DNA methylation in elderly women, suggesting an effect of age on DNA methylation response. Richardson (331) reviewed studies that support the theory that DNA methylation decreases with age. Some mechanisms that may contribute to this decrease are endogenous changes such as altered expression of DNA methyltransferases, and exogenous changes such as dietary factors, drugs, and UV light (332). Stern et al. (12) reported significantly higher labeled methyl group incorporation in individuals with the TT compared to the CC genotype in an observational study that included older individuals (25 to 75 y). The mean ages for the TT and CC genotype groups were 52 and 49 y, respectively. Also, the mean plasma folate values of these individuals (CC: 23.8 nmol/L vs TT: 21.3 nmol/L) were very similar to the post-depletion values in the present study. Perhaps increased age and the

PAGE 145

131 chronically low folate values in these subjects may have affected the DNA methylation of these individuals. Although both genotype groups experienced an increase in the mean percent change of [ 3 H]methyl group acceptance during depletion in the present study, only subjects with the CC genotype had an increase in [ 3 H]methyl group acceptance during repletion. This is in contrast to the subjects with the TT genotype who had a significant decrease in percent change in [ 3 H]methyl group acceptance during repletion. Another method available to indirectly assess global DNA methylation is the cytosine-extension assay developed by Pogribny et al. (295). This method is based on the selective use of methylation-sensitive restriction enzymes that leave a 5 guanine overhang after DNA cleavage followed by single nucleotide primer extension with [ 3 H]dCTP (295). It has advantages over the methyl acceptance assay in that it has the ability to simultaneously detect both global and CpG island methylation in one assay; requires less DNA; and is independent of DNA integrity (295). This analytical method was evaluated in the present study to assess global DNA methylation but was not used on study samples due to the large intra-and interassay variations. In addition to the indirect methods available to determine DNA methylation (e.g., [ 3 H]methyl group acceptance assay and cytosine-extension assay), newer methods have been developed to measure methylcytosine and cytosine directly using liquid chromatography and mass spectrometry (LC/MS) (13). Friso et al. (13) first developed an LC/MS method to directly measure methylcytosine in subjects with the TT and CC genotypes for the MTHFR 677CT polymorphism in an observational study of an Italian population. Plasma folate and quantities of methylcytosine were significantly lower in subjects with the TT genotype compared to the CC genotype. An important

PAGE 146

132 limitation to this study is that it was observational data and folate intake was not controlled. Overall folate concentrations (serum and red blood cell folate) were much lower than that reported in the present study, which likely reflects the consumption of unfortified foods by the Italian population. A different LC/MS analytical procedure was developed by our research group for determination of methylcytosine (mCyt) and cytosine (Cyt). Our method differed from the method used by Friso et al. (13) in that we used MS/MS detection in order to detect the bases. Another difference between the analytical procedures was the fact that our method did not involve the use of internal standards. In contrast, external standards were used to quantify the mCyt and Cyt and calculate the ratio of mCyt/total Cyt (tCyt). Although a significant difference in mCyt/tCyt ratios between genotype groups was not detected at any time point in the present study of young women, there was a trend for subjects with serum folate concentrations below the baseline median (40 nmol/L) to have a lower mCyt/tCyt ratio than subjects with serum folate concentrations above the median. The inability to detect differences between genotype groups may have been a result of the relatively small sample size and high folate intake of our population compared to the Italian population (n = 292) with chronically low folate intakes evaluated in the observational study reported by Friso et al. (13). In response to folate repletion, a significant increase in the percent change in mCyt/tCyt ratio in subjects with the TT genotype was detected. This is consistent with the results by Friso et al. (13) who found significantly more methylcytosine in subjects with the TT genotype with plasma folate concentrations in the highest tertile compared to the lowest tertile. This indicates that as folate status is improved in these individuals,

PAGE 147

133 DNA methylation is restored. A change in mCyt/tCyt ratio during repletion in subjects with the CC genotype was not detected, which is also consistent with data by Friso et al. (13) who did not find a significant difference in DNA methylation among tertiles of plasma folate in subjects with the CC genotype. Based on all of the global DNA methylation indicators, subjects with the TT genotype seemed to have been more negatively affected by folate depletion and more positively affected by folate repletion. A large percentage of women of reproductive age in countries other than the US and Canada consume unfortified diets that provide a mean folate intake of ~200 g/d (82,333,334) which has been demonstrated to be inadequate to maintain normal folate status (287). Chronic consumption of these low folate diets by women of reproductive age with the TT genotype for the MTHFR 677CT polymorphism may increase their risk of impaired pregnancy outcome should pregnancy occur (304). Since initiation of the folic acid fortification program in the United States and Canada, typical folic acid intakes have increased by ~200 g/d (275,278), which probably accounts for the overall higher mean serum folate concentration at baseline (47 nmol/L) in the present study. The pre-depletion serum folate and plasma homocysteine concentrations of subjects in this study agree with other reports of folate status of young women in the US during the post-fortification period (303,317). In addition, 41% of the women in this study (9 TT, 8 CC) were taking a folic acid containing supplement prior to the start of the study. This may have contributed to the higher overall folate values at baseline. These high values differ with the folate status in countries that do not practice folic acid fortification, where blood folate concentrations are much lower than those in the United States. For example, in a recent population-based study in the Netherlands (82), the mean serum folate and plasma

PAGE 148

134 homocysteine concentrations were 7.5 nmol/L and 13.6 mol/L, respectively, compared to the overall means of 47 nmol/L and 6.6 mol/L observed in the present study. In addition, subjects with the TT genotype for the MTHFR 677CT polymorphism in the Netherlands study had a mean plasma homocysteine concentration of 17.1 mol/L (82) compared to 7.1 mol/L in the present study. Individuals with chronically elevated homocysteine concentrations may be at an increased risk for cardiovascular disease (335), NTDs (138), or pregnancy complications (139). In conclusion, subjects with the TT genotype responded more negatively to folate depletion (115 g DFE/d) than subjects with the CC genotype as indicated by the lower serum folate concentration post depletion and the trend for an increased plasma homocysteine concentration post-depletion. In addition, subjects with the TT genotype responded less positively to repletion with the current RDA for folate (400 g DFE/d) than subjects with the CC genotype as evidenced by the significantly lower red blood cell folate concentration post-repletion and the lack of a significant decrease in plasma homocysteine concentration post-repletion. Plasma homocysteine concentration also tended to be higher in subjects with the TT genotype post-repletion. Therefore, subjects with the TT genotype may require more daily folate to maintain status than subjects with the CC genotype. In addition, the data suggest that changes in global DNA methylation of subjects with the TT genotype responded more favorably to repletion than subjects with the CC genotype.

PAGE 149

CHAPTER 6 SUMMARY A 14 wk metabolic study was conducted to determine differences in the folate status and DNA methylation response of women of childbearing age with the CC and TT genotypes for the MTHFR 677CT polymorphism to folate depletion (115 g DFE/d) and repletion with the current RDA for folate (400 g DFE/d). Serum and red blood cell folate and plasma homocysteine concentration, [ 3 H]methyl group incorporation, and mCyt/tCyt ratio were evaluated in response to folate depletion and repletion. The findings of the current study indicate that women of reproductive age with the TT genotype for the MTHFR 677CT polymorphism who consume low-folate diets are at greater risk for impaired folate status than women with the CC genotype. Individuals with the TT genotype for the MTHFR 677CT polymorphism who are folate deficient are more likely to develop an increased homocysteine concentration, a risk factor for NTDs and pregnancy complications, than individuals with the CC genotype (7,139). These data suggest that women of childbearing age with the TT genotype may require higher folate intakes to maintain normal folate status and prevent an elevation in plasma homocysteine concentration than women with the CC genotype. These data also suggest that DNA methylation of women with the TT genotype may respond more positively to increased folate intake compared to women with the CC genotype. The results of this study provide important new data that can be considered in future revisions of the folate DRI in which genotype differences are considered. 135

PAGE 150

APPENDIX A VITAMIN-MINERAL SUPPLEMENT COMPOSITION Table A-1. Vitamin/mineral RDA, five-day diet average, and supplement calculations. Vitamin/mineral RDA for young women 5 d average from diet Amount in supplement Vitamin A (RE) 800 799 0.8 Vitamin D (g) 5 5.7 0 Vitamin E (g) 15 12.5 2.5 Vitamin K (g) 65 64.8 0.2 Vitamin C (mg) 75 80.5 0 Thiamin (mg) 1.1 0.8 0.3 Riboflavin (mg) 1.1 1.2 0 Niacin (mg) 14 15.2 0 Pantothenic Acid (mg) 5 3.4 1.6 Vitamin B 6 (mg) 1.3 1.3 0.04 Folate (g DFE) 400 115 0 Vitamin B 12 (g) 2.4 2.4 0 Calcium (mg) 1000 830 170 Phosphorus (mg) 700 1138 0 Magnesium (mg) 310 232 77.8 Iron (mg) 15 9.1 5.9 Zinc (mg) 12 7.5 4.5 Copper (g) 3 1.1 1.9 Selenium (g) 55 84.9 0 Sodium (mg) 2400 3064 0 Potassium (mg) 2000 1810 190 RE = retinol equivalents; g = micrograms; mg = milligrams 136

PAGE 151

APPENDIX B CHOLINE CONTENT OF DIET Table B-1. First analysis of dietary choline content. Day Breakfast & Snack Lunch & Snack Dinner & Snack Average Daily Total (mg) A 62 111 187 359 B 35 84 111 229 C 39 106 135 280 D 34 99 103 235 E 58 60 209 327 Daily Average: 286 57 mg Table B-2. Second analysis of dietary choline content. Day Breakfast & Snack Lunch & Snack Dinner & Snack Average Daily Total (mg) A 66 103 150 320 B 82 86 98 266 C 43 77 105 225 D 66 107 121 294 E 75 74 162 310 Daily Average: 283 38 mg 137

PAGE 152

138 APPENDIX C POLYMERASE CHAIN REACTION GEL 198 bp 175 bp ct ct ct ct cc tt cc cc cc tt cc 2 3 4 5 6 7 8 9 10 11 12 Lane Figure C-1. Sample gel with amplified DNA fragments. Amplified DNA fragments were separated on an agarose gel and visualized by the use of ethidium bromide. Individuals with the MTHFR 677CT CC genotype have DNA fragments that are 198 base pairs long. This forms a single band as seen in Lane 2. Individuals with the MTHFR 677CT TT genotype (Lane 3) have DNA fragments that are 175 and 23 base pairs long (the 23 base pair band is not shown above). When the gene contains both C and T alleles, as seen in individuals heterozygous for the MTHFR 677CT polymorphism (Lane 4), three bands appear: one 198 base pairs long, one 175 base pairs long, and one 23 base pairs long (the 23 base pair band is not shown above).

PAGE 153

LIST OF REFERENCES 1. Bailey, L. B., Moyers, S., and Gregory, J.F. (2001) Folate. In: Present Knowledge in Nutrition (Bowman, B. A. a. R., R.M., ed.), pp. 214-229. ILSI Press, Washington, D.C. 2. Bailey, L. B. & Gregory, J. F. (1999) Polymorphisms of methylenetetrahydrofolate reductase and other enzymes: metabolic significance, risks and impact on folate requirement. J Nutr 129: 919-922. 3. Frosst, P., Blom, H. J., Milos, R., Goyette, P., Sheppard, C. A., Matthews, R. G., Boers, G. J., den Heijer, M., Kluijtmans, L. A., van den Heuvel, L. P. & et al. (1995) A candidate genetic risk factor for vascular disease: a common mutation in methylenetetrahydrofolate reductase. Nat Genet 10: 111-113. 4. Botto, L. & Yang, Q. (2000) 5,10-Methylenetetrahydrofolate reductase gene variants and congenital anomalies: a HuGE review. Am. J. Epidemiol. 151: 862-877. 5. Yamada, K., Chen, Z., Rozen, R. & Matthews, R. G. (2001) Effects of common polymorphisms on the properties of recombinant human methylenetetrahydrofolate reductase. PNAS 98: 14853-14858. 6. Bagley, P. J. & Selhub, J. (1998) A common mutation in the methylenetetrahydrofolate reductase gene is associated with an accumulation of formylated tetrahydrofolates in red blood cells. Proc Natl Acad Sci U S A 95: 13217-13220. 7. Jacques, P. F., Bostom, A. G., Williams, R. R., Ellison, R. C., Eckfeldt, J. H., Rosenberg, I. H., Selhub, J. & Rozen, R. (1996) Relation between folate status, a common mutation in methylenetetrahydrofolate reductase, and plasma homocysteine concentrations. Circulation 93: 7-9. 8. Adams, R. L. (1990) DNA methylation. The effect of minor bases on DNA-protein interactions. Biochem J 265: 309-320. 9. Jones, P. A. & Takai, D. (2001) The role of DNA methylation in mammalian epigenetics. Science 293: 1068-1070. 139

PAGE 154

140 10. Pogribny, I. P., Basnakian, A. G., Miller, B. J., Lopatina, N. G., Poirier, L. A. & James, S. J. (1995) Breaks in genomic DNA and within the p53 gene are associated with hypomethylation in livers of folate/methyl-deficient rats. Cancer Res 55: 1894-1901. 11. Blount, B. C., Mack, M. M., Wehr, C. M., MacGregor, J. T., Hiatt, R. A., Wang, G., Wickramasinghe, S. N., Everson, R. B. & Ames, B. N. (1997) Folate deficiency causes uracil misincorporation into human DNA and chromosome breakage: implications for cancer and neuronal damage. Proc Natl Acad Sci U S A 94: 3290-3295. 12. Stern, L. L., Mason, J. B., Selhub, J. & Choi, S. W. (2000) 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 9: 849-853. 13. Friso, S., Choi, S. W., Girelli, D., Mason, J. B., Dolnikowski, G. G., Bagley, P. J., Olivieri, O., Jacques, P. F., Rosenberg, I. H., Corrocher, R. & Selhub, J. (2002) A common mutation in the 5,10-methylenetetrahydrofolate reductase gene affects genomic DNA methylation through an interaction with folate status. Proc Natl Acad Sci U S A 99: 5606-5611. 14. Lynberg, M. C. & Edmonds, L. D. (1994) State use of birth defects surveillancs. Birth outcomes: from data to action. In: CDC Public Health Surveillance for Women, Infants, and Children, pp. 217-229. Centers for Disease Control and Prevention, Atlanta, GA. 15. van der Put, N. M., Steegers-Theunissen, R. P., Frosst, P., Trijbels, F. J., Eskes, T. K., van den Heuvel, L. P., Mariman, E. C., den Heyer, M., Rozen, R. & Blom, H. J. (1995) Mutated methylenetetrahydrofolate reductase as a risk factor for spina bifida. Lancet 346: 1070-1071. 16. Shields, D. C., Kirke, P. N., Mills, J. L., Ramsbottom, D., Molloy, A. M., Burke, H., Weir, D. G., Scott, J. M. & Whitehead, A. S. (1999) 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 64: 1045-1055. 17. Shane, B. (1995) Folate chemistry and metabolism. In: Folate in Health and Disease (Bailey, L. B., ed.), pp. 1-22. Marcel Dekker, New York. 18. Suitor, C. W. & Bailey, L. B. (2000) Dietary folate equivalents: interpretation and application. J Am Diet Assoc 100: 88-94. 19. Gregory, J. F. (1989) Chemical and nutritional aspects of folate research: analytical procedures, methods of folate synthesis, stability, and bioavailability of dietary folates. Adv Food Nutr Res 33: 1-101.

PAGE 155

141 20. FDA (1996) Food standards: amendment of standards of identity for enriched grain products to require addition of folic acid. Fed Regist 61: 8781-8789. 21. Lewis, C. J., Crane, N. T., Wilson, D. B. & Yetley, E. A. (1999) Estimated folate intakes: data updated to reflect food fortification, increased bioavailability, and dietary supplement use. Am J Clin Nutr 70: 198-207. 22. Gregory, J. F. (1995) The Bioavailability of Folate. In: Folate in Health and Disease (Bailey, L. B., ed.), pp. 195-235. Marcel Dekker, New York, NY. 23. Sauberlich, H. E., Kretsch, M. J., Skala, J. H., Johnson, H. L. & Taylor, P. C. (1987) Folate requirement and metabolism in nonpregnant women. Am J Clin Nutr 46: 1016-1028. 24. Sauberlich, H. E. (1985) Bioavailability of vitamins. Prog Food Nutr Sci 9: 1-33. 25. Gregory, J. F. (1997) Bioavailability of folate. Eur J Clin Nutr 51 Suppl 1: S54-59. 26. Pfeiffer, C. M., Rogers, L. M., Bailey, L. B. & Gregory, J. F. (1997) Absorption of folate from fortified cereal-grain products and of supplemental folate consumed with or without food determined by using a dual-label stable-isotope protocol. Am J Clin Nutr 66: 1388-1397. 27. IOM (1998) Folate. In: Dietary Reference Intakes, pp. 196-305. National Academy Press, Washington D.C. 28. Bailey, L. B. (1998) Dietary reference intakes for folate: the debut of dietary folate equivalents. Nutr Rev 56: 294-299. 29. Halsted, C. H., Ling, E. H., Luthi-Carter, R., Villanueva, J. A., Gardner, J. M. & Coyle, J. T. (1998) Folylpoly-gamma-glutamate carboxypeptidase from pig jejunum. Molecular characterization and relation to glutamate carboxypeptidase II. J Biol Chem 273: 20417-20424. 30. Reisenauer, A. M., Krumdieck, C. L. & Halsted, C. H. (1977) Folate conjugase: two separate activities in human jejunum. Science 198: 196-197. 31. Zimmerman, J., Gilula, Z., Selhub, J. & Rosenberg, I. H. (1989) Kinetic analysis of the effect of luminal pH on transport of folic acid in the small intestine. Int J Vitam Nutr Res 59: 151-156. 32. Said, H. M., Chatterjee, N., Haq, R. U., Subramanian, V. S., Ortiz, A., Matherly, L. H., Sirotnak, F. M., Halsted, C. & Rubin, S. A. (2000) Adaptive regulation of intestinal folate uptake: effect of dietary folate deficiency. Am J Physiol Cell Physiol 279: C1889-1895. 33. Steinberg, S. E. (1984) Mechanisms of folate homeostasis. Am J Physiol 246: G319-324.

PAGE 156

142 34. Stokstad, E. L. (1990) Historical Perspectives on Key Advances in the Biochemistry and Physiology of Folates. In: Folic Acid Metabolism in Heakth and Disease (Picciani, M. F., Stokstad, E. L. & Gregory, J. F., eds.), pp. 1-15. Wiley-Liss, New York, NY. 35. Finnell, R. H., Spiegelstein, O., Wlodarczyk, B., Triplett, A., Pogribny, I. P., Melnyk, S. & James, J. S. (2002) DNA methylation in Folbp1 knockout mice supplemented with folic acid during gestation. J Nutr 132: 2457S-2461S. 36. Chango, A., Emery-Fillon, N., de Courcy, G. P., Lambert, D., Pfister, M., Rosenblatt, D. S. & Nicolas, J. P. (2000) A polymorphism (80G->A) in the reduced folate carrier gene and its associations with folate status and homocysteinemia. Mol Genet Metab 70: 310-315. 37. Piedrahita, J. A., Oetama, B., Bennett, G. D., van Waes, J., Kamen, B. A., Richardson, J., Lacey, S. W., Anderson, R. G. & Finnell, R. H. (1999) Mice lacking the folic acid-binding protein Folbp1 are defective in early embryonic development. Nat Genet 23: 228-232. 38. Antony, A. C. (1996) Folate receptors. Annu Rev Nutr 16: 501-521. 39. Hresko, R. C., Kruse, M., Strube, M. & Mueckler, M. (1994) Topology of the Glut 1 glucose transporter deduced from glycosylation scanning mutagenesis. J Biol Chem 269: 20482-20488. 40. Zhao, R., Russell, R. G., Wang, Y., Liu, L., Gao, F., Kneitz, B., Edelmann, W. & Goldman, I. D. (2001) Rescue of embryonic lethality in reduced folate carrier-deficient mice by maternal folic acid supplementation reveals early neonatal failure of hematopoietic organs. J Biol Chem 276: 10224-10228. 41. De Marco, P., Calevo, M. G., Moroni, A., Merello, E., Raso, A., Finnell, R. H., Zhu, H., Andreussi, L., Cama, A. & Capra, V. (2003) Reduced folate carrier polymorphism (80A-->G) and neural tube defects. Eur J Hum Genet 11: 245-252. 42. Herbert, V. (1962) Minimal daily adult folate requirement. Arch Intern Med 110: 649-652. 43. Shane, B. & Stokstad, E. L. (1985) Vitamin B12-folate interrelationships. Annu Rev Nutr 5: 115-141. 44. Garrow, T. A. (2001) Betaine-Dependent Remethylation. In: Homocysteine in Health and Disease (Carmel, R. & Jacobsen, D. W., eds.), pp. 145-152. Cambridge University Press, Cambridge, UK. 45. Horne, D. W., Cook, R. J. & Wagner, C. (1989) Effect of dietary methyl group deficiency on folate metabolism in rats. J Nutr 119: 618-621.

PAGE 157

143 46. Varela-Moreiras, G., Selhub, J., daCosta, K.-A. & Zeisel, S. H. (1992) Effect of chronic choline deficiency in rats on liver folate content and distribution. J Nutr Biochem 3: 519-522. 47. Varela-Moreiras, G., Ragel, C. & Perez de Miguelsanz, J. (1995) Choline deficiency and methotrexate treatment induces marked but reversible changes in hepatic folate concentrations, serum homocysteine and DNA methylation rates in rats. J Am Coll Nutr 14: 480-485. 48. Lai, C. & Shields, P. G. (1999) The role of interindividual variation in human carcinogenesis. J Nutr 129: 552S-555S. 49. Kang, S. S., Wong, P. W., Zhou, J. M., Sora, J., Lessick, M., Ruggie, N. & Grcevich, G. (1988) Thermolabile methylenetetrahydrofolate reductase in patients with coronary artery disease. Metabolism 37: 611-613. 50. Goyette, P., Sumner, J. S., Milos, R., Duncan, A. M., Rosenblatt, D. S., Matthews, R. G. & Rozen, R. (1994) Human methylenetetrahydrofolate reductase: isolation of cDNA mapping and mutation identification. Nat Genet 7: 551. 51. Goyette, P., Frosst, P., Rosenblatt, D. S. & Rozen, R. (1995) Seven novel mutations in the methylenetetrahydrofolate reductase gene and genotype/phenotype correlations in severe methylenetetrahydrofolate reductase deficiency. Am J Hum Genet 56: 1052-1059. 52. Sibani, S., Leclerc, D., Weisberg, I. S., O'Ferrall, E., Watkins, D., Artigas, C., Rosenblatt, D. S. & Rozen, R. (2003) Characterization of mutations in severe methylenetetrahydrofolate reductase deficiency reveals an FAD-responsive mutation. Hum Mutat 21: 509-520. 53. Rozen, R. (1997) Genetic predisposition to hyperhomocysteinemia: deficiency of methylenetetrahydrofolate reductase (MTHFR). Thromb Haemost 78: 523-526. 54. Matthews, R. G., Vanoni, M. A., Khani, S., Hainfield, J. F. & Wall, J. (1984) Flavins and flavoproteins : proceedings of the Eighth International Symposium. Flavins and Flavoproteins, Brighton, England. 55. Sumner, J., Jencks, D. A., Khani, S. & Matthews, R. G. (1986) Photoaffinity labeling of methylenetetrahydrofolate reductase with 8-azido-S-adenosylmethionine. J Biol Chem 261: 7697-7700. 56. Matthews, R. G., Sheppard, C. & Goulding, C. (1998) Methylenetetrahydrofolate reductase and methionine synthase: biochemistry and molecular biology. Eur J Pediatr 157 Suppl 2: S54-59. 57. Weisberg, I., Tran, P., Christensen, B., Sibani, S. & Rozen, R. (1998) A second genetic polymorphism in methylenetetrahydrofolate reductase (MTHFR) associated with decreased enzyme activity. Mol Genet Metab 64: 169-172.

PAGE 158

144 58. Guenther, B. D., Sheppard, C. A., Tran, P., Rozen, R., Matthews, R. G. & Ludwig, M. L. (1999) The structure and properties of methylenetetrahydrofolate reductase from Escherichia coli suggest how folate ameliorates human hyperhomocysteinemia. Nat Struct Biol 6: 359-365. 59. Matthews, R. G. (2002) Methylenetetrahydrofolate reductase: A common human polymorphism and its biochemical implications. Chem Rec 2: 4-12. 60. Hustad, S., Ueland, P. M., Vollset, S. E., Zhang, Y., Bjorke-Monsen, A. L. & Schneede, J. (2000) Riboflavin as a Determinant of Plasma Total Homocysteine: Effect Modification by the Methylenetetrahydrofolate Reductase C677T Polymorphism. Clin Chem 46: 1065-1071. 61. McNulty, H., McKinley, M. C., Wilson, B., McPartlin, J., Strain, J., Weir, D. G. & Scott, J. M. (2002) Impaired functioning of thermolabile methylenetetrahydrofolate reductase is dependent on riboflavin status: implications for riboflavin requirements. Am J Clin Nutr 76: 436-441. 62. Jacques, P. F., Kalmbach, R., Bagley, P. J., Russo, G. T., Rogers, G., Wilson, P. W., Rosenberg, I. H. & Selhub, J. (2002) 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 132: 283-288. 63. Moat, S. J., Ashfield-Watt, P. A., Powers, H. J., Newcombe, R. G. & McDowell, I. F. (2003) Effect of riboflavin status on the homocysteine-lowering effect of folate in relation to the MTHFR (C677T) genotype. Clin Chem 49: 295-302. 64. Rozen, R. (2002) Methylenetetrahydrofolate reductase: a link between folate and riboflavin? Am J Clin Nutr 76: 301-302. 65. Brattstrom, L., Wilcken, D. E., Ohrvik, J. & Brudin, L. (1998) Common methylenetetrahydrofolate reductase gene mutation leads to hyperhomocysteinemia but not to vascular disease: the result of a meta-analysis. Circulation 98: 2520-2526. 66. Stevenson, R. E., Schwartz, C. E., Du, Y. Z. & Adams, M. J., Jr. (1997) Differences in methylenetetrahydrofolate reductase genotype frequencies, between Whites and Blacks. Am J Hum Genet 60: 229-230. 67. Zittoun, J., Tonetti, C., Bories, D., Pignon, J. M. & Tulliez, M. (1998) Plasma homocysteine levels related to interactions between folate status and methylenetetrahydrofolate reductase: a study in 52 healthy subjects. Metabolism 47: 1413-1418.

PAGE 159

145 68. van der Put, N. M., Gabreels, F., Stevens, E. M., Smeitink, J. A., Trijbels, F. J., Eskes, T. K., van den Heuvel, L. P. & Blom, H. J. (1998) A second common mutation in the methylenetetrahydrofolate reductase gene: an additional risk factor for neural-tube defects? Am J Hum Genet 62: 1044-1051. 69. Viel, A., Dall'Agnese, L., Simone, F., Canzonieri, V., Capozzi, E., Visentin, M. C., Valle, R. & Boiocchi, M. (1997) Loss of heterozygosity at the 5,10-methylenetetrahydrofolate reductase locus in human ovarian carcinomas. Br J Cancer 75: 1105-1110. 70. Friedman, G., Goldschmidt, N., Friedlander, Y., Ben-Yehuda, A., Selhub, J., Babaey, S., Mendel, M., Kidron, M. & Bar-On, H. (1999) A common mutation A1298C in human methylenetetrahydrofolate reductase gene: association with plasma total homocysteine and folate concentrations. J Nutr 129: 1656-1661. 71. Chango, A., Boisson, F., Barbe, F., Quilliot, D., Droesch, S., Pfister, M., Fillon-Emery, N., Lambert, D., Fremont, S., Rosenblatt, D. S. & Nicolas, J. P. (2000) The effect of 677C-->T and 1298A-->C mutations on plasma homocysteine and 5,10-methylenetetrahydrofolate reductase activity in healthy subjects. Br J Nutr 83: 593-596. 72. Dekou, V., Whincup, P., Papacosta, O., Ebrahim, S., Lennon, L., Ueland, P. M., Refsum, H., Humphries, S. E. & Gudnason, V. (2001) The effect of the C677T and A1298C polymorphisms in the methylenetetrahydrofolate reductase gene on homocysteine levels in elderly men and women from the British regional heart study. Atherosclerosis 154: 659-666. 73. Bailey, L. B., Duhaney, R. L., Maneval, D. R., Kauwell, G. P., Quinlivan, E. P., Davis, S. R., Cuadras, A., Hutson, A. D. & Gregory, J. F. (2002) Vitamin B-12 status is inversely associated with plasma homocysteine in young women with C677T and/or A1298C methylenetetrahydrofolate reductase polymorphisms. J Nutr 132: 1872-1878. 74. Kang, S. S., Passen, E. L., Ruggie, N., Wong, P. W. & Sora, H. (1993) Thermolabile defect of methylenetetrahydrofolate reductase in coronary artery disease. Circulation 88: 1463-1469. 75. Engbersen, A. M., Franken, D. G., Boers, G. H., Stevens, E. M., Trijbels, F. J. & Blom, H. J. (1995) Thermolabile 5,10-methylenetetrahydrofolate reductase as a cause of mild hyperhomocysteinemia. Am J Hum Genet 56: 142-150. 76. Deloughery, T. G., Evans, A., Sadeghi, A., McWilliams, J., Henner, W. D., Taylor, L. M., Jr. & Press, R. D. (1996) Common mutation in methylenetetrahydrofolate reductase. Correlation with homocysteine metabolism and late-onset vascular disease. Circulation 94: 3074-3078.

PAGE 160

146 77. Harmon, D. L., Woodside, J. V., Yarnell, J. W., McMaster, D., Young, I. S., McCrum, E. E., Gey, K. F., Whitehead, A. S. & Evans, A. E. (1996) The common 'thermolabile' variant of methylene tetrahydrofolate reductase is a major determinant of mild hyperhomocysteinaemia. Qjm 89: 571-577. 78. Malinow, M. R., Nieto, F. J., Kruger, W. D., Duell, P. B., Hess, D. L., Gluckman, R. A., Block, P. C., Holzgang, C. R., Anderson, P. H., Seltzer, D., Upson, B. & Lin, Q. R. (1997) The effects of folic acid supplementation on plasma total homocysteine are modulated by multivitamin use and methylenetetrahydrofolate reductase genotypes. Arterioscler Thromb Vasc Biol 17: 1157-1162. 79. Gudnason, V., Stansbie, D., Scott, J., Bowron, A., Nicaud, V. & Humphries, S. (1998) C677T (thermolabile alanine/valine) polymorphism in methylenetetrahydrofolate reductase (MTHFR): its frequency and impact on plasma homocysteine concentration in different European populations. EARS group. Atherosclerosis 136: 347-354. 80. Nelen, W. L., Blom, H. J., Thomas, C. M., Steegers, E. A., Boers, G. H. & Eskes, T. K. (1998) 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 128: 1336-1341. 81. Ashfield-Watt, P. A., Pullin, C. H., Whiting, J. M., Clark, Z. E., Moat, S. J., Newcombe, R. G., Burr, M. L., Lewis, M. J., Powers, H. J. & McDowell, I. F. (2002) Methylenetetrahydrofolate reductase 677C->T genotype modulates homocysteine responses to a folate-rich diet or a low-dose folic acid supplement: a randomized controlled trial. Am J Clin Nutr 76: 180-186. 82. de Bree, A., Verschuren, W. M., Bjorke-Monsen, A. L., van der Put, N. M., Heil, S. G., Trijbels, F. J. & Blom, H. J. (2003) Effect of the methylenetetrahydrofolate reductase 677C-->T mutation on the relations among folate intake and plasma folate and homocysteine concentrations in a general population sample. Am J Clin Nutr 77: 687-693. 83. Husemoen, L. L., Thomsen, T. F., Fenger, M., Jorgensen, H. L. & Jorgensen, T. (2003) Contribution of thermolabile methylenetetrahydrofolate reductase variant to total plasma homocysteine levels in healthy men and women. Inter99 (2). Genet Epidemiol 24: 322-330. 84. Kauwell, G. P., Wilsky, C. E., Cerda, J. J., Herrlinger-Garcia, K., Hutson, A. D., Theriaque, D. W., Boddie, A., Rampersaud, G. C. & Bailey, L. B. (2000) Methylenetetrahydrofolate reductase mutation (677C-->T) negatively influences plasma homocysteine response to marginal folate intake in elderly women. Metabolism 49: 1440-1443.

PAGE 161

147 85. Molloy, A. M., Mills, J. L., Kirke, P. N., Ramsbottom, D., McPartlin, J. M., Burke, H., Conley, M., Whitehead, A. S., Weir, D. G. & Scott, J. M. (1998) 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 78: 155-159. 86. Molloy, A. M., Daly, S., Mills, J. L., Kirke, P. N., Whitehead, A. S., Ramsbottom, D., Conley, M. R., Weir, D. G. & Scott, J. M. (1997) Thermolabile variant of 5,10-methylenetetrahydrofolate reductase associated with low red-cell folates: implications for folate intake recommendations. Lancet 349: 1591-1593. 87. Guinotte, C. L., Burns, M. G., Axume, J. A., Hata, H., Urrutia, T. F., Alamilla, A., McCabe, D., Singgih, A., Cogger, E. A. & Caudill, M. A. (2003) Methylenetetrahydrofolate reductase 677CT variant modulates folate status response to controlled Folate intakes in young women. J. Nutr. 133: 1272-1280. 88. Kang, S. S., Wong, P. W., Susmano, A., Sora, J., Norusis, M. & Ruggie, N. (1991) Thermolabile methylenetetrahydrofolate reductase: an inherited risk factor for coronary artery disease. Am J Hum Genet 48: 536-545. 89. Kang, S. S. & Wong, P. W. (1996) Genetic and nongenetic factors for moderate hyperhomocyst(e)inemia. Atherosclerosis 119: 135-138. 90. Kluijtmans, L. A., van den Heuvel, L. P., Boers, G. H., Frosst, P., Stevens, E. M., van Oost, B. A., den Heijer, M., Trijbels, F. J., Rozen, R. & Blom, H. J. (1996) 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 58: 35-41. 91. Kluijtmans, L. A., Kastelein, J. J., Lindemans, J., Boers, G. H., Heil, S. G., Bruschke, A. V., Jukema, J. W., van den Heuvel, L. P., Trijbels, F. J., Boerma, G. J., Verheugt, F. W., Willems, F. & Blom, H. J. (1997) Thermolabile methylenetetrahydrofolate reductase in coronary artery disease. Circulation 96: 2573-2577. 92. Morita, H., Taguchi, J., Kurihara, H., Kitaoka, M., Kaneda, H., Kurihara, Y., Maemura, K., Shindo, T., Minamino, T., Ohno, M., Yamaoki, K., Ogasawara, K., Aizawa, T., Suzuki, S. & Yazaki, Y. (1997) Genetic polymorphism of 5,10-methylenetetrahydrofolate reductase (MTHFR) as a risk factor for coronary artery disease. Circulation 95: 2032-2036. 93. Wilcken, D. E., Wang, X. L., Sim, A. S. & McCredie, R. M. (1996) Distribution in healthy and coronary populations of the methylenetetrahydrofolate reductase (MTHFR) C677T mutation. Arterioscler Thromb Vasc Biol 16: 878-882.

PAGE 162

148 94. Szczeklik, A., Sanak, M., Jankowski, M., Dropinski, J., Czachor, R., Musial, J., Axenti, I., Twardowska, M., Brzostek, T. & Tendera, M. (2001) Mutation A1298C of methylenetetrahydrofolate reductase: risk for early coronary disease not associated with hyperhomocysteinemia. Am J Med Genet 101: 36-39. 95. Chen, Z., Karaplis, A. C., Ackerman, S. L., Pogribny, I. P., Melnyk, S., Lussier-Cacan, S., Chen, M. F., Pai, A., John, S. W., Smith, R. S., Bottiglieri, T., Bagley, P., Selhub, J., Rudnicki, M. A., James, S. J. & Rozen, R. (2001) Mice deficient in methylenetetrahydrofolate reductase exhibit hyperhomocysteinemia and decreased methylation capacity, with neuropathology and aortic lipid deposition. Hum Mol Genet 10: 433-443. 96. Rothenbacher, D., Fischer, H. G., Hoffmeister, A., Hoffmann, M. M., Marz, W., Bode, G., Rosenthal, J., Koenig, W. & Brenner, H. (2002) Homocysteine and methylenetetrahydrofolate reductase genotype: association with risk of coronary heart disease and relation to inflammatory, hemostatic, and lipid parameters. Atherosclerosis 162: 193-200. 97. Moyers, S. & Bailey, L. B. (2001) Fetal malformations and folate metabolism: review of recent evidence. Nutr Rev 59: 215-224. 98. Chen, J., Giovannucci, E. L. & Hunter, D. J. (1999) MTHFR polymorphism, methyl-replete diets and the risk of colorectal carcinoma and adenoma among U.S. men and women: an example of gene-environment interactions in colorectal tumorigenesis. J Nutr 129: 560S-564S. 99. Chen, J., Giovannucci, E., Kelsey, K., Rimm, E. B., Stampfer, M. J., Colditz, G. A., Spiegelman, D., Willett, W. C. & Hunter, D. J. (1996) A methylenetetrahydrofolate reductase polymorphism and the risk of colorectal cancer. Cancer Res 56: 4862-4864. 100. Ma, J., Stampfer, M. J., Giovannucci, E., Artigas, C., Hunter, D. J., Fuchs, C., Willett, W. C., Selhub, J., Hennekens, C. H. & Rozen, R. (1997) Methylenetetrahydrofolate reductase polymorphism, dietary interactions, and risk of colorectal cancer. Cancer Res 57: 1098-1102. 101. Toffoli, G., Gafa, R., Russo, A., Lanza, G., Dolcetti, R., Sartor, F., Libra, M., Viel, A. & Boiocchi, M. (2003) Methylenetetrahydrofolate reductase 677 C-->t polymorphism and risk of proximal colon cancer in north Italy. Clin Cancer Res 9: 743-748. 102. Levine, A. J., Siegmund, K. D., Ervin, C. M., Diep, A., Lee, E. R., Frankl, H. D. & Haile, R. W. (2000) The methylenetetrahydrofolate reductase 677C-->T polymorphism and distal colorectal adenoma risk. Cancer Epidemiol Biomarkers Prev 9: 657-663.

PAGE 163

149 103. Marugame, T., Tsuji, E., Kiyohara, C., Eguchi, H., Oda, T., Shinchi, K. & Kono, S. (2003) Relation of plasma folate and methylenetetrahydrofolate reductase C677T polymorphism to colorectal adenomas. Int J Epidemiol 32: 64-66. 104. Skibola, C. F., Smith, M. T., Kane, E., Roman, E., Rollinson, S., Cartwright, R. A. & Morgan, G. (1999) Polymorphisms in the methylenetetrahydrofolate reductase gene are associated with susceptibility to acute leukemia in adults. PNAS 96: 12810-12815. 105. Wiemels, J. L., Smith, R. N., Taylor, G. M., Eden, O. B., Alexander, F. E., Greaves, M. F. & United Kingdom Childhood Cancer Study Investigators, n. (2001) Methylenetetrahydrofolate reductase (MTHFR) polymorphisms and risk of molecularly defined subtypes of childhood acute leukemia. PNAS 98: 4004-4009. 106. Franco, R. F., Simoes, B. P., Tone, L. G., Gabellini, S. M., Zago, M. A. & Falcao, R. P. (2001) The methylenetetrahydrofolate reductase C677T gene polymorphism decreases the risk of childhood acute lymphocytic leukaemia. Br J Haematol 115: 616-618. 107. McGlynn, K. A., Wang, L. & Patrick-Acevedo, N. Y. (2000) Methylenetetrahydrofolate reductase and methionine synthase, folate, alcohol and breast cancer. Proc Am Assoc Canc Res 41: 588. 108. Campbell, I. G., Baxter, S. W., Eccles, D. M. & Choong, D. Y. (2002) Methylenetetrahydrofolate reductase polymorphism and susceptibility to breast cancer. Breast Cancer Res 4: R14. 109. Sharp, L., Little, J., Schofield, A. C., Pavlidou, E., Cotton, S. C., Miedzybrodzka, Z., Baird, J. O. C., Haites, N. E., Heys, S. D. & Grubb, D. A. (2002) Folate and breast cancer: the role of polymorphisms in methylenetetrahydrofolate reductase (MTHFR). Cancer Letters 181: 65-71. 110. Semenza, J. C., Delfino, R. J., Ziogas, A. & Anton-Culver, H. (2003) Breast cancer risk and methylenetetrahydrofolate reductase polymorphism. Breast Cancer Res Treat 77: 217-223. 111. Esteller, M., Garcia, A., Martinez-Palones, J. M., Xercavins, J. & Reventos, J. (1997) Germ line polymorphisms in cytochrome-P450 1A1 (C4887 CYP1A1) and methylenetetrahydrofolate reductase (MTHFR) genes and endometrial cancer susceptibility. Carcinogenesis 18: 2307-2311. 112. Shen, H., Xu, Y., Zheng, Y., Qian, Y., Yu, R., Qin, Y., Wang, X., Spitz, M. R. & Wei, Q. (2001) Polymorphisms of 5,10-methylenetetrahydrofolate reductase and risk of gastric cancer in a Chinese population: a case-control study. Int J Cancer 95: 332-336.

PAGE 164

150 113. Miao, X., Xing, D., Tan, W., Qi, J., Lu, W. & Lin, D. (2002) Susceptibility to gastric cardia adenocarcinoma and genetic polymorphisms in methylenetetrahydrofolate reductase in an at-risk Chinese population. Cancer Epidemiol Biomarkers Prev 11: 1454-1458. 114. Song, C., Xing, D., Tan, W., Wei, Q. & Lin, D. (2001) Methylenetetrahydrofolate reductase polymorphisms increase risk of esophageal squamous cell carcinoma in a Chinese population. Cancer Res 61: 3272-3275. 115. Chen, J., Ma, J., Stampfer, M. J., Palomeque, C., Selhub, J. & Hunter, D. J. (2002) Linkage disequilibrium between the 677C>T and 1298A>C polymorphisms in human methylenetetrahydrofolate reductase gene and their contributions to risk of colorectal cancer. Pharmacogenetics 12: 339-342. 116. Crott, J. W., Mashiyama, S. T., Ames, B. N. & Fenech, M. F. (2001) Methylenetetrahydrofolate reductase C677T polymorphism does not alter folic acid deficiency-induced uracil incorporation into primary human lymphocyte DNA in vitro. Carcinogenesis 22: 1019-1025. 117. Crott, J. W., Mashiyama, S. T., Ames, B. N. & Fenech, M. (2001) The effect of folic acid deficiency and MTHFR C677T polymorphism on chromosome damage in human lymphocytes in vitro. Cancer Epidemiol Biomarkers Prev 10: 1089-1096. 118. Roe, D. A. (1991) Evaluation of publicly available scientific evidence regarding certain nutrient-disease relationships. 1. Folic acid and neural tube defects, pp. 1-29, A21-A28. Life Sciences Research Office, Bethesda, MD. 119. Botto, L. D., Moore, C. A., Khoury, M. J. & Erickson, J. D. (1999) Neural-tube defects. N Engl J Med 341: 1509-1519. 120. Hibbard, B. M. (1964) The role of folic acid in pregnancy with particular reference to anaemia, abruption and abortion. J Obstet Gynaecol Br Commonw 71: 529-542. 121. Smithells, R. W., Sheppard, S., Schorah, C. J., Seller, M. J., Nevin, N. C., Harris, R., Read, A. P. & Fielding, D. W. (1980) Possible prevention of neural-tube defects by periconceptional vitamin supplementation. Lancet 1: 339-340. 122. Laurence, K. M., James, N., Miller, M. H., Tennant, G. B. & Campbell, H. (1981) Double-blind randomised controlled trial of folate treatment before conception to prevent recurrence of neural-tube defects. Br Med J (Clin Res Ed) 282: 1509-1511. 123. (1991) Prevention of neural tube defects: results of the Medical Research Council Vitamin Study. MRC Vitamin Study Research Group. Lancet 338: 131-137. 124. Czeizel, A. E. & Dudas, I. (1992) Prevention of the first occurrence of neural-tube defects by periconceptional vitamin supplementation. N Engl J Med 327: 1832-1835.

PAGE 165

151 125. Berry, R. J., Li, Z., Erickson, J. D., Li, S., Moore, C. A., Wang, H., Mulinare, J., Zhao, P., Wong, L. Y., Gindler, J., Hong, S. X. & Correa, A. (1999) Prevention of neural-tube defects with folic acid in China. China-U.S. Collaborative Project for Neural Tube Defect Prevention [corrected; erratum to be published]. N Engl J Med 341: 1485-1490. 126. Kirke, P. N., Molloy, A. M., Daly, L. E., Burke, H., Weir, D. G. & Scott, J. M. (1993) Maternal plasma folate and vitamin B12 are independent risk factors for neural tube defects. Q J Med 86: 703-708. 127. Daly, L. E., Kirke, P. N., Molloy, A., Weir, D. G. & Scott, J. M. (1995) Folate levels and neural tube defects. Implications for prevention. Jama 274: 1698-1702. 128. Moore, L. L., Bradlee, M. L., Singer, M. R., Rothman, K. J. & Milunsky, A. (2003) Folate intake and the risk of neural tube defects: an estimation of dose-response. Epidemiology 14: 200-205. 129. Mornet, E., Muller, F., Lenvoise-Furet, A., Delezoide, A. L., Col, J. Y., Simon-Bouy, B. & Serre, J. L. (1997) Screening of the C677T mutation on the methylenetetrahydrofolate reductase gene in French patients with neural tube defects. Hum Genet 100: 512-514. 130. Bjorke-Monsen, A. L., Ueland, P. M., Schneede, J., Vollset, S. E. & Refsum, H. (1997) Elevated plasma total homocysteine and C677T mutation of the methylenetetrahydrofolate reductase gene in patients with spina bifida. Qjm 90: 593-596. 131. de Franchis, R., Buoninconti, A., Mandato, C., Pepe, A., Sperandeo, M. P., Del Gado, R., Capra, V., Salvaggio, E., Andria, G. & Mastroiacovo, P. (1998) The C677T mutation of the 5,10-methylenetetrahydrofolate reductase gene is a moderate risk factor for spina bifida in Italy. J Med Genet 35: 1009-1013. 132. Koch, M. C., Stegmann, K., Ziegler, A., Schroter, B. & Ermert, A. (1998) Evaluation of the MTHFR C677T allele and the MTHFR gene locus in a German spina bifida population. Eur J Pediatr 157: 487-492. 133. Lucock, M., Daskalakis, I., Hinkins, M. & Yates, Z. (2001) An examination of polymorphic genes and folate metabolism in mothers affected by a spina bifida pregnancy. Mol Genet Metab 73: 322-332. 134. Rampersaud, E., Melvin, E. C., Siegel, D., Mehltretter, L., Dickerson, M. E., George, T. M., Enterline, D., Nye, J. S. & Speer, M. C. (2003) Updated investigations of the role of methylenetetrahydrofolate reductase in human neural tube defects. Clin Genet 63: 210-214.

PAGE 166

152 135. Parle-McDermott, A., Mills, J. L., Kirke, P. N., O'Leary, V. B., Swanson, D. A., Pangilinan, F., Conley, M., Molloy, A. M., Cox, C., Scott, J. M. & Brody, L. C. (2003) Analysis of the MTHFR 1298A-->C and 677C-->T polymorphisms as risk factors for neural tube defects. J Hum Genet 48: 190-193. 136. Isotalo, P. A., Wells, G. A. & Donnelly, J. G. (2000) Neonatal and fetal methylenetetrahydrofolate reductase genetic polymorphisms: an examination of C677T and A1298C mutations. Am J Hum Genet 67: 986-990. 137. Volcik, K. A., Blanton, S. H., Tyerman, G. H., Jong, S. T., Rott, E. J., Page, T. Z., Romaine, N. K. & Northrup, H. (2000) Methylenetetrahydrofolate reductase and spina bifida: evaluation of level of defect and maternal genotypic risk in Hispanics. Am J Med Genet 95: 21-27. 138. Mills, J. L., McPartlin, J. M., Kirke, P. N., Lee, Y. J., Conley, M. R., Weir, D. G. & Scott, J. M. (1995) Homocysteine metabolism in pregnancies complicated by neural-tube defects. Lancet 345: 149-151. 139. Vollset, S. E., Refsum, H., Irgens, L. M., Emblem, B. M., Tverdal, A., Gjessing, H. K., Monsen, A. L. & Ueland, P. M. (2000) Plasma total homocysteine, pregnancy complications, and adverse pregnancy outcomes: the Hordaland Homocysteine study. Am J Clin Nutr 71: 962-968. 140. Eskes, T. K. (1994) Possible basis for primary prevention of birth defects with folic acid. Fetal Diagn Ther 9: 149-154. 141. Steegers-Theunissen, R. P., Boers, G. H., Trijbels, F. J., Finkelstein, J. D., Blom, H. J., Thomas, C. M., Borm, G. F., Wouters, M. G. & Eskes, T. K. (1994) Maternal hyperhomocysteinemia: a risk factor for neural-tube defects? Metabolism 43: 1475-1480. 142. Rosenquist, T. H., Ratashak, S. A. & Selhub, J. (1996) Homocysteine induces congenital defects of the heart and neural tube: effect of folic acid. Proc Natl Acad Sci U S A 93: 15227-15232. 143. VanAerts, L. A., Blom, H. J., Deabreu, R. A., Trijbels, F. J., Eskes, T. K., Copius Peereboom-Stegeman, J. H. & Noordhoek, J. (1994) Prevention of neural tube defects by and toxicity of L-homocysteine in cultured postimplantation rat embryos. Teratology 50: 348-360. 144. Hansen, D. K., Grafton, T. F., Melnyk, S. & James, S. J. (2001) Lack of embryotoxicity of homocysteine thiolactone in mouse embryos in vitro. Reprod Toxicol 15: 239-244. 145. Greene, N. D. E., Dunlevy, L. P. E. & Copp, A. J. (2003) Homocysteine is embryotoxic but does not cause neural tube defects in mouse embryos. Anat Embryol 206: 185-191.

PAGE 167

153 146. Wenstrom, K. D., Johanning, G. L., Owen, J., Johnston, K. E., Acton, S., Cliver, S. & Tamura, T. (2000) Amniotic fluid homocysteine levels, 5,10-methylenetetrahydrafolate reductase genotypes, and neural tube closure sites. Am J Med Genet 90: 6-11. 147. Molloy, A. M., Mills, J. L., McPartlin, J., Kirke, P. N., Scott, J. M. & Daly, S. (2002) Maternal and fetal plasma homocysteine concentrations at birth: the influence of folate, vitamin B12, and the 5,10-methylenetetrahydrofolate reductase 677C-->T variant. Am J Obstet Gynecol 186: 499-503. 148. Wyszynski, D. F. & Diehl, S. R. (2000) Infant C677T mutation in MTHFR, maternal periconceptional vitamin use, and risk of nonsyndromic cleft lip. Am J Med Genet 92: 79-80. 149. Tolarova, M. (1982) Periconceptional supplementation with vitamins and folic acid to prevent recurrence of cleft lip. Lancet 2: 217. 150. Czeizel, A. E., Toth, M. & Rockenbauer, M. (1996) Population-based case control study of folic acid supplementation during pregnancy. Teratology 53: 345-351. 151. Czeizel, A. E., Timar, L. & Sarkozi, A. (1999) Dose-dependent effect of folic acid on the prevention of orofacial clefts. Pediatrics 104: e66. 152. Hayes, C., Werler, M. M., Willett, W. C. & Mitchell, A. A. (1996) Case-control study of periconceptional folic acid supplementation and oral clefts. Am J Epidemiol 143: 1229-1234. 153. Wong, W. Y., Eskes, T. K., Kuijpers-Jagtman, A. M., Spauwen, P. H., Steegers, E. A., Thomas, C. M., Hamel, B. C., Blom, H. J. & Steegers-Theunissen, R. P. (1999) Nonsyndromic orofacial clefts: association with maternal hyperhomocysteinemia. Teratology 60: 253-257. 154. Mills, J. L. (1999) Folate and oral clefts: where do we go from here? New directions in oral clefts research. Teratology 60: 251-252. 155. Mills, J. L., Kirke, P. N., Molloy, A. M., Burke, H., Conley, M. R., Lee, Y. J., Mayne, P. D., Weir, D. G. & Scott, J. M. (1999) Methylenetetrahydrofolate reductase thermolabile variant and oral clefts. Am J Med Genet 86: 71-74. 156. Gaspar, D. A., Pavanello, R. C., Zatz, M., Passos-Bueno, M. R., Andre, M., Steman, S., Wyszynski, D. F. & Matiolli, S. R. (1999) 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 87: 197-199. 157. Martinelli, M., Scapoli, L., Pezzetti, F., Carinci, F., Carinci, P., Stabellini, G., Bisceglia, L., Gombos, F. & Tognon, M. (2001) C677T variant form at the MTHFR gene and CL/P: a risk factor for mothers? Am J Med Genet 98: 357-360.

PAGE 168

154 158. Prescott, N. J., Winter, R. M. & Malcolm, S. (2003) Maternal MTHFR genotype contributes to the risk of non-syndromic cleft lip and palate. J Med Genet 39: 368-369. 159. Shotelersuk, V., Ittiwut, C., Siriwan, P. & Angspatt, A. (2003) Maternal 677CT/1298AC genotype of the MTHFR gene as a risk factor for cleft lip. J Med Genet 40. 160. Shaw, G. M., Rozen, R., Finnell, R. H., Todoroff, K. & Lammer, E. J. (1998) Infant C677T mutation in MTHFR, maternal periconceptional vitamin use, and cleft lip. Am J Med Genet 80: 196-198. 161. Shaw, G. M., Todoroff, K., Finnell, R. H., Rozen, R. & Lammer, E. J. (1999) Maternal vitamin use, infant C677T mutation in MTHFR, and isolated cleft palate risk. Am J Med Genet 85: 84-85. 162. Blanton, S. H., Kolle, B. S., Hecht, J. T., Mulliken, J. B. & Martin, E. R. (2000) No evidence supporting MTHFR as a risk factor in the development of familial NSCLP. Am J Med Genet 92: 370-371. 163. Blanton, S. H., Patel, S., Hecht, J. T. & Mulliken, J. B. (2002) MTHFR is not a risk factor in the development of isolated nonsyndromic cleft lip and palate. Am J Med Genet 110: 404-405. 164. Shaw, G. M., O'Malley, C. D., Wasserman, C. R., Tolarova, M. M. & Lammer, E. J. (1995) Maternal periconceptional use of multivitamins and reduced risk for conotruncal heart defects and limb deficiencies among offspring. Am J Med Genet 59: 536-545. 165. Scanlon, K. S., Ferencz, C., Loffredo, C. A., Wilson, P. D., Correa-Villasenor, A., Khoury, M. J. & Willett, W. C. (1998) Preconceptional folate intake and malformations of the cardiac outflow tract. Baltimore-Washington Infant Study Group. Epidemiology 9: 95-98. 166. Boot, M. J., Steegers-Theunissen, R. P., Poelmann, R. E., Van Iperen, L., Lindemans, J. & Gittenberger-De Groot, A. C. (2003) Folic acid and homocysteine affect neural crest and neuroepithelial cell outgrowth and differentiation in vitro. Dev Dyn 227: 301-308. 167. Storti, S., Vittorini, S., Lascone, M. R., Sacchelli, M., Collavoli, A., Ripoli, A., Cocchi, G., Biagini, A. & Clerico, A. (2003) Association between 5,10-methylenetetrahydrofolate reductase C677T and A1298C polymorphisms and conotruncal heart defects. Clin Chem Lab Med 41: 276-280.

PAGE 169

155 168. James, S. J., Pogribna, M., Pogribny, I. P., Melnyk, S., Hine, R. J., Gibson, J. B., Yi, P., Tafoya, D. L., Swenson, D. H., Wilson, V. L. & Gaylor, D. W. (1999) Abnormal folate metabolism and mutation in the methylenetetrahydrofolate reductase gene may be maternal risk factors for Down syndrome. Am J Clin Nutr 70: 495-501. 169. Hobbs, C. A., Sherman, S. L., Yi, P., Hopkins, S. E., Torfs, C. P., Hine, R. J., Pogribna, M., Rozen, R. & James, S. J. (2000) Polymorphisms in genes involved in folate metabolism as maternal risk factors for Down syndrome. Am J Hum Genet 67: 623-630. 170. O'Leary, V. B., Parle-McDermott, A., Molloy, A. M., Kirke, P. N., Johnson, Z., Conley, M., Scott, J. M. & Mills, J. L. (2002) MTRR and MTHFR polymorphism: link to Down syndrome? Am J Med Genet 107: 151-155. 171. Stuppia, L., Gatta, V., Gaspari, A. R., Antonucci, I., Morizio, E., Calabrese, G. & Palka, G. (2002) C677T mutation in the 5,10-MTHFR gene and risk of Down syndrome in Italy. Eur J Hum Genet 10: 388-390. 172. Yanamandra, K., Bocchini Jr, J. A. & Thurmon, T. F. (2003) Absence of association of fetal MTHFR C677T polymorphism with prenatal Down syndrome pregnancies. Eur J Hum Genet 11: 5. 173. Prasmusinto, D., Skrablin, S., Hofstaetter, C., Fimmers, R. & van der Ven, K. (2002) The methylenetetrahydrofolate reductase 677 C-->T polymorphism and preeclampsia in two populations. Obstet Gynecol 99: 1085-1092. 174. Sohda, S., Arinami, T., Hamada, H., Yamada, N., Hamaguchi, H. & Kubo, T. (1997) Methylenetetrahydrofolate reductase polymorphism and pre-eclampsia. J Med Genet 34: 525-526. 175. Lopez-Quesada, E., Vilaseca, M. A. & Lailla, J. M. (2003) Plasma total homocysteine in uncomplicated pregnancy and in preeclampsia. Eur J Obstet Gynecol Reprod Biol 108: 45-49. 176. Vollset, S. E., Refsum, H., Tverdal, A., Nygard, O., Nordrehaug, J. E., Tell, G. S. & Ueland, P. M. (2001) Plasma total homocysteine and cardiovascular and noncardiovascular mortality: the Hordaland Homocysteine Study. Am J Clin Nutr 74: 130-136. 177. Laivuori, H., Kaaja, R., Ylikorkala, O., Hiltunen, T. & Kontula, K. (2000) 677 C-->T polymorphism of the methylenetetrahydrofolate reductase gene and preeclampsia. Obstet Gynecol 96: 277-280. 178. Lissak, A., Sharon, A., Fruchter, O., Kassel, A., Sanderovitz, J. & Abramovici, H. (1999) Polymorphism for mutation of cytosine to thymine at location 677 in the methylenetetrahydrofolate reductase gene is associated with recurrent early fetal loss. Am J Obstet Gynecol 181: 126-130.

PAGE 170

156 179. Arruda, V. R., von Zuben, P. M., Chiaparini, L. C., Annichino-Bizzacchi, J. M. & Costa, F. F. (1997) The mutation Ala677-->Val in the methylene tetrahydrofolate reductase gene: a risk factor for arterial disease and venous thrombosis. Thromb Haemost 77: 818-821. 180. Nelen, W. L., Blom, H. J., Steegers, E. A., den Heijer, M., Thomas, C. M. & Eskes, T. K. (2000) Homocysteine and folate levels as risk factors for recurrent early pregnancy loss. Obstet Gynecol 95: 519-524. 181. Nelen, W. L., Blom, H. J., Steegers, E. A., den Heijer, M. & Eskes, T. K. (2000) Hyperhomocysteinemia and recurrent early pregnancy loss: a meta-analysis. Fertil Steril 74: 1196-1199. 182. Nelen, W. L., Steegers, E. A., Eskes, T. K. & Blom, H. J. (1997) Genetic risk factor for unexplained recurrent early pregnancy loss. Lancet 350: 861. 183. Zetterberg, H., Regland, B., Palmer, M., Ricksten, A., Palmqvist, L., Rymo, L., Arvanitis, D. A., Spandidos, D. A. & Blennow, K. (2002) Increased frequency of combined methylenetetrahydrofolate reductase C677T and A1298C mutated alleles in spontaneously aborted embryos. Eur J Hum Genet 10: 113-118. 184. Quere, I., Bellet, H., Hoffet, M., Janbon, C., Mares, P. & Gris, J. C. (1998) A woman with five consecutive fetal deaths: case report and retrospective analysis of hyperhomocysteinemia prevalence in 100 consecutive women with recurrent miscarriages. Fertil Steril 69: 152-154. 185. Grandone, E., Margaglione, M., Colaizzo, D., d'Addedda, M., D'Andrea, G., Pavone, G. & Di Minno, G. (1998) Methylene tetrahydrofolate reductase (MTHFR) 677T-->C mutation and unexplained early pregnancy loss. Thromb Haemost 79: 1056-1057. 186. Holmes, Z. R., Regan, L., Chilcott, I. & Cohen, H. (1999) The C677T MTHFR gene mutation is not predictive of risk for recurrent fetal loss. Br J Haematol 105: 98-101. 187. Banerjee, R. V., Johnston, N. L., Sobeski, J. K., Datta, P. & Matthews, R. G. (1989) Cloning and sequence analysis of the Escherichia coli metH gene encoding cobalamin-dependent methionine synthase and isolation of a tryptic fragment containing the cobalamin-binding domain. J Biol Chem 264: 13888-13895. 188. Leclerc, D., Campeau, E., Goyette, P., Adjalla, C. E., Christensen, B., Ross, M., Eydoux, P., Rosenblatt, D. S., Rozen, R. & Gravel, R. A. (1996) Human methionine synthase: cDNA cloning and identification of mutations in patients of the cblG complementation group of folate/cobalamin disorders. Hum Mol Genet 5: 1867-1874.

PAGE 171

157 189. Tsai, M. Y., Bignell, M., Yang, F., Welge, B. G., Graham, K. J. & Hanson, N. Q. (2000) Polygenic influence on plasma homocysteine: association of two prevalent mutations, the 844ins68 of cystathionine beta-synthase and A(2756)G of methionine synthase, with lowered plasma homocysteine levels. Atherosclerosis 149: 131-137. 190. Silaste, M. L., Rantala, M., Sampi, M., Alfthan, G., Aro, A. & Kesaniemi, Y. A. (2001) Polymorphisms of key enzymes in homocysteine metabolism affect diet responsiveness of plasma homocysteine in healthy women. J Nutr 131: 2643-2647. 191. Kluijtmans, L. A., Young, I. S., Boreham, C. A., Murray, L., McMaster, D., McNulty, H., Strain, J. J., McPartlin, J., Scott, J. M. & Whitehead, A. S. (2003) Genetic and nutritional factors contributing to hyperhomocysteinemia in young adults. Blood 101: 2483-2488. 192. Jacques, P. F., Bostom, A. G., Selhub, J., Rich, S., Curtis Ellison, R., Eckfeldt, J. H., Gravel, R. A. & Rozen, R. (2003) Effects of polymorphisms of methionine synthase and methionine synthase reductase on total plasma homocysteine in the NHLBI Family Heart Study. Atherosclerosis 166: 49-55. 193. van der Put, N. M., van der Molen, E. F., Kluijtmans, L. A., Heil, S. G., Trijbels, J. M., Eskes, T. K., Van Oppenraaij-Emmerzaal, D., Banerjee, R. & Blom, H. J. (1997) Sequence analysis of the coding region of human methionine synthase: relevance to hyperhomocysteinaemia in neural-tube defects and vascular disease. Qjm 90: 511-517. 194. Harmon, D. L., Shields, D. C., Woodside, J. V., McMaster, D., Yarnell, J. W., Young, I. S., Peng, K., Shane, B., Evans, A. E. & Whitehead, A. S. (1999) Methionine synthase D919G polymorphism is a significant but modest determinant of circulating homocysteine concentrations. Genet Epidemiol 17: 298-309. 195. den Heijer, M., Koster, T., Blom, H. J., Bos, G. M., Briet, E., Reitsma, P. H., Vandenbroucke, J. P. & Rosendaal, F. R. (1996) Hyperhomocysteinemia as a risk factor for deep-vein thrombosis. N Engl J Med 334: 759-762. 196. Salomon, O., Rosenberg, N., Zivelin, A., Steinberg, D. M., Kornbrot, N., Dardik, R., Inbal, A. & Seligsohn, U. (2001) Methionine synthase A2756G and methylenetetrahydrofolate reductase A1298C polymorphisms are not risk factors for idiopathic venous thromboembolism. Hematol J 2: 38-41. 197. Yates, Z. & Lucock, M. (2002) Methionine synthase polymorphism A2756G is associated with susceptibility for thromboembolic events and altered B vitamin/thiol metabolism. Haematologica 87: 751-756. 198. Chen, J., Giovannucci, E., Hankinson, S. E., Ma, J., Willett, W. C., Spiegelman, D., Kelsey, K. T. & Hunter, D. J. (1998) A prospective study of methylenetetrahydrofolate reductase and methionine synthase gene polymorphisms, and risk of colorectal adenoma. Carcinogenesis 19: 2129-2132.

PAGE 172

158 199. Swanson, D. A., Liu, M. L., Baker, P. J., Garrett, L., Stitzel, M., Wu, J., Harris, M., Banerjee, R., Shane, B. & Brody, L. C. (2001) Targeted disruption of the methionine synthase gene in mice. Mol Cell Biol 21: 1058-1065. 200. Christensen, B., Arbour, L., Tran, P., Leclerc, D., Sabbaghian, N., Platt, R., Gilfix, B. M., Rosenblatt, D. S., Gravel, R. A., Forbes, P. & Rozen, R. (1999) Genetic polymorphisms in methylenetetrahydrofolate reductase and methionine synthase, folate levels in red blood cells, and risk of neural tube defects. Am J Med Genet 84: 151-157. 201. Zhu, H., Wicker, N. J., Shaw, G. M., Lammer, E. J., Hendricks, K., Suarez, L., Canfield, M. & Finnell, R. H. (2003) Homocysteine remethylation enzyme polymorphisms and increased risks for neural tube defects. Mol Genet Metab 78: 216-221. 202. Doolin, M. T., Barbaux, S., McDonnell, M., Hoess, K., Whitehead, A. S. & Mitchell, L. E. (2002) Maternal genetic effects, exerted by genes involved in homocysteine remethylation, influence the risk of spina bifida. Am J Hum Genet 71: 1222-1226. 203. Leclerc, D., Wilson, A., Dumas, R., Gafuik, C., Song, D., Watkins, D., Heng, H. H., Rommens, J. M., Scherer, S. W., Rosenblatt, D. S. & Gravel, R. A. (1998) Cloning and mapping of a cDNA for methionine synthase reductase, a flavoprotein defective in patients with homocystinuria. Proc Natl Acad Sci U S A 95: 3059-3064. 204. Wilson, A., Leclerc, D., Rosenblatt, D. S. & Gravel, R. A. (1999) Molecular basis for methionine synthase reductase deficiency in patients belonging to the cblE complementation group of disorders in folate/cobalamin metabolism. Hum Mol Genet 8: 2009-2016. 205. Wilson, A., Platt, R., Wu, Q., Leclerc, D., Christensen, B., Yang, H., Gravel, R. A. & Rozen, R. (1999) A common variant in methionine synthase reductase combined with low cobalamin (vitamin B12) increases risk for spina bifida. Mol Genet Metab 67: 317-323. 206. Devlin, A. M., Ling, E. H., Peerson, J. M., Fernando, S., Clarke, R., Smith, A. D. & Halsted, C. H. (2000) Glutamate carboxypeptidase II: a polymorphism associated with lower levels of serum folate and hyperhomocysteinemia. Hum Mol Genet 9: 2837-2844. 207. Vargas-Martinez, C., Ordovas, J. M., Wilson, P. W. & Selhub, J. (2002) The glutamate carboxypeptidase gene II (C>T) polymorphism does not affect folate status in the Framingham Offspring cohort. J Nutr 132: 1176-1179.

PAGE 173

159 208. Fodinger, M., Dierkes, J., Skoupy, S., Rohrer, C., Hagen, W., Puttinger, H., Hauser, A. C., Vychytil, A. & Sunder-Plassmann, G. (2003) Effect of Glutamate Carboxypeptidase II and Reduced Folate Carrier Polymorphisms on Folate and Total Homocysteine Concentrations in Dialysis Patients. J Am Soc Nephrol 14: 1314-1319. 209. Ulrich, C. M., Bigler, J., Bostick, R., Fosdick, L. & Potter, J. D. (2002) Thymidylate Synthase Promoter Polymorphism, Interaction with Folate Intake, and Risk of Colorectal Adenomas. Cancer Res 62: 3361-3364. 210. Skibola, C. F., Smith, M. T., Hubbard, A., Shane, B., Roberts, A. C., Law, G. R., Rollinson, S., Roman, E., Cartwright, R. A. & Morgan, G. J. (2002) Polymorphisms in the thymidylate synthase and serine hydroxymethyltransferase genes and risk of adult acute lymphocytic leukemia. Blood 99: 3786-3791. 211. De Marco, P., Calevo, M. G., Moroni, A., Arata, L., Merello, E., Cama, A., Finnell, R. H., Andreussi, L. & Capra, V. (2001) Polymorphisms in genes involved in folate metabolism as risk factors for NTDs. Eur J Pediatr Surg 11: S14-17. 212. Fenech, M. & Ferguson, L. R. (2001) Vitamins/minerals and genomic stability in humans. Mutat Res 475: 1-6. 213. Balaghi, M. & Wagner, C. (1993) DNA methylation in folate deficiency: use of CpG methylase. Biochem Biophys Res Commun 193: 1184-1190. 214. Kim, Y. I., Pogribny, I. P., Basnakian, A. G., Miller, J. W., Selhub, J., James, S. J. & Mason, J. B. (1997) Folate deficiency in rats induces DNA strand breaks and hypomethylation within the p53 tumor suppressor gene. Am J Clin Nutr 65: 46-52. 215. Titenko-Holland, N., Jacob, R. A., Shang, N., Balaraman, A. & Smith, M. T. (1998) Micronuclei in lymphocytes and exfoliated buccal cells of postmenopausal women with dietary changes in folate. Mutat Res 417: 101-114. 216. Pennisi, E. (2001) Behind the scenes of gene expression. Science 293: 1064-1067. 217. Costello, J. F. & Plass, C. (2001) Methylation matters. J Med Genet 38: 285-303. 218. Li, E., Bestor, T. H. & Jaenisch, R. (1992) Targeted mutation of the DNA methyltransferase gene results in embryonic lethality. Cell 69: 915-926. 219. Okano, M., Bell, D. W., Haber, D. A. & Li, E. (1999) DNA methyltransferases Dnmt3a and Dnmt3b are essential for de novo methylation and mammalian development. Cell 99: 247-257. 220. Gardiner-Garden, M. & Frommer, M. (1987) CpG islands in vertebrate genomes. J Mol Biol 196: 261-282.

PAGE 174

160 221. Kafri, T., Ariel, M., Brandeis, M., Shemer, R., Urven, L., McCarrey, J., Cedar, H. & Razin, A. (1992) Developmental pattern of gene-specific DNA methylation in the mouse embryo and germ line. Genes Dev 6: 705-714. 222. Jones, P. A. (1996) DNA methylation errors and cancer. Cancer Res 56: 2463-2467. 223. Bestor, T. H. (2000) The DNA methyltransferases of mammals. Hum Mol Genet 9: 2395-2402. 224. Jahner, D., Stuhlmann, H., Stewart, C. L., Harbers, K., Lohler, J., Simon, I. & Jaenisch, R. (1982) De novo methylation and expression of retroviral genomes during mouse embryogenesis. Nature 298: 623-628. 225. Bestor, T. H. (1992) Activation of mammalian DNA methyltransferase by cleavage of a Zn binding regulatory domain. Embo J 11: 2611-2617. 226. Klimasauskas, S., Kumar, S., Roberts, R. J. & Cheng, X. (1994) HhaI methyltransferase flips its target base out of the DNA helix. Cell 76: 357-369. 227. Bird, A. P. (1986) CpG-rich islands and the function of DNA methylation. Nature 321: 209-213. 228. Kass, S. U., Pruss, D. & Wolffe, A. P. (1997) How does DNA methylation repress transcription? Trends Genet 13: 444-449. 229. Cooney, C. A., Dave, A. A. & Wolff, G. L. (2002) Maternal methyl supplements in mice affect epigenetic variation and DNA methylation of offspring. J Nutr 132: 2393S-2400S. 230. Jones, P. L., Veenstra, G. J., Wade, P. A., Vermaak, D., Kass, S. U., Landsberger, N., Strouboulis, J. & Wolffe, A. P. (1998) Methylated DNA and MeCP2 recruit histone deacetylase to repress transcription. Nat Genet 19: 187-191. 231. Robertson, K. D. (2002) DNA methylation and chromatin unraveling the tangled web. Oncogene 21: 5361-5379. 232. Niculescu, M. D. & Zeisel, S. H. (2002) Diet, methyl donors and DNA methylation: interactions between dietary folate, methionine and choline. J Nutr 132: 2333S-2335S. 233. Scott, J. M. (1999) Folate and vitamin B12. Proc Nutr Soc 58: 441-448. 234. Alonso-Aperte, E. & Varela-Moreiras, G. (1996) Brain folates and DNA methylation in rats fed a choline deficient diet or treated with low doses of methotrexate. Int J Vitam Nutr Res 66: 232-236.

PAGE 175

161 235. Kim, Y. I., Christman, J. K., Fleet, J. C., Cravo, M. L., Salomon, R. N., Smith, D., Ordovas, J., Selhub, J. & Mason, J. B. (1995) Moderate folate deficiency does not cause global hypomethylation of hepatic and colonic DNA or c-myc-specific hypomethylation of colonic DNA in rats. Am J Clin Nutr 61: 1083-1090. 236. MacGregor, J. T., Wehr, C. M., Hiatt, R. A., Peters, B., Tucker, J. D., Langlois, R. G., Jacob, R. A., Jensen, R. H., Yager, J. W., Shigenaga, M. K., Frei, B., Eynon, B. P. & Ames, B. N. (1997) 'Spontaneous' genetic damage in man: evaluation of interindividual variability, relationship among markers of damage, and influence of nutritional status. Mutat Res 377: 125-135. 237. Jacob, R. A., Gretz, D. M., Taylor, P. C., James, S. J., Pogribny, I. P., Miller, B. J., Henning, S. M. & Swendseid, M. E. (1998) Moderate folate depletion increases plasma homocysteine and decreases lymphocyte DNA methylation in postmenopausal women. J Nutr 128: 1204-1212. 238. Rampersaud, G. C., Kauwell, G. P., Hutson, A. D., Cerda, J. J. & Bailey, L. B. (2000) Genomic DNA methylation decreases in response to moderate folate depletion in elderly women. Am J Clin Nutr 72: 998-1003. 239. Fowler, B. M., Giuliano, A. R., Piyathilake, C., Nour, M. & Hatch, K. (1998) Hypomethylation in cervical tissue: is there a correlation with folate status? Cancer Epidemiol Biomarkers Prev 7: 901-906. 240. Fenech, M., Aitken, C. & Rinaldi, J. (1998) Folate, vitamin B12, homocysteine status and DNA damage in young Australian adults. Carcinogenesis 19: 1163-1171. 241. Jhaveri, M. S., Wagner, C. & Trepel, J. B. (2001) Impact of Extracellular Folate Levels on Global Gene Expression. Mol Pharmacol 60: 1288-1295. 242. van den Veyver, I. B. (2002) Genetic effects of methylation diets. Annu. Rev. Nutr. 22: 255-282. 243. Wilson, M. J., Shivapurkar, N. & Poirier, L. A. (1984) Hypomethylation of hepatic nuclear DNA in rats fed with a carcinogenic methyl-deficient diet. Biochem J 218: 987-990. 244. Wainfan, E., Dizik, M., Stender, M. & Christman, J. K. (1989) Rapid appearance of hypomethylated DNA in livers of rats fed cancer-promoting, methyl-deficient diets. Cancer Res 49: 4094-4097. 245. Christman, J. K., Sheikhnejad, G., Dizik, M., Abileah, S. & Wainfan, E. (1993) Reversibility of changes in nucleic acid methylation and gene expression induced in rat liver by severe dietary methyl deficiency. Carcinogenesis 14: 551-557.

PAGE 176

162 246. Hoffman, D. R., Marion, D. W., Cornatzer, W. E. & Duerre, J. A. (1980) S-Adenosylmethionine and S-adenosylhomocystein metabolism in isolated rat liver. Effects of L-methionine, L-homocystein, and adenosine. J Biol Chem 255: 10822-10827. 247. James, S. J., Melnyk, S., Pogribna, M., Pogribny, I. P. & Caudill, M. A. (2002) Elevation in S-Adenosylhomocysteine and DNA Hypomethylation: Potential Epigenetic Mechanism for Homocysteine-Related Pathology. J. Nutr. 132: 2361S-2366. 248. Yi, P., Melnyk, S., Pogribna, M., Pogribny, I. P., Hine, R. J. & James, S. J. (2000) Increase in plasma homocysteine associated with parallel increases in plasma S-adenosylhomocysteine and lymphocyte DNA hypomethylation. J Biol Chem 275: 29318-29323. 249. Miller, J. W., Nadeau, M. R., Smith, J., Smith, D. & Selhub, J. (1994) Folate-deficiency-induced homocysteinaemia in rats: disruption of S-adenosylmethionine's co-ordinate regulation of homocysteine metabolism. Biochem J 298 ( Pt 2): 415-419. 250. Caudill, M. A., Wang, J. C., Melnyk, S., Pogribny, I. P., Jernigan, S., Collins, M. D., Santos-Guzman, J., Swendseid, M. E., Cogger, E. A. & James, S. J. (2001) Intracellular S-adenosylhomocysteine concentrations predict global DNA hypomethylation in tissues of methyl-deficient cystathionine beta-synthase heterozygous mice. J Nutr 131: 2811-2818. 251. Chiang, P. K. (1998) Biological effects of inhibitors of S-adenosylhomocysteine hydrolase. Pharmacol Ther 77: 115-134. 252. Medina, M., Urdiales, J. L. & Amores-Sanchez, M. I. (2001) Roles of homocysteine in cell metabolism: old and new functions. Eur J Biochem 268: 3871-3882. 253. Baylin, S. B. & Herman, J. G. (2000) DNA hypermethylation in tumorigenesis: epigenetics joins genetics. Trends Genet 16: 168-174. 254. Ehrlich, M. (2002) DNA methylation in cancer: too much, but also too little. Oncogene 21: 5400-5413. 255. Wainfan, E. & Poirier, L. A. (1992) Methyl groups in carcinogenesis: effects on DNA methylation and gene expression. Cancer Res 52: 2071s-2077s. 256. Sibani, S., Melnyk, S., Pogribny, I. P., Wang, W., Hiou-Tim, F., Deng, L., Trasler, J., James, S. J. & Rozen, R. (2002) Studies of methionine cycle intermediates (SAM, SAH), DNA methylation and the impact of folate deficiency on tumor numbers in Min mice. Carcinogenesis 23: 61-65.

PAGE 177

163 257. Dianov, G. L., Timchenko, T. V., Sinitsina, O. I., Kuzminov, A. V., Medvedev, O. A. & Salganik, R. I. (1991) Repair of uracil residues closely spaced on the opposite strands of plasmid DNA results in double-strand break and deletion formation. Mol Gen Genet 225: 448-452. 258. Pogribny, I. P., Muskhelishvili, L., Miller, B. J. & James, S. J. (1997) Presence and consequence of uracil in preneoplastic DNA from folate/methyl-deficient rats. Carcinogenesis 18: 2071-2076. 259. Duthie, S. J. & Hawdon, A. (1998) DNA instability (strand breakage, uracil misincorporation, and defective repair) is increased by folic acid depletion in human lymphocytes in vitro. Faseb J 12: 1491-1497. 260. Melnyk, S., Pogribna, M., Miller, B. J., Basnakian, A. G., Pogribny, I. P. & James, S. J. (1999) Uracil misincorporation, DNA strand breaks, and gene amplification are associated with tumorigenic cell transformation in folate deficient/repleted Chinese hamster ovary cells. Cancer Lett 146: 35-44. 261. Duthie, S. J., Grant, G. & Narayanan, S. (2000) Increased uracil misincorporation in lymphocytes from folate-deficient rats. Br J Cancer 83: 1532-1537. 262. Tucker, J. D. & Preston, R. J. (1996) Chromosome aberrations, micronuclei, aneuploidy, sister chromatid exchanges, and cancer risk assessment. Mutat Res 365: 147-159. 263. Sutherland, G. R. (1979) Heritable fragile sites on human chromosomes I. Factors affecting expression in lymphocyte culture. Am J Hum Genet 31: 125-135. 264. Everson, R. B., Wehr, C. M., Erexson, G. L. & MacGregor, J. T. (1988) Association of marginal folate depletion with increased human chromosomal damage in vivo: demonstration by analysis of micronucleated erythrocytes. J Natl Cancer Inst 80: 525-529. 265. Fenech, M. & Rinaldi, J. (1994) The relationship between micronuclei in human lymphocytes and plasma levels of vitamin C, vitamin E, vitamin B12 and folic acid. Carcinogenesis 15: 1405-1411. 266. Odagiri, Y. & Uchida, H. (1998) Influence of serum micronutrients on the incidence of kinetochore-positive or -negative micronuclei in human peripheral blood lymphocytes. Mutat Res 415: 35-45. 267. Fenech, M. & Rinaldi, J. (1995) A comparison of lymphocyte micronuclei and plasma micronutrients in vegetarians and non-vegetarians. Carcinogenesis 16: 223-230. 268. Fenech, M. F., Dreosti, I. E. & Rinaldi, J. R. (1997) Folate, vitamin B12, homocysteine status and chromosome damage rate in lymphocytes of older men. Carcinogenesis 18: 1329-1336.

PAGE 178

164 269. Fenech, M. (1997) The advantages and disadvantages of the cytokinesis-block micronucleus method. Mutat Res 392: 11-18. 270. Crott, J. & Fenech, M. (2001) Preliminary study of the genotoxic potential of homocysteine in human lymphocytes in vitro. Mutagenesis 16: 213-217. 271. Fenech, M. (2002) Micronutrients and genomic stability: a new paradigm for recommended dietary allowances (RDAs). Food Chem Toxicol 40: 1113-1117. 272. Fenech, M. (2001) The role of folic acid and Vitamin B12 in genomic stability of human cells. Mutat Res 475: 57-67. 273. CDC (1992) Recommendations for the use of folic acid to reduce the number of cases of spina bifida and other neural tube defects. Morb Mortal Wkly Rep 41: 1-7. 274. CDC (2002) Folate status in women of childbearing age, by race/ethnicity United States, 1999-2000. MMWR Morb Mortal Wkly Rep 51: 808-810. 275. Choumenkovitch, S. F., Jacques, P. F., Nadeau, M. R., Wilson, P. W., Rosenberg, I. H. & Selhub, J. (2001) Folic acid fortification increases red blood cell folate concentrations in the Framingham study. J Nutr 131: 3277-3280. 276. Honein, M. A., Paulozzi, L. J., Mathews, T. J., Erickson, J. D. & Wong, L. Y. (2001) Impact of folic acid fortification of the US food supply on the occurrence of neural tube defects. Jama 285: 2981-2986. 277. Choumenkovitch, S. F., Selhub, J., Wilson, P. W., Rader, J. I., Rosenberg, I. H. & Jacques, P. F. (2002) Folic Acid Intake from Fortification in United States Exceeds Predictions. J Nutr 132: 2792-2798. 278. Quinlivan, E. P. & Gregory, J. F. (2003) Effect of food fortification on folic acid intake in the United States. Am J Clin Nutr 77: 221-225. 279. NIH (2001) NIH Policy on Reporting Race and Ethnicity Data: Subjects in Clinical Research. 280. Ford, E. S. & Bowman, B. A. (1999) Serum and red blood cell folate concentrations, race, and education: findings from the third National Health and Nutrition Examination Survey. Am J Clin Nutr 69: 476-481. 281. Jacques, P. F., Rosenberg, I. H., Rogers, G., Selhub, J., Bowman, B. A., Gunter, E. W., Wright, J. D. & Johnson, C. L. (1999) Serum total homocysteine concentrations in adolescent and adult Americans: results from the third National Health and Nutrition Examination Survey. Am J Clin Nutr 69: 482-489. 282. Baggott, J. E. (1999) Serum folate and homocysteine concentrations in large population samples of US ethnic and racial groups. Am J Clin Nutr 70: 937; author reply 937-940.

PAGE 179

165 283. Jacques, P. F., Rosenberg, I. H., Rogers, G., Selhub, J., Wright, J. D. & Johnson, C. L. (1999) Reply to JE baggott. Am J Clin Nutr 70: 939-940. 284. Lindenbaum, J., Healton, E. B., Savage, D. G., Brust, J. C., Garrett, T. J., Podell, E. R., Marcell, P. D., Stabler, S. P. & Allen, R. H. (1995) Neuropsychiatric disorders caused by cobalamin deficiency in the absence of anemia or macrocytosis. 1988. Nutrition 11: 181; discussion 180, 182. 285. Herbert, V. (1987) Recommended dietary intakes (RDI) of folate in humans. Am J Clin Nutr 45: 661-670. 286. Jacob, R. A., Wu, M. M., Henning, S. M. & Swendseid, M. E. (1994) Homocysteine increases as folate decreases in plasma of healthy men during short-term dietary folate and methyl group restriction. J Nutr 124: 1072-1080. 287. O'Keefe, C. A., Bailey, L. B., Thomas, E. A., Hofler, S. A., Davis, B. A., Cerda, J. J. & Gregory, J. F.d (1995) Controlled dietary folate affects folate status in nonpregnant women. J Nutr 125: 2717-2725. 288. Ubbink, J. B., Becker, P. J., Vermaak, W. J. & Delport, R. (1995) Results of B-vitamin supplementation study used in a prediction model to define a reference range for plasma homocysteine. Clin Chem 41: 1033-1037. 289. Gunter, E. W., Bowman, B. A., Caudill, S. P., Twite, D. B., Adams, M. J. & Sampson, E. J. (1996) Results of an international round robin for serum and whole-blood folate. Clin Chem 42: 1689-1694. 290. Herbert, V. (1987) Making sense of laboratory tests of folate status: folate requirements to sustain normality. Am J Hematol 26: 199-207. 291. Shane, B., Tamura, T. & Stokstad, E. L. (1980) Folate assay: a comparison of radioassay and microbiological methods. Clin Chim Acta 100: 13-19. 292. Rasmussen, K. & Moller, J. (2000) Total homocysteine measurement in clinical practice. Ann Clin Biochem 37 ( Pt 5): 627-648. 293. Rasmussen, K. & Moller, J. (2001) Methodologies of Testing. In: Homocysteine in Health and Disease (Carmel, R. & Jacobsen, D. W., eds.), pp. 199-209. Cambridge University Press, Cambridge, UK. 294. Capdevila, A. & Wagner, C. (1998) Measurement of plasma S-adenosylmethionine and S-adenosylhomocysteine as their fluorescent isoindoles. Anal Biochem 264: 180-184. 295. Pogribny, I., Yi, P. & James, S. J. (1999) A sensitive new method for rapid detection of abnormal methylation patterns in global DNA and within CpG islands. Biochem Biophys Res Commun 262: 624-628.

PAGE 180

166 296. Fujiwara, H. & Ito, M. (2002) Nonisotopic cytosine extension assay: A highly sensitive method to evaluate CpG island methylation in the whole genome. Anal Biochem 307: 386. 297. Cooney, C. A., Wise, C. K. & Poirier, L. A. (1997) An improved sample preparation method for the quantitative HPLC determination of 5-methyl-deoxycytidine in animal tissue. J Liquid Chromat 20: 129-193. 298. Frommer, M., McDonald, L. E., Millar, D. S., Collis, C. M., Watt, F., Grigg, G. W., Molloy, P. L. & Paul, C. L. (1992) A genomic sequencing protocol that yields a positive display of 5-methylcytosine residues in individual DNA strands. Proc Natl Acad Sci U S A 89: 1827-1831. 299. Tice, R. R., Agurell, E., Anderson, D., Burlinson, B., Hartmann, A., Kobayashi, H., Miyamae, Y., Rojas, E., Ryu, J. C. & Sasaki, Y. F. (2000) Single cell gel/comet assay: guidelines for in vitro and in vivo genetic toxicology testing. Environ Mol Mutagen 35: 206-221. 300. Basnakian, A. G. & James, S. J. (1996) Quantification of 3'OH DNA breaks by random oligonucleotide-primed synthesis (ROPS) assay. DNA Cell Biol 15: 255-262. 301. Barbaux, S., Kluijtmans, L. A. & Whitehead, A. S. (2000) Accurate and rapid "multiplex heteroduplexing" method for genotyping key enzymes involved in folate/homocysteine metabolism. Clin Chem 46: 907-912. 302. Ulvik, A. & Ueland, P. M. (2001) Single nucleotide polymorphism (SNP) genotyping in unprocessed whole blood and serum by real-time PCR: application to SNPs affecting homocysteine and folate metabolism. Clin Chem 47: 2050-2053. 303. Esfahani, S. T., Cogger, E. A. & Caudill, M. A. (2003) Heterogeneity in the prevalence of methylenetetrahydrofolate reductase gene polymorphisms in women of different ethnic groups. J Am Diet Assoc 103: 200-207. 304. Scholl, T. O. & Johnson, W. G. (2000) Folic acid: influence on the outcome of pregnancy. Am J Clin Nutr 71: 1295S-1303S. 305. Herbert, V. (1963) A palatable diet for producing experimental folate deficiency in man. Am J Clin Nutr 12: 17-20. 306. IOM (1998) Choline. In: Dietary Reference Intakes, pp. 390-422. National Academy Press, Washington, D.C. 307. Kauwell, G. P., Lippert, B. L., Wilsky, C. E., Herrlinger-Garcia, K., Hutson, A. D., Theriaque, D. W., Rampersaud, G. C., Cerda, J. J. & Bailey, L. B. (2000) Folate status of elderly women following moderate folate depletion responds only to a higher folate intake. J Nutr 130: 1584-1590.

PAGE 181

167 308. Martin, J. I., Landen, W. O., Jr., Soliman, A. G. & Eitenmiller, R. R. (1990) Application of a tri-enzyme extraction for total folate determination in foods. J Assoc Off Anal Chem 73: 805-808. 309. Koc, H., Mar, M. H., Ranasinghe, A., Swenberg, J. A. & Zeisel, S. H. (2002) Quantitation of choline and its metabolites in tissues and foods by liquid chromatography/electrospray ionization-isotope dilution mass spectrometry. Anal Chem 74: 4734-4740. 310. Blakley, R. L. (1969) The Biochemistry of Folic Acid and Related Pteridines. Wiley, New York, NY. 311. Tamura, T., Freeberg, L. E. & Cornwell, P. E. (1990) Inhibition of EDTA of growth of Lactobacillus casei in the folate microbiological assay and its reversal by added manganese or iron. Clin Chem 36: 1993. 312. Sauberlich, H. E., Dowdy, R. P. & Skala, J. H. (1974) Folacin (Folic Acid, Pteroylmonoglutamic Acid, Folate). In: Laboratory Tests for the Assessment of Nutritional Status. CRC Press, Cleveland, OH. 313. Vester, B. & Rasmussen, K. (1991) High performance liquid chromatography method for rapid and accurate determination of homocysteine in plasma and serum. Eur J Clin Chem Clin Biochem 29: 549-554. 314. Renbaum, P. & Razin, A. (1992) Mode of action of the Spiroplasma CpG methylase M.SssI. FEBS Lett 313: 243-247. 315. Crain, P. F. (1990) Preparation and enzymatic hydrolysis of DNA and RNA for mass spectrometry. Methods Enzymol 193: 782-790. 316. Cox, D. R. & Stuart, A. (1955) Some quick sign tests for trend in location and dispersion. Biometrika 42: 80-95. 317. Caudill, M. A., Le, T., Moonie, S. A., Esfahani, S. T. & Cogger, E. A. (2001) Folate status in women of childbearing age residing in Southern California after folic acid fortification. J Am Coll Nutr 20: 129-134. 318. Choi, S. W., Friso, S., Dolnikowski, G. G., Bagley, P. J., Edmondson, A. N., Smith, D. E. & Mason, J. B. (2003) Biochemical and molecular aberrations in the rat colon due to folate depletion are age-specific. J Nutr 133: 1206-1212. 319. Herrmann, W., Quast, S., Ulrich, M., Schultze, H., Bodis, M. & Geisel, J. (1999) Hyperhomocysteinemia in high-aged subjects: relation of B-vitamins, folic acid, renal function and the methylenetetrahydrofolate reductase mutation. Atherosclerosis 144: 91-101.

PAGE 182

168 320. Selhub, J., Jacques, P. F., Wilson, P. W., Rush, D. & Rosenberg, I. H. (1993) Vitamin status and intake as primary determinants of homocysteinemia in an elderly population. Jama 270: 2693-2698. 321. Ueland, P. M., Refsum, H. & Brattstrom, L. (1992) Plasma homocysteine and cardiovascular disease. In: Atherosclerotic vascular disease, hemostasis, and endothelial function (Francis, R. B. J., ed.), pp. 183-236. Marcel Dekker, Inc., New York. 322. Guttormsen, A. B., Ueland, P. M., Svarstad, E. & Refsum, H. (1997) Kinetic basis of hyperhomocysteinemia in patients with chronic renal failure. Kidney Int 52: 495-502. 323. Jacob, R. A., Jenden, D. J., Allman-Farinelli, M. A. & Swendseid, M. E. (1999) Folate nutriture alters choline status of women and men fed low choline diets. J Nutr 129: 712-717. 324. Trimble, K. C., Molloy, A. M., Scott, J. M. & Weir, D. G. (1993) The effect of ethanol on one-carbon metabolism: increased methionine catabolism and lipotrope methyl-group wastage. Hepatology 18: 984-989. 325. Zeisel, S. H., Da Costa, K. A., Franklin, P. D., Alexander, E. A., Lamont, J. T., Sheard, N. F. & Beiser, A. (1991) Choline, an essential nutrient for humans. Faseb J 5: 2093-2098. 326. Steenge, G. R., Verhoef, P. & Katan, M. B. (2003) Betaine supplementation lowers plasma homocysteine in healthy men and women. J Nutr 133: 1291-1295. 327. Schwahn, B. C., Chen, Z., Laryea, M. D., Wendel, U., Lussier-Cacan, S., Genest, J., Jr., Mar, M. H., Zeisel, S. H., Castro, C., Garrow, T. & Rozen, R. (2003) Homocysteine-betaine interactions in a murine model of 5,10-methylenetetrahydrofolate reductase deficiency. Faseb J 17: 512-514. 328. Homocysteine Lowering Trialists' Collaboration (1998) Lowering blood homocysteine with folic acid based supplements: meta-analysis of randomised trials. BMJ 316: 894-898. 329. Stampfer, M. J., Malinow, M. R., Willett, W. C., Newcomer, L. M., Upson, B., Ullmann, D., Tishler, P. V. & Hennekens, C. H. (1992) A prospective study of plasma homocyst(e)ine and risk of myocardial infarction in US physicians. Jama 268: 877-881. 330. Boushey, C. J., Beresford, S. A., Omenn, G. S. & Motulsky, A. G. (1995) A quantitative assessment of plasma homocysteine as a risk factor for vascular disease. Probable benefits of increasing folic acid intakes. Jama 274: 1049-1057. 331. Richardson, B. (2003) Impact of aging on DNA methylation. Ageing Res Rev 2: 245-261.

PAGE 183

169 332. Richardson, B. C. (2002) Role of DNA methylation in the regulation of cell function: autoimmunity, aging and cancer. J Nutr 132: 2401S-2405S. 333. Food Standards Agency (2002) Draft Report of the Expert Group on Vitamins and Minerals. Office of National Statistics MRC Human Nutrition Research, United Kingdom. 334. Irish Universities Nutrition Alliance (2001) IUNA North/South Ireland Food Consumption Survey: Summary Report on Food and Nutrient Intakes, Anthropometry, Attitudinal Data, and Physical Activity Patterns. Food Safety Promotion Board, Dublin, Ireland. 335. Refsum, H., Ueland, P. M., Nygard, O. & Vollset, S. E. (1998) Homocysteine and cardiovascular disease. Annu Rev Med 49: 31-62.

PAGE 184

BIOGRAPHICAL SKETCH Karla Pagn Shelnutt was born in Rio Piedras, Puerto Rico on August 6, 1975. She grew up in Coral Springs, Florida where she graduated from Marjory Stoneman Douglas High School in 1993. She graduated with high honors from the University of Florida in 1997 with a Bachelor of Science degree in Food Science and Human Nutrition (with a specialization in dietetics). In June of 1998, she completed her dietetic internship at University of Alabama at Birmingham and passed her dietetic registration examination later that year. In April 2000, she was awarded a Master of Science degree in Clinical Nutrition also from the University of Alabama at Birmingham. After working as a clinical dietitian for four months at West Boca Medical Center in Boca Raton, Florida, she entered the doctoral program in the Food Science and Human Nutrition Department at the University of Florida in August of 2000. She graduated with a doctoral degree in August 2003 and plans to work as a Postdoctoral Research Associate with Dr. Gail Kauwell at the University of Florida. 170


Permanent Link: http://ufdc.ufl.edu/UFE0001199/00001

Material Information

Title: Effect of the Methylenetetrahydrofolate Reductase 677C to T Polymorphism on Folate status and DNA methylation response in young women
Physical Description: Mixed Material
Creator: Shelnutt, Karla Pagan ( Author, Primary )
Publication Date: 2003
Copyright Date: 2003

Record Information

Source Institution: University of Florida
Holding Location: University of Florida
Rights Management: All rights reserved by the source institution and holding location.
System ID: UFE0001199:00001

Permanent Link: http://ufdc.ufl.edu/UFE0001199/00001

Material Information

Title: Effect of the Methylenetetrahydrofolate Reductase 677C to T Polymorphism on Folate status and DNA methylation response in young women
Physical Description: Mixed Material
Creator: Shelnutt, Karla Pagan ( Author, Primary )
Publication Date: 2003
Copyright Date: 2003

Record Information

Source Institution: University of Florida
Holding Location: University of Florida
Rights Management: All rights reserved by the source institution and holding location.
System ID: UFE0001199:00001


This item has the following downloads:


Full Text












EFFECT OF THE METHYLENETETRAHYDROFOLATE REDUCTASE 677C-T
POLYMORPHISM ON FOLATE STATUS AND DNA METHYLATION RESPONSE
IN YOUNG WOMEN















By

KARLA PAGAN SHELNUTT


A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL
OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT
OF THE REQUIREMENTS FOR THE DEGREE OF
DOCTOR OF PHILOSOPHY

UNIVERSITY OF FLORIDA


2003

































This dissertation is dedicated to my aunt Nellie J. Mendez whose successful career as a
dietitian inspired me to study nutrition. Although her career ended early as a result of
health complications, she continues to incorporate good nutrition into her daily life. I
hope that one day I will leave my mark in nutrition research as she has done in
community nutrition.















ACKNOWLEDGMENTS

I would like to thank the members of my supervisory committee (Lynn B. Bailey,

Ph.D.; Gail P. A. Kauwell, Ph.D.; Jesse F. Gregory III, Ph.D.; and Michael S. Kilberg,

Ph.D.). I would like to extend special thanks to my major professor Dr. Bailey for giving

me the opportunity to work with her on such an important and exciting project. Her

guidance over the years has been invaluable to me. She has always assisted me in any

capacity that I needed and treated me as a colleague instead of a student. She is an

amazing scientist who is successful in balancing her family life with her career. I also

would like to acknowledge Dr. Gail Kauwell for all of her scientific knowledge and

moral support. Without her expertise, this project would not have been successful. I

would also like to thank all of the members of our research group (Carrie Chapman,

Aisha Cuadras, David Maneval, Karen Novak, and Angeleah Browdy, Ph.D.) who were

essential to the completion of this project. Further thanks are extended to George

Henderson, Ph.D.; Eoin Quinlivan, Ph.D.; and Steve Davis, Ph.D. for their technical

assistance and advice. I would like to thank the staff at the General Clinical Research

Center for their assistance with the feeding study, and especially Doug Theriaque for his

essential assistance with the statistical analyses.

Special thanks are extended to the many friends I have made over the years who

have constantly encouraged me throughout this process, especially Carrie Chapman,

Suzie Cole, Britton McPherson, and Sara Rathman. The prayers of these amazing

women have been crucial during this experience.









I wish to thank my family for all of their love and support over the years. I wish to

especially thank my parents, who continue to pray for me on a daily basis. Their constant

words of encouragement have provided guidance to me when I felt I had lost my way.

Last but not least, I would like to thank my husband and best friend, David Shelnutt. His

unconditional love has sustained me through the toughest parts of graduate school. His

help over the years has given me the opportunity to be successful. I look forward to

being able to spend the rest of our lives continuing to support each other in all of our

endeavors.
















TABLE OF CONTENTS

page

A C K N O W L E D G M E N T S ................................................................................................. iii

LIST OF TABLES .................................................... ....... .. .............. viii

LIST OF FIGURES ......... ......................... ...... ........ ............ ix

L IST O F A B B R E V IA TIO N S ................................................................... .....................x

A B S T R A C T .............................................. ..........................................x iii

CHAPTER

1 IN TR OD U CTION ............................................... .. ......................... ..

H ypotheses ................................................ 3
Specific A im ................................................................. 3

2 BACKGROUND AND LITERATURE REVIEW ..................... ........................... 4

F o late ....................................................... 4
C h e m istry .............................................................................................................. 4
D dietary Sources ...............................................................4
B ioavailability ............................................................. .5
A absorption .................................................................................................. ....... 7
T ra n sp o rt .......................................................................................... ............. 8
Biochem ical Functions .............................................................. .. .............10
Folate-Related Polym orphism s ................................ ................. ............ 17
Methylenetetrahydrofolate Reductase (MTHFR) ...............................................17
M ethionine Synthase (M S) .................................. .......................... ......... 41
M ethionine Synthase Reductase (M SR)................................... ......... ......44
O their P olym orphism s ..............................................................................45
D N A Stability ................................. ....................... ..... 47
DN A M ethylation ...................................................................... ........47
D N A Strand B reaks ............................................ ............... ............... 61
M icronuclei Form action .................................................................................. 64
D ietary R reference Intakes (D RIs) ........................................ ......................... 65
Folate Status in Women of Reproductive Age ..........................................................67
E effect of Fortification .............................................. .............................. 67


v









Effect of Ethnicity .................. ............................ ......... .. ........ .... 68
Folate D efficiency .......................................... ........... ... .. ............ 69
F olate Statu s A ssessm ent......................................... .............................................70
A analytical M methodology ..................................................................... ..................72
Blood Folate Analysis ................................................................. .... ... 72
Plasma Homocysteine and SAM/SAH Ratio Analysis .......................................73
D N A Stability .......... ..... .......................................................................... ........ 74
Genotype Determ nation .................................. .....................................77
R research Significance......................................................................... .................. 77

3 STUDY DESIGN AND METHODS ............................................................. 80

Subject Screening and Description...... ................... ..................80
S tu d y D e sig n .......................................................................................... ............... 8 1
General Clinical Research Center (GCRC) Protocol...............................................82
Dietary Treatment and Supplementation Description .........................................83
Sam ple Collection and Processing......................................... ......................... 86
A nalytical M ethods............ ................................................................ ........ .. ...... .. 87
Food Folate Extraction .............................................. .............................. 87
Food C holine A analysis ............................................... ............................. 88
Supplem ental Folic A cid ......................................................... ............... 88
M icrobiological A ssay ................................................... ........... ............... 89
Plasma Homocysteine Concentration............................................. 90
D N A E xtraction .......... ..... .............................................................. .. .... ..... .. 9 1
D N A Q uantitation .......................................... ................. .. ...... 91
MTHFR Genotype Determination..................... .... ......................... 92
M ethyl Acceptance Assay ..................................... .......................93
Liquid Chromatography/Mass Spectrometry/Mass Spectrometry Assay ..........94
S statistic al A n aly sis ................................................................................................ 9 7

4 R E SU L T S ............... .............. ........... ... .............................................101

Folate C ontent of M enus .................................. .............. .................................... 101
Serum Folate Concentration ................................ ......... .... ............... 101
Red Blood Cell Folate Concentration......................... ......................... 105
Homocysteine Concentration..................... ....... ............................. 108
DNA M ethylation .............................. .. ................. ........ .. .. ............ 112

DISCUSSION AND CONCLUSIONS .....................................................................118

SUM M ARY ........... ........ ........................................... 135

APPENDIX

A VITAMIN-MINERAL SUPPLEMENT COMPOSITION ..................................... 136

B CH OLINE CON TEN T OF D IET ............................................................................137









C POLYMERASE CHAIN REACTION GEL.................................138

L IST O F R E F E R E N C E S ........................................................................ .................... 139

B IO G R A PH IC A L SK E T C H .................................................................. ....................170
















LIST OF TABLES


Table page

2-1 Compartmentalization of enzymes and reactions.........................................16

3-1 Five-day cycle m enu ...................... ...... ............ ........................... 84

3-2 Selective reaction monitoring fragmentation table ...............................................96

4-1 Folate content of meals and total daily intake (tg DFE) ............................101

4-2 Serum folate concentration...................... ....... ............................. 103

4-3 Red blood cell folate concentration..................... .... .......................... 106

4-4 Hom ocysteine concentration ................................... ...................................... 110

4-5 DNA [3H]methyl group acceptance .............................................. ............... 113

4-6 Percentage (% ) of methylated cytosine......... .................................... ............ 113

4-7 Percent (%) change in DNA [3H]methyl group acceptance................................113

4-8 Raw change in m Cyt/tCyt ratio.............................. .................... ...... ......... 114

A-1 Vitamin/mineral RDA, five-day diet average, and supplement calculations.........136

B-l First analysis of dietary choline content.......................... ..... .... ........... 137

B-2 Second analysis of dietary choline content............. ...................... .............137
















LIST OF FIGURES

Figure page

2 -1 S tru ctu re o f fo late ................... ............ ...................................... ...................... .. .4

2-2 Folate m etabolism .................................................. ........ .. ........ .... 11

2-3 H istidine m etabolism ......... ........................................................ ...... .... ... ..13

2-4 Compartmentalization of folate pools ......................................... ...............15

3-1 Study design ............................................................... ... .... ......... 81

3-2 LC-M S/M S analysis of a standard ........................................ ........ ............... 96

3-3 LC-M S/M S analysis of a hydrolyzed sample .................................. ............... 97

4-1 Weekly unadjusted mean serum folate concentrations .......................................102

4-2 Mean serum folate concentration by genotype ............................................... 104

4-3 Weekly unadjusted mean red blood cell folate concentrations............................105

4-4 Mean red blood cell folate concentration by genotype .......................................107

4-5 Weekly unadjusted mean plasma homocysteine concentrations............................108

4-6 Mean plasma homocysteine concentration by genotype...................................111

4-7 Weekly unadjusted mean serum folate and plasma homocysteine
concentrations by genotype. ..................................................................... ......... 112

4-8 Percent change in [3H]methyl group acceptance...................................................116

4-9 Percent change in mCyt/tCyt ratio .............. .. ...................... ....117

C-1 Sample gel with amplified DNA fragments .................. .............................. 138
















LIST OF ABBREVIATIONS


Abbreviation


AI
ALL
ANCOVA
ANOVA
AP
BHMT
CBC-D
CDC
cDNA
CHES
CpG
cpm
CV
d
dCTP
DFE
DHF
DNA
DNMT
dNTP
dpm
DRI
dsDNA
dTMP
dTTP
dUMP
dUTP
EAR
EDTA
FAD
FBPs
FDA
FIGLU
FMN
g
GCPII
GCRC


Meaning


adequate intake
acute lymphoid leukemia
analysis of covariance
analysis of variance
apyrimidinic
betaine:homocysteine methyltransferase
complete blood counts with differentials
Centers for Disease Control and Prevention
complementary DNA
2-[N-cyclohexylamino]ethanesulfonic acid
cytosine-guanine dinucleotide
counts per minute
coefficient of variation
day
deoxycytosine triphosphate
dietary folate equivalents
dihydrofolate
deoxyribonucleic acid
DNA methyltransferase
deoxynucleotide triphosphate
disintegrations per minute
dietary reference intakes
double-stranded DNA
deoxythymidylate monophosphate
deoxythymidylate triphosphate
deoxyuridylate monophosphate
deoxyuridylate triphosphate
estimated average requirement
ethylenediaminetetraacetic acid
flavin adenine dinucleotide
folate binding proteins
Food and Drug Administration
formiminoglutamate
flavin mononucleotide
gram
glutamate carboxypeptidase II
General Clinical Research Center









h
HCG
HDAC
HEPES
HPLC
IOM
kDa
kg
L
LC/ESI-IDMS

LC/MS
LC-MS/MS
LS
LSES
mCyt
mg
min
ml
mmol
mo
MRC
MS
MS/MS
MSR
MTHFR
NaOH
NEB
ng
NHANES
nmol
NTDs

PCR
RA
Rb
RDA
RFC
RNA
ROPS
s
SAH
SAM
SD
SEA
SHMT


hour
human chorionic gonadotropin
histone deacetylase
hydroxyethylpiperazine-N'-[2-ethanesulfonic acid]
high performance liquid chromatography
Institute of Medicine
kilodalton
kilograms
liter
liquid chromatography/electrospray ionization-isotope
dilution mass spectrometry
liquid chromatography/mass spectrometry
liquid chromatography-mass spectrometry/mass spectrometry
least squares
low socioeconomic status
methylcytosine
milligrams
minute
milliliter
millimole
month
Medical Research Council
methionine synthase
mass spectrometry/mass spectrometry
methionine synthase reductase
methylenetetrahydrofolate reductase
sodium hydroxide
New England Biolabs
nanogram
National Health and Nutrition Examination Survey
nanomole
neural tube defects
old
polymerase chain reaction
radiobinding assay
retinoblastoma protein
recommended dietary allowance
reduced folate carrier
ribonucleic acid
random oligonucleotide primed synthesis
seconds
S-adenosylhomocysteine
S-adenosylmethionine
standard deviation
socioeconomically advantaged
serine hydroxymethyltransferase









SNP single nucleotide polymorphism
SST serum separator tubes
tCyt total cytosine
TE tris EDTA
THF tetrahydrofolate
UL upper limit
VTE venous thromboembolism
wk week
y year
Itg microgram
[l microliter
[tmol micromole















Abstract of Dissertation Presented to the Graduate School
of the University of Florida in Partial Fulfillment of the
Requirements for the Degree of Doctor of Philosophy

EFFECT OF THE METHYLENETETRAHYDROFOLATE REDUCTASE 677C-T
POLYMORPHISM ON FOLATE STATUS AND DNA METHYLATION RESPONSE
IN YOUNG WOMEN

By

Karla Pagan Shelnutt

August 2003

Chair: Lynn B. Bailey
Major Department: Food Science and Human Nutrition

Methylenetetrahydrofolate reductase (MTHFR) is a key enzyme in folate

metabolism. A C->T substitution at base pair 677 of MTHFR may impair folate status

when folate intake is marginal, a particular concern for women of reproductive age.

Folate status response to a controlled folate depletion-repletion feeding protocol (14 wk)

was investigated in young women (20 to 30 y) with (TT) (n = 19) and without (CC) (n =

22) the MTHFR 677C--T variant. Subjects consumed a moderately folate-deplete diet

(115 |tg DFE/d) for 7 wk, followed by folate repletion for 7 wk with the current RDA for

this population (400 |tg DFE/d).

Overall serum folate decreased (P < 0.0001) during depletion and increased (P <

0.0001) during repletion with lower (P = 0.03) post-depletion serum folate in women

with the TT versus CC genotype (LS mean 15.3 + 1.2 vs 19.5 + 1.3 nmol/L,

respectively). Folate status was low (serum folate < 13.6 nmol/L) in more women with









the TT (59%) versus the CC (15%) genotype post-depletion. Red blood cell folate for all

subjects decreased during depletion (P < 0.0001) and repletion (P = 0.02) with lower (P =

0.04) red blood cell folate in women with the TT versus the CC genotype post-repletion

(LS mean 1036 53 vs 1203 53 nmol/L, respectively). Homocysteine increased (P <

0.0001) during depletion and decreased for subjects with the CC (P = 0.02) but not the

TT (P = 0.47) genotype during repletion. Homocysteine tended to differ by genotype

post-depletion [10.5 3.3 (TT) vs 8.9 1.9 (CC) (P = 0.09)] and post-repletion [8.9 1.6

(TT) vs 7.2 1.9 (CC) (P = 0.08)]. Overall percent change in [3H]methyl group

acceptance tended to increase during depletion (P = 0.08). Women with the TT genotype

had an increase in raw and percent change in methylated cytosine during repletion (P <

0.05). Except for global DNA methylation, these data suggest that young women with

the MTHFR 677 TT genotype respond more negatively to folate depletion and less

positively to repletion with the current RDA compared to women with the CC genotype.














CHAPTER 1
INTRODUCTION

Folate is an essential vitamin that functions to accept and transfer one-carbon units

in nucleotide synthesis, amino acid interconversions including methionine regeneration,

and the formation of S-adenosylmethionine (SAM)-the primary methylating agent in the

body. An important reaction in folate metabolism is the reduction of

5,10-methylenetetrahydrofolate (5,10-methyleneTHF) to 5-methyltetrahydrofolate

(5-methylTHF) by the 5,10-methylenetetrahydrofolate reductase (MTHFR) enzyme.

This reaction generates folate in the form of 5-methylTHF, essential for the remethylation

of homocysteine to methionine. The conversion of homocysteine to methionine is

essential for the synthesis of SAM, which is the methyl donor in > 100 physiological

reactions, including methylation of DNA, RNA, and membrane phospholipids as

previously reviewed (1).

Emerging science involving single-nucleotide polymorphisms offers new insight

and a more precise understanding of how individual genetic variations influence folate-

dependent metabolic pathways (2). Single-nucleotide polymorphisms, genetic variations

present in > 1% of the population, can act alone or synergistically with nutritional

deficiencies to accelerate and accentuate metabolic pathologies. A very common

polymorphism and one that has been studied extensively is a C-to-T substitution at base

pair 677 in the gene that codes for MTHFR (3). The MTHFR 677C->T variant is

prevalent in the overall population: -12% homozygous (TT); and -50% heterozygous

(CT) (4). This variant results in an alanine-to-valine substitution in the enzyme resulting









in a marked reduction in enzyme stability that can be enhanced by the addition of

5-methylTHF (5). In individuals who are homozygous for the MTHFR 677C->T variant

(TT) the impaired enzyme stability is associated with reduced methylated blood folate (6)

and elevated plasma homocysteine concentration that can be significantly improved in

response to improved folate status (7).

Different studies have provided evidence that methylation of DNA plays a role in

genome stability and gene expression (8,9). The potential for impaired folate status

coupled with the MTHFR 677 TT genotype to negatively influence genome stability

(10,11), including DNA methylation (12,13), provided one incentive for the present

study. The effect of folate inadequacy in individuals with the TT genotype on indicators

of global DNA methylation has not been previously evaluated in a metabolic study in

which nutrient intake was carefully controlled. The present study was designed to

monitor changes in DNA methylation by MTHFR genotype in conjunction with other

folate status indicators in response to a folate depletion-repletion protocol.

An estimated 120,000-150,000 infants are born with a major birth defect in the US

each year, representing more than 3% of all live births (14). Emerging scientific

evidence involving single nucleotide polymorphisms such as the MTHFR 677C->T

variant, including data from the present study, may lead to a precise understanding of

how individual genetic variations influence folate-dependent metabolic pathways and

potentially increase birth defect risk. Various studies have shown that the MTHFR 677

TT genotype is a significant risk factor for neural tube defects when folate intake is not

sufficient to maintain metabolic homeostasis (15,16). Since the combined presence of the

MTHFR 677C->T variant and low folate status has been associated with increased risk









for birth defects, the present study was designed to address the specific aim in females of

reproductive age.

Hypotheses

1. Consumption of a low-folate diet will result in more significant impairment in
indicators of folate status and global DNA methylation in young women with the
homozygous MTHFR 677C->T TT genotype compared to those with the CC
genotype.

2. Consumption of the current Recommended Dietary Allowance for folate (400 [tg
dietary folate equivalents/d) following a low-folate diet will not be sufficient to
significantly improve folate status or global DNA methylation in females with the
TT compared to the CC genotype.

Specific Aim

The specific aim of the present study is to determine the combined effects of the

MTHFR 677C-T variant and dietary folate depletion-repletion on indicators of folate

status response (serum and red blood cell folate and plasma homocysteine

concentrations) and global DNA methylation ([3H]methyl group acceptance and

methylcytosine/total cytosine ratio) in women of reproductive age.















CHAPTER 2
BACKGROUND AND LITERATURE REVIEW

Folate

Chemistry

Folate is a general term that refers to naturally occurring food folate and synthetic

folic acid found in supplements and fortified foods. Synthetic folic acid is yellow with a

molecular weight of 441.4. Folate is made up of three parts that are all necessary for

vitamin activity. A pteridine bicyclic ring is linked by a methylene group to

para-aminobenzoic acid whose carboxy group is peptide bound to the a-amino group of

glutamate to form folate (17) (Figure 2-1).

H2N N N

N 0 H
OH N )-C-N-C-CH2CH 2-COOH
H H
COOH
p-a am no
pterdi ne benzoic acid glutamate


Figure 2-1. Structure of folate. Modified from Bailey, L. B., Moyers, S., and Gregory,
J.F. (2001) Folate. In: Present Knowledge in Nutrition (Bowman, B. A. and
Russell, R.M., ed.), p. 214. ILSI Press, Washington, D.C.

Dietary Sources

Mammals are unable to synthesize folate de novo and must therefore obtain it from

the diet. Folate is not considered to be prevalent in the food supply. Two different types

of folate are available for consumption: folate found naturally in the food supply and

synthetic folic acid added to enriched food products and supplements.









Naturally occurring food folate. Some good sources of naturally occurring food

folates are orange juice, dark green leafy vegetables, asparagus, strawberries, peanuts,

and legumes such as kidney and lima beans (18). Folate concentration in raw foods is

higher than in cooked foods due to the folate losses incurred by cooking (19). Variability

of losses associated with cooking is attributable to differences in oxygen exposure,

ascorbic acid content, and the amount of water present (19). The folates that occur

naturally in food are the fully reduced tetrahydrofolates and usually have 5 to 7

glutamates in the polyglutamate side chain (19).

Folic acid in fortified foods. Another major source of folate is synthetic folic acid

used in enriched foods and fortified, ready-to-eat breakfast cereals. Effective January 1,

1998, the U.S. Food and Drug Administration (FDA) mandated the fortification of all

cereal grain products labeled as enriched (e.g., bread, pasta, flour, rice) with 140 ptg folic

acid per 100 g of product (20). Thousands of food items have been affected under this

mandate (21). A large number of ready-to-eat breakfast cereals contain folic acid in

varying amounts. Most ready-to-eat cereals are fortified to provide 25% the Daily Value

of folate, with some brands providing four times this amount (18).

Bioavailability

Folate bioavailability refers to the overall efficiency of utilization of the vitamin,

including absorption, transport, metabolism, catabolism, and excretion (1,22). The

bioavailability of food folate differs greatly from the bioavailability of synthetic folic

acid.

Food folate bioavailability. The bioavailability of folate in different foods varies

considerably because of differences in digestion, absorption, and metabolism. Some

possible digestion/absorption issues that can affect folate bioavailability include altered









pH, conjugase activity, and transit time (22). The effect of different food components on

folate utilization further complicates the issue by trapping the folate in the food matrix or

inhibiting conjugase activity (22). Food processing can cause oxidative damage of

folates (22) and can account for 50 to 95% of any folate lost (23). Alcohol and

pharmaceuticals also can inhibit the absorption of folate (22). All of these factors

attribute to the high variability of folate bioavailability, which can be as high as 96% and

as low as 25% (24).

Synthetic folic acid bioavailability. Synthetic folic acid is much more

bioavailable than food folate. When folic acid is administered without food it is 100%

bioavailable (25). Sauberlich et al. (23) estimated that when compared to synthetic folic

acid, the bioavailability of food folate is no more than 50% available. Pfeiffer et al. (26)

examined the bioavailability of 13C5-labeled folic acid administered in apple juice when

given with or without food. Consumption of the folic acid supplement with food slightly

decreases its bioavailability by approximately 15% (26). The Dietary Reference Intake

Committee of the Institute of Medicine (IOM) used these data to derive an estimated

bioavailability of folic acid when consumed with food (85%) (27). This estimate of folic

acid bioavailability was used by the IOM as a basis for the new term, dietary folate

equivalents (DFEs), which was used to express the 1998 Dietary Reference Intakes

(DRIs).

Dietary folate equivalents (DFEs). The DFEs convert all forms of dietary folate,

including synthetic folic acid in fortified products, to an amount that is equivalent to food

folate (28). The rationale for the DFEs is that folic acid consumed with food is only 85%

bioavailable and naturally occurring food folate is only 50% bioavailable. Using these









estimates of folate bioavailability, folic acid consumed with food is estimated to be 1.7

times (i.e., 85/50) more available than natural food folate (27). The total folate content

can be calculated by multiplying the micrograms of synthetic folic acid by 1.7 and then

adding this number to the number of micrograms of food folate present in the meal (18).

To compare folic acid with food folate the following conversions are used: 1 [tg DFE = 1

[tg food folate = 0.5 [tg folic acid taken on an empty stomach = 0.6 [tg folic acid taken

with meals. When estimating food folate intake only, no adjustment is needed (27).

Absorption

Before absorption of ingested food folate can occur, it must be hydrolyzed to the

monoglutamate form by glutamate carboxypeptidase II (EC 3.4.17.21) (29) also called

pteroylpolyglutamate hydrolase or folate conjugase. This enzyme is located in the jejunal

brush border membrane and has an optimum pH of 6.5 to 7.0 (30). Under normal

conditions, monoglutamyl folate is transported across the intestinal membrane by a

saturable, pH-dependent carrier-mediated intestinal folate carrier (31). The expression of

this intestinal folate carrier may be upregulated in response to folate deficiency (32).

When folate concentrations are high, a nonsaturable mechanism involving passive

diffusion transports folate across the intestinal membrane. This mechanism may be more

important in the absorption of supplemental folic acid as opposed to food folate (22).

Most folate is converted to the reduced forms, dihydrofolate (DHF) or tetrahydrofolate

(THF), by dihydrofolate reductase before portal blood entry. This mechanism is

saturable and large amounts of oxidized folic acid have been found in the plasma and

urine of individuals ingesting 400-800 [tg/d of folic acid (22). Further metabolism to a

methylated or formylated form of folate also may occur in the mucosal cells before portal

blood entry (22).









Transport

Once the folate is absorbed in the monoglutamate form, it travels in the portal

circulation, mostly as 5-methylTHF, to the liver where it is reduced and conjugated for

retention. The major forms of folate in the liver are 5-methylTHF and 10-formylTHF,

which can be secreted into the bile and reabsorbed via enterohepatic circulation (33).

Approximately two-thirds of folates in the plasma are protein bound, 50% of which are

bound to albumin. The remaining one-third is tightly bound to folate binding protein

(34).

Folate binding proteins (FBPs). Membrane associated FBPs, also known as

folate receptors, transport folate across membranes from circulating blood into cells, are

highly localized and expressed in specific tissues and cells (35), and have a higher

affinity for oxidized folic acid than for reduced folates (36). The FBPs are essential for

normal embryonic development and can cause embryonic lethality if the gene that codes

for FBP is knocked out in mice (37). The FBP knockout mice have been brought to term

with folinic acid supplementation and were born with normal phenotypes (35). Although

the FBP has been characterized, little is known about how cells obtain their folate.

Antony (38) reviewed two possible mechanisms by which cells take up folate. The first

is the well-known process of endocytosis, and the second is referred to as potocytosis. In

potocytosis the FBPs are gathered in clusters at the plasma membrane, which move into

caveolae and concentrate circulating folates. Following sudden acidification of the

caveloae, the folates dissociate from the folate receptors and are transported into the

cytoplasm by anion channels. This is not the only mechanism for folate transport (38).

This membrane-associated FBP has extensive homology with the plasma FBP (17).









Reduced folate carrier (RFC). A RFC also exists that transports reduced folates

into the cell (36). The RFC protein is the same folate carrier protein that is expressed in

the intestine. The RFCs are located in all tissues and cells (35) and have a higher affinity

for reduced folates than oxidized folic acid. They function to form channels in the

plasma membrane through which reduced folates, mostly 5-methylTHF, can cross into

the cytoplasm of the cell (39). Zhao et al. (40) determined the consequence of

inactivation of the RFC in embryonic RFC knockout mice who died in utero. They also

observed that near normal development of RFC knockout mice could be attained with

daily supplementation of the heterozygous pregnant dams with folic acid, but these pups

died within 12 d of birth. Zhao et al. (40) concluded that the folic acid supplementation

was sufficient in utero, but that there was probably insufficient folate in the mother's

milk to sustain life.

A polymorphism of the RFC gene that results in a G--A substitution at base pair

80 replaces arginine with histidine (36) in the protein. This polymorphism has not been

found to affect plasma folate or homocysteine concentrations in adults but may be more

important to the developing embryo (36). De Marco et al. (41) observed that neural tube

defect (NTD)-affected Italian children with the homozygous variant genotype (GG) had a

significantly higher risk for NTDs than control children with the normal genotype (AA).

In addition, mothers with the GG genotype had a higher risk of giving birth to an NTD-

affected child compared to mothers with the AA genotype (40).

Folates enter cells in the monoglutamate form and if in the oxidized form must be

reduced by dihydrofolate reductase to THF and polyglutamated by folylpoly-y-glutamate

synthetase. Conjugation retains folates in the cells and is required for participation in









one-carbon metabolism. The different coenzymatic forms of folate are all derivatives of

THF and can contain a methyl (CH3), methylene (-CH3-), methenyl (-CH=), formyl

(-CH=O), or formimino (-CH=NH) group (1). Tetrahydrofolate accepts the one-carbon

moieties, which become bonded to the N5 or N10 atoms or both. The polyglutamate

folates must be hydrolyzed before release from the tissues into circulation with

y-glutamylhydrolase converting the folates back to the monoglutamate form (17). The

total body content of folate is approximately 15 to 30 mg with the liver containing 50%

of the body's folate stores (42).

Biochemical Functions

Folate's main biochemical functions are to accept and transfer one-carbon units

involved in amino acid metabolism, purine and pyrimidine synthesis, and the formation

of SAM, the main methyl donor in > 100 reactions. Tetrahydrofolate is the main

acceptor molecule and ultimately ends up in pathways required for nucleotide

biosynthesis or methylation reactions. An overview of the different folate pathways is

presented in Figure 2-2.

Amino acid metabolism. Folate functions as an intermediate in the metabolism of

the amino acids serine, glycine, methionine, homocysteine, and histidine. The

metabolism of serine, glycine, methionine, and homocysteine are closely related.

Tetrahydrofolate is converted to 5,10-methyleneTHF via serine

hydroxymethyltransferase (SHMT) (Figure 2-2, reaction 3). Serine is converted to

glycine in the process, and both reactions are reversible. Pyridoxal phosphate is required

as a cofactor for this reaction. The degradation of glycine requires THF and NAD+. The

P-carbon of serine provides the most one-carbon units in all mammalian systems. These













DNA Synthesis


IformylTHF ----- 5-methylTHI/ Homoyst



Purine Synthesis + THF Cystathionine

1. dihydrofolate reductase 8. methionine synthase
2. thymidylate synthase 9. S-adenosylmethionine synthase Cysteine
3. serine hydroxymethyltransferase 10. cellular methyltransferases
4. cyclohydrolase 11. S-adenosylhomocysteine hydrolase
5. formate-activating enzyme 12. cystahionine 3 -synthase
6. transformylases 13. cystathionase
7. methylenetetrahydrofolate 14. betaine:homocysteine
reductase methyltransferase
Figure 2-2. Folate metabolism. Adapted from Bailey, L. B., Moyers, S., and Gregory, J.F. (2001) Folate. In: Present Knowledge in
Nutrition (Bowman, B.A. and Russell, R.M., ed.), p. 217. ILSI Press, Washington, D.C.









one-carbon units get transferred to THF (43). After 5,10-methyleneTHF is formed, it is

reduced to 5-methylTHF by the irreversible reaction requiring the enzyme MTHFR

(Figure 2-2, reaction 7). The substrate and cofactor for methionine synthase, the enzyme

that remethylates homocysteine to methionine, is 5-methylTHF (Figure 2-2, reaction 8)

and requires vitamin B12, or cobalamin, as a coenzyme. A vitamin B12 deficiency can

lead to a secondary folate deficiency that investigators have termed the "methyl trap".

Methionine synthase depends on vitamin B12 as a coenzyme and its activity is reduced

during a vitamin B 12 deficiency. This results in the accumulation of 5-methylTHF,

which gets "trapped" in this form and results in the decreased availability of THF and all

of the other forms of folate. This can lead to megaloblastic anemia due to insufficient

folate coenzymes available for DNA synthesis (17).

Homocysteine can also be remethylated to methionine by betaine:homocysteine

methyltransferase (BHMT) (Figure 2-2, reaction 14), which is a betaine-dependent

reaction found in the pathway of choline oxidation. Choline is a methyl rich compound

present in food that is first oxidized to betaine and then demethylated by BHMT (44).

Choline also can be synthesized by the body in the SAM dependent methylation of

phosphatidylethanolamine to form phosphatidylcholine, which can be further catabolized

to choline. Choline oxidation primarily takes place in the liver and kidney. Choline is

first oxidized to betaine aldehyde via choline oxidase. Betaine aldehyde is then oxidized

to betaine via betaine aldehyde dehydrogenase (44). Betaine can then be used in the

BHMT reaction. The interdependence of choline and folate has been demonstrated in a

variety of studies with rats. Rats fed a choline or choline and methionine deficient diet









for 2 wk to 12 mo had decreased hepatic folate stores (45-47). Adverse effects from a

choline deficiency were reversed within 2 wk with adequate dietary choline (47).

Histidine is the other amino acid dependent on folate metabolism. The deamination

of histidine produces urocanic acid, which becomes formiminoglutamate (FIGLU) after

further metabolism. Formiminoglutamate loses the formimino group via formimino

transferase to THF to produce N5-formiminoTHF. Glutamic acid is the final product of

this reaction. The pathway of histidine metabolism is shown in Figure 2-3.

THF


Histidine



FIGLU
Formimino transferase

Glutamic Acid


5-formiminoTHF


Figure 2-3. Histidine metabolism.

Purine and pyrimidine synthesis. The nucleotide biosynthesis pathway utilizes

the 5,10-methyleneTHF coenzyme. Pyrimidine synthesis involves the formation of

deoxythymidylate monophosphate (dTMP) from deoxyuridylate monophosphate (dUMP)

via thymidylate synthase (Figure 2-2, reaction 2), which requires 5,10-methyleneTHF as

a coenzyme. This is the rate-limiting step in the cell cycle and allows DNA replication to

continue. Once thymidylate synthase utilizes 5,10-methyleneTHF as a coenzyme, it is

oxidized to DHF and then regenerated to THF via dihydrofolate reductase (Figure 2-2,

reaction 1). Purine synthesis depends on the conversion of 5,10-methyleneTHF to









10-formylTHF via 10-formylTHF synthetase (Figure 2-2, reaction 4). The folate

coenzyme is used to donate formyl groups to different positions of the purine ring and

depends on a format activating enzyme (Figure 2-2, reaction 5).

S-Adenosyl methionine (SAM) synthesis. The formation of SAM is dependent

on the metabolism of homocysteine and methionine. Under normal conditions,

methionine synthase transfers the methyl group from 5-methylTHF to vitamin B12 and

then to homocysteine. In this reaction, 5-methylTHF is reconverted to THF. The newly

synthesized methionine can then be converted to SAM by SAM synthase (Figure 2-2,

reaction 9) when combined with adenosine. S-Adenosylmethionine can then act to

donate methyl groups in over 100 different methylation reactions, including DNA and

RNA methylation. Once a methyltransferase uses the methyl group provided by SAM, it

is converted to S-adenosylhomocysteine (SAH) (Figure 2-2, reaction 10).

S-Adenosylhomocysteine is hydrolyzed to homocysteine and adenosine via the reversible

SAH hydrolase (Figure 2-2, reaction 11). Homocysteine can be remethylated to

methionine or converted to cysteine through the transulfuration pathway where

cystathionine P-synthase (CBS) (Figure 2-2, reaction 12) converts homocysteine to

cystathionine, a reaction that is dependent on pyridoxal phosphate and serine.

Cystathionase (Figure 2-2, reaction 13) converts cystathionine to cysteine.

Cytosolic folate metabolism differs from mitochondrial folate metabolism. The

composition of folate pools within the cytosol and mitochondria varies among different

tissues (1). In the cytosol, one source of one-carbon groups is provided by format from

the mitochondria and cytosol that is converted to 10-formylTHF by 10-formyl synthetase

(Figure 2-4, enzyme C3). It serves as a one-carbon source in the cytosol for purine












Cytosol
CO THF
C1 (
10-formylTHF -



Purines + THF 5,10- methenylTH]


Mitochondrial Matrix


Figure 2-4. Compartmentalization of folate pools between the cytosol and mitochondria. Reproduced from Bailey, L. B., Moyers, S.,
and Gregory, J.F. (2001) Folate. In: Present Knowledge in Nutrition (Bowman, B.A. and Russell, R.M., ed.), p. 218
Figure 3. ILSI Press, Washington, D.C.









Table 2-1. Compartmentalization of enzymes and reactions.
Reaction Enzyme
C1:M1 5,10-MethyleneTHF dehydrogenase
C2:M2 5,10-MethyleneTHF cyclohydrogenase
C3:M3 10-FormylTHF synthetase
C4 Glycine N-methyltransferase
C5:M5 Serine hydroxymethyltransferase
C6 Methionine synthase
C7 Glycinamide ribonucleotide transformylase
C8 Phosphoribosylamino-imidazole carboxamide
transformylase
C9 5,10-MethyleneTHF reductase
C10 Thymidylate synthase
M11 Dimethylglycine dehydrogenase
M12 Sarcosine dehydrogenase
M13 Glycine cleavage system
C14:M14 10-FormylTHF dehydrogenase
Reproduced from Bailey, L. B., Moyers, S., and Gregory, J.F. (2001) Folate. In: Present
Knowledge in Nutrition (Bowman, B.A. and Russell, R.M., ed.), p. 218 Table 2. ILSI
Press, Washington, D.C.

synthesis. 10-FormylTHF can be further reduced to 5,10-methenylTHF by

5,10-methenylTHF cyclohydrolase (Figure 2-4, enzyme C2) and reduced once again by

5,10-methyleneTHF dehydrogenase to 5,10-methyleneTHF for pyrimidine synthesis.

The enzyme MTHFR (Figure 2-4, enzyme C9) reduces 5,10-methyleneTHF to

5-methylTHF for homocysteine remethylation to methionine in the cytosol. The

interconversion of serine and glycine not only provides one-carbon units to the cytosol,

but also to the mitochondria. The difference between cytosolic and mitochondrial folate

metabolism can be explained by the form of SHMT (Figure 2-4, enzymes C5:M5), the

enzyme used to catalyze this reaction, since cytosolic SHMT is different than

mitochondrial SHMT. Glycine cleavage occurs only in the mitochondria. The final

folate-dependent steps of choline catabolism occur in the mitochondria.









The biochemical functions of folate are collectively referred to as one-carbon

metabolism (1). Although each function is independently important, they are all

dependent on the cellular concentration of THF and each other to function.

Folate-Related Polymorphisms

Single nucleotide variations in the genome can result in genes that code for

enzymes with different activity. If a genetic variation is present in > 1% of the

population, it is considered a genetic polymorphism (48). Polymorphisms affecting one-

carbon metabolism are of special interest to investigators due to their frequency in the

population and their effects on disease risk and developmental abnormalities. The focus

of folate research has shifted from evaluating the effect of severe deficiency on clinical

indicators and blood folate concentrations to identifying functional indicators of disease

risk and how they are affected by genetic polymorphisms. The next section will include a

synopsis of recent research findings and key research questions related to polymorphisms

affecting folate metabolic function.

Methylenetetrahydrofolate Reductase (MTHFR)

Although folate metabolism involves over 30 genes, enzymes, and transporters, the

most extensively studied polymorphism affects MTHFR, the enzyme that catalyzes the

reduction of 5,10-methyleneTHF to 5-methylTHF (3). In 1988 Kang and colleagues (49)

discovered a thermolabile variant of MTHFR that was positively correlated to increased

cardiovascular risk and increased homocysteine concentrations. In 1994, Goyette et al.

(50) isolated a complementary DNA (cDNA) copy of human MTHFR that was 22

kilobases long consisting of 11 exons. Isolation of this cDNA enabled them to identify

nine mutations in the gene in patients with severe MTHFR deficiency (50,51). A more

recent report increased this number to 33 severe mutations identified in patients with









severe MTHFR deficiency (52). Frosst et al. (3) identified a homozygous variant of this

gene that resulted in decreased enzyme activity and increased thermolability in vitro.

This autosomal recessive variation results in a C--T substitution at base pair 677 that

replaces alanine with valine in the enzyme. Any decrease in MTHFR activity from the

677C->T variant will result in lower 5-methylTHF available to donate a methyl group for

the remethylation of homocysteine to methionine. This can result in elevated

homocysteine concentrations in individuals with the homozygous variant (TT) genotype

in contrast to the heterozygous variant (CT) or homozygous normal (CC) MTHFR

677C->T genotype. The homozygous TT variant is termed "thermolabile" because of a

significant decrease in residual activity compared to CC controls after heat inactivation at

460C for 5 min (49). Frosst et al. (3) compared specific activity and residual activity after

heating for 5 min at 460C. They found that subjects with the TT genotype had a specific

activity < 50% of that of the subjects with the CC genotype and a residual activity < 35%

after heating. Rozen (53) found a 50-60% reduction in residual activity among

individuals with the TT genotype at 370C and a 65% decrease at 460C compared to

individuals with the CC genotype.

Matthews et al. (54) reported that the MTHFR gene codes for an enzyme with two

identical 77 kDa subunits, each consisting of a 40 kDa N-terminal domain and a 37 kDa

C-terminal domain. In addition, Frosst et al. (3) found a 70 kDa subunit, which they

attributed to the presence of isozymes. Sumner et al. (55) discovered that SAM binds to

the C-terminal domain, which led them to implicate this domain as the regulatory region

because SAM is an allosteric inhibitor of MTHFR. Matthews et al. (56) reviewed the

similarities between the N-terminal domain of the human MTHFR and smaller proteins









of enteric bacteria that catalyze the same reaction. They explained that since SAM does

not regulate the bacteria proteins, the N-terminal domain must be the catalytic region.

This is important because the C--T substitution occurs in exon 4 of the N-terminal

domain (57). Matthews et al. (56) purified MTHFR from Escherichia coli. They

reported that the enzyme was a flavoprotein with flavin adenine dinucleotide (FAD) as its

cofactor, which is reduced by NADH and NADPH and then reduces 5,10-methyleneTHF

to 5-methylTHF. Guenther et al. (58) characterized the normal (CC) and variant (TT)

MTHFR enzymes from Escherichia coli, which revealed that the variant enzyme loses its

flavin cofactor more readily than the normal enzyme. They reported that the catalytic

domain shared by all MTHFRs is a barrel that binds FAD, and that the variant does not

cause decreased activity of the enzyme directly but instead decreases enzyme stability,

which facilitates the dissociation of FAD and decreases enzyme activity. They found that

folate supplementation confers protection by increasing the affinity of MTHFR for FAD.

Matthews (59) hypothesized that the exact mechanism of this protection lies in the

positioning of the folate cofactor with respect to the barrel. Folate may bind with the

pteridine ring stacked above the flavin, thus stabilizing the enzyme and preventing FAD

dissociation. Yamada et al. (5) were the first to characterize human MTHFR by using a

baculovirus expression system. They identified important differences between the

mammalian enzyme and the prokaryotic enzyme. For example, the prokaryotic enzyme

lacks a regulatory domain and uses NADH as a reductant while the mammalian enzyme

has a regulatory domain and uses NADPH as a reductant. Another difference is that

prokaryotic MTHFR is a tetramer in contrast to the mammalian MTHFR, which is a

dimer. Their results agree with the previous findings of Guenther et al. (58) that the









decreased activity of the enzyme is due to decreased protein stability resulting from FAD

dissociation. This loss of FAD leads to dissociation of the dimer into monomers and

reduction in activity (59) and has been found to occur 10 times faster in the homozygous

variant (TT) compared to the normal (CC) enzyme (58). Yamada et al. (5) concluded

that individuals with the TT genotype for the MTHFR 677C->T polymorphism are at

higher risk for increased homocysteine concentrations if their folate concentrations are

low. This increased risk is due to less remethylation of homocysteine to methionine as a

result of the decreased MTHFR stability. Decreased MTHFR stability will lead to even

more dissociation of FAD and result in less enzyme activity. The in vitro studies

performed by this group showed that increasing folate, SAM, and FAD concentrations

protected against loss of activity in the affected human MTHFR enzyme.

The association between riboflavin status and homocysteine concentration has been

evaluated. Hustad et al. (60) reported a significant inverse dose-response association

between plasma riboflavin and homocysteine concentrations in their combined MTHFR

677 CT/TT group. When plasma riboflavin was separated into quartiles, they observed a

1.4 kmol/L higher homocysteine concentration in the lowest compared to the highest

quartile. They concluded that riboflavin is an independent determinant of plasma

homocysteine concentration regardless of folate status. McNulty et al. (61) separated the

riboflavin status of their population into tertiles and found that among subjects in the

lowest tertile, the mean homocysteine concentration of subjects with the TT genotype

was approximately twice the concentration in either the CT or CC genotype groups.

There were no significant differences in mean homocysteine concentrations between

genotype groups in either the medium or high tertiles of riboflavin status. They









concluded that unlike folate status, riboflavin status is only a predictor of homocysteine

concentration in subjects with the TT genotype, and that a riboflavin-dependent

mechanism may contribute to the interrelationship between folate and plasma

homocysteine concentrations in this group (61). Jacques et al. (62) further clarified the

relationship between riboflavin and homocysteine concentrations by reporting that the

association between these indices in their population was only significant in those with

lower folate status. When further stratified by genotype, the interaction between folate

and riboflavin was only significant in subjects with the TT genotype. Only those with

low folate status and the TT genotype showed a significant association between

riboflavin and homocysteine concentrations. In contrast to the previously described

studies, Moat et al. (63) found an inverse association between riboflavin and

homocysteine concentrations in individuals with both the CC and TT genotypes, but

further studies are warranted to clarify this discrepancy. Rozen (64) suggested that

additional studies to determine the role of folate in the riboflavin-homocysteine

association would be useful before drawing conclusions.

An extensive review by Botto and Yang (4) provides a description of the

population frequency for the MTHFR 677C-T polymorphism. They estimated that

approximately 22% of Hispanics, 12% of Caucasians, 11% of Asians, and 1 to 2% of

African Americans are homozygous for the variant. The variant has been found in -35%

of alleles (3,65). The frequency of the homozygous variant varies among racial and

ethnic groups. The lowest prevalence is among blacks and Africans (-2% and 0%).

Stevenson et al. (66) reported the frequency of the TT genotype in 151 whites and 146









blacks from South Carolina and found a frequency of 13% and 0%, respectively, which

agrees with other reports (4).

The effect of the TT genotype can be quantitated by evaluating the change in

cellular composition of the folate derivatives (6). If 5,10-methyleneTHF is not reduced

to 5-methylTHF, the methylene group of 5,10-methyleneTHF can be donated to dUMP to

form dTMP or oxidized to 5,10-methenylTHF, which ultimately is converted to

10-formylTHF for the synthesis of purines (6). One would expect a decrease in MTHFR

activity to result in an increase in formylated folates and a decrease in methylated folates.

Bagley and Selhub (6) provided data to support this hypothesis by evaluating the effect of

the MTHFR genotype on the form of folate within red blood cells. Of eight subjects with

the CC genotype, all had 100% methylated folates. Of the 10 subjects with the TT

genotype, eight had formylated folates ranging from 0 to 59% of the total folate content.

Genotype did not affect total red blood cell folate concentration. Zittoun et al. (67) also

evaluated the effect of the TT genotype on methylfolates and reported that in subjects

with the CC genotype, 71% of the red blood cell folate concentration was methylfolate

compared to 66% and 27% in individuals with the CT and TT genotypes, respectively.

After separating their study population into high and low red blood cell folate status,

Friso et al. (13) found significantly less methylated folates in subjects with the TT

genotype, regardless of folate status. These data support the findings of Bagley and

Selhub (6).

Another more recently discovered MTHFR polymorphism, 1298A->C, results in

an A--C base pair substitution at base pair 1298 that replaces glutamic acid with alanine

in the C-terminal regulatory domain of the enzyme resulting in a homozygous variant









(CC), heterozygous variant (AC), or normal (AA) genotype (57,68). The MTHFR

1298A->C polymorphism was first discovered by Viel et al. (69) in patients with ovarian

carcinomas, but was not characterized until later (57,68). Yamada et al. (5) were unable

to distinguish between the properties of the variant and normal enzymes after baculovirus

expression. They determined that the homozygous variant enzyme did not affect

catalytic function alone or in combination with the 677 TT genotype, and that it appeared

thermostable and unaffected by changes in in vitro folate concentration. This is in

contrast to the 60% decrease in activity observed in lymphocyte extracts with the

homozygous variant genotype (57)

This MTHFR 1298A->C polymorphism was reported to affect 10% of the

Canadian population with an allele frequency of 33%. Neither individuals with the CC

variant nor individuals with the heterozygous variant genotype have higher homocysteine

concentrations or lower folate concentrations (68,70-72) than normal individuals. The

risk appears when the 677C->T variant is combined with the 1298A-C variant.

Individuals heterozygous for both polymorphisms have been reported to have reduced

MTHFR activity based on observed reductions in folate concentration and increased

homocysteine concentration in population groups (68).

An association between vitamin B12 status and the 677C->T and 1298A-C

MTHFR genotypes was recently reported by Bailey et al. (73). Individuals with the

homozygous MTHFR 677C->T variant had higher plasma homocysteine concentrations

than any other genotype combination, including those heterozygous for both mutations.

They reported a negative association between plasma vitamin B 12 concentration and

serum folate concentration with plasma homocysteine concentration, which was









dependent on MTHFR genotype. Plasma homocysteine concentration significantly

decreased as vitamin B 12 concentration increased in individuals heterozygous for both

MTHFR polymorphisms in contrast to the individuals with the normal genotypes whose

plasma vitamin B 12 concentration was not associated with changes in homocysteine

concentration. This study was the first to report an inverse correlation between plasma

vitamin B 12 concentration and plasma homocysteine concentration in individuals

heterozygous for the MTHFR 1298A->C polymorphism even when vitamin B12 was

within the normal range.

The population frequency of the MTHFR 677C->T and 1298A-C polymorphisms

in the MTHFR gene far exceeds the percentage required to be defined as a genetic

polymorphism (i.e., > 1% of the population) (48) and has been reported to affect

individual folate and homocysteine concentrations. Further study of these genetic

variations in different populations may determine whether individuals with the

homozygous variants require higher intakes of different vitamins.

Folate, MTHFR, and homocysteine. When Kang et al. (49) tested the

thermolability of MTHFR, they found that two of their six subjects who had the

thermolabile enzyme had hyperhomocysteinemia and deficient plasma folate

concentrations, while the other four subjects had normal homocysteine and folate

concentrations. This is the first study to show a possible relationship between deficient

folate status and hyperhomocysteinemia in subjects with the thermolabile enzyme.

Several subsequent studies have provided data that individuals with the TT genotype

have higher homocysteine concentrations than the individuals with the CT or CC

genotypes (65,74-83). Frosst et al. (3) concluded that the MTHFR TT genotype is the









most important genetic determinant for moderate hyperhomocysteinemia. Jacques et al.

(7) clarified that hyperhomocysteinemia associated with the MTHFR 677C->T variant is

dependent on folate status, and results from Kauwell et al. (84) support this conclusion.

Folate status also has been shown to be impaired in individuals with the TT genotype

compared to controls (67,77,85). Specifically, pregnant and nonpregnant women

homozygous for the MTHFR 677C->T variant have been found to have lower red cell

folate concentrations than women with the CC genotype (86). Kauwell et al. (84)

demonstrated using a controlled folate dietary protocol with elderly women that subjects

with the TT genotype had significantly lower folate concentrations and higher

homocysteine concentrations after 7 wk of depletion than subjects with the CT or CC

genotypes. Subjects with the TT genotype had the greatest reduction in homocysteine

concentration after 7 wk of folic acid supplementation with 415 .g/d than subjects with

the CT or CC genotype. Guinotte et al. (87) used a similar dietary protocol in a

metabolic study of Mexican American women. Subjects with the TT genotype had lower

serum and red blood cell folate concentrations throughout depletion (7 wk) and repletion

(7 wk) with 400 tg DFE/d. Homocysteine concentrations did not differ between

genotypes throughout depletion, but subjects with the TT genotype had higher

homocysteine concentrations throughout repletion with 400 [tg DFE/d.

Ashfield-Watt et al. (81) examined the influence of the MTHFR 677C->T genotype

on folate status response in adults aged 18 to 65 who were given dietary intake advice.

Subjects were assigned to one of three dietary intake advice groups for 4 mo each: advice

to exclude folate-rich and fortified foods, advice to consume a folate-rich diet (400 pg

folate/d), and advice to take a daily folic acid supplement (400 .g folic acid/d). Subjects









with the TT genotype had the greatest increase in homocysteine and decrease in folate

concentrations after 4 mo of folate exclusion. Folate supplementation with fortified

foods or supplements were effective in lowering plasma homocysteine concentrations of

subjects with the TT genotype to a significantly greater extent than that of the subjects

with the CT and CC genotypes. Dietary advice to consume a folate-rich diet and advice

to take a daily folic acid supplement effectively increased folate concentrations in

subjects with the TT genotype above the baseline concentrations of subjects with the CC

genotype, but the response was not significantly greater than that observed in the other

genotype groups. The investigators recommended folate intakes between 400 and 600 tg

folate/d (575 to 830 [tg DFE/d) for individuals with the TT genotype to maintain normal

homocysteine concentrations. Although this was not a controlled metabolic study, it

demonstrated the beneficial effects of folate intake on folate status and homocysteine

concentrations.

In a recently published observational study, de Bree et al. (82) determined plasma

folate and homocysteine concentrations by genotype in a Dutch population. They

observed that subjects with the TT genotype had lower plasma folate and higher plasma

homocysteine concentrations than subjects with the CT and CC genotypes. The

difference in folate status indicators between MTHFR genotypes is well established in

observational data.

MTHFR and chronic disease. Kang et al. (49) were the first to suggest that the

homozygous MTHFR 677C-T variant may be a risk factor for cardiovascular disease.

Since then conflicting results have been reported. Some studies have found a positive

association between MTHFR, hyperhomocysteinemia, and vascular disease (74,75,88-92)









while others have not found this association (76,93,94). Brattstrom et al. (65) conducted

a meta-analysis to determine the relationship between cardiovascular disease risk and

MTHFR genotype. They reported that although individuals with the TT genotype had an

average 25% higher homocysteine concentration than individuals with the CC genotype,

the variant is not associated with an increase in cardiovascular disease risk. A possible

reason for the lack of association between cardiovascular disease risk and individuals

with the TT genotype is that there was insufficient power in the individual studies.

Chen et al. (95) developed a MTHFR knockout mouse model to evaluate the in vivo

pathogenic mechanisms of MTHFR deficiency. Plasma homocysteine concentrations

were 1.6-fold and 10-fold higher in heterozygous and homozygous knockout mice

compared to controls. Both the heterozygous and homozygous knockout mice had

significantly decreased SAM concentrations and/or increased SAH concentrations and

abnormal lipid deposition in the aorta. Homozygous knockout mice were

developmentally retarded with cerebellar pathology. Although this study evaluated the

effect of MTHFR deficiency in mice, it is reflective of possible abnormalities in humans

with the TT genotype.

Only one study has linked homozygosity for the 1298A->C variant with a higher

risk for early-onset coronary artery disease independent of homocysteine concentration

(94). Rothenbacher et al. (96) did not find an association between the MTHFR 1298 CC

genotype and risk for coronary heart disease. Most other studies suggest that this

polymorphism is benign unless it is combined with the homozygous MTHFR 677C->T

variant (97).









MTHFR and cancer. Various research groups have evaluated the effect of genetic

interactions and cancer. Chen et al. (98) reviewed their previously published studies with

respect to MTHFR and colorectal cancer. They found an inverse association between the

homozygous TT variant and risk of colorectal cancer in two different case-control studies

conducted in the Health Professional's Follow-up Study (99) and Physician's Health

Study (100). This protective effect was diminished by a high alcohol and low methionine

intake. Ma et al. (100) found that there was no protective effect if folate status was low,

but if folate status was adequate, risk for colorectal cancer was reduced by 50% in

individuals with the TT genotype versus the CC genotype. Toffoli et al. (101) reported a

significantly reduced risk of developing proximal colon cancer in Italian subjects with the

TT genotype compared to the CC or CT genotypes. Other studies only have observed a

protective effect in subjects with the TT genotype with high plasma folate concentrations

compared to subjects with the CC or CT genotypes with low plasma folate concentrations

(102,103). Studies also have shown a protective effect of the homozygous TT mutation

related to acute lymphocytic leukemia (ALL) in children and adults (104-106). In

contrast, the homozygous TT variant appears to be associated with an increased risk for

breast (107-110), endometrial (111), gastric (112,113), and esophageal (114) cancers in

certain populations, although folate status of the subjects was not reported in any of the

studies.

Few studies have investigated the influence of the MTHFR 1298A-C

polymorphism on cancer risk. Chen et al. (115) reported that subjects with the

homozygous variant (CC) had a decreased risk for colorectal cancer compared to the

normal genotype (AA), but it was a less substantial independent risk factor than the









MTHFR 677C->T polymorphism. Sharp et al. (109) investigated the role of the MTHFR

1298A->C polymorphism on breast cancer risk and observed a reduced risk in

individuals with the CC genotype compared to the AA genotype. No effect was reported

for the CC genotype on gastric cancer risk (113).

Matthews et al. (59) hypothesized that the reason a potentially harmful

polymorphism is so common in humans is this protective effect against colon cancer and

certain ALLs. The reason for this protective effect is hypothesized to be that under low

folate conditions, the polymorphism decreases the flux of 5,10-methyleneTHF into

5-methylTHF and instead this coenzyme can be used to convert dUMP to dTMP, which

is the rate limiting step in cell synthesis (99). Crott et al. (116,117) tested this hypothesis

in vitro and did not find a significant decrease in uracil misincorporation (116) or

micronuclei formation (117) in human lymphocytes taken from individuals with the TT

genotype compared to those with the CC genotype. The investigators argued that their

results are due to differences between in vitro and in vivo conditions.

Folate, MTHFR, and neural-tube defects (NTDs). Neural tube defects are

congenital abnormalities resulting from the malformation of the brain and/or spinal cord

or from a failure of the skeleton to cover them, resulting in a protrusion (118).

Development of these malformations occurs prior to day 28 of gestation, before many

women know they are pregnant. Most groups studying NTDs generally only include

anencephaly (failure of the brain to develop) or spina bifida (exposure of the spinal cord

due to defective closure of the neural tube), but NTDs also include encephalocele,

craniorachischisis, and iniencephaly. Each year anencephaly or spina bifida occur in 1 in

1,000 pregnancies in the US and roughly 300,000 or more births worldwide (119). This









section will focus on key studies linking folate status, MTHFR, and homocysteine with

NTD risk.

During the 1960's Hibbard suggested that folate deficiency might be a cause for

congenital birth defects (120). This hypothesis was supported by many subsequent

studies. Smithells et al. (121) and Laurence et al. (122) conducted studies to evaluate the

possible protective effect of folic acid supplementation against NTDs in high-risk

mothers who already had an NTD-affected pregnancy. In both studies, the risk for a

subsequent NTD-affected pregnancy was significantly decreased in the supplemented

compared to that of the nonsupplemented mothers. These preliminary findings prompted

a large randomized placebo-controlled intervention trial to determine whether it was folic

acid alone that reduced the risk of NTDs. The Medical Research Council (MRC)

Vitamin Study launched this trial in 1983 and ended it early because they found that 72%

of NTDs were prevented with 4 mg folic acid supplementation daily in women who had a

previous NTD-affected pregnancy (123). Czeizel and Dudas (124) found that

periconceptional folic acid supplementation (800 [g/d) could prevent the first occurrence

of NTDs in a Hungarian population group compared to subjects receiving no folic acid.

Berry and colleagues (125) investigated the effect of periconceptional folic acid

supplementation in Chinese women from areas with high rates of NTDs (northern region

of China) and low rates of NTDs (southern region of China). They observed that 400

[tg/d of folic acid reduced the occurrence of NTDs from 4.8 to 1.0 per 1000 pregnancies

in the northern region of China and 1.0 to 0.6 in the southern region of China. Kirke et

al. (126) evaluated blood samples from 56,049 pregnant women on their first clinic visit.

Of this group 81 had NTD-affected babies and their blood folate concentrations were









compared to 247 control females with normal pregnancy outcomes. The median plasma

vitamin B 12 (pmol/L), serum folate (ng/ml), and red blood cell folate (ng/ml)

concentrations in the mothers with NTD-affected pregnancies compared to the control

mothers were 243 and 296, 3.5 and 4.6, and 269 and 338, respectively. They concluded

that both vitamin B 12 and folate were independent risk factors for NTDs. Using data

from the Kirke et al. (126) study, Daly et al. (127) were able to stratify the serum and red

blood cell folate concentrations into quintiles. They found a greater than 8-fold increase

in risk of NTDs in those women with red blood cell folate concentrations < 150 ng/ml

versus those with red blood cell folate concentrations > 906 ng/ml. They concluded that

NTD risk is reduced as red blood cell folate concentrations increase even in the normal

range. Moore et al. (128) observed that NTD risk declines in a dose-responsive manner

according to supplemental folic acid intake, dietary folate intake, and total folate intake.

They reported that for every additional 500 tg DFE/d consumed, the prevalence of NTDs

decreased by 0.78 cases per 1000 pregnancies. In addition, compared to women in the

lowest quintile of folate intake (0 to 149 DFE/d), the prevalence of NTDs decreased by

34%, 30%, 56%, and 77% among the offspring of women consuming 150 to 399, 400 to

799, 800 to 1199, and > 1200 DFE/d, respectively (128).

This strong link between NTDs and folate has prompted investigations to determine

whether polymorphisms of genes that code for enzymes involved in folate metabolism

have an effect on NTD risk. van der Put et al. (15) were one of the first groups to

associate the MTHFR 677C->T variant with NTD risk. Botto and Yang (4) provided an

extensive review of case-control studies evaluating the association between the

homozygous MTHFR 677C->T variant and risk for spina bifida. Shields et al. (16)









obtained blood samples from a large number of NTD-affected Irish individuals and their

parents to determine if there was any genetic association with NTDs. They concluded

that the homozygous TT genotype in embryos is the most important risk factor for NTD

pregnancy outcome in this population. They also stated that the homozygous TT

genotype might not play such a large role in other populations with more adequate folate

status. Despite the many positive associations found between individuals with the TT

genotype and NTD risk, not all study results support this correlation (129-133). Lucock

et al. (133) hypothesized that another genetic insult of folate metabolism may underlie the

condition. Recently, Rampersaud et al. (134) reported a significantly higher frequency of

the TT genotype in 175 Caucasian NTD-affected subjects compared to controls, but the

investigators suggested that an additional gene may be responsible for an increase in

NTD risk. After careful consideration of published data related to the MTHFR 677C->T

polymorphism and NTD risk, the IOM concluded that only approximately 15% of NTD

risk is attributable to the homozygous variant of MTHFR (27). Molloy et al. (85)

investigated the role of maternal MTHFR TT genotype on the risk of having an NTD-

affected pregnancy. Although they found a greater frequency of mothers of affected

children with the TT genotype than control mothers, they did not find an effect of the TT

genotype on red blood cell folate concentration. Therefore they concluded that maternal

risk factors for NTDs are not explained by the TT genotype for MTHFR. Instead,

maternal folate status may be the most important risk factor.

Few studies have evaluated the association between the 1298A-C polymorphism

and risk of NTDs (4). van der Put et al. (68) did not find a significant risk for NTDs in

individuals with the 1298A-C homozygous or heterozygous variant genotypes;









however, they observed a significantly increased risk for individuals with combined

heterozygosity for the 1298A-C and 677C->T polymorphisms. They concluded that

this combined heterozygosity accounts for a proportion ofNTDs not explained by the TT

genotype and should therefore be considered a risk factor for NTDs. Parle-McDermott et

al. (135) did not find an association between NTDs and the MTHFR 1298A->C

polymorphism. Because most of the MTHFR and NTD studies have focused on living

infants and parents, Isotalo et al. (136) collected fetal tissue samples from spontaneously

aborted and terminated pregnancies to compare genotypes to neonatal umbilical cord

samples for controls. They found all possible genotype combinations including the

677CT/1298CC and 677TT/1298CC genotypes in the fetal tissue, which were absent in

the neonatal control group. The presence of NTDs could not be determined in the fetal

tissue group, but due to the presence of the 677CT/1298CC and 677TT/1298CC

genotypes in fetal tissue, they concluded that the combined MTHFR variants contributed

to decreased fetal viability. Volcik et al. (137) found a higher rate of the homozygous

1298A-C variant in mothers of an NTD-affected pregnancy in a Hispanic population,

although it was not significant. Their results also contrast those of Isotalo et al. (136) in

that they identified all possible genotype combinations in living individuals, including the

677CT/1298CC and 677TT/1298CC genotypes. It is possible that ethnic differences

between the populations studied resulted in different genotypes in the population and,

therefore, differences in results between the two studies.

Homocysteine concentrations also have been found to be higher in mothers of

NTD-affected infants (138,139), but the mechanism is unknown. Based on

epidemiological studies, Eskes et al. (140) and Steegers-Theunissen et al. (141)









hypothesized that homocysteine may have a direct teratogenic effect on the embryo. To

test this hypothesis, Rosenquist et al. (142) treated avian embryos with incremental

physiologic doses of homocysteine and observed the development of NTDs. They

concluded that homocysteine is in fact teratogenic to the embryo. In contrast,

homocysteine exposure to mammalian embryos (rat or mouse) did not increase the

incidence of NTDs (143-145). Greene et al. (145) concluded that elevated homocysteine

concentrations in NTD-affected pregnancies are more likely a marker of abnormal folate

or methionine metabolism. More recently, high homocysteine concentrations have been

found in NTD-affected individuals with the TT genotype. Bjorke-Monsen et al. (130)

found significantly higher homocysteine concentrations in NTD-affected patients versus

controls, but this increase was confined to patients with the CT or TT genotype.

Wenstrom et al. (146) collected amniotic fluid from NTD-affected pregnancies and

controls and found significantly more patients with TT genotypes in the NTD-affected

fluid versus controls. They also found significantly higher amniotic fluid homocysteine

concentrations in fluid from cases compared to controls. High homocysteine

concentrations due to a polymorphism in the gene coding for MTHFR may be a risk

factor for NTDs (139,147).

The association of folate status, MTHFR polymorphisms, and homocysteine

concentrations with NTD risk has been widely studied. Although results have been

conflicting, the general opinion is that each factor plays a role in the occurrence ofNTDs

with folate status being the greatest contributor to risk. MTHFR polymorphisms may

affect folate status and increase homocysteine concentrations resulting in an increased

risk associated with the MTHFR polymorphism. Future studies need to further evaluate









this possibility in order to achieve the goal of preventing the maximal number of folic

acid preventable NTDs.

Folate, MTHFR, and fetal malformations. Although there is significant

evidence indicating a protective effect of folic acid supplementation in the prevention of

NTDs, studies are beginning to link poor folate nutriture with other fetal malformations.

Folate is involved in DNA synthesis and methionine production, both of which are

crucial for normal embryonic development. Low folate status may contribute to

embryonic malformations for the following reasons: decreased cell division,

homocysteine-associated vascular events, impaired maternal-to-fetal folate transfer, or

homocysteine-associated neurotoxicity, as reviewed by Moyers and Bailey (97). An

estimated 120,000 to 150,000 infants with fetal malformations are born in the United

States yearly (97). This section will focus on the possible role folate status and MTHFR

polymorphisms may have on the risk for cleft lip and palate, congenital heart defects, and

Down syndrome.

Cleft lip with or without cleft palate is one of the more common malformations

seen at birth and occurs in every 1/500 to 1/1000 births (148). Early studies documented

the beneficial effect of high dose folic acid supplementation (10 mg) in the reduction of

recurrent cleft lip with or without cleft palate (149). Two to 3 mg of folic acid also have

been shown to significantly protect against cleft lip with or without cleft palate in a

Hungarian population (150). The dose of folic acid used is important because 0.8 mg of

folic acid in a multivitamin was not found to be protective against cleft lip with or

without cleft palate in a Hungarian population (124,151). Hayes et al. (152) did not find

a protective effect of periconceptional folic acid supplementation of mothers against the









risk of oral clefts in infants in a North American population. The conflicting data on

whether or not folic acid supplementation confers a protective effect against oral clefts in

the North American population are reviewed by Moyers and Bailey (97). Wong et al.

(153) determined the prevalence of hyperhomocysteinemia in mothers of infants with

orofacial clefts and concluded that maternal hyperhomocysteinemia is a risk factor for

having a child with an orofacial cleft. Mills (154) listed some possible limitations to the

study by Wong et al. (153), including unexpectedly lower homocysteine concentrations

in the controls compared to cases, differences in vitamin status between cases and

controls, and the vitamin status of the population during the study compared to

pregnancy. Mills (154) also listed some possible explanations including a metabolic

defect in vitamin B6 and folate metabolism. Additional studies are needed to further

define the relationship between maternal homocysteine concentrations and orofacial

clefts.

Because MTHFR is directly involved in folate metabolism, there may be an

association between the homozygous MTHFR 677C->T variant and cleft lip with or

without cleft palate. An Irish group determined the prevalence of the TT genotype in

their population with cleft lip with or without cleft palate. They found that the

homozygous 677C-T variant was significantly more prevalent in an Irish population

with cleft palate only; however, periconceptional maternal vitamin use was not reported

(155). In contrast, Gaspar et al. (156) did not find an increased prevalence of individuals

with the TT genotype in their cleft lip patients with or without cleft palate. The presence

of the TT genotype in mothers of children affected with cleft lip with or without cleft

palate also has been shown to be significantly more prevalent when compared to controls









(157). A subsequent study reported a significantly higher frequency of the TT genotype

in mothers with affected children who were also affected themselves compared to healthy

mothers with affected children (158). In addition, Shotelersuk et al. (159) did not

observe an association between patient genotypes and occurrence of cleft lip or palate,

but did detect a significantly higher frequency of compound heterozygotes (MTHFR

677CT and 1298AC) in mothers of patients. In a series of studies, an American group

determined that infants with the TT genotype are not at increased risk for cleft lip (160)

or cleft palate (161). Neither study indicated an interaction between infant genotype and

maternal multivitamin use on the malformation occurrence. Reanalysis of the data by

Wyszynski et al. (148) revealed significant differences in the risk of clefting between

patients with the TT compared to the CC genotypes in patients whose mothers were not

supplemented. The effect of the TT genotype of patients with supplemented mothers

decreased the risk for orofacial clefts slightly, but not significantly. They concluded that

based on their reanalysis of the data that periconceptional vitamin supplementation may

protect against clefting (148). In another series of studies in an American population, no

association was observed between familial (162) or isolated (163) nonsyndromic cleft lip

and palate and the TT genotype. The relationship between cleft lip with or without cleft

palate and MTHFR needs further study.

Birth defects involving the heart include transposition of the great arteries,

conotruncal heart defects, atrial septal defects, and others. Conotruncal heart defects are

one of the more prevalent defects and occur in every 4/10,000 births. They are disorders

that involve the neural crest cells that are ultimately incorporated into the aortic arch

vessels, truncal outflow tract, and vessel walls, as reviewed by Moyers and Bailey (97).









There are three main observational studies that provide evidence regarding the

relationship between folate and heart defects (150,164,165). Briefly, high doses (2-3 mg)

and low doses (i.e., multivitamin or fortified cereal) of folic acid supplementation

significantly reduced the risk for heart defects in one Hungarian and two American

studies. Boot et al. (166) attributed the development of conotruncal heart defects to the

abnormal differentiation of neural crest cells in the presence of high homocysteine

concentrations in a recently published in vitro study. Although the inverse association

between folate supplementation and reduced risk for heart defects has been the focus of

investigations, the role of MTHFR on conotruncal heart defect risk has been evaluated in

only one study. Storti et al. (167) determined the MTHFR 677 and 1298 genotypes in an

Italian population of affected children and their parents. Although increased odds ratios

for heart defects were detected for different combinations of MTHFR polymorphisms in

mothers and affected children, none of the odds ratios were significant. More studies

evaluating the role of MTHFR polymorphisms on the risk of conotruncal heart defects

are needed.

Down syndrome is a genetic disorder that results from the presence of three copies

of chromosome 21 (trisomy 21). This extra chromosome is a result of abnormal

chromosome segregation during meiosis, also known as meiotic nondisjunction. Ninety-

five percent of Down syndrome cases are maternal, with nondisjunction occurring in the

oocyte before conception. Down syndrome occurs in every 1/600 live births and in 1/150

conceptions. It is estimated that approximately 80% of all trisomy 21 conceptions results

in spontaneous abortion. It is a major public health concern and is the leading cause of

mental retardation (168). James and colleagues (168) hypothesized that altered folate









metabolism in mothers of affected children with the TT genotype may affect their DNA

methylation and result in nondisjunction that leads to Down syndrome. They found that

the presence of the MTHFR 677C->T variant on one or both alleles in these mothers

significantly increased the risk of having a child with Down syndrome. Their results

were supported by Hobbs et al. (169). Other studies have not reported an association

between maternal MTHFR genotype and Down Syndrome (170-172). Differences in the

ethnicity among population groups may have contributed to the differences in results.

Controlled intervention clinical trials will need to be performed before preconceptional

folic acid supplementation is recommended to reduce the risk of Down syndrome.

The possibility of preventing major malformations with folic acid supplementation

has directed the focus of research to the influence of folate status and MTHFR genotype

on fetal malformations. More studies are needed to develop concrete evidence in support

of folic acid supplementation as a preventive measure against fetal malformations.

Homocysteine, MTHFR, and pregnancy outcome. Although the main focus of

folate research involves NTDs and fetal malformations, evidence has been reported for

the role of homocysteine and MTHFR polymorphisms in preeclampsia and early

pregnancy loss. Preeclampsia and recurrent early pregnancy loss are very serious

complications of pregnancy. Some recent findings for preeclampsia and early pregnancy

loss will be discussed in this section.

Preeclampsia is defined as pregnancy-induced proteinuric hypertension with onset

of clinical symptoms beyond 20 wk gestation (173). The cause of preeclampsia is

thought to be due to increased resistance to uterine artery blood flow (174) and has been

associated with elevated homocysteine concentrations (139). L6pez-Quesada et al. (175)









observed a 7.7-fold increased risk for preeclampsia in pregnant women with

hyperhomocysteinemia in the third trimester (> 10.5 [tmol/L) compared to normal

pregnant controls. The role of homocysteine as a risk factor for vascular events has been

well established (176). Because the homozygous MTHFR 677C->T variant is associated

with elevated homocysteine concentrations in individuals with low folate status (3),

investigators have implicated this polymorphism in mothers as a risk factor for

preeclampsia. There have been positive associations (174) and negative associations

(177) found between mothers with the homozygous TT variant and the development of

preeclampsia. Prasmusinto et al. (173) did not associate this polymorphism in mothers or

infants with an increased risk for preeclampsia. Although reported data are not

consistent, possible explanations for the variation include the differences in the

population groups and subsequent variation in the frequency of the MTHFR

polymorphism.

Recurrent early pregnancy loss is another serious problem in pregnancy. Although

still unresolved, the mechanism responsible for this pregnancy complication has been a

topic of intense study. There is evidence to support the hypothesis that abnormal

procoagulant activity may be a causative factor for early pregnancy loss (178). As

mentioned previously, high homocysteine concentrations are associated with various

cardiovascular diseases, including increased prothrombotic tendency (179). Since high

homocysteine concentrations may promote thrombotic events, they also may play a

significant role in early pregnancy loss. Nelen and colleagues (180,181) concluded that

elevated homocysteine concentrations were a risk factor for recurrent early pregnancy

loss in a case-controlled study (180) and a meta-analysis (181). Any factor that may









increase homocysteine concentrations also may increase the risk for early pregnancy loss.

Results from different studies have supported an association between the MTHFR 677

TT genotype and risk for recurrent early pregnancy loss (80,182). Zetterberg et al. (183)

determined the genotype of spontaneously aborted embryos for the 677C->T and

1298A->C MTHFR polymorphisms and found a high prevalence of MTHFR

polymorphic genotypes. Regardless of the population, homozygosity for the 677C-T

variant has been found to be a risk factor for early pregnancy loss (80,182,184). Other

studies have been unable to confirm this association in different populations (184-186).

More studies are needed on the 1298A-C polymorphism to draw any conclusions.

In summary, risks for preeclampsia and recurrent early pregnancy loss are

additional examples of pregnancy complications that may be reduced with folic acid

supplementation. The possible increased need for folate in individuals homozygous for

the MTHFR polymorphisms needs to be substantiated further prior to recommending

doses of folic acid for the prevention of these pregnancy complications.

Methionine Synthase (MS)

Another widely studied enzyme involved in folate metabolism is methionine

synthase (MS), which catalyzes the conversion of homocysteine to methionine. This

enzyme requires 5-methylTHF and cobalamin to function properly. Bacterial MS has

been characterized and studied extensively (187). It consists of 3372 nucleotides and a

molecular weight of 123,640. Escherichia coli MS studies have shown the enzyme to be

a modular protein consisting of four different regions. The module residing on the

N-terminal is the one responsible for binding homocysteine. The second module binds

5-methylTHF, the third binds cobalamin, and the fourth binds SAM (56). Leclerc et al.









(188) were the first to clone human MS cDNA, which shares approximately 58% identity

with Escherichia coli, and to describe specifics about its localization, expression, and

partial characterization. They reported on a gene that codes for a protein containing 3795

base pairs encoding a polypeptide of 1265 amino acids located near the telomeric region

of the long arm on chromosome band 1q43. They also identified an A--G transition at

base pair 2756 that converted an aspartic acid into a glycine in patients with MS activity

deficiency. They hypothesized that this MS deficiency could be associated with mild

hyperhomocysteinemia.

MS and chronic disease. The role of the MS 2756A-G variant in chronic disease

has been investigated. Individuals with the homozygous 2756A-G variant have been

reported to have lower fasting plasma homocysteine concentrations (189-191). Jacques

et al. (192) found no evidence for an association between homocysteine and the MS

2756A-G variant, van der Put et al. (193) sequenced the entire coding region of MS in

eight individuals with hyperhomocysteinemia (four NTD patients and four vascular

disease patients) to determine whether mutations in this gene were involved in

homocysteine-related diseases. They reported that there was no association between the

MS 2756A--G substitution and hyperhomocysteinemia in their population. In fact, they

also detected a slightly lower homocysteine concentration in patients homozygous for the

variant compared to heterozygotes, as described previously. It was hypothesized that

when the strong helix breaker glycine was present in the enzyme instead of the moderate

helix breaker aspartic acid at the position near the vitamin B12 binding site, enzyme

function may be modified. In contrast, Harmon et al. (194) reported that this

polymorphism was associated significantly with homocysteine concentrations in their









population of Irish males, with the AA genotype having higher homocysteine

concentrations. Silaste et al. (190) investigated the effect of two 5 wk dietary

interventions of low and high folate intake to determine whether genetic variations in

enzymes involved in homocysteine metabolism affect the responsiveness of folate status

to naturally occurring food folate. They reported that individuals with the G allele for the

MS variant had a greater decrease in homocysteine concentrations and lower

homocysteine concentrations during the high folate period than individuals with the

normal enzyme. Since hyperhomocysteinemia also is a risk factor for venous

thromboembolism (VTE) an association with the MS 2756A->G variant was

investigated, den Heijer et al. (194) and Salomon et al. (195,196) did not find and

association between the GG variant and VTE. In contrast, Yates and Lucock (197)

reported a protective effect for the presence of the G allele in relationship to risk for

VTE.

Although the association of the MS 2756A->G variant and cancer has been

investigated, no significant correlation was found (198). Swanson et al. (199) developed

a MS knockout mouse model to study the pathophysiology associated with a severe MS

deficiency. Heterozygous knockout mice had a 50-60% decrease in enzyme activity and

slightly elevated plasma homocysteine and methionine concentrations while complete

omission of MS resulted in embryonic lethality, proving that MS activity is essential for

early embryonic development in mice.

MS and NTDs. The possible association between a polymorphism affecting MS

and risk for NTDs has been the focus of investigations. van der Put et al. (193) did not

find an increased prevalence of the MS 2756 GG genotype in NTD cases compared to









controls or in mothers of children with NTDs compared to controls. Unlike the

homozygous MTHFR 677C->T variant, it is hypothesized that the MS 2756 GG

genotype confers some sort of protection against NTDs compared to the normal enzyme.

Christensen et al. (200) did not find the MS 2756 GG genotype in any of their cases, but

10% of their controls had the 2756 GG genotype. In contrast, Zhu et al. (201) observed a

higher frequency of the G allele in NTD cases compared to controls. The GG genotype

was detected in very low frequencies in cases. Doolin et al. (202) found an increased risk

for spina bifida-affected pregnancy in mothers with the GG genotype compared to the

AA genotype. They emphasized the importance of considering maternal and embryonic

genotypes when evaluating risk for spina bifida. The role of the MS 2756A-G

polymorphism on NTD risk is not well defined and needs further investigation.

Methionine Synthase Reductase (MSR)

Associated with MS is MSR, a flavoprotein responsible for the reductive activation

required for the maintenance of MS once cobalamin becomes oxidized over time.

Leclerc et al. (203) isolated the cDNA clone for the human MSR gene and found that it

had three binding sites for FMN, FAD, and NADPH, which are required for the reduction

of MS. The gene consists of 2094 base pairs encoding a polypeptide of 698 amino acids

with a molecular mass of 77.7 kDa. It is localized on human chromosome 5. Wilson et

al. (204) discovered a variation in the gene coding for MSR in homocystinuric patients

with severe MS deficiency. This polymorphism results in a A--G transition at base pair

66, which converts an isoleucine to methionine in the protein. Individuals can have the

homozygous variant (GG), heterozygous variant (AG), or normal (AA) genotype for this

MSR 66A->G variation.









MSR and NTDs. The association between the MSR 66A->G variant and risk for

NTDs has been evaluated. Wilson et al. (205) found that cases and mothers of cases had

almost a 2-fold increase in risk for spina bifida compared to controls, although not

significant. These investigators observed that when low cobalamin status was coupled

with the MSR GG genotype, it conferred an even greater risk for NTDs in the children

and mothers. In addition, a combination of homozygous MTHFR 677C->T and MSR

66A->G variants in both children and mothers conferred a 4-fold increase in risk, a

greater increase than for each individual variant. These results were supported by Zhu et

al. (201) who observed that the G allele was associated with increased risk for spina

bifida. Doolin et al. (202) reported that the risk of having a child with spina bifida

increased with increasing maternal G alleles. These results link the homozygous MSR

66A->G variant with an increased risk of spina bifida, but more studies are needed to

make any concrete associations.

Other Polymorphisms

Although much of folate research focuses on the preceding polymorphisms, other

polymorphisms have been identified that have not received as much attention. Devlin et

al. (206) identified a polymorphism in the gene coding for the conjugase enzyme located

in the intestine that results in 53% less activity than the normal enzyme. This enzyme

termed GCPII is responsible for cleaving polyglutamates into monoglutamates.

Glutamate carboxypeptidase II can have a C--T transition at base pair 1561, which

replaces histidine with tyrosine at codon 475, codes for a 750 amino acid polypeptide,

and is located on chromosome 11. Devlin et al. (206) associated the 1561C-T transition

in GCPII with low serum folate concentrations and higher homocysteine concentrations









in a healthy English population. More studies are needed to corroborate these findings

because dietary intakes were not assessed at the time of sample collection, and recent

intakes of folate, vitamin B6, vitamin B 12 can affect folate and homocysteine

concentrations. Any supplements taken by the subjects containing any of these B

vitamins also could have had an effect. Vargas-Martinez et al. (207) concluded that the

polymorphism is not associated with plasma folate or homocysteine concentrations after

determining the GCPII 1516C-T genotype in subjects from the Framingham Offspring

Study. They took many factors into account known to affect folate and homocysteine

concentrations and, after adjusting for these factors, still did not find an association. A

possible explanation for variations between these studies are the factors accounted for by

Vargas-Martinez et al. (207). Vargas-Martinez et al. (207) concluded that this

polymorphism has no effect on folate and homocysteine concentrations when

confounding factors are taken into account. Fodinger et al. (208) observed that GCPII

was a predictor of red blood cell folate but not homocysteine concentrations in chronic

dialysis patients.

Other polymorphisms that may have an impact on folate status are the thymidylate

synthase promoter region polymorphisms (209,210) and folate receptor polymorphisms

(37,211). Multiple studies have illustrated the complexity of the problem of how a

genetic variation associated with different diseases and birth defects can be modified by

nutrient intake. A perfect example of this comes from the previously discussed study by

Jacques et al. (7), which reported that homocysteine concentrations were higher in

individuals with the homozygous MTHFR 677C->T variant only when plasma folate

concentration was below the median. There was no difference in homocysteine between









individuals with the TT and CC genotype with plasma folate concentrations at or above

the median. Adequate folate status also can provide protection against certain cancers in

individuals with the TT genotype, while inadequate folate status can add to the risk for

these cancers (100).

DNA Stability

For years scientists have known that exposure to chemical mutagens and

carcinogens can cause gene mutations and chromosome damage. Only recently has there

been more and more evidence linking dietary factors to similar damage (212). Currently,

a large body of evidence supports the conclusion that folate deficiency has a negative

impact on DNA stability. The proposed mechanisms by which folate deficiency impairs

DNA structure include: DNA methylation, uracil misincorporation, DNA strand breaks,

and micronuclei formation (11,213-215). This section will summarize research findings

related to each mechanism to provide a basis for understanding the links between folate

status and DNA stability.

DNA Methylation

Until recently, not much was known about gene regulation. It is now recognized

that it is not just the primary sequence of the DNA that determines whether or not a gene

is expressed. Gene transcription depends on modifications referred to as epigenetics,

which is defined as any DNA modification that regulates gene activity without changing

the primary DNA sequence and can persist through one or more generations (216). It is

these epigenetic modifications that allow cells to adapt to external stimuli from the

environment and from regulatory molecules within the body without having to change the

primary DNA sequence (216). DNA methylation is one epigenetic modification that is

critical for normal development of cells and organs (217). Li et al. (218) reported that a









mutation of a key methylating enzyme, Dnmtl, produced a recessive, lethal phenotype.

A few years later, Okano et al. (219) confirmed these findings in a study that targeted two

other DNA methyltransferases, Dnmt3a and Dnmt3b. Deletion of Dnmt3a produced

mice that appeared normal at birth but failed to grow normally and died after 4 wk. In

contrast, deletion of Dnmt3b did not produce any viable mice at birth. Deletion of both

DNA methyltransferases produced an embryonic lethal phenotype. These series of

studies illustrate the importance of methylation for the normal development of mammals.

Before the function of DNA methylation can be discussed, the specifics of

methylation will be reviewed briefly. As reviewed by Costello and Plass (217), the

majority of 5'-methylcytosine is present in cytosine-guanine (CpG) dinucleotides.

Clusters of CpGs also exist and are referred to as CpG islands. Gardiner-Garden and

Frommer (220) originally defined these CpG islands as regions of DNA from 200 base

pairs to several kilobases in length with a CpG frequency approximately five times

greater than the whole genome. They also determined that these CpG islands comprise 1

to 2% of the genome. It is estimated that 3 to 6% of total mammalian cytosine bases are

methylated and that 70% of mammalian CpG dinucleotides are methylated (8). The exact

mechanism that determines the pattern of DNA methylation is unknown. Methylation

assays using polymerase chain reaction amplification have shown that mouse sperm

DNA is primarily methylated at all non-CpG island sites located throughout the genome

in contrast to mouse oocyte DNA, which is primarily unmethylated at these same

nonCpG island sites (221). Early in the life of the embryo, the CpG sites of germ and

somatic cells are demethylated and then de novo methylation reestablishes a methylation

pattern, although differently in germ cells versus somatic cells (221). Different









mechanisms can change DNA methylation patterns during different developmental

stages. The spontaneous deamination of a methylated cytosine produces thymine, which

can be replicated and result in a TG base pair mismatch and subsequent TA transition.

This transition is more difficult for the cell to repair (222) and can result in the loss of

heritable methylation patterns (8).

There are three DNA methyltransferases that have been discovered which transfer a

methyl group from SAM to the cytosine in CpG dinucleotides: Dnmtl, Dnmt3a and

Dnmt3b (223). These enzymes are necessary for embryonic development (218,219) and

possess de novo methylation capabilities (i.e., methylation of a CpG sequence opposite an

unmethylated CpG sequences) (224). Dnmtl has a higher affinity for hemimethylated

DNA and therefore works mostly as a maintenance enzyme (225). The exact mechanism

as to how the 5-position of cytosine becomes methylated is still unknown. It is

hypothesized that cytosine is everted from the DNA helix and inserted deep within the

active site of the DNA methyltransferases where a methyl group is transferred (226). The

mechanism of DNA methylation involves the methylated parent strand as a guide. The

daughter DNA strand is methylated by one of the maintenance DNA methyltransferases

shortly after replication, resulting in the exact methylation pattern present in the parent

strand. DNA methylation patterns are preserved throughout many rounds of replication

by DNA methyltransferases (8).

DNA methylation has many functions. Stabilization of DNA by preventing

cleavage by nucleases is one proposed function for DNA (8). The primary function,

however, is as a gene silencer. Genes make up a small portion of the genome, while the

rest is made up of introns, repetitive elements, and potentially active transposable









elements. In order for genes to be successfully expressed, all of the noncoding DNA

needs to be silenced. Mammals appear to have evolved to use methylation as a

mechanism to silence this DNA. The post-transcriptional addition of a methyl group to

the 5-position of cytosine alters the DNA-protein interactions, which in turn keeps the

DNA from being transcribed (9). CpG islands often contain the promoter regions near

the 5' end of genes. If these promoter regions are unmethylated, this denotes active

transcription (227). If they are methylated, transcription is suppressed. As technology

has improved, scientists have determined that CpGs located within these CpG islands are

mostly unmethylated while CpGs located outside these islands are generally methylated

(217). Costello and Plass (217) suggested that these patterns of methylation may act to

separate the genome into transcriptionally active and inactive areas. These specific

methylation patterns are upheld through DNA replication in order to promote and

maintain the transcription of specific genes (228). Methylation within these promoter

regions can stop transcription. Cooney et al. (229) reported that dietary methyl group

supplementation of rats has a significant effect on DNA methylation and their subsequent

methylation-dependent phenotype by successfully changing the coat color of the yellow

agouti mouse.

Kass et al. (228) reviewed how methylation in the promoter region represses

transcription. The most obvious mechansim would be to prevent the transcription factors

and proteins from binding to the DNA. However, this cannot be the only explanation

because there is transcription machinery that will bind to DNA despite its methylation.

Jones et al. (230) reported that the binding of the methyl CpG binding complex 2 to

methylated promoter regions recruits transcriptional repressors. These complexes contain









histone deacetylases (HDAC 1, HDAC2), which function to deacetylate lysine residues in

histone tails, which are associated with DNA and result in the compaction of chromatin.

Once the chromatin becomes compacted, it is transcriptionally inactive. Therefore, DNA

methylation also has a repressive effect on chromatin, serves to inactivate one of the two

X chromosomes in females during development, and determines the expression of

imprinted genes (217). Similar machinery is used to establish methylation patterns.

Robertson (231) reviewed two models for how DNA methylation patterns may be

established in somatic cells. The first model incorporates the use of HDACs, ATP-

dependent remodeling complexes, and DNA methyltransferases. Histones destined for

silencing are first deacetylated by HDACs. The chromatin remodeling enzymes can now

move the nucleosomes that are wrapped by DNA from side to side in an ATP-dependent

manner to allow the DNA methyltransferases access to its target DNA sites. These

enzymes also may create a specific area on the DNA that is recognized by the DNA

methyltransferases. Once DNA is methylated, the methyl-CpG binding proteins bind the

methylated cytosines and further repress transcription as described above. The second

model involves the maintenance methylase (DNMT1) and the retinoblastoma protein, Rb,

which is a protein involved in transcriptional regulation of chromatin (231). In resting

cells, Rb is associated with DNMT1 and inhibits its catalytic activity to prevent any

aberrant methylation of the genome. Early in cell division (S phase) both proteins

colocalize with the replication foci. In late S phase Rb is no longer colocalized with the

replication foci and, instead, HDAC colocalizes with DNMT1, which is now active. It is

possible that Rb departs from the replication foci when hypermethylated regions are

replicated (231).









The main methyl donor used by the DNA methyltransferases is SAM. DNA

methylation is dependent on the availability of SAM; therefore, anything that affects the

supply of SAM may have an effect on DNA methylation. In humans, the main dietary

sources of methyl groups that are transferred to SAM are folate, choline, and methionine

(232). SAM is the main methyl donor in over 100 reactions. The methyl group

requirement normally exceeds the supply available in food, but the difference is usually

made up by the synthesis of methyl groups utilizing folate coenzymes (213). Most of the

5-methylTHF is regenerated through the one-carbon cycle illustrated in Figure 2-2, but a

small amount is lost through urinary excretion, skin, bile, and catabolism. If this folate is

not replaced, it could decrease the methylation cycle resulting in lower amounts of SAM

available for methylation and increased homocysteine concentrations (233).

Early studies investigating the role of folate in methylation used rats as a

mammalian model. Balaghi and Wagner (213) found that after 4 wk of feeding a folate

deficient diet to rats, methylation of hepatic DNA was significantly reduced in folate

depleted rats compared to controls. Alonso-Aperte and Varela-Moreiras (234) observed

that administration of the folate antagonist methotrexate in rats produced a folate

deficiency compared to controls and was associated with significant hypomethylation of

brain DNA. Kim et al. (235) investigated global and protooncogene specific (c-myc)

DNA methylation. As a model for conditions proceeding colorectal neoplasia in rats and

humans, the effect of moderate folate depletion over a longer period of time was

evaluated. After 15 and 24 wk, the folate deplete rats had a significantly lower plasma

folate concentration than the control rats. There was no significant difference in global

methylation after 15 or 24 wk. They also did not find a significant difference in









methylation of c-myc between the two groups after 15 or 24 wk of folate depletion. It

was hypothesized that the SAM/SAH ratios were not low enough, the strain of rat used

was resistant to hypomethylation, or the betaine pathway compensated for any folate loss

and prevented significant hypomethylation. Kim et al. (214) investigated the effect of a

more severe folate deficiency resulting from antibiotic treatment on global DNA

methylation and strand breaks and methylation and strand breaks within a specific

sequence of the p53 tumor suppressor gene. Folate deficient rats at 4 and 6 wk had

significantly lower plasma folate concentrations than the controls and significantly more

strand breaks within the p53 gene than control rats. They also had reduced DNA

methylation of this gene that obtained significance at 6 wk. The investigators concluded

that a dietary folate deficiency could have a negative effect within critical regions of the

p53 tumor suppressor gene. Deficient rats fed for 6 wk also had significantly higher

global strand break accumulation than controls and rats fed for 4 wk, indicating a time

dependence of strand breaks. They did not find a significant difference between global

methylation between the two groups at 4 or 6 wk, strengthening their previous

observations.

Although the preliminary rat studies were important in making a connection

between folate status and DNA methylation, the links to clinical implications in humans

were unclear. Studies correlating low folate status in humans with increases in

micronuclei formation and uracil misincorporation were reported, which supported using

them as new folate status indicators in addition to the traditional blood folate values

(11,236). Jacob et al. (237) evaluated DNA methylation as a potential new biomarker for

folate status. This study was designed to assess the correlation between impaired folate









status and DNA methylation in humans. Ten, healthy postmenopausal women lived in a

metabolic unit for the 13 wk of this study and consumed a low folate diet and varying

amounts of synthetic folic acid to provide intakes ranging from 56 to 516 tg/d during this

depletion-repletion study. Days 6 to 41 were designed to provide a moderately-deplete

folate diet to evaluate the effect of low folate intakes not associated with overt clinical

signs of deficiency. Mean plasma folate concentrations dropped significantly from

baseline to post-depletion. Lymphocyte DNA hypomethylation was determined using a

methyl acceptance assay, whereby acceptance of [3H]methyl groups is inversely

associated with methylation. The investigators observed that the marginal folate

deficiency induced in these postmenopausal women was associated with reduced DNA

methylation, which was reversed with folate supplementation. This was the first study to

show that folate intake affects DNA methylation in humans and reflects the results seen

in the aforementioned rat studies (213,214). Rampersaud et al. (238) investigated the

effects of controlled folate intake on global genomic DNA methylation in leukocytes of

elderly women. Thirty-three healthy, elderly women consumed a depletion-repletion diet

consisting of a low folate diet (118 [tg/d) for 7 wk and repletion with 200 or 415 [tg/d for

7 wk. Blood samples were taken weekly and leukocyte DNA methylation was

determined using the methyl acceptance assay (213). Moderate folate depletion in

elderly women was severe enough to be associated with increased [3H]methyl

incorporation in vitro, which reflects decreased methylation in vivo. Decreased

methylation was evident in these women at wk 7, when they were also found to have the

lowest folate status. DNA methylation did not increase after 7 wk of repletion with 200

or 415 tg folate/d, which suggests that the repletion period may not have been not long









enough to increase the methylation. The results from this study show that low folate

status may significantly reduce DNA methylation and support the use of DNA

methylation as a functional folate status indicator. The investigators stressed that results

related to DNA methylation status within specific cells may not relate to whole body

methylation.

Epidemiological data also support a correlation between folate status and DNA

methylation. Fowler et al. (239) found a significant, inverse correlation between cervical

tissue folate concentration and DNA methylation and serum folate concentration and

DNA methylation. An inverse correlation between folate status and DNA methylation

has not been found by all investigators. Fenech et al. (240) measured the folate and

lymphocyte DNA methylation status of young Australian men and women. Volunteers

were required to consume their normal diets and were assigned to eat a bowl of fortified

or unfortified bran cereal once/d plus a placebo or a vitamin supplement for 24 wk. The

folate group received 2700 ptg folic acid/d and 27 ptg vitamin B12/d (cereal + tablet),

while the placebo group received no folic acid. No significant differences in DNA

methylation were detected after treatment relative to baseline for either group or any

correlations between DNA methylation and folate status. This is not surprising because

this was not a folate depletion protocol, and all volunteers had normal status throughout

the study. It is only when folate status is impaired that the availability of methyl groups

decreases and there is a decrease in genomic methylation (233).

The effect of folate on methylation and subsequent transcription was evaluated in

cell culture experiments. Jhaveri et al. (241) performed experiments with human

nasopharyngeal epidermoid carcinoma KB cells to determine whether folate deplete









media affects the transcription of genes. They found that eight genes responded to a

variation of folate concentrations. Three genes were upregulated and five genes were

downregulated in cells grown in folate deficienct media. H-cadherin, a protein involved

with cell adhesion molecules, was one of the genes down regulated in folate deficient

media. This down regulation was associated with hypermethylation of the CpG island

that contained the promoter region. These data indicated that decreased folate positively

and negatively influences the expression of certain genes, so that folate deficiency affects

specific genes rather than global methylation. Regardless of whether experiments were

in vitro or in vivo, results support a role for folate in DNA methylation and transcription.

Methionine is an essential amino acid that is the immediate precursor to SAM in

the methionine cycle and is required for protein synthesis (242). It is well known that

methyl deficient diets (i.e., choline and methionine deficient) can cause global

hypomethylation in rats (243,244) and that it can occur within 1 wk of starting the diet

(244,245). Choline deficient diets also have been reported to induce significant

hypomethylation of brain DNA compared to controls (234).

SAM/SAH regulation of DNA methylation. Once SAM is used as a substrate for

methylation, SAH is formed within the active site of the methyltransferase enzyme. Most

methyltransferases have a higher binding affinity for SAH than SAM; therefore, excess

SAH can result in strong product inhibition of these methyltransferases, which may lead

to decreased DNA methylation (246). In a review by James and colleagues (247), three

defense mechanisms against toxic SAH concentrations are discussed. They suggest that

SAH can be bound to proteins, exported into the plasma, or hydrolyzed by SAH

hydrolase, which degrades SAH into homocysteine and adenosine.









In order for methyltransferases to work efficiently, SAH concentrations need to be

regulated intracellularly, which is primarily accomplished by SAH hydrolase.

Homocysteine can then be remethylated to form methionine or shuttled down the

transulfuration pathway (248). Previous studies have implicated decreased availability of

SAM as a limiting cofactor for methyltransferases, and therefore, decreased cellular

methylation (249).

The SAM/SAH ratio has been used to predict reduced cellular methylation, with

SAM as the main effector (249). Yi et al. (248) estimated whether SAM or SAH had a

greater impact on global DNA methylation in healthy, adult women. They found that an

increase in homocysteine concentration in these women correlated with a significant

increase in SAH concentration but had no relationship with SAM. A strong, correlation

between homocysteine concentration and the SAM/SAH ratio was detected, and this

decreased ratio was associated with an increased SAH concentration rather than

decreased SAM concentration. Lymphocyte DNA hypomethylation increased

significantly with increasing concentrations of SAH, but had no relationship with SAM.

This study was the first to show that moderate elevations in homocysteine concentration

are in fact associated with increases in SAH and decreases in lymphocyte DNA

methylation. These investigators suggested that there might be another mechanism of

homocysteine pathogenicity via SAH inhibition of DNA methyltransferases. Instead of

limiting DNA methyltransferase activity and subsequently causing reduced DNA

methylation, it was hypothesized that low SAM concentrations may instead affect DNA

methylation by decreasing thymidine and purine synthesis via increased activity of

MTHFR (248).









A combined genetic and dietary approach in a mouse model was used by Caudill et

al. (250) to investigate the effect of changes in plasma homocysteine and intracellular

SAM, SAH, and the SAM/SAH ratio on global DNA methylation in different tissues.

Mice were normal or heterozygous for CBS deficiency and were randomized into a

methyl-deficient diet group or the control group for 24 wk. The combined results for

different tissues indicated that an increase in SAH alone or in conjunction with a decrease

in SAM was associated with a decrease in DNA methylation. A decrease in SAM alone

was not sufficient to affect DNA methylation. In addition, a decrease in the SAM/SAH

ratio was associated with reduced DNA methylation only when associated with an

increase in SAH. The negative effects increased SAH concentrations and decreased

DNA methylation may have on altered gene expression and chromatin formation have

been reviewed (251,252). James et al. (247) hypothesized that SAH-mediated reduced

DNA methylation as a result of increased homocysteine concentrations may increase

DNA damage from homocysteine-induced free radicals. Further research is needed to

determine whether this is a possibility.

Genetic polymorphisms and methylation. Individuals homozygous for the

MTHFR 677C->T variant have diminished methylation capabilities. It has been shown

that the distribution of folate forms was altered in the red blood cells of individuals with

the TT genotype (6). These individuals had formylated folates in addition to methylated

folates in contrast to individuals with the CC genotype who exclusively had methylated

folates. Genotype did not affect total red blood cell folate content. This finding

prompted Stern et al. (12) to use a methyl acceptance assay to evaluate whether this

decrease in methylated folates affected DNA methylation in these individuals. They









found that individuals with the TT genotype had significantly decreased DNA

methylation compared to individuals with the CC genotype, and that this methylation was

directly correlated with red blood cell folate concentration. This was the first study to

report that the TT genotype may be associated with epigenetic alterations. These findings

were supported by a second observational study by the same research group in a larger

population. Friso et al. (13) evaluated the effect of folate status and MTHFR 677C->T

TT genotype on DNA methylation in an Italian population. They directly measured

methylated cytosines using a liquid chromatography/mass spectrometry method. DNA

methylation in individuals with the TT genotype was approximately 50% that of

individuals with the CC genotype when plasma folate concentration was below the

median for this population group. There was no significant difference in methylation

between the individuals with the TT and CC genotypes when plasma folate concentration

was above the median. An inverse relationship between DNA methylation and

homocysteine concentration also was observed in this study. An important limitation of

this study is that dietary folate intake was not controlled. These results have been

corroborated using a MTHFR knockout mouse model. Chen et al. (95) found lower

global DNA methylation in heterozygous and homozygous knockout mice compared to

control mice. This body of evidence supports the conclusion that the homozygous

MTHFR 677C->T variant is associated with reduced DNA methylation when folate

status is low.

Methylation and cancer. Recent evidence has shown that cancer is a process that

is modified by DNA mutations and epigenetic mechanisms (253). In normal cells, CpG

islands are hypomethylated and located in the promoter regions of 40-50% of genes.









Many cancer genes are being discovered that are hypermethylated in the promoter region

(253). The role of hypermethylation of DNA in cancer involves the silencing of genes by

hypermethylation of promoter regions. Although the information on hypermethylation is

extensive, a review by Ehrlich (254) argues that there has been inadequate attention given

to global hypomethylation of DNA in cancer. The evidence for global hypomethylation

in carcinogenesis is considerable. Rats fed methyl deficient diets had hypomethylated

liver DNA that was associated with increased mRNAs for protooncogenes (255). After 1

mo of consumption of a diet adequate in methyl groups, a reversal in methylation and

protooncogene expression was observed (255). Rats fed methyl deficient diets also had

hypomethylated p53 tumor suppressor genes (10,214), which could enhance tumor

production. Sibani et al. (256) reported that reduced DNA methylation in preneoplastic

intestinal cells was directly associated with tumor multiplicity, which was increased

under low folate conditions. It was hypothesized that if hypermethylation and

hypomethylation of DNA can be altered, there may be an opportunity to influence tumor

production.

Caution is warranted when interpreting data regarding hypermethylation and

hypomethylation of DNA. These terms denote more or less methylation of DNA relative

to some standard. When studying cancer, normal tissue is the standard (254). A different

standard has to be developed for each tissue studied because methylation is species and

tissue-specific. Cell types also have to be considered because tissues are comprised of a

mixture of cells (254). As reviewed by Ehrlich (254), DNA hypomethylation has been

found in leukemia, liver and prostate, and cervical cancer, and DNA hypermethylation

has been found in colon, kidney, esophageal, and pancreatic cancer.









DNA Strand Breaks

Another measure of genomic stability is the number of strand breaks in the DNA.

Strand breaks were initially found to be associated with reduced DNA methylation

resulting from a methyl group deficiency. Pogribny et al. (10) fed rats a diet deficient in

the methyl donors methionine, choline, and folate and found that genomic strand breaks

increased with increasing DNA hypomethylation. They also showed that increased

methylation protected the DNA from enzyme-induced strand breaks. These results were

supported by a subsequent study from the same group (214) who found increased DNA

strand breaks with prolonged folate deficiency. They hypothesized that DNA

hypomethylation secondary to folate deficiency may induce strand breaks by changing

the conformation of the chromatin and protein-protein interactions. These changes make

the DNA more susceptible to DNA-damaging agents or endonucleases (214).

Although methyl group deficiency was the first explanation for strand breaks,

recent research also has implicated uracil misincorporation as a source for strand breaks.

As seen in Figure 2-2, the enzyme thymidylate synthase requires 5,10-methyleneTHF as

a coenzyme for the conversion of dUMP to dTMP for DNA synthesis and repair. A

folate deficiency will limit the amount of coenzyme available for this conversion and

cause a buildup of dUMP with subsequent dUTP misincorporation into DNA by DNA

polymerase resulting in a UA base pair (11). Uracil is excised from DNA by uracil

deglycosylase, which can cause breaks in the DNA if there is insufficient dTMP available

for repair. If two breaks are located within 12 base pairs from each other on the DNA it

could result in a double-stranded DNA break (257). Another mechanism of uracil

misincorporation involves spontaneous deamination of nonmethylated cytosine residues

to uracil, which can result in a UG base pair and a cytosine to thymine transition mutation









if the uracil is not excised before replication. Uracil misincorporation is not a problem

unless the capacity for uracil excision is exceeded (258).

Uracil excision is an important DNA repair mechanism involving different

enzymes. Uracil DNA glycosylase recognizes a conformational change produced by the

misincorporated uracil and binds to the DNA. The uracil is flipped out of the double

helix into the active site of the enzyme and cleaved. This leaves an apyrimidinic (AP)

site, which is cleaved by an AP endonuclease. Deoxyribophosphodiesterase removes the

5'-phosphate group, DNA polymerase inserts the correct nucleotide, and DNA ligase

seals the gaps in the DNA (258).

Cell culture studies have confirmed that folate deficiency causes increased uracil

misincorporation and DNA strand breakage. Duthie and Hawdon (259) used single cell

gel electrophoresis with uracil deglycosylase on human lymphocytes to determine DNA

damage in cells in a variety of experiments. Stimulation of cells in folate deficient media

resulted in no growth compared to cells in folate-rich media that grew 6-fold in 8 d. The

effect of folate deficiency was graded for cell growth, with cells growing normally in 100

ng/ml, poorly in 10 ng/ml, and not at all in 1 ng/ml. Uracil misincorporation and DNA

strand breakage was significantly greater in cells grown in a folate deficient medium

compared to cells grown with folic acid. Finally, cells grown in folate deficient media

for 5 d were unable to repair oxidant-induced DNA damage as efficiently as controls.

Melnyk et al. (260) used Chinese hampster ovary cells to determine the effect of folate on

DNA damage. Cell growth was inhibited in folate deficient media. They expanded the

experiments and measured intracellular nucleotide concentrations, finding significantly

increased dUTP and decreased dTTP concentrations and an increased dUTP/dTTP ratio









during folate deprivation. Uracil misincorporation, AP sites, and DNA strand breaks all

increased with increasing duration of folate depletion. These cell culture studies

supported an association between poor folate status and increased DNA damage.

Pogribny et al. (258) found significantly increased uracil misincorporation and AP sites

after just 3 wk of feeding rats a methyl deficient diet low in methionine, choline and

folate. DNA strand breaks also were increased but progressed more slowly with

significant results appearing after 9 wk of feeding and continued to increase with

prolonged feeding. Duthie et al. (261) separated their rats into a control group, a folate

deficient group, a choline/methionine deficient group, and a folate/choline/methionine

deficient group and determined DNA integrity at 4, 8, and 10 wk. Lymphocyte DNA

strand breaks were higher in all groups compared to controls after just 4 wk. The greatest

amount of strand breakage was seen in the folate/choline/methionine deficient group.

The choline/methionine deficient group had more DNA strand breaks than the folate

deficient group at 4 wk, indicating that methyl group deficiency has a greater effect on

strand breaks than folate deficiency alone. Uracil misincorporation was highest in the

folate deficient group, with no uracil misincorporation seen in the choline/methionine

group, indicating the specificity of this biomarker for folate deficiency. The investigators

concluded that DNA strand breaks are more affected by methyl-donor status than folate

status, and that uracil misincorporation is more affected by folate status than methyl-

donor status.

The extent of uracil misincorporation in humans has been evaluated. After

separation of subjects in normal and deficient folate groups at baseline, Blount et al. (11)

supplemented all subjects with 5 mg of folic acid for 8 wk. DNA uracil concentrations









were reduced in all subjects after supplementation with folic acid. The greatest decrease

was seen in subjects with the lowest folate status at baseline. Folate deficient subjects

had the greatest chromosome breakage as measured by micronucleated reticulocytes and

erythrocytes at baseline compared with controls. Folate supplementation reduced

chromosome breakage in folate deficient subjects but had no effect on folate replete

controls.

Independent of whether strand breaks are formed from DNA hypomethylation, low

methyl-donor status, or uracil misincorporation, they lead to chromosome damage and

are associated with increased risk for cancer. Adequate folate intake is essential for the

prevention of chromosome damage and may reduce cancer risk.

Micronuclei Formation

DNA strand breaks ultimately lead to micronuclei, which are formed by the loss of

whole chromosomes or portions of chromosomes from daughter nuclei at mitosis and

form small, independent nuclei within the cytoplasm of a cell (262). Although

micronuclei appear in almost every cell type, cells from the hematopoietic system are

most widely used because of their ease of collection (262). Micronuclei formation is a

marker of genetic damage that is used to assess different risk factors for their genotoxic

capabilities. The role of folic acid in initiating the formation of micronucei was first

investigated in early in vitro experiments that were able to induce micronuclei formation

with folic acid deficient media (263). They also found dose-dependent protection from

micronuclei formation with increasing folic acid concentrations in the media of cultured

lymphocytes (263). Everson et al. (264) supported these findings by reviewing a case

study of a subject with an elevated frequency of micronucleated cells, which returned to

normal after folic acid supplementation.









Many subsequent studies have substantiated the role of folic acid in micronucleated

cells (215,236,265,266). Differences have been found between men and women, with

folate having an effect on micronuclei formation in women, but not in men (267).

Fenech et al. (268) evaluated the association between folate status and micronuclei

formation in older men and found a significant increase in micronuclei formation in men

with a folate deficiency without any clinical manifestations. They also conducted an

intervention study in older men to determine whether folic acid supplements could affect

the genetic damage rate in lymphocytes. They did not find a decrease in micronuclei

formation with folic acid intakes of up to 10 times the DRI for 16 wk (268).

Homocysteine also has been reported to be an independent risk factor for micronuclei

formation, although more studies are needed to confirm these results (269,270).

Quantifying micronuclei is an easy and fast method to measure genetic damage. It

is an assay that can be used in vitro and in vivo to evaluate the influence of factors like

environmental toxins and radiation, on cell carcinogenicity (262). This method,

combined with other quantitative and qualitative techniques, can provide good insight

relative to the etiology of cancer in mammals.

Dietary Reference Intakes (DRIs)

The DRIs are recommendations for intakes of specific nutrients and include the

Recommended Dietary Allowances (RDA), Estimated Average Requirements (EAR),

Tolerable Upper Limits (UL), and Adequate Intakes (AI). The most recent

recommendations were published in 1998 by the National Academy of Science IOM

(27). Previous recommendations were established to prevent clinical deficiencies of each

nutrient. The basis for the 1998 IOM DRI recommendations changed from preventing

clinical deficiencies to ensuring optimum health (27). The EAR for folate for women









between the ages of 19 to 50 is 320 [g DFE/d and is calculated as the amount of folate

needed to meet the requirements of 50% of this population. The RDA for this group,

which is based on the EAR, is 400 [tg DFE/d and is set to cover the needs of 97 to 98% of

these individuals (27). The IOM also recommends that all women of childbearing age

consume 400 tg/d of folic acid in the form of supplements and/or fortified foods in

addition to the daily diet to reduce the risk of NTDs. An UL of 1,000 tg/d of synthetic

folic acid was set for this age to avoid a delay in the diagnosis of a vitamin B12

deficiency that could otherwise lead to neurological damage from a masked vitamin B12

deficiency (27).

Folate plays an important role in genomic stability. Fenech (271) argues that

because DNA damage increases the risk for degenerative diseases and aging, the basis of

dietary intake recommendations should shift to defining optimal intakes of nutrients to

prevent DNA damage. Vitamin B12 and folate are the two nutrients with the greatest

effects on DNA stability (239). Fenech et al. (240) performed a dietary intervention

study to evaluate DNA damage and concluded that an intake of 700 tg folic acid/d and 7

[tg vitamin B12/d was sufficient to minimize chromosome damage, which are amounts

that greatly exceed the current DRIs for these nutrients. The results of different

intervention studies in humans that indicate DNA damage is minimized when red blood

cell folate concentration is > 700 nmol/L were evaluated in a review by Fenech (272).

This red blood cell folate concentration is associated with folate intakes greater than the

current DRIs. This review also lists different intakes recommended by various

investigators to minimize genomic instability that range from 228 to 10,000 [tg folic









acid/d. Fenech (271) concluded that there is a need for an international collaboration to

establish DRIs for nutrients to enhance genomic stability.

Folate Status in Women of Reproductive Age

Effect of Fortification

Earlier research associating the intake of folic acid with NTD risk reduction

prompted the Public Health Service in 1992 to recommend that all women of

childbearing age consume 400 tg/d of folic acid (273). Since then there have been

monumental efforts to help women in this age group meet this recommendation,

including the mandate by the FDA in the US in 1996 to fortify all enriched cereal grain

products with folic acid.

The fortification of the food supply with folic acid has had a positive effect on the

folate status of the population as a whole. In order to assess the benefit of fortification,

the CDC (274) compared serum and red blood cell folate concentration for women of

childbearing age who participated in the 1999 National Health and Nutrition Examination

Survey (NHANES 1999) to women of childbearing age who participated in the Third

National Health and Nutrition Examination Survey (NHANES III, 1988-1994). The

serum and red blood cell folate concentrations from the NHANES III were 6.3 and 181

ng/ml and increased significantly to 16.2 and 663 ng/ml, respectively, in NHANES 1999.

Choumenkovitch et al. (275) also evaluated the effect of fortification in a cross-sectional

study with participants in the Framingham Offspring Cohort. They compared the mean

red blood cell folate concentration of subjects before fortification to red blood cell folate

concentration of subjects after fortification and observed a 38% increase post

fortification.









Fortification also has had a positive effect clinically. Honein et al. (276) reported a

19% reduction in NTD prevalence in the United States, although they concede that other

factors also may have contributed to this decrease. The FDA predicted that fortification

would increase the average folate consumption by 100 [tg/d. Studies by Choumenkovitch

et al. (277) and Quinlivan and Gregory (278) determined that intakes are approximately

twice the predicted level.

Effect of Ethnicity

Ethnicity may affect the folate status and homocysteine concentration of certain

populations. Specifically, certain Hispanic groups may be the most affected. The Office

of Management and Budget's revised minimum standards for reporting race, which have

been adopted by the NIH, define two ethnic categories: Hispanic and non-Hispanic (279).

Hispanic ethnic groups can be further divided into the subcategories Cuban, Mexican,

Puerto Rican, South or Central American, or other Spanish culture or origin, regardless of

race (279). Ford and Bowman (280) published data from NHANES III on serum and red

blood cell folate concentration in non-Hispanic white, non-Hispanic black, and Mexican

American men and women. Both Mexican American men and women had significantly

lower serum and red blood cell folate concentrations compared to non-Hispanic white

men and women even after controlling for dietary folate intake. In addition, Jacques et

al. (281) published data from NHANES III on homocysteine concentrations in non-

Hispanic white, non-Hispanic black, and Mexican American men and women. Mexican

American women had significantly lower homocysteine concentrations compared to non-

Hispanic white and non-Hispanic black women. In response to the publication of these

two studies, Baggott (282) questioned how lower serum folate concentrations could lead

to lower homocysteine concentrations in Mexican Americans, which is inconsistent with









a large body of literature that reports that lower serum folate concentrations are

associated with higher homocysteine concentrations. Jacques et al. (283) responded by

suggesting that lifestyle factors and racial and genetic differences may influence the

homocysteine concentration of this population, and that the lower homocysteine

concentrations are not a result of higher serum folate and vitamin B 12 concentrations.

The basis for the lower serum and red blood cell folate and homocysteine concentrations

in Mexican Americans needs further study.

Folate Deficiency

There is a progressive sequence of events associated with the development of folate

deficiency. The first event is a reduction in serum folate concentration, which can occur

within 1 to 3 wk. Serum folate concentration is considered to be a short-term indicator of

folate status because it is most indicative of recent intake. After a longer period of folate

deficiency, red blood cell folate concentration will begin to drop. Red blood cell folate

concentration is considered a long-term indicator of folate status. Homocysteine

concentration begins to increase around the same time that red blood cell folate

concentration decreases (284).

After blood folate status is compromised for a period of time, the clinical

manifestations of folate deficiency develop. Neutrophil hypersegmentation causes a state

of macrocytosis without anemia. Eventually, macrocytic, megaloblastic anemia will

develop with decreases in hemoglobin, hematocrit, and red blood cell folate

concentrations (284). A decrease in folate availability will cause a reduction in the DNA

cycle, ultimately decreasing cell division, and resulting in the formation of the large,

immature red blood cells observed in megaloblastic anemia (233).









Folate Status Assessment

Folate status assessment can be separated into quantitative assessment and

functional assessment. The two quantitative indicators are serum and red blood cell

folate concentrations. The functional indicators include plasma homocysteine,

SAM/SAH ratio, DNA methylation, uracil misincorporation, and DNA strand breaks.

Quantitative measures of folate status can provide an indication to the clinician of

the patient's folate status. Because serum folate is a more sensitive indicator of recent

intake, an isolated serum folate measurement may not differentiate between a temporary

reduction in serum folate and chronic folate deficiency (27). Repeated serum folate

measurements over time will provide more information as to the status of the individual.

A serum folate concentration less than 3 ng/ml (6.8 nmol/L) is reflective of a negative

folate balance in the individual at the time the sample was taken (285). Red blood cell

folate concentration is the primary indicator chosen to determine folate status because it

reflects tissue stores of the vitamin (27). Red blood cells only take up folate during

erythropoiesis and have a lifespan of 120 d; therefore, they are indicative of long-term

folate status (27). The IOM recognizes a red blood cell folate concentration of < 140

ng/ml (305 nmol/L) as a deficient red blood cell folate concentration and has based this

number on a review of studies in which this value was associated with clinical indicators

of deficiency (27). Although red blood cell folate concentration is a better indicator of

folate status than serum folate concentration, both serum and red blood cell folate

concentrations taken together can provide a good measure of folate status.

The functional indicators of folate status become affected as a result of changes in

the quantitative indicators. Plasma homocysteine concentration becomes elevated during

a folate deficiency because methionine synthase, which requires folate as a cofactor, is









unable to remethylate homocysteine to methionine. As mentioned previously, different

studies have provided evidence that homocysteine concentration rises as folate

concentration decreases. Specifically, Jacob et al. (286) reported that folate depletion in

young men led to a rise in homocysteine concentration. O'Keefe et al. (287) found an

inverse relationship between serum and red blood cell folate concentrations and plasma

homocysteine concentration in women. Although there are differences in the cut off

values for plasma homocysteine, any value > 14 tmol/L is generally considered high

(27). Ubbink et al. (288) suggested a reference range of 4.9 to 11.7 [tmol/L. Regardless

of gender, low folate concentrations result in increased homocysteine concentrations.

High homocysteine concentrations lead to an increase in SAH concentration via the

reversible enzyme SAH hydrolase (Fig. 2-2, reaction 11). Increases in SAH without a

concurrent increase in SAM results in a decreased SAM/SAH ratio. Balaghi and Wagner

(213) found significantly decreased SAM/SAH ratios in rats fed a folate deficient diet for

4 wk compared to controls. There are currently no published reports related to

SAM/SAH ratios in humans fed folate-deficient diets.

DNA methylation, uracil misincorporation, and DNA strand breaks are also

functional indicators of folate status but cannot be used in the same way as plasma

homocysteine concentration. Each indicator has been discussed in detail under the DNA

stability section. Briefly, a folate deficiency can affect DNA methylation by limiting the

amount of methionine available for the production of SAM. Low folate status causes a

decrease in the formation of dTTP from dUTP and results in uracil misincorporation.

Finally, uracil misincorporation ultimately leads to DNA strand breaks as a result of

DNA repair. There are no generalized norms for these indicators because of individual









variation in DNA and variation in the assessment of these methods. These indicators are

only useful when comparing subjects to a control group or to themselves after some kind

of treatment within the same study.

Analytical Methodology

A variety of methods have been used to assess folate status. Not all research

groups use the same methods, and even within a given method, protocols may vary

between labs. These differences limit the ability to directly compare results from

different studies without an evaluation of interlaboratory differences (289). Although

there are many methods used, this section reviews some of the more commonly used

methods.

Blood Folate Analysis

Serum and red blood cell folate concentrations can be measured using a

microbiological assay or a radiobinding assay. The microbiological assay is the most

widely accepted method for determining folate concentrations in blood, urine, tissue, and

food samples. The test organism used is Lactobacillus case subspecies rhamnosis. This

organism metabolizes the greatest number of folate derivatives, including polyglutamted

residues with up to three glutamates (290). The growth of the organism is calculated by

measuring the turbidity of the medium, which is directly proportional to the amount of

folate in the sample. This assay cannot be used on samples containing antibiotics or

methotrexate because they will inhibit the growth of the organism.

The radiobinding assay (RA) also is used to measure folate concentrations of

samples because of its speed and the fact that it is not affected by pharmaceuticals. A

folate binding protein attached to microbeads and [125I]-labeled folic acid or

methyltetrahydrofolate are used to quantitate serum or red blood cell folate concentration.









Unlabeled folate competes with labeled folate for binding to the folate binding protein.

Samples are centrifuged and bound folates and microbeads precipitated. The supernatant

is discarded and bound labeled folate is counted in a scintillation counter. The decrease

in radioactivity is proportional to the folate concentration in the sample. There have been

problems associated with using RA assays. Folate values for the first 4 y of the

NHANES III were determined using a RA, and results had to be adjusted to correct

values after it was determined that the RA results were 30% too high due to the problems

with the standards produced by the manufacturer of the assay kit (27).

Although the microbiological assay tends to be time consuming and tedious, it has

been the preferred method for quantitative analysis of folate status. The RA is a quicker

method and tend to be lower than values obtained using the microbiological method

(291). A round robin comparison of different lab techniques for the measurement of

serum and red blood cell folate concentration was conducted by the CDC (289). Five

different analytical techniques were used by 20 different laboratories worldwide to assess

the folate concentrations of six serum and six whole blood pools. They reported overall

CVs for the serum folate pools and the whole blood pools of 27.6% and 35.7%,

respectively. They also reported a 2- to 9-fold difference in concentrations of the pools

within the different methods. These results support the fact that folate values cannot be

compared between labs unless interlaboratory variations are considered.

Plasma Homocysteine and SAM/SAH Ratio Analysis

Measurement of plasma homocysteine concentration is generally limited to total

homocysteine concentration. Before total homocysteine concentration can be measured,

the disulfides must be chemically reduced. The preferred approach to homocysteine

quantitation is an HPLC method because of the relatively low cost of chemicals and









solvents, the availability of equipment, and the existence of a fully automated assay

(292). Some other automated methods include gas chromatography-mass spectrometry,

liquid chromatography electrospray tandem mass spectrometry, and immunoassay (293).

HPLC methods for homocysteine utilize reversed-phase columns, which have

hydrophobic chains that protrude and retain the hydrophobic molecules of interest. These

molecules are eluted in order of hydrophobicity using a mobile phase. Once eluted,

homocysteine can be detected fluorescently, electrochemically, or colorimetrically. A

similar technique also can be used to measure SAM and SAH concentrations (294).

DNA Stability

The most common ways to assess genomic stability is by evaluating DNA

methylation, uracil misincorporation, and DNA strand breaks. Because analysis of these

indicators differs between laboratories, results cannot be compared between different

studies. Comparisons can be made only within studies between controls and

experimental subjects.

DNA methylation. The original assay used to determine DNA methylation

involved the use of [3H]SAM and a bacterial DNA methylase that only methylates at the

5-position of cytosines. Incorporation of labeled methyl groups is inversely related to the

extent of DNA methylation in the sample (213). Some limitations of this method are that

damaged DNA can interfere with the methylase, and that DNA strand breaks or abasic

sites can give a false positive (295). Pogribny et al. (295) developed a cytosine-extension

method based on the use of a methylation-sensitive restriction enzyme that leaves a

5'guanine overhang followed by single nucleotide primer extension with [3H]dCTP. The

extent of [3H]dCTP incorporation also is inversely proportional to the DNA methylation

in the sample. The cytosine-extension method is less subject to error than the methyl









acceptance assay because the integrity of the DNA does not influence the method, which

can be applied to nanogram quantities of DNA (295). Fujiwara and Ito (296) made one

modification of the cytosine extension protocol to circumvent the use of radioactivity.

Biotinylated dCTP was added to the digested DNA by Taq polymerase, which was

visualized with a streptavidin-alkaline phosphatase reaction.

Friso et al. (13) developed a liquid chromatography/mass spectrometry method that

allowed direct quantitation of methylated cytosine residues. DNA is enzymatically

hydrolyzed with sequential digestion to nucleotides and separated into the four bases.

Methylated cytosine elutes 2 min after cytosine and can be quantitated. This method is

more accurate than the previous two methods in determining DNA methylation because it

is quantitative rather than semiquantitative, it measures methylated cytosines directly

rather than indirectly, and it has small intra- and inter-assay CVs. A similar method by

Cooney et al. (297) involves the direct measurement of 5-methyldeoxycytidine after

sequential enzymatic digestion and HPLC separation. A less utilized method involves

bisulfite-induced modification of genomic DNA to convert cytosine to uracil but

methylated cytosines remain nonreactive (298).

Uracil misincorporation. A high degree of uracil misincorporation can lead to

DNA instability when DNA repair enzymes remove the uracil and leave single-strand

breaks that could result in the less repairable double-stranded breaks (11). The original

assay used to determine uracil misincorporation employed gas chromatography/mass

spectrometry in negative chemical ionization mode after DNA was digested with uracil

deglycosylase. This allowed for direct measurement of uracil in the DNA sample.









The comet assay can also be used to determine uracil misincorporation (259). The

comet assay involves the use of a microscope slide covered with agarose and cells. These

slides are washed with uracil deglycosylase to excise the uracil. Slides are subjected to

electrophoresis and stained. The DNA sample resembles a comet, with the intact DNA in

the head, and any DNA fragments in the tail. Uracil misincorporation is related to the

fluorescence in the tail. More recent assays convert misincorporated uracils to DNA

strand breaks with addition of the endonuclease Exo III after uracil excision with uracil

deglycosylase. DNA strand breaks can then be quantified using a comet assay (299) or

by random oligonucleotide-primed synthesis (ROPS) assay (258,260).

DNA strand breaks. There are many different methods available to quantify DNA

strand breaks. Some of the older methods include alkali elution, DNA unwinding assays,

and unscheduled DNA synthesis. Agarose gel electrophopresis, terminal

deoxynucleotide transferase, nick translation, and the ROPS assay are more current

methods of detection. Most of these assays do not distinguish 3'OH from 5'OH ends and

require large amounts of DNA (300). Basnakian and James (300) developed a ROPS

assay based on random oligo-nucleotide synthesis catalyzed by Klenow fragment

polymerase in order to detect low frequency 3'OH DNA strand breaks. After

denaturation and renaturation of the DNA, the single-stranded DNA serves as its own

template for extension using [32P]-labeled dNTPs. Incorporation of labeled dNTPs is

proportional to the number of strand breaks in the DNA sample. The strength of this

assay is that it only requires nanogram concentrations of DNA and it can detect single-

stranded and double-stranded DNA breaks.









Genotype Determination

With the discovery of single nucleotide polymorphisms (SNPs), determining

subject genotypes for these SNPs has become common practice. The most common

method of genotyping involves polymerase chain reaction (PCR) to amplify the desired

region of DNA. Once the region is amplified, specific restriction enzymes are added

depending on the SNP being studied. Fragments are then separated by electrophoresis on

an agarose gel. There are several SNPs associated with folate metabolism. In order to

simplify the search for multiple SNPs in one sample, Barbaux et al. (301) developed a

method that allows genotyping of four SNPs on one gel. This "heteroduplexing" method

involves the use of a heteroduplexing generator instead of restriction enzymes. The

generator is a synthetic DNA molecule identical in sequence to the SNP of choice with a

microdeletion adjacent to the polymorphic site (301). The generator combined with PCR

technology enables multiple genotyping in a single-tube reaction that can be separated on

a single gel. Recently, Ulvik and Ueland (302) developed a method utilizing real-time

PCR to genotype for multiple folate related SNPs in whole blood or serum in one tube

with the goal of eliminating the DNA purification step. Advances in technology keep

improving current methods and will one day enable all SNPs associated with folate

metabolism to be identified in the most efficient and cost friendly manner.

Research Significance

The MTHFR 677C-T polymorphism affects a large percentage of the population

with an estimated frequency of -12% for the TT genotype with considerable variation

between different ethnic groups (4,303). Blood folate concentrations are reduced (80,86),

homocysteine concentrations increased (7,65,67,82), and DNA methylation diminished

(13) in individuals with the homozygous TT genotype for the MTHFR 677C->T variant.









It is well recognized that impaired folate status is associated with abnormal fetal growth

and development (304) and increased risk of pregnancy complications (139), and that

periconceptional folic acid supplementation significantly reduces the risk ofNTDs (119).

The metabolic basis of these observations has not been definitively established but may

relate to folate's role in nucleotide biosynthesis (1), DNA methylation (218,219), and/or

maintenance of normal homocysteine concentrations (142). Since the combined presence

of the MTHFR 677C-T variant and low folate status has been associated with increased

risk for birth defects, the present study was designed to address the specific aim in

females of reproductive age. Guinotte et al. (87) recently published a metabolic study in

this age group with a similar design as the present study to investigate differences in

response to folate depletion and repletion with the RDA by MTHFR genotype in young

women of Mexican American descent. An important difference between these studies is

the ethnicity of the study groups, a factor that has been shown to significantly affect

folate status and homocysteine concentration (280,281). The present study is the first

controlled metabolic study performed in predominantly non-Hispanic women of

childbearing age to determine whether there are differences in response to folate

depletion and repletion between MTHFR genotypes. In addition, the effect of controlled

folate intake on DNA methylation in women with the CC or TT MTHFR 677C->T

genotype has not been reported in this age group. Although DNA methylation was found

to be significantly reduced after depletion and reversed with folate supplementation in

post-menopausal women fed a folate-controlled diet (237), these researchers did not

evaluate the effect of genotype on methylation. Similarly, the effect of MTHFR

genotype on DNA methylation was was not considered in a group of elderly women fed a






79


folate-controlled depletion diet (238). The present study is the first to report the effect of

folate depletion-repletion on global DNA methylation in women of childbearing age

based on MTHFR 677C-T genotype. In addition, there are no genotype-specific DRI

recommendations, and data are insufficient to determine whether individuals with the TT

genotype require more folate to maintain normal folate status than individuals with the

CC genotype. Data from this study can be considered when making future revisions of

the RDA for folate.














CHAPTER 3
STUDY DESIGN AND METHODS

Subject Screening and Description

After approval of the study protocol by the University of Florida Institutional

Review Board, nonpregnant, healthy women 20 to 30 y old were recruited from the

Gainesville, FL area by distributing flyers and placing announcements in local papers.

Inclusion criteria were normal blood chemistry, blood folate concentrations, body weight

(within 120% of ideal body weight), and health status as determined by medical history.

Exclusion criteria were chronic alcohol consumption or use of tobacco or any

medications. Approximately 3500 women were screened initially over the phone by a

research nurse. Women who seemed to meet the initial screening criteria (n = 379)

reported to our lab to have their blood drawn to determine genotype status for the

MTHFR 677C->T polymorphism. Only women with the normal CC or homozygous

variant TT genotypes were eligible for the study. Forty-six women were selected to

participate in the study. All subjects screened and enrolled in the study provided signed

informed consent and agreed to participate for the duration of either a 7 wk or 14 wk

study and to comply with the study protocol. This study was performed in conjunction

with another study whose subjects did not complete the repletion phase of the study.

Forty-one women (22 CC, 19 TT) completed the depletion phase of the study (7 wk), and

20 women (10 CC, 10 TT) completed the entire depletion-repletion protocol (14 wk).

Serum and red blood cell folate, and plasma vitamin B 12, pyridoxal phosphate, and









homocysteine concentrations were normal at baseline for all subjects (i.e., > 7 nmol/L,

>317 nmol/L, > 125 pmol/L, > 20 nmol/L, and < 14 [tmol/L, respectively).

Study Design

Subjects adhered to a depletion-repletion feeding protocol divided into two

consecutive periods of 49 d (7 wk) each (Fig. 3-1). Subjects consumed a low-folate diet

providing 115 20 pg dietary folate equivalents (DFE)/d during the first 7 wk of the

study. The repletion diet consisted of a combination of the depletion diet plus folic acid

and provided 400 tg DFE/d [115 + 285 pg DFE (168 tg folic acid X 1.7 = 285 [g DFE)]

(27). This controlled metabolic feeding study was conducted in the General Clinical

Research Center (GCRC) at Shands Hospital at the University of Florida in Gainesville,

FL.





Depletion Repletion
115 tg DFE/d 400 tg DFE/d




A A A A A A A A



0 1 2 3 4 5 6 7 8 9 10 11 12 13 14



Blood collection A CBC-D and HCG 0 Blood chemistry and
leukocyte collection
Figure 3-1. Study design.

Fasting blood samples were obtained at weekly intervals to determine changes in

serum and red blood cell folate and homocysteine concentrations. A blood chemistry









profile was performed at baseline (wk 0), post-depletion (wk 7) and post-repletion (wk

14) to monitor health status (Quest Diagnostic Laboratories; Gainesville, FL).

Leukocytes were collected at baseline (wk 0), post-depletion (wk 7), and post-repletion

(wk 14) for DNA extraction. Complete blood counts with differentials (CBC-D) were

performed biweekly and these measurements were evaluated throughout the study to

monitor hematological indices. In the event of a reduction in hematocrit to < 30%, 50 mg

of iron as ferrous fumarate from a time-released Ferro-Sequels caplet (Inverness

Medical, Inc; Waltham, MA) was provided with dinner until values reverted to > 30%.

Additionally, biweekly samples were obtained for quantitative analysis of serum human

chorionic gonadotropin (HCG) (Quest Diagnostic Laboratories; Gainesville, FL) in order

to detect a pregnancy very early during the folate depletion phase and throughout the

repletion phase. Subjects were instructed to use barrier methods for contraception if they

were sexually active, and they were informed of the potential risks to maternal and fetal

health posed by consumption of the low-folate diet.

General Clinical Research Center (GCRC) Protocol

Breakfast was consumed at the GCRC between 6:30 am and 7:30 am for the

duration of the study. Subjects were provided with a "take-out" lunch and snacks each

day and returned to the GCRC between 5:00 pm and 6:00 pm for dinner. They were

required to eat all foods and only those foods provided to them for the duration of the

study. Subjects were permitted to take all meals for 2 d away from the GCRC and were

supplied with all of the menu items packed for transport with detailed instructions on

reheating food items. Compliance with the protocol was monitored through close daily

contact by the research team and weekly evaluation of changes in serum folate

concentration.









Dietary Treatment and Supplementation Description

An experimental diet that provided a limited quantity of folate using a variety of

palatable entrees and accompaniments was developed and analyzed. The diet consisted

of a 5-d menu cycle as shown in Table 3-1. Many commercially prepared food items

could not be used in this study as a consequence of the 1996 FDA mandate that all

enriched foods be fortified with folic acid (20). To keep the folate content to a minimum,

customized recipes were developed and prepared with unenriched flour obtained from

Kansas State University (Manhattan, KS) and other low-folate ingredients. Recipes were

modified in the GCRC kitchen and taste-tested at laboratory meetings to select the

recipes that were the most palatable. Foods made with the unenriched flour included

waffles, pancakes, blueberry muffins, pita bread, biscuits, brownies, cookies, cakes,

toppings, and pizza crust. A limited selection of canned, low-folate vegetables were used

in this study, and each of these was boiled 3 times with the cooking liquid being

discarded after each boiling to help leach endogenous folate from the food (305). All

foods were weighed to 0.05 g to ensure that each subject received equal amounts of

food and that portion sizes remained constant throughout the study.

The macronutrient and micronutrient contents of the diet, excluding folate content,

were estimated using the Minnesota Nutrient Data System (Version 4.03; Nutrition

Coordinating Center at the University of Minnesota; Minneapolis, MN). According to

this analysis, the diet provided 2358 kilocalories distributed as 11% protein, 62%

carbohydrate, and 27% fat. Folate content was determined by laboratory analysis

involving a trienzyme extraction procedure followed by a microbiological assay using

Lactobacillus case (L. case) as described below. Analyses indicated that the diet

contained an average of 115 20 |tg DFE/d.









Table 3-1. Five-day cycle menu
Day 1 Day 2 Day 3 Day 4 Day 5

Breakfast


Shredded wheat
Skim milk
Raisins
Brown sugar
Grape juice


Sandwich,
Pita pocket*
Honey ham
Cheese
Mayonnaise
Applesauce
Doritos
Soda


Enchiladas,
Chicken
Cheese
Enchilada
sauce*
Corn tortillas
Green beans
Blueberry tart
Ice cream
Cranberry juice


Waffle*
Syrup
Peaches, canned
Cranberry juice


Tuna salad
Crackers
Pears, canned
Corn chips
Soda


Chicken pot pie,
Chicken
Potatoes
Carrots
Green beans
Sauce
Margarine
Crust
Green beans
Orange sherbet
Grape juice


Cornflakes
Skim milk
Raisins
Apple juice


Lunch
Sandwich,
Biscuit*
Baked ham
Cheese
Peaches, canned
Popcorn
Soda


Dinner
Tacos,
Seasoned beef
Taco sauce
Sour cream
Cheese
Corn tortilla
shells
Carrots
Orange sherbet
Cranberry juice


Snacks
Fruit cocktail, Applesauce Fruit cocktail,
canned Shortbread canned
Ginger cookies rounds Snickerdoodle
Caramel Popcorn cookies*
popcorn Apple juice Brownie*
Apple juice Skim milk Grape juice
Skim milk Skim milk
*Menu item prepared with unenriched flour


Pancake*
Syrup
Hash browns
Applesauce
Cranapple juice


Sandwich,
Pita pocket*
Turkey breast
Cheese
Mayonnaise
Fruit cocktail,
canned
Corn chips
Soda


Pizza,
Crust*
Sauce
Mozzarella
cheese
Turkey ham
chunks
Marinated green
beans
Apple crisp*
Cranberry juice


Pears, canned
Chocolate chip
cookies*
Caramel
popcorn
Grape juice
Skim milk


Blueberry
muffin*
Skim milk
Pears, canned
Apple juice


Baked stuffed
potato
Turkey ham
chunks
Fruit cocktail
canned
Popcorn
Soda


BBQ chicken,
Chicken
BBQ sauce
Margarine
Mashed
potatoes
Green beans
Chocolate
pudding
Shortbread
bar
Cranapple
juice


Peaches,
canned
Oatmeal
raisin
cookies*
Pound cake*
Skim milk


fMenu item boiled three times to minimize folate content









Subjects consumed a custom-formulated, folic acid free supplement (Westlab

Pharmacy; Gainesville, FL) with breakfast and dinner to provide the 1998 Recommended

Dietary Allowance (RDA) for all other nutrients except choline. The vitamin-mineral

supplement composition is presented in Appendix A. The composition of the custom-

formulated supplement was determined by comparing the computerized micronutrient

analysis of the low-folate diet to the 1998 RDA for all nutrients except choline.

Additionally, the micronutrient analysis was adjusted for the loss of water-soluble

vitamins due to boiling the vegetables 3 times by assuming 100% loss and subtracting

those amounts from the total value obtained by computer analysis. Nutrients other than

folate and choline present in the diet at less than 100% the 1998 RDA were included in

the supplement. The choline content of the diet was analyzed as discussed below and

found to provide 283 mg/d, which is 67% of the Adequate Intake (AI) for choline (425

mg/d) (306). In addition, a separate calcium supplement (Citracal Mission Pharmacal;

San Antonio, TX) provided 200 mg of calcium as calcium citrate to provide calcium that

was not included in the supplement.

Subjects' body weights were maintained within 5% of baseline. Subjects who lost

more than 5% of their initial body weight were provided with foods with relatively little

or no nutrient value aside from calories (i.e., margarine, candy, Jello, and sweetened

beverages). If weight loss was not sustained by these measures, the carbohydrate-based

caloric supplement Moducal (Mead Johnson Nutritionals; Evansville, IN) was added to

beverages. For weight gain exceeding 5% of initial body weight, margarine was

excluded from food preparation, and only unsweetened beverages were permitted.









Caffeinated beverages were limited to two 12 oz carbonated beverages or one cup of

coffee/d plus one 12 oz beverage.

Sample Collection and Processing

Weekly fasting blood samples were collected throughout the 14 wk study by a

registered nurse using a # 23 gauge needle and a 1/2 inch butterfly (Vacutainer Blood

Collection Sets; Becton Dickinson, Vacutainer Systems; Franklin Lakes, NJ). All blood

samples were processed within 1 h of collection, as previously described (238,307). A

total of 15 blood collections per subject were obtained during the 14 wk study period.

Blood samples were treated with extra precaution during collection, processing, and

storage to ensure protection from light by wrapping blood collection tubes in foil,

processing under yellow lights (Sylvania Gold; Danvers, MA), and storing in cardboard

boxes, respectively.

Blood for serum samples was collected in 8.3 ml SST gel and clot activator tubes

(Vacutainer; Becton Dickinson, Rutherford, NJ) and kept at room temperature for 30 to

60 min to allow time for clotting. Serum was obtained by centrifuging the SST gel clot

activator tubes at 650 x g for 15 min at 210C (International Equipment Company; Model

HN-S II Centrifuge, Needham Heights, MA). Supernatant sera were mixed with sodium

ascorbate (1 mg/ml), aliquoted into 200 [l samples, and stored at -300C until analysis.

Whole blood was collected in 7 ml tubes containing K3 ethylenediaminetetraacetic

acid (EDTA) (Vacutainer; Becton Dickinson, Rutherford, NJ). Blood for plasma

homocysteine was kept on ice until processing. A small aliquot of whole blood held at

room temperature was used for hematocrit determination and another portion diluted 20-

fold in 1 mg/ml ascorbic acid was aliquotted into 200 [l samples and frozen for