{"Affiliation":[{"label":"Affiliation","value":"Land and Food Systems, Faculty of","attrs":{"lang":"en","ns":"http:\/\/vivoweb.org\/ontology\/core#departmentOrSchool","classmap":"vivo:EducationalProcess","property":"vivo:departmentOrSchool"},"iri":"http:\/\/vivoweb.org\/ontology\/core#departmentOrSchool","explain":"VIVO-ISF Ontology V1.6 Property; The department or school name within institution; Not intended to be an institution name."}],"AggregatedSourceRepository":[{"label":"Aggregated Source Repository","value":"DSpace","attrs":{"lang":"en","ns":"http:\/\/www.europeana.eu\/schemas\/edm\/dataProvider","classmap":"ore:Aggregation","property":"edm:dataProvider"},"iri":"http:\/\/www.europeana.eu\/schemas\/edm\/dataProvider","explain":"A Europeana Data Model Property; The name or identifier of the organization who contributes data indirectly to an aggregation service (e.g. Europeana)"}],"Campus":[{"label":"Campus","value":"UBCV","attrs":{"lang":"en","ns":"https:\/\/open.library.ubc.ca\/terms#degreeCampus","classmap":"oc:ThesisDescription","property":"oc:degreeCampus"},"iri":"https:\/\/open.library.ubc.ca\/terms#degreeCampus","explain":"UBC Open Collections Metadata Components; Local Field; Identifies the name of the campus from which the graduate completed their degree."}],"Creator":[{"label":"Creator","value":"Mujica Coopman, Maria Fernanda","attrs":{"lang":"en","ns":"http:\/\/purl.org\/dc\/terms\/creator","classmap":"dpla:SourceResource","property":"dcterms:creator"},"iri":"http:\/\/purl.org\/dc\/terms\/creator","explain":"A Dublin Core Terms Property; An entity primarily responsible for making the resource.; Examples of a Contributor include a person, an organization, or a service."}],"DateAvailable":[{"label":"Date Available","value":"2023-09-30T07:00:00Z","attrs":{"lang":"en","ns":"http:\/\/purl.org\/dc\/terms\/issued","classmap":"edm:WebResource","property":"dcterms:issued"},"iri":"http:\/\/purl.org\/dc\/terms\/issued","explain":"A Dublin Core Terms Property; Date of formal issuance (e.g., publication) of the resource."}],"DateIssued":[{"label":"Date Issued","value":"2020","attrs":{"lang":"en","ns":"http:\/\/purl.org\/dc\/terms\/issued","classmap":"oc:SourceResource","property":"dcterms:issued"},"iri":"http:\/\/purl.org\/dc\/terms\/issued","explain":"A Dublin Core Terms Property; Date of formal issuance (e.g., publication) of the resource."}],"Degree":[{"label":"Degree (Theses)","value":"Doctor of Philosophy - PhD","attrs":{"lang":"en","ns":"http:\/\/vivoweb.org\/ontology\/core#relatedDegree","classmap":"vivo:ThesisDegree","property":"vivo:relatedDegree"},"iri":"http:\/\/vivoweb.org\/ontology\/core#relatedDegree","explain":"VIVO-ISF Ontology V1.6 Property; The thesis degree; Extended Property specified by UBC, as per https:\/\/wiki.duraspace.org\/display\/VIVO\/Ontology+Editor%27s+Guide"}],"DegreeGrantor":[{"label":"Degree Grantor","value":"University of British Columbia","attrs":{"lang":"en","ns":"https:\/\/open.library.ubc.ca\/terms#degreeGrantor","classmap":"oc:ThesisDescription","property":"oc:degreeGrantor"},"iri":"https:\/\/open.library.ubc.ca\/terms#degreeGrantor","explain":"UBC Open Collections Metadata Components; Local Field; Indicates the institution where thesis was granted."}],"Description":[{"label":"Description","value":"Maternal folate, riboflavin, betaine, choline, and vitamins B-6 and B-12 (B-12) concentrations, also known as methyl nutrients, have an interrelated role in fetal growth and DNA methylation. To date, the relationship between individual maternal methyl nutrient concentrations and neonatal anthropometric outcomes have shown conflicting results, while the interrelationship of maternal methyl nutrient concentrations and their association with DNA methylation levels of fetal growth and obesity-related genes in the offspring is unknown. \r\nThe overall goal of my thesis was to provide novel evidence of the interrelationship of maternal methyl nutrients during early pregnancy i.e., <20 weeks of gestation and their association with neonatal anthropometric outcomes and fetal growth and obesity programming in Canadian mother-offspring dyads. To address this goal, firstly, the relationship between concentrations of betaine, a methyl donor nutrient, and total homocysteine (tHcy), an intermediate metabolite of methylation reactions, was tested in 723 pregnant women at early pregnancy. Betaine was inversely associated with tHcy (\u03b2=-0.21\u00b5mol\/L; 95%CI -0.34, -0.07\u00b5mol\/L). Furthermore, the relationship between maternal methyl nutrient patterns and neonatal anthropometric outcomes was explored. Methyl nutrient patterns were mainly characterized by maternal B-12 biomarkers and betaine. Only second-trimester B-12 pattern was inversely associated with head circumference (HC) (\u03b2=-0.13cm; 95%CI -0.24, -0.03cm) and HC z-score (\u03b2=-0.09; 95%CI -0.09, -0.01). Lastly, whether DNA methylation levels of CpG sites associated with IGF-2, HIF-3\u03b1, RXRA, LEP, LEP-R, DNMT-1, DNMT-3A, and DNMT-3B genes, measured in infants, differed by maternal red blood cell (RBC) folate concentration \u22641360 nmol\/L and RBC folate >1360 nmol\/L was tested, and whether these CpG sites were associated with maternal methyl nutrient patterns. Infant DNA methylation levels did not significantly differ by maternal folate concentration and were not significantly associated with maternal methyl nutrient patterns. In summary, these results indicated that betaine and total B-12 were the main drivers of maternal methyl nutrients patterns in these folate-replete populations. However, the lack of association between maternal methyl nutrient patterns with neonatal anthropometric outcomes and DNA methylation levels of fetal growth and obesity-related genes in infants suggests the need for a better understanding of the role of these nutrients in fetal programming and growth.","attrs":{"lang":"en","ns":"http:\/\/purl.org\/dc\/terms\/description","classmap":"dpla:SourceResource","property":"dcterms:description"},"iri":"http:\/\/purl.org\/dc\/terms\/description","explain":"A Dublin Core Terms Property; An account of the resource.; Description may include but is not limited to: an abstract, a table of contents, a graphical representation, or a free-text account of the resource."}],"DigitalResourceOriginalRecord":[{"label":"Digital Resource Original Record","value":"https:\/\/circle.library.ubc.ca\/rest\/handle\/2429\/75762?expand=metadata","attrs":{"lang":"en","ns":"http:\/\/www.europeana.eu\/schemas\/edm\/aggregatedCHO","classmap":"ore:Aggregation","property":"edm:aggregatedCHO"},"iri":"http:\/\/www.europeana.eu\/schemas\/edm\/aggregatedCHO","explain":"A Europeana Data Model Property; The identifier of the source object, e.g. the Mona Lisa itself. This could be a full linked open date URI or an internal identifier"}],"FullText":[{"label":"Full Text","value":"   MATERNAL METHYL NUTRIENTS, OBESITY PROGRAMMING, AND NEONATAL ANTHROPOMETRIC OUTCOMES by Maria Fernanda Mujica Coopman B.Sc. Universidad de Chile (Chile), 2010  M.Sc. Universidad de Chile (Chile), 2015 A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF  DOCTOR OF PHILOSOPHY  in THE FACULTY OF GRADUATE AND POSTDOCTORAL STUDIES (Human Nutrition)  THE UNIVERSITY OF BRITISH COLUMBIA (Vancouver)  August 2020  \u00a9 Maria Fernanda Mujica Coopman, 2020  ii    The following individuals certify that they have read, and recommend to the Faculty of Graduate and Postdoctoral Studies for acceptance, the dissertation entitled: Maternal methyl nutrients, obesity programming, and neonatal anthropometric outcomes  submitted by Maria Fernanda Mujica Coopman in partial fulfillment of the requirements for the degree of Doctor of Philosophy In Human Nutrition  Examining Committee:   Dr. Yvonne Lamers, Human Nutrition (UBC) Supervisor   Dr. Catherine Field, Department of Agricultural, Food and Nutritional Science (University of Alberta) Supervisory Committee Member  Dr. Ronaldo Cerri, Applied Animal Biology (UBC) University Examiner Dr. Alexander Beristain, Reproductive Sciences (UBC) University Examiner Dr. Stephanie Atkinson, Department of Pediatrics (McMaster University) External University Examiner iii  Abstract  Maternal folate, riboflavin, betaine, choline, and vitamins B-6 and B-12 (B-12) concentrations, also known as methyl nutrients, have an interrelated role in fetal growth and DNA methylation. To date, the relationship between individual maternal methyl nutrient concentrations and neonatal anthropometric outcomes have shown conflicting results, while the interrelationship of maternal methyl nutrient concentrations and their association with DNA methylation levels of fetal growth and obesity-related genes in the offspring is unknown.  The overall goal of my thesis was to provide novel evidence of the interrelationship of maternal methyl nutrients during early pregnancy i.e., <20 weeks of gestation and their association with neonatal anthropometric outcomes and fetal growth and obesity programming in Canadian mother-offspring dyads. To address this goal, firstly, the relationship between concentrations of betaine, a methyl donor nutrient, and total homocysteine (tHcy), an intermediate metabolite of methylation reactions, was tested in 723 pregnant women at early pregnancy. Betaine was inversely associated with tHcy (\u03b2=-0.21\u00b5mol\/L; 95%CI -0.34, -0.07\u00b5mol\/L). Furthermore, the relationship between maternal methyl nutrient patterns and neonatal anthropometric outcomes was explored. Methyl nutrient patterns were mainly characterized by maternal B-12 biomarkers and betaine. Only second-trimester B-12 pattern was inversely associated with head circumference (HC) (\u03b2=-0.13cm; 95%CI -0.24, -0.03cm) and HC z-score (\u03b2=-0.09; 95%CI -0.09, -0.01). Lastly, whether DNA methylation levels of CpG sites associated with IGF-2, HIF-3\u03b1, RXRA, LEP, LEP-R, DNMT-1, DNMT-3A, and DNMT-3B genes, measured in infants, differed by maternal red blood cell (RBC) folate concentration \u22641360 nmol\/L and RBC folate >1360 nmol\/L was tested, and whether these CpG sites were associated with maternal methyl nutrient patterns. Infant DNA methylation levels did not significantly differ iv  by maternal folate concentration and were not significantly associated with maternal methyl nutrient patterns. In summary, these results indicated that betaine and total B-12 were the main drivers of maternal methyl nutrients patterns in these folate-replete populations. However, the lack of association between maternal methyl nutrient patterns with neonatal anthropometric outcomes and DNA methylation levels of fetal growth and obesity-related genes in infants suggests the need for a better understanding of the role of these nutrients in fetal programming and growth.     v  Lay Summary  Maternal B-vitamins, choline, methionine, and betaine, also known as methyl nutrients, participate in interrelated pathways required for fetal growth and provision of methyl groups for DNA methylation. DNA methylation is an important biological process, especially during early pregnancy, because it contributes to \u2018program\u2019 gene expression for later in life. However, the relationship of combined maternal methyl nutrient concentrations with DNA methylation levels of genes related to fetal growth and obesity, and with the baby\u2019s size at birth is unknown. The overall goal of my research was to investigate the association of combined methyl nutrient concentrations at early pregnancy with baby\u2019s birth size, and DNA methylation levels of genes related to fetal growth and obesity. My findings suggest that B-12 and betaine characterized maternal methyl nutrients patterns; however, there no association between maternal methyl nutrient patterns and baby\u2019s birth size and DNA methylation of fetal growth and obesity genes.    vi  Preface This dissertation was prepared according to the requirements for a Ph.D. thesis by The University of British Columbia (UBC) Faculty of Graduate and Postdoctoral Studies. This research was the result of a collaborative effort. The sample analyses and data collection presented in this dissertation were conducted in the UBC Nutritional Biomarker Laboratory led by Dr Yvonne Lamers, the Newborn Screening Laboratory at the British Columbia (BC) Children\u2019s Hospital, and the Pathology Laboratory at St Paul\u2019s Hospital in Vancouver, BC, as well as at the Li Ka Shing Centre for Research at the University of Alberta in Edmonton, AB, and the University of Calgary in Calgary, AB. Maternal and neonatal data included in Chapters 2 and 3 were retrieved from the BC Perinatal Data Registry. DNA methylation data included in Chapters 4 and 5 were analyzed at the UBC Department of Medical Genetics. All projects and associated methods were approved by the UBC \/ Children's and Women's Health Centre of BC Research Ethics Board (certificates H17-03141 and H15-00820).  A version of Chapter 2 has been published. Maria F Mujica-Coopman, Amy Tan, Theresa H Schroder, Graham Sinclair, Hilary D Vallance and Yvonne Lamers. Serum betaine and dimethylglycine are higher in South Asian compared with European pregnant women in Canada, with betaine and total homocysteine inversely associated in early and mid-pregnancy, independent of ethnicity. J Nutr. 2019 Dec 1;149(12):2145-2155. I was responsible for design, and data analysis, as well I wrote the first version of the manuscript. Dr. Theresa H Schroder, former PhD student in Dr. Yvonne Lamers\u2019 lab, and I with assistance by staff members and undergraduate research students performed the sample analyses. Amy Tan led the acquisition of maternal and neonatal data from the BC Perinatal Data Registry. Dr. Graham Sinclair, Dr. Hilary D Vallance and Amy Tan contributed to data interpretation and manuscript revisions. Dr. Yvonne vii  Lamers was the supervisory author on this project and was involved throughout the project in study design, concept formation, method development, and manuscript composition. Preliminary results of this chapter were presented at the 11th International Conference on Homocysteine & One-Carbon Metabolism, Aarhus, Denmark (2017), and at the BC Children\u2019s Hospital Research Institute Research Trainee Day, Vancouver, Canada (2018).  A version of Chapter 3 is being prepared for submission to a peer-reviewed journal. I was responsible for the study design, and data analysis, as well I wrote the first version of the manuscript. Dr. Theresa H Schroder, former PhD student in Dr. Yvonne Lamers\u2019 lab, and I with assistance by staff members and undergraduate research students performed the biomarker analyses. Amy Tan led the acquisition of maternal and neonatal data from the BC Perinatal Data Registry. Amy Tan, Dr. Graham Sinclair, Dr. Andre Mattman, and Dr. Hilary D Vallance contributed to data interpretation and manuscript revisions. Dr. Yvonne Lamers was the supervisory author on this project and was involved throughout the project in concept formation, method development, and manuscript composition. Preliminary results of this chapter were presented at the Folic Acid, Vitamin B12, and One-Carbon Metabolism FASEB Science Research Conference, Halifax, Canada (2018), at the Nutrition 2018 Annual Meeting of the American Society for Nutrition, Boston, United States (2018), and at the Canadian National Perinatal Research and DoHAD Meeting, Mont-Tremblant, Canada (2019).  For Chapter 4, I was the lead investigator and I lead the data analysis. Dr. Catherine Field (University of Alberta), Dr. Rhonda Bell (University of Alberta), Dr. Nicole Letourneau (University of Calgary) and Dr. Deborah Dewey (University of Calgary) of the Alberta Pregnancy Outcomes and Nutrition (APrON) Team were co-investigators in this project. Staff in Dr. Catherine Field\u2019s lab had previously performed the folate biomarker analysis in the Li Ka viii  Shing Centre for Research at the University of Alberta. DNA methylation analyses had been conducted by staff at Dr. Michael Kobor\u2019s lab in the UBC Department of Medical Genetics. Maggie Fu, a PhD student in Dr. Michael Kobor\u2019s lab, contributed with compilation and analysis of DNA methylation data. Dr. Yvonne Lamers was the supervisory author on this project and was involved throughout the project in concept formation, method development, and chapter composition.  For Chapter 5, I was the lead investigator and I lead the data analysis. Dr. Catherine Field, Dr. Rhonda Bell, Dr. Nicole Letourneau and Dr. Deborah Dewey of the APrON Team were co-investigators in this project. Staff in Dr. Yvonne Lamers\u2019 lab and I performed the B-6, riboflavin, choline, betaine and related biomarker analysis. Serum total vitamin B12 was measured at the BC Children's Hospital Pathology Laboratory. Biomarker data for folate and holotranscobalamin had been previously collected by staff in Dr. Catherine Field\u2019s lab in the Li Ka Shing Centre for Research at the University of Alberta. DNA methylation analyses had been conducted by staff at Dr. Michael Kobor\u2019s lab in the UBC Department of Medical Genetics. Dr. Yvonne Lamers was the supervisory author on this project and was involved throughout the project in concept formation, method development, and chapter composition. Preliminary results of Chapter 5 were presented at the 12th International Conference on One-Carbon Metabolism, B-Vitamins and Homocysteine, Tarragona, Spain (2019).  Dr. Catherine Field (University of Alberta), Dr. Zhaoming Xu (UBC Human Nutrition Program) and Dr. Patricia Janssen (UBC School of Population and Public Health) served in their role as supervisory committee members and contributed to the review of my project proposal, the interpretation of project outcomes, and the review of this thesis.   ix  Table of Contents  Abstract .......................................................................................................................................... iii Lay Summary .................................................................................................................................. v Preface ............................................................................................................................................ vi Table of Contents ........................................................................................................................... ix List of Figures ............................................................................................................................. xvii List of Abbreviations .................................................................................................................. xviii Acknowledgements .................................................................................................................... xxiii Dedication ................................................................................................................................... xxv Chapter 1: INTRODUCTION, LITERATURE REVIEW, AND OBJECTIVES AND HYPOTHESES .............................................................................................................................. 1 1.1 Introduction ...................................................................................................................... 1 1.2 Literature review .............................................................................................................. 4 1.2.1 B-vitamins, one-carbon metabolism, and methylation pathways ................................ 5 1.2.2 Maternal methyl nutrient concentrations during pregnancy ...................................... 10 1.2.2.1 Interaction between folate-B12 and betaine dependent homocysteine remethylation pathways during pregnancy ........................................................................ 14 1.2.3 Role of methyl nutrients in early stages of pregnancy ............................................... 17 1.2.4 Suggested role of DNA methylation in embryonic and fetal programming .............. 22 1.2.4.1 Methyl nutrients, DNA methylation and obesity programming ........................ 25 1.2.4.2 Obesity related genes ......................................................................................... 27 1.2.4.2.1 Retinoid X-receptor \u03b1 gene .......................................................................... 27 x  1.2.4.2.2 Leptin gene ................................................................................................... 28 1.2.4.3 Hypoxia inducible factor gene ........................................................................... 29 1.2.5 Maternal methyl nutrients and related metabolites and their relationship with neonatal anthropometric outcomes ......................................................................................... 31 1.2.5.1 Indicators of neonatal anthropometric outcomes ............................................... 31 1.2.5.2 Betaine and neonatal anthropometric outcomes ................................................. 33 1.2.5.3 Choline and neonatal anthropometric outcomes ................................................ 34 1.2.5.4 Folate and folic acid and neonatal anthropometric outcomes ............................ 36 1.2.5.5 Homocysteine and neonatal anthropometric outcomes ...................................... 42 1.2.5.6 Vitamin B-12, B-6, riboflavin and neonatal anthropometric outcomes ............. 43 1.3 Summary and rationale ................................................................................................... 51 1.4 Research objectives and hypothesis ............................................................................... 52 Chapter 2: MATERNAL SERUM BETAINE CONCENTRATION AND ITS RELATIONSHIP WITH TOTAL HOMOCYSTEINE CONCENTRATION IN EUROPEAN AND SOUTH ASIAN PREGNANT WOMEN ................................................. 56 2.1 Introduction .................................................................................................................... 56 2.2 Study design and methods .............................................................................................. 58 2.2.1 Sample size ................................................................................................................. 59 2.2.2 Maternal demographic, obstetric and neonatal information ....................................... 60 2.2.3 Biochemical analyses ................................................................................................. 61 2.2.3.1 Blood processing ................................................................................................ 61 2.2.3.2 Biomarker quantitation ....................................................................................... 61 2.2.4 Statistical analyses ...................................................................................................... 63 xi  2.3 Results ............................................................................................................................ 64 2.3.1 Maternal characteristics .............................................................................................. 64 2.3.2 Betaine, DMG, and related metabolite concentrations across weeks of gestation and by ethnicity ............................................................................................................................. 64 2.3.3 Association between maternal betaine and total homocysteine concentrations in early pregnancy ............................................................................................................................... 70 2.4 Discussion ...................................................................................................................... 72 Chapter 3: FIRST- AND SECOND-TRIMESTER MATERNAL METHYL NUTRIENT PATTERNS AND THEIR ASSOCIATION WITH NEONATAL ANTHROPOMETRIC OUTCOMES IN CANADIAN MOTHER-NEWBORN DYADS ........................................... 78 3.1 Introduction .................................................................................................................... 78 3.2 Study design and methods .............................................................................................. 80 3.2.1 Maternal and neonatal information ............................................................................ 80 3.2.1.1 Maternal data ...................................................................................................... 80 3.2.1.2 Neonatal data ...................................................................................................... 81 3.2.2 Biochemical analyses ................................................................................................. 82 3.2.2.1 Biomarker quantitation ....................................................................................... 82 3.2.3 Statistical analyses ...................................................................................................... 82 3.2.3.1 Principal components analysis ........................................................................... 83 3.3 Results ............................................................................................................................ 89 3.3.1 Maternal characteristics ................................................................................................. 89 3.3.1 Neonatal characteristics .............................................................................................. 89 3.3.2 Maternal methyl nutrient concentrations .................................................................... 90 xii  3.3.3 Maternal methyl nutrient patterns .............................................................................. 94 3.3.4 Association between first-trimester maternal methyl nutrient pattern scores and neonatal anthropometric outcomes ......................................................................................... 99 3.3.5 Association between second-trimester maternal methyl nutrient pattern scores and neonatal anthropometric outcomes ......................................................................................... 99 3.4 Discussion .................................................................................................................... 104 Chapter 4: COMPARISON BETWEEN DNA METHYLATION LEVELS IN CPG SITES OF FETAL GROWTH AND OBESITY-RELATED GENES IN 3-MONTH OLD INFANTS OF MOTHERS WITH RBC FOLATE CONCENTRATION WITHIN AND ABOVE THE REFERENCE RANGE IN EARLY PREGNANCY ..................................... 110 4.1 Introduction .................................................................................................................. 110 4.2 Study design and methods ............................................................................................ 112 4.2.1 Maternal lifestyle, demographic and anthropometric information ........................... 113 4.2.2 Maternal supplement intake ..................................................................................... 113 4.2.3 Infant demographic and anthropometric information .............................................. 114 4.2.4 Infant DNA methylation information ....................................................................... 114 4.2.5 Biochemical analyses ............................................................................................... 116 4.2.5.1 Blood processing .............................................................................................. 116 4.2.5.2 Biomarker quantification .................................................................................. 116 4.2.6 Statistical analysis .................................................................................................... 116 4.3 Results .......................................................................................................................... 119 4.3.1 Maternal characteristics ............................................................................................ 119 4.3.1 Infant characteristics ................................................................................................ 120 xiii  4.3.2 Comparison of DNA methylation levels of CpG sites associated with fetal growth and obesity-related and DNA methyltransferase genes in offspring by maternal RBC folate concentration ........................................................................................................................ 122 4.4 Discussion .................................................................................................................... 125 Chapter 5: RELATIONSHIP BETWEEN MATERNAL METHYL NUTRIENT PATTERNS IN EARLY PREGNANCY AND INFANT DNA METHYLATION LEVELS OF FETAL GROWTH, OBESITY-RELATED GENES AND METHYLTRANSFERASE CODING GENES. ..................................................................................................................... 131 5.1 Introduction .................................................................................................................. 131 5.2 Study design and participants ....................................................................................... 133 5.2.1 Biochemical analyses ............................................................................................... 133 5.2.1.1 Blood processing .............................................................................................. 133 5.2.1.2 Biomarker quantification .................................................................................. 133 5.2.2 Statistical analysis .................................................................................................... 135 5.2.2.1 Principal component analysis ........................................................................... 135 5.3 Results .......................................................................................................................... 138 5.3.1 Maternal characteristics ............................................................................................ 138 5.3.2 Maternal methyl nutrient patterns ............................................................................ 140 5.3.1 Infant characteristics ................................................................................................ 142 5.3.1 Association of maternal methyl nutrient patterns and DNA methylation levels of fetal growth and obesity-related genes in the offspring ....................................................... 142 5.4 Discussion .................................................................................................................... 144 Chapter 6: GENERAL DISCUSSION, CONCLUSIONS AND FUTURE DIRECTIONS 152 xiv  6.1 General discussion of the key findings ........................................................................ 152 6.1.1 Summary of my Ph.D. research work ...................................................................... 152 6.2 Discussion of key findings ........................................................................................... 156 6.2.1 Maternal betaine concentration and its relationship with tHcy concentration in a folate-replete population ...................................................................................................... 156 6.2.2 Early-pregnancy methyl nutrient patterns are characterized by betaine and B-12 and riboflavin related metabolites in Canadian pregnant women ............................................... 159 6.2.3 Early-pregnancy methyl nutrient patterns were associated with HC, while DNA methylation levels of fetal growth and obesity-related genes in the infants were not associated with maternal methyl nutrient patterns or maternal folate status at early pregnancy ............................................................................................................................. 165 6.3 Overall strengths and limitations .................................................................................. 167 6.4 Future research directions ............................................................................................ 171 6.4.1 Periconceptional methyl nutrient status in women and DNA methylation levels of fetal growth and obesity-related genes in the offspring ....................................................... 171 6.4.2 Obesity programming and its consequences later in life .......................................... 172 6.4.3 Paternal methyl nutrient concentrations and DNA methylation in the offspring ..... 173 6.4.4 Derivation of maternal methyl nutrient patterns in non-folate-replete populations . 174 6.5 Significance and contribution of research .................................................................... 175 References .................................................................................................................................. 177 Appendices ................................................................................................................................. 217 A 1. Maternal and neonatal information obtained from Prenatal Genetic Screening Program and British Columbia Perinatal Data Registry ......................................................................... 217 xv   List of Tables Table 1.1 Longitudinal studies reporting on the relationship between maternal methyl nutrient concentrations and neonatal anthropometric outcomes ................................................................. 47 Table 1.2 Longitudinal studies reporting on the relationship between maternal methyl nutrient concentrations and DNA methylation of fetal growth and obesity related genes in the newborns 50 Table 2.1 Maternal and neonatal characteristics of apparently healthy European and South Asian pregnant women and their offspring1 ............................................................................................. 65 Table 2.2 Serum biomarker concentrations in the first and second trimester of pregnancy by ethnicity1 ......................................................................................................................................... 69 Table 2.3 Association of maternal serum betaine and total homocysteine concentrations across weeks of gestation in 495 apparently healthy European and South Asian pregnant using longitudinal regression models1 ..................................................................................................... 71 Table 3.1 Newborn characteristics1 ................................................................................................ 91 Table 3.2 First and second-trimester methyl nutrient and related metabolite concentrations of apparently healthy Canadian pregnant women in the total sample and in those women included in the first or second-trimester PCA1 ................................................................................................. 93 Table 3.3. Loadings of methyl nutrients for the first and second-trimester maternal methyl nutrient patterns .............................................................................................................................. 95 Table 3.4. Characteristics of apparently healthy Canadian pregnant women included in first or second trimester PCA1 ................................................................................................................... 97 Table 3.5 Characteristics of newborns of pregnant women included in the first or second trimester PCA 1 .............................................................................................................................................. 98 xvi  Table 3.6 Associations of maternal methyl nutrient patterns in the first and second trimester of apparently healthy Canadian pregnant women with neonatal anthropometric outcomes ............ 101 Table 3.7 Associations of maternal methyl nutrient patterns in the first and second trimester of apparently healthy Canadian pregnant women with neonatal anthropometric outcomes using Kramer\u2019s growth charts ................................................................................................................ 103 Table 4.1 CpG sites associated with fetal growth and obesity-related genes included in this study ...................................................................................................................................................... 115 Table 4.2 Maternal characteristics of the total sample of pregnant women and those with RBC folate concentration \u22641360 nmol\/L and RBC folate concentration >1360 nmol ......................... 120 Table 4.3 Infant characteristics of the total sample of infants and those of pregnant women with RBC folate concentration \u22641360 nmol\/L and RBC folate concentration >1360 nmol ................ 121 Table 4.4 Comparison of DNA methylation levels of CpG sites associated with fetal growth and obesity-related genes in infants of mothers with RBC folate \u22641360 nmol\/L and RBC folate >1360 nmol\/L at early pregnancy1 .......................................................................................................... 123 Table 5.1 Maternal characteristics of all women with infant DNA methylation data by tissue and those included in the PCA1........................................................................................................... 139 Table 5.2 Maternal biomarker concentrations of all women with infant DNA methylation data by tissue and those included in the PCA1 .......................................................................................... 141 Table 5.3 Loadings of methyl nutrients for early-pregnancy methyl nutrient patterns ................ 141 Table 5.4 Infants characteristics of offspring with DNA methylation data by tissue and those included in the PCA1 .................................................................................................................... 143 xvii  List of Figures Figure 1.1 Schematic of one carbon metabolism ............................................................................. 9 Figure 1.2 Schematic of pregnancy stages and DNA methylation  ................................................ 21 Figure 2.1 Maternal serum betaine concentration across weeks of gestation in European and South Asian pregnant women. ....................................................................................................... 66 Figure 2.2 Maternal serum DMG concentration across weeks of gestation in European and South Asian pregnant women ................................................................................................................... 66 Figure 2.3 First and second-trimester betaine concentration in European and South Asian pregnant women. ............................................................................................................................ 67 Figure 2.4 First and second trimester DMG concentration in European and South Asian pregnant women. ........................................................................................................................................... 68 Figure 3.1  Scree plot of eigenvalues of derived first-trimester principal components ................. 85 Figure 3.2 Scree plot of eigenvalues of derived second-trimester principal components .............. 85 Figure 3.3. Flowchart showing serum samples analyzed and sequence of biomarkers determined for each trimester of pregnancy ...................................................................................................... 92 Figure 3.4. Retained maternal methyl nutrient patterns in the first trimester ................................ 96 Figure 3.5. Retained maternal methyl nutrient patterns in the second trimester ............................ 96 Figure 5.1 Scree plot of eigenvalues of derived early-pregnancy principal components ............ 137 xviii  List of Abbreviations AGA Appropriate for gestational age  AGRP  Agouti-related protein APrON study Alberta Pregnancy Outcomes and Nutrition study  B-6  Vitamin B-6  B-12  Vitamin B-12 BC  BADH British Columbia  Betaine aldehyde dehydrogenase   BCPDR British Columbia Perinatal Data Registry  BHMT  Betaine-homocysteine S-methyltransferase BL  Birth length BMI  BECs Body mass index  Buccal epithelial cells  BW  Birth weight  CB  Cord blood CBS  Cystathionine \u03b2-synthase CGL  CH3  CHDH   Cystathionine \uf067-lyase  Methyl group   Choline dehydrogenase CpG  5'-C-phosphate-G-3'site CV  DHF  Coefficient of variation  Dihydrofolate  DHFR Dihydrofolate reductase xix   DMG   Dimethylglycine  DMGDH Dimethylglycine dehydrogenase   DMR  Differentially methylated region DNA  Deoxyribonucleic acid DNMT-1  DNA methyltransferase 1 DNMT-3A  DNA methyltransferase-3A  DNMT-3B  DNA methyltransferase-3B DNMT-3L  DNA methyltransferase-3L  DNMT-1o  DOHaD DNA methyltransferase-1o  Developmental Origins of Health and Disease   FAD  FAS  Flavin adenine dinucleotide   Fatty acid synthase FGR  fMet-RNA Fetal growth restriction  Formylmethionyl-tRNA  FMN  Flavin mononucleotide  FSA Forward Sortation Area  FTHFS  Formate-tetrahydrofolate ligase  GA  Gestational age GCS Glycine cleavage system  GD  Gestational diabetes  GNMT Glycine N-methyltransferase  GUSTO Growing Up in Singapore Towards Healthy Outcomes study xx  HC  HIF Head circumference  Hypoxia inducible factor  HoloTC  Holotranscobalamin HTN Hypertensive disorder of pregnancy  IGF-1  Insulin growth factor-1 IGF-2  Insulin growth factor-2 IQR  Kg Interquartile range  Kilogram  LBW  Low birth weight   LC-MS\/MS Liquid chromatography-tandem mass spectrometry  LEP   Leptin gene LGA  Large for gestational age Mg  Milligram  MoBa  Norwegian Mother and Child Cohort Study MUAC mid\u2013upper arm circumference  MS Methionine synthase  \u03b1-MSH  Melanocyte-stimulating hormone 5-MTHF  5-methyltetrahydrofolate MTHFR  Methylenetetrahydrofolate reductase NADP   NADPH  NEST Nicotinamide adenine dinucleotide phosphate  Nicotinamide adenine dinucleotide phosphate hydrogen  Newborn Epigenetics STudy  xxi  NTD  Neural tube defect OR Odds ratio  PC  Phosphatidylcholine PCA Principal component analysis  PCA 1 First trimester principal component analysis  PCA 2 Second trimester principal component analysis   PE Phosphatidylethanolamine  PEMT Phosphatidylethanolamine-N-methyltransferase  PI  Ponderal index   PL Pyridoxal  PLP  Pyridoxal 5\u2019-phosphate PMNS  POMC Pune Maternal Nutrition Study  Pro-opiomelanocortin  PREFORM  PREnatal Folic acid exposure on DNA methylation study PSBC Perinatal Services BC R Methyl group acceptor   RBC  Red blood cell RXRA Retinoid X-receptor \u03b1  SAH  S-adenosylhomocysteine SAM  S-adenosylmethionine  SD  Standard deviation  SDH  SE Sarcosine dehydrogenase  Standard error xxii    SGA  SHMT1  SHMT2  SOGC Small for gestational age  Cytosolic serine hydroxymethyltransferase  Mitochondrial serine hydroxymethyltransferase  Society of Obstetricians and Gynaecologists of Canada  SWS UK Southampton Women\u2019s Survey  tHcy  Total homocysteine THF  Tetrahydrofolate Total B-12  TS Total vitamin B-12   Thymidylate synthase  UK United Kingdom   U.S.A.  United States of America WHO  Wks  World Health Organization  Weeks   xxiii  Acknowledgements Firstly, I would like sincerely to thank my supervisor, Dr. Yvonne Lamers for providing me endless opportunities during the past 5 years. I will be forever grateful for receiving your guidance and mentorship since we started working together for my internship in 2014. I really admire your passion for nutrition and, I deeply admire your human capacity, always trying to support my progress and development of my research and human skills, and I feel completely lucky to have you as my supervisor. You have been a source of inspiration and a wonderful role model, and I can say that I have become a better researcher and person after working with you. Thank you so much for always believing in me, even in the most challenging moments. Also, I am very grateful to my committee member, Dr. Catherine Field, who has encouraged me to follow my passion and interest in research, always providing positive comments and feedback. Despite your numerous commitments, you have always made time to talk and provide me guidance about future opportunities. I feel deeply lucky to have the opportunity to receive your guidance and work with you.  I would also like to thank the members of my supervisory committee members, Dr. Zhaoming Xu and Dr. Patricia Janssen for their support and guidance throughout my graduate studies. I am very grateful for the continuous dedication and encouragement of all the members of my supervisory committee.  I feel very lucky for working with all the students, research assistants, and technicians in Dr. Yvonne Lamers\u2019 Lab. I would like to extend special thanks to my fellow trainees and friends Amy Tan, Zubair, Amy McMahon, Larisse, Ricky, Clementine, and Nadia for their friendship, constant support, encouragement and for making a really good experience to work with you during the last years (The best collaborative and supportive team in the world!).  xxiv    I would also like to acknowledge the funding sources for my studies including the 4-Year Doctoral Fellowship (4YF), the Graduate Student Awards, and the Nutritional Research Fellowship from the University of British Columbia. Also, I would like to thank to the University of British Columbia, the Faculty of Land and Food Systems, the BC Children's Hospital Research Institute and the Women\u2019s Health Research Cluster for the travel awards. Lastly, special thanks are owed to my wonderful family and friends including my mother, Gabriela, Horacio, and Fresia who never stopped believing in me. To Sandra, for her constant support, patience, and for being unconditionally present during these years. Also, I would like to thank my \u201cCanadian family\u201d Alejandra, Jose Miguel, Javiera, Mari and Pascal for supporting me all this time, and for their generous friendship and love. Lastly, I would like to thank my unconditional Chilean friends, Trinidad, Tamara, Pablo, Andrea, Francisco, Fernando, Camilo for their love and bringing all the happiness to my life.             xxv  Dedication      Para Gabriela, Sandra y Paloma    1  Chapter 1: INTRODUCTION, LITERATURE REVIEW, AND OBJECTIVES AND HYPOTHESES  1.1 Introduction  Pregnancy is characterized as a vulnerable period of life due to the increased rate of cellular differentiation and proliferation (1), where maternal factors, including nutritional status, have a crucial role in ensuring a healthy pregnancy and optimal fetal development (2,3). Maternal B-vitamin adequacy, for example, adequate folate concentration, has been associated with a lower risk of adverse pregnancy outcomes such as small for gestational age (SGA) and preeclampsia (4\u20136). B-vitamins such as folate, riboflavin, vitamin B-6 (B-6) and vitamin B-12 (B-12) participate, with betaine, choline, and methionine, in one-carbon metabolism. In one-carbon metabolism, methyl groups and other one-carbon units are transferred through the folate and methionine cycles to ultimately contribute to nucleotide formation and cellular proliferation (7). Additionally, methylation reactions are central to one-carbon metabolism as is the remethylation of homocysteine, an intermediary metabolite of one-carbon metabolism, to methionine, and the subsequent formation of S-adenosyl methionine (SAM), the universal methyl donor, which is key for deoxyribonucleic acid (DNA) methylation (7). Due to their interdependent role in one-carbon metabolism and methylation reactions, the aforementioned nutrients are often described as methyl nutrients.  The essential role of methyl nutrients, for example that of folate, in the early stages of pregnancy (i.e., <20 weeks of gestation) has been described since the 1990s. Findings of two clinical trials published in the early 1990s showed that periconceptional supplementation with 4000\u00b5g of folic acid\/day, which is the synthetic form of folate, from at least 1 month before until  2  2 months after conception, reduced the risk of a neural tube defects (NTD) recurrence in 70% in pregnant women with prior NTD-affected pregnancy (8), whereas periconceptional folic acid supplementation with a dose of 800\u00b5g\/day reduced the primary occurrence of NTDs by nearly 70% (9). In 1998, the Institute of Medicine recommended that all women capable of becoming pregnant consume 400\u00b5g of folic acid\/day from supplements or fortified foods (10). The current recommendation for periconceptional folic acid supplementation in Canada states that all women with a low risk of NTD capable of being pregnant should consume 400\u00b5g of folic acid\/day from 2-3 months before conception until lactation (11). However, the most common dose of folic acid in prenatal supplements available on the market is 1000\u00b5g of folic acid (12,13). The high dose and prolonged use of prenatal supplements have led to high folate status in Canadian childbearing aged (14,15) and pregnant (16) women. It has been suggested that the use of high-dose folic acid supplements throughout pregnancy may be associated with adverse health outcomes in the offspring such as dermatitis, asthma or allergic diseases; however, the currently available evidence is only associational and not conclusive (17).  Potential mechanisms underlying the association between prenatal folic acid supplementation and pregnancy outcomes, for example the prevention of NTDs, are not fully understood. Given the role of folate as the key methyl donor, it has been suggested that folate plays an essential role in the provision of methyl groups for DNA methylation reactions and thereby in the programming of the neural tube closure (18,19). DNA methylation is one of the most stable epigenetic mechanisms, which has a crucial role in gene expression (18). Alterations of DNA methylation levels during early stages of development (e.g., in gametogenesis and embryogenesis stages) may lead to long-term health consequences in the offspring (20,21). Recent studies have suggested that DNA methylation of fetal growth and obesity related-genes  3  may underlie the relationship between maternal methyl nutrient status and neonatal outcomes (20,22), such as birth weight (BW), birth length (BL), head circumference (HC), and body composition. However, limited evidence is available regarding the relationship of maternal methyl nutrient status and DNA methylation levels of fetal growth and obesity-related genes in the offspring. A prospective study conducted in North American pregnant women showed that red blood cell (RBC) folate concentration, a long-term indicator of folate status, in the second quartile, measured at 12 weeks of gestation was positively associated with DNA methylation levels of the insulin growth factor-2 (IGF-2) gene in the offspring, which is a well-known imprinted gene that plays a key role in fetal growth programming (22). Furthermore, it has been reported that pre-pregnancy and maternal dietary methyl nutrient intake (e.g., folate and betaine) is associated with DNA methylation levels of obesity-related genes such as leptin gene (LEP) in the offspring (23,24); however, the potential relationship between maternal methyl nutrient status and DNA methylation levels of obesity-related genes (i.e., LEP) in the offspring is currently unknown. Evidence from pregnancy studies has presented conflicting results of the relationship between individual maternal methyl nutrients (e.g., maternal folate concentration) with neonatal anthropometric outcomes (i.e., BW, BL and HC). Maternal RBC folate concentrations in early (i.e., <20 weeks of gestation) and late (i.e., \u226520 weeks of gestation) pregnancy have been positively associated with BW (25\u201327), whereas only one of several studies has shown a positive association between early-pregnancy plasma folate concentrations and BW (28). Early and late pregnancy total homocysteine (tHcy) concentration, an intermediate metabolite of the remethylation reaction in one-carbon metabolism, has been negatively associated with BW and HC (25,28\u201333). Early and late-pregnancy plasma folate, B-12, and B-6, in the form of pyridoxal  4  5\u2019-phosphate (PLP), concentrations have shown no significant association with neonatal anthropometric outcomes (34\u201338), while late-pregnancy betaine concentration has been negatively associated with BW and neonatal fat mass (39). However, the relationship between combined maternal methyl nutrient status and neonatal anthropometric outcomes has not been explored yet.  Given the interrelated role of methyl nutrients in one-carbon metabolism which is key for nucleotide synthesis and the provision of methyl groups for methylation reactions, a better understanding of the role of early pregnancy methyl nutrient status on DNA methylation levels of obesity and fetal growth-related genes in the offspring and neonatal anthropometric outcomes is still required. This review provides an overview of the one-carbon metabolism and related methylation pathways. In addition, it presents current evidence of the variation of maternal methyl nutrient concentrations throughout pregnancy. Furthermore, this review provides a summary of the role of methyl nutrients in early pregnancy, and the current evidence on the relationship between maternal methyl nutrient diet, maternal methyl nutrient status and DNA methylation of fetal growth and obesity-related genes. Finally, the current research related to the relationship between individual maternal methyl nutrient concentration (e.g., maternal folate concentration) and neonatal anthropometric outcomes is discussed.   1.2 Literature review This literature review aims to provide background information on the role of B-vitamins, methionine, choline and, betaine, also known as methyl nutrients, in one-carbon metabolism, followed by an overview of methyl nutrient concentrations during pregnancy. Furthermore, the literature review presents a summary of the current knowledge of the role of methyl nutrients on  5  fetal development, and evidence of the relationship between maternal methyl nutrients status and DNA methylation levels of fetal growth and obesity-related genes is provided. Finally, the relationship between maternal methyl nutrients status and neonatal anthropometric outcomes is discussed, along with the rationale behind the research project.   1.2.1 B-vitamins, one-carbon metabolism, and methylation pathways Folate, B-12, B-6, riboflavin, betaine and choline, also known as methyl nutrients, have interdependent roles in maternal and fetal metabolism, because of their interrelated functions in one-carbon metabolism. Their interrelated pathways are essential for the formation of phospholipids and protein, DNA and neurotransmitter synthesis, membrane synthesis and cell formation, and methylation reactions including DNA methylation. In brief, folate, B-12 and B-6 are essential for the formation of DNA, and have important roles in cognitive and neurological development (40,41). Additionally, folate, B-12, riboflavin and B-6 participate in protein and lipid metabolism (7,42,43). Choline is an essential nutrient, and a key component of cell membrane structure and neurotransmitter synthesis and lipid metabolism, which makes it a crucial factor in fetal neurodevelopment during early pregnancy (44,45). Furthermore, betaine, a nutrient obtained from choline oxidation and diet, is a methyl donor nutrient and the most important osmolyte in cells playing a protective role in embryonic cells (46\u201348).  Specifically, in addition to other functions, the aforementioned B-vitamins, betaine, and choline participate and interact in multiple one-carbon metabolism reactions, which are compartmentalized in the nucleus, mitochondria and cytosol of human cells (49). In the cytosol, the one-carbon metabolism reactions are part of interdependent biosynthetic pathways, which have as primary functions 1) nucleotide synthesis (i.e., purine and thymidylate synthesis), and 2)  6  remethylation of homocysteine to methionine (Figure 1.1). In the cytosol, folate functions as a co-substrate in the transfer of one-carbon units (7,50). Folate enters the one-carbon metabolism primarily as 5-methyltetrahydrofolate (5-MTHF); however, it can also enter as tetrahydrofolate (THF). Folic acid, the synthetic form of folate, requires conversion to dihydrofolate (DHF) and subsequently to THF by the enzyme dihydrofolate reductase upon digestion. THF can be converted to 10-formyl-THF for purine synthesis, or to 5,10-methylene-THF for the conversion of deoxyuridylate to thymidylate. 5,10-methylene-THF is also the precursor of 5-MTHF and is converted to 5-MTHF by the enzyme methylenetetrahydrofolate reductase (MTHFR) with riboflavin serving as cofactor in the form of flavin adenine dinucleotide (FAD) (41).  5-MTHF transfers its one-carbon unit to homocysteine to regenerate methionine, one of the most important methyl donor nutrients. This remethylation reaction requires B-12 as a cofactor (40). Methionine is the precursor of S-adenosylmethionine (SAM), the primary methylation agent in DNA and protein methylation. Variations in intracellular SAM concentration may lead to alterations in DNA methylation and consequently, in the regulation of gene expression (51). SAM is required in over 100 transmethylation reactions forming methylated products and S-adenosylhomocysteine (SAH). The latter is subsequently cleaved to homocysteine, which is either remethylated to methionine or enters the transsulfuration pathway to form cysteine by the action of the enzymes, cystathionine \u03b2-synthase (CBS) and cystathionine \uf067-lyase (CGL), using B-6, in the form of PLP, as a cofactor (42). Furthermore, riboflavin in its flavin mononucleotide (FMN) form is key for the conversion of B-6 into its coenzyme PLP form. Cysteine is a key nutrient for glutathione and taurine synthesis, both of which have a crucial role in intracellular redox balance (52).   7  It has been reported that approximately 40% of cellular folate is located  in the mitochondria (53). The primary functions of one-carbon metabolism in the mitochondria are 1) to generate one-carbon units for the cytosol one-carbon metabolism, 2) to produce glycine, and 3) to produce formylmethionyl-tRNA (fMet-tRNA), which is used for protein synthesis (7,50). In the mitochondria, the folate form THF is converted to 5,10-methylene-THF by several catabolic reactions that involve serine, glycine, and dimethylglycine (DMG), as well as the condensation of THF with formaldehyde (7). After 5,10-methylene-THF is converted to 5,10-methenyl-THF and, subsequently, oxidized to 10-formyl-THF. Later, 10-formyl-THF is converted to fMet-tRNA and formate (7,54), with formate being the source of one-carbon units for one-carbon metabolism in the cytosol (7,55).  In addition to B-vitamins, choline and betaine also participate in several reactions of the one-carbon metabolism. In the liver, free choline can be used for the synthesis of phosphatidylcholine (PC) by the CDP-pathway. Once PC is formed it can be used for the synthesis of sphingomyelin, a critical compound of the membrane in the brain. Furthermore, the liver-enzyme phosphatidylethanolamine-N-methyltransferase (PEMT) can synthesize PC from phosphatidylethanolamine (PE) through three transmethylation reactions using SAM (56). Additionally, free choline can be released from the synthesized PC by the action of the enzymes phospholipase and phosphodiesterase. In the mitochondria, free choline can be oxidized to produce betaine, which is the methyl donor for the folate and B-12 independent remethylation pathway occurring in the liver and kidneys (57). In the mitochondria and cytosol, betaine donates a methyl group to homocysteine, obtaining methionine and DMG as metabolic products, by the action of the betaine-homocysteine S-methyltransferase (BHMT) (44,49), while in the mitochondria, DMG can be converted to sarcosine and subsequently, to glycine by the action of  8  the mitochondrial enzymes dimethylglycine dehydrogenase (DMGDH) and sarcosine dehydrogenase (SDH), respectively (57). In summary, folate, B-12, B-6, and riboflavin, in addition to choline, betaine, and methionine participate in interrelated pathways that are crucial for nucleotide formation, cellular division, tissue differentiation and methylation reactions which make them key nutrients for the provision of methyl groups for stable epigenetic mechanisms such as DNA methylation (42,58,59). 9   Figure 1.1 Schematic of one carbon metabolism  5-MTHF, 5-methyltetrahydrofolate; BADH, betaine aldehyde dehydrogenase; BHMT, betaine-homocysteine methyltransferase; CBS, cystathionine-\u03b2-synthase; CGL, cystathionine gamma-lyase; CH3, methyl group; CHDH, choline dehydrogenase; DHF, dihydrofolate; DHFR, dihydrofolate reductase; DMG, dimethylglycine; DMGDH, dimethylglycine dehydrogenase; DNMT-1, DNA methyltransferase-1; DNMT-3, DNA methyltransferase-3; FAD, flavin adenine dinucleotide; fMet-RNA, formylmethionyl-tRNA; FMN, flavin mononucleotide; FTHFS, formate-tetrahydrofolate ligase; GCS, glycine cleavage system; GNMT, Glycine N-methyltransferase; MS, methionine synthase; MTHFR; methylenetetrahydrofolate reductase; NADP, nicotinamide adenine dinucleotide phosphate; NADPH, nicotinamide adenine dinucleotide phosphate hydrogen; PLP, pyridoxal-5'-phosphate; R, methyl acceptor; SAM, S-adenosylmethionine; SAH, S-adenosylhomocysteine; SDH, sarcosine dehydrogenase; SHMT1, cytosolic serine hydroxymethyltransferase; SHMT2, mitochondrial serine hydroxymethyltransferase; THF, tetrahydrofolate; TS, thymidylate synthase. 10   1.2.2 Maternal methyl nutrient concentrations during pregnancy  Periconceptional folic acid supplementation, defined as folic acid supplementation from at least 1 month before until 2 months after conception, has been recommended to prevent the occurrence of NTDs (60). Results of two clinical trials published in the early 1990s, showed that periconceptional folic acid supplementation (4000\u00b5g\/day) reduced the risk of NTDs in 70% in all pregnant women with a prior NTD-affected pregnancy (8), whereas periconceptional folic acid supplementation (800\u00b5g\/day) reduced the primary occurrence of NTDs by ~100% (9). In 1998, the Institute of Medicine indicated that all women capable of becoming pregnant should consume 400\u00b5g of folic acid\/day from supplements or fortified foods (10). However, in North America, only 50% of pregnant women were reported to consume folic acid supplements in the periconceptional period (61), and 49% had an unplanned pregnancy (62). Because of the early development of the neural tube of the fetus (i.e., the neural tube closes before 4 weeks of gestation), mandatory folic acid fortification (150\u00b5g\/100g) of white flour and pasta was implemented in Canada in 1998 (63) as a public health strategy to increase the consumption of folic acid intake in women.  Longitudinal studies conducted in pregnant women who do not regularly consume folic acid supplements and from countries without mandatory folic acid fortification have shown that plasma and RBC folate concentrations, which are short- and long-term indicators of folate status, respectively, are higher at early pregnancy (i.e., <20 weeks of gestation) compared to late pregnancy (i.e., >20 weeks of gestation)  (64,65). For example, Velzing et al. (65) reported that plasma folate concentration increased from 16.4 nmol\/L at 9 weeks of gestation to 19.6 nmol\/L at 16 weeks of gestation in West African pregnant women. However, it is known that after this early  11  period, plasma, serum and RBC folate concentrations decline across weeks of gestation in late pregnancy, with this change being more pronounced in women who only consume folic acid supplements in the first trimester (64,66).  A different situation has been reported in pregnant women who take folic acid supplements beyond the first trimester (i.e., > 12 weeks of gestation), and reside in countries with mandatory folic acid fortification (e.g., Canadian population). Specifically, in Canada, the Society of Obstetricians and Gynaecologists of Canada (SOGC) recommends that all women who could become pregnant should take folic acid supplements (400\u00b5g\/day) starting 2-3 months before pregnancy until the end of lactation (11); however, it has been reported that most Canadian pregnant women take 1000\u00b5g\/day of folic acid at some time during the first three months of pregnancy, which has been associated with high folate concentration (i.e., RBC folate >1360 nmol\/L) (16,67,68).  A study conducted in pregnant women with all of them reporting prenatal multivitamin supplement use in Vancouver, British Columbia (BC) showed no changes in plasma folate concentration between early (35.9 nmol\/L, 16 weeks of gestation) and late pregnancy (35.8 nmol\/L, 36 weeks of gestation) (36). In pregnant women living in Toronto, RBC folate concentration increased from 2423 nmol\/L at 12-16 weeks of gestation to 2862 nmol\/L at delivery in folic acid supplement users (93%), while RBC folate concentration decreased from 2343 nmol\/L to 2262 nmol\/L in non-folic acid supplement users (7%) (16).  Maternal B-6, measured as PLP, and total vitamin B-12 (total B-12) concentrations continuously decline from the first to the third trimester of pregnancy (36,69\u201373). In Canada, the PREnatal Folic acid exposure on DNA methylation in the newborn infant (PREFORM) study showed that maternal PLP concentration dropped from 107 nmol\/L (95% CI 98-116 nmol\/L) at  12  early pregnancy (i.e., 12-16 weeks of gestation) to 58 nmol\/L (95% CI 53-62 nmol\/L) at delivery (37-42 weeks of gestation) (71), while maternal total B-12 concentration decreased from 219 pmol\/L (95% CI 210-229 pmol\/L) to 169 pmol\/L (95% CI 162-176 pmol\/L) (72). Similarly, Wu et al. (36) found that maternal total B-12 significantly declined from 259 pmol\/L (IQR 199-351 pmol\/L) at 16 weeks of gestation to 206 pmol\/L (IQR 150-284 pmol\/L) at 36 weeks of gestation. In a Spanish pregnancy study, maternal total B-12 concentration decreased from 293 pmol\/L in the preconception period to 267 pmol\/L, 230 pmol\/L and 198pmol\/L  at 8, 20 and 32 weeks of gestation, respectively (73).  Similar to PLP and total B-12 concentrations during pregnancy, maternal betaine concentration has also been shown to significantly decrease during early pregnancy (i.e., <20 weeks of gestation) (64,65). Velzing et al. (65) found that maternal betaine concentration decreased from 16.3 \u00b5mol\/L to 10.3 \u00b5mol\/L between 6 to 16 weeks of gestation. Consistently, betaine concentration dropped from ~21 \u00b5mol\/L to ~13 \u00b5mol\/L between <12 to 24-27 weeks of gestation in Spanish pregnant women (64). Data on maternal riboflavin status is limited. To date, only two studies, both conducted in the Netherlands, have reported maternal riboflavin status. In 1995, a pregnancy study in Caucasian women showed that whole blood riboflavin concentration slightly increase during the last weeks of pregnancy (74). Later, Cikot et al. (69) also found that plasma riboflavin concentration increased in late pregnancy (i.e., 32 weeks of gestation). However, further pregnancy studies are still needed to understand changes in plasma and whole blood riboflavin concentrations across trimesters of pregnancy.  Potential mechanisms underlying the decline in maternal folate, betaine, B-6 and total B-12 concentrations across trimesters are likely related to the increased maternal-fetal nutrient transport (75), as well as hemodilution due to an increase of maternal blood volume (1,76). An  13  increase in glomerular filtration has been reported during pregnancy, and this may accelerate folate and betaine catabolism and clearance (77,78). In addition, an increase in activity of a non-specific tissue alkaline phosphatase occurs during pregnancy, which suggests that PLP is converted to pyridoxal (PL), decreasing PLP concentration (79,80). Furthermore, inadequate dietary folate and B-6 intake may also affect folate and B-6 related biomarker concentrations (81\u201383). In Canada, folic acid supplement use has been indicated as the main predictor of folate concentration in childbearing-aged women (15); therefore, the extended use of folic acid supplements throughout pregnancy likely explains the increase of maternal folate status in Canadian pregnant women (16,67,68). In contrast to the pregnancy-related changes of circulating folate, B-6, total B-12 and betaine concentrations, maternal free choline concentration increases across gestational weeks (65). From the totality of water-soluble choline forms, i.e., free choline, phosphocholine and glycerophosphocholine, and lipid-soluble choline forms, i.e., PC and sphingomyelin, only free choline and PC are transported to the fetus. In humans, cord blood free choline concentration was reported to be 3-4 times higher than maternal plasma free choline concentration, which suggests a preferential transport of free choline from the mother to the fetus (84). The increase of maternal free choline concentration throughout gestation may be explained by the stimulated endogenous synthesis of choline. It has been reported that the fetus has a limited capacity to synthesize PC by the CDP-pathway (REF), and is thus dependent on the maternal choline supply. The PEMT coding gene contains estrogen response elements whereby estrogen can modulate PEMT expression (85), which may promote the synthesis of PC under increasing estrogen concentration, such as pregnancy (86). A recent randomized controlled trial showed that in pregnant women consuming a supplemented diet that contained 480 or 930 mg of total choline\/day (i.e., 100 or  14  550 mg of choline chloride plus 380 mg of dietary choline, respectively), their total choline intake, i.e., the sum of the water-soluble, lipid soluble and supplemental forms of choline, was preferentially used for PC synthesis by the PEMT and cytidine diphosphate-choline pathways (87). Subsequently, PC can be converted to free choline, which may partially explain the increasing free choline concentration across trimesters.  1.2.2.1 Interaction between folate-B12 and betaine dependent homocysteine remethylation pathways during pregnancy High maternal plasma concentration of tHcy, the intermediary metabolite of the methionine cycle, in early (i.e., <20 weeks of gestation) and late pregnancy (i.e., >20 weeks of gestation) has been associated with adverse pregnancy outcomes such as preeclampsia (88), SGA (89), premature delivery (90), and impaired neural cell development of the fetus (91).  Maternal plasma folate concentration, as well as folic acid supplementation, are negatively associated with plasma tHcy concentration in early and late pregnancy (65,92). Recently, betaine has become a nutrient of interest due to its role as a methyl donor in the alternative folate-independent homocysteine remethylation pathway, and especially during early fetal development, under low folate concentration (64,93) or when folate-dependent pathways are impaired (94), betaine is used as a methyl donor nutrient. In contrast, it has been suggested that under folate-replete conditions, and due to the ubiquity of the folate-dependent remethylation pathway, folate is the most used methyl donor nutrient for homocysteine remethylation (64,66).  Results from a study conducted in a sample of Spanish pregnant women from a non-folate-replete population, showed that plasma betaine concentration predicted tHcy concentration in pregnant women with low folate concentration (defined as folate concentration <11.4 nmol\/L)  15  during late pregnancy (>24 weeks of gestation) (64). Similar results were shown by Wallace et al. (66) in a study conducted in pregnant women living in the Republic of Seychelles, an archipelago located in the Indian ocean; betaine concentration at 28 weeks of gestation was negatively associated with tHcy concentration only in women with low methionine concentration (defined as methionine concentration <23.8 \u00b5mol\/L at 28 weeks of gestation). Different results were reported for the relationship between maternal betaine and tHcy concentrations in pregnant women from folate-replete populations. A pregnancy study conducted in Vancouver by Wu et al. (36), in which 80% of the participants self-identified as Europeans, showed no significant association between betaine and tHcy concentrations at 16 and 32 weeks of gestation.  Potential explanations for the inverse association between tHcy and betaine concentrations under low folate and methionine concentrations are likely related to the upregulation of BHMT under low folate conditions and tandem-regulation of methionine in BHMT activity (93,95). While, the extended use of folic acid supplementation throughout pregnancy and lactation in Canadian pregnant women (16,68) and the potential subsequent inhibitory effect on the expression of BHMT in the liver (96), as well as the suggested spared use of betaine as a methyl donor (97,98) resulting from high folate status (i.e., RBC folate concentration >1360 nmol\/L) (16,67), may explain the lack of a significant association between betaine and tHcy concentrations in this folate-replete population.  Current knowledge from pregnancy studies of the association of circulating betaine and tHcy concentrations is limited to relatively few pregnancy studies. Most of the studies were conducted in populations from non-mandatory folic acid fortified countries in which ~70% of the participants used periconceptional folic acid supplementation, and thus where pregnant women presumably consume smaller amounts of folic acid-fortified foods, than in countries where  16  mandatory folic acid fortification is implemented and there is a higher compliance of folic acid supplement use in childbearing-aged and pregnant women, and mandatory folic acid fortification is implemented. Additionally, only one study has explored the relationship between maternal betaine and tHcy concentrations at early and late pregnancy in pregnant women from a mandatory folic acid fortified population, specifically in Vancouver, Canada. However, this study only explored the relationship between betaine and tHcy concentrations at 16 weeks of gestation and included a sample of pregnant women of primarily (80%) European ethnicity.  According to Statistics Canada, the South Asian population made up the largest ethnic minority in BC in 2011 (99). A recent study published by Jeruszka-Bielak et al. (100) found that pregnant women of South Asian ethnicity residing in Metro Vancouver had a higher risk of B-12 deficiency compared to those of European ethnicity, in late pregnancy (second or third trimester). Whether low maternal total-B12 concentration impairs the use of folate in the folate-B-12 dependent remethylation pathway of homocysteine to methionine and promotes the use of betaine as a methyl donor is unknown in this population. To date, it is also unknown whether maternal betaine concentration may differ between European and South Asian pregnant women residing in Canada. Results of the 2007-2009 Canadian Health Measures Survey indicated that total B-12 but not folate concentration was a significant predictor of tHcy concentration in adults. However, the national data did not include pregnant women (101). Given the role of betaine as a methyl donor in the remethylation pathway which is critical for the synthesis of SAM, the universal methyl donor, further pregnancy studies exploring the relationship between early-pregnancy betaine and tHcy concentrations in populations with high folate status are needed to understand if betaine becomes a key nutrient for the remethylation of homocysteine to methionine in folate-replete populations.   17   1.2.3 Role of methyl nutrients in early stages of pregnancy In the first trimester of pregnancy (Figure 1.2), a crucial physiological process - the differentiation of three embryonic germ layers (i.e., endoderm, mesoderm, and ectoderm) - occurs and this determines the structure and function of fetal organs (1). In addition, from the extraembryonic mesoderm, placenta is formed. In vitro and animal studies have shown that betaine preferentially accumulates in fertilized eggs by a specific transporter, the sodium\/imino-acid transporter-1 (102,103), while the role of betaine as a methyl donor during early stages of development has been confirmed by the presence of BHMT activity during the blastocyst stage in a mouse model (47). In mice, the inhibition of folate intracellular transport increases in utero mortality due to the early damage and failure of hematopoietic organs, and the use of folic acid supplementation prevented the offspring death (104). Additionally, it has been observed that a folic acid deficient diet in mice from embryonic day to 17-19 days impaired mitosis and cellular differentiation, and increased cellular apoptosis in brain (105). Findings of a study conducted in female mice showed that the consumption of a folate deficient diet (i.e., 0mg of folic acid\/kg) before and during gestation was associated with a lower fetal\/placental weight ratio (REF), and inhibited the mTOR intracellular placental pathway (REF), which is associated with cellular growth (REF). Additionally, a maternal folate deficient diet decreased the activity of the placental amino acid transporter (i.e., system A and L) (106), which suggest that maternal folate status is also key for an adequate placenta development.  Furthermore, Lambrot et al. (107) showed that a low paternal dietary folate intake in mice negatively impacted the fertility rate and the implantation of the embryo into the uterus, and increased the presence of anatomical abnormalities such as hydrocephalus in the litter. In  18  addition, results of the study indicated that a folate-deficient paternal diet altered the sperm epigenome at the DNA methylation level. In humans, it is well-known that folate has a vital role at early stages of pregnancy. Periconceptional folic acid supplementation has been shown to significantly decrease the occurrence of NTDs in the offspring (8,9). Furthermore, findings of a study conducted by Daly et al.(108) indicated that women with RBC folate concentration <340 nmol\/L, measured ~15 weeks of gestation, have at least eightfold more risk of NTDs compared to those with RBC folate >906 nmol\/L. Neural tube closure occurs before 4 weeks of gestation; therefore, these findings indicate that folate is involved in early stages of cellular differentiation and nervous system development. Similar to the role of folate, a dietary choline-deficient diet (i.e., 0g chloride choline \/kg from embryonic day to 17-19 day) in pregnant mice decreased cellular proliferation of progenitor cells and significantly increased the apoptosis cellular rate in the brain of the offspring, compared to offspring of pregnant mice that received a choline-supplemented diet (i.e., 4.95g choline chloride\/kg) (109). Additionally, it has been reported that offspring of dams who received a choline-deficient diet (i.e., 0.3g choline\/kg) for 6 weeks had a higher incidence of ventricular septal defects, compared to offspring of dams receiving a control diet (110). Regarding the extra embryonic tissue, choline supplementation (5.6g of choline chloride\/kg diet) during pregnancy has also been shown to reduce the expression of proinflammatory factors such as tumor necrosis factor-\u03b1 and interleukin-8 in a sex-dependent manner, in placenta tissue of murine (111). Additionally, choline supplementation decreased the transcription of the glucose transporter-1 and increased the transcription of the glucose transport-3 and DHA transporters, suggesting that maternal choline supply is important not only for embryonic development, but also for an adequate supply of other critical nutrients, such as DHA, to the embryo and fetus (112).   19  Furthermore, in humans, findings from a case-control study conducted in North American women indicated that lower serum total choline concentration (i.e., <2.71 mmol\/L), which represents the sum of the choline esters converted to free choline, measured between 15-18 weeks of gestation, was associated with a higher risk of NTDs compared to those with higher serum total choline concentration (i.e.,  \u22653.21 mmol\/L)(113), which also suggest a potential critical role of total choline concentration at early stages of development.  Riboflavin also has a key role in early embryogenesis. Embryos from dams fed a riboflavin-deficient diet for 6 weeks had a significantly higher occurrence of heart defects such as ventricular septal defects and lower ventricular wall thickness, compared to those receiving a control diet (110). These results were consistent with the findings of a pregnancy case-control study, in which the mothers of children with congenital heart disease had a significantly lower dietary riboflavin intake (1.32 mg\/day) compared to the control group (1.41 mg\/day) (114). Possible mechanisms underlying the relationship between riboflavin-deficient status and heart abnormalities are associated with the coenzyme role of riboflavin in the forms of FAD and FMN. Both FAD and FMN participate in the mitochondrial electron transport chain. It has been previously reported that alterations in mitochondrial DNA may cause defects in cardiac muscle (115,116). Additionally, DNA methylation may potentially underlie the relationship between riboflavin and embryo development. A recent study conducted by Azzi et al. (117) showed that usual dietary riboflavin intake during the year before pregnancy was positively correlated (r=0.02, P<0.05) with the DNA methylation index of differentially methylated regions (DMR) of ZAC1 gene, a gene involved in fetal growth and metabolism, in leukocytes of cord blood (118,119).  20  In addition to folate, choline, betaine and riboflavin, the essential amino acid methionine is also vital during early stages of embryogenesis. At the 2 cell-embryo stage, 190-350 fmoles\/per embryo\/hour of methionine are up taken by the fertilized egg (3-4% are converted to SAM) while at morula stage this amount increases to 650 fmoles\/embryo\/hours (5% converted to SAM)(120). The decrease in methionine uptake and high homocysteine concentration (by the action of an inhibitor of the methionine cycle) impairs embryo development suggesting that adequate SAM and methionine concentrations are crucial for blastocyst development (120). Similar of the role of methionine, in vitro and animal studies have also shown that B-12 and B-6 are critically needed for early embryonic development. The disturbance of the uptake of B-12 by the genetic deletion of cubilin, a peripheral membrane protein expressed in epithelial cells such as intestinal cells, in mice was shown to result in embryonic lethality (121). Additionally, homozygous cubilin knockout embryos were developmentally delayed and had abnormal organ differentiation (121). Furthermore, findings of a retrospective pregnancy study showed that first-trimester total B-12 concentration was inversely associated with DNA methylation levels measured of angiogenesis-related genes (e.g., VEGF), measured in placenta, which suggest that maternal B-12 may be involved in the programming of placental development(122).  In bovine cells cultured with 250 \u00b5M of pyridoxine, a form of B-6, showed higher blastocyst formation and cell division rate compared to the control (i.e., without pyridoxine) (123). Also, B-6 supplementation significantly decreased the apoptotic rate in early stages of cellular division suggesting that B-6, among the other methyl nutrients, is vital for adequate early fetal development.  21    Figure 1.2 Schematic of pregnancy stages and DNA methylation 1  1Black line represents de novo DNA methylation in embryo stage; blue dotted line represents de novo DNA methylation of the male genome; red dotted line represents de novo DNA methylation of the female genome. 22  1.2.4 Suggested role of DNA methylation in embryonic and fetal programming According to the Barker\u2019s fetal programming theory established in the 1950s, and later expanded to the Developmental Origins of Health and Disease (DOHaD) theory, environmental and maternal lifestyle factors such as maternal nutritional status, maternal smoking, alcohol consumption, drug use, and maternal weight during critical periods of development such as pregnancy, can have long-term health consequences in the offspring (124).  Findings from retrospective cohort studies indicated that low birth weight (LBW) was associated with death related to ischemic heart disease later in life (125,126). Furthermore, offspring of survivors of the Dutch famine in 1944 who were exposed to intrauterine famine (i.e., maternal energy intake ~600 kcal\/day), showed higher rates of obesity and cardiovascular diseases as adults (127). Additionally, offspring of pregnant women who were exposed to famine had a lower BL and higher ponderal index (PI) compared to offspring of pregnant women who were not exposed to famine (128).  While the mechanisms are not fully understood, epigenetic changes are suggested to be involved in fetal programming (20,129). Epigenetic changes refer to heritable DNA modifications, which do not change the DNA sequence (130). Many types of epigenetic mechanisms have been identified including methylation, acetylation, phosphorylation, as well as chromatin modifications (130). Results of the Dutch famine cohort study indicated that offspring who were exposure to intrauterine famine had significantly lower DNA methylation levels in CpG sites associated with IGF-2, a gene involved in fetal growth, compared to offspring that were not exposed to famine during the prenatal period (131). DNA methylation is a complex epigenetic mechanism, which involves the symmetrical incorporation of methyl groups into specific DNA strand sites (5-position of cytosine residues, defined as 5-methyl cytosine or 5mC)  23  within 5\u2019-C-phosphate-G-3\u2019 (CpG) dinucleotides (132). CpG dinucleotides sites from CpG islands are usually unmethylated and are commonly located in the promoter region of the genes and other regulatory regions. Due to their proximity to the promoter region of the genes, CpG islands play a key role regulating gene expression, thus being crucial for tissue and pattern differentiation in early development. There are three different types of CpG sites: low, intermediate and high-density CpG sites. Each of these respond differently to DNA methylation (133). For example, the methylation of high-density CpG sites is not always associated with gene suppression, while the methylation of intermediate density CpG sites may result in gene silencing.  The transfer of the methyl group from SAM, the universal methyl donor in cells obtained from one-carbon metabolism, to a DNA strand is conducted by the enzyme DNA methyltransferase (DNMT) that comes in 3 isoforms, i.e., DNMT-1, DNMT-3A, and DNMT-3B (132). The DNA strand and the methyl group are attached by a covalent bond, which promotes the recruitment of methyl-CpG binding proteins. It has been suggested that the recruitment of CpG binding proteins inhibits gene expression by blocking the binding of transcription factors, as well as by modifying the histone structure.  An adequate DNA methylation process during the first stages of pregnancy is crucial since DNA methylation patterns are preserved across multiple cellular division cycles, regulating gene expression across the lifespan (58). During early stages of development (i.e., fertilization and implantation stages), demethylation and posterior remethylation of DNA strands occurs in the fertilized egg (Figure 1.2). The demethylation process differs between the paternal and maternal gametes. Shortly after fertilization, paternal and maternal gametes undergo active demethylation. It has been indicated that the ten-eleven translocated family of dioxygenase  24  enzymes convert 5mC to 5-hydroxy-methyl cytosine (5hmC), the oxidized form, and these are further converted to 5-formylcytosine and 5-carboxycytosine which undergo base excision repair (134,135) coupled with the action of thymine DNA glycosylase, generating abasic sites and regenerating unmodified cytosines (136,137). It has been shown that maternal gametes are protected from active DNA demethylation by the presence of the germ cell protein PGCs7\/Stella, while in paternal gamete this protein only protects imprinted genes (138). However, more research is needed to fully understand the role of PGCs7\/Stella in the protection of DNA methylation and demethylation at early stages of pregnancy. In maternal and paternal gametes, a passive demethylation process also occurs during the pre-implantation stage, mainly by the nuclear exclusion of DNMT-1 (139). Additionally, DNMT-1 has a lower affinity with 5hmC compared to 5mC, which may promote its exclusion from the nucleus, as well as a passive demethylation process (133).  De novo DNA methylation is conducted exclusively by DNMT-3A and DNMT-3B (18). A third de novo DNMT-3L has been described, apparently playing a role in the activation of DNMT-3A and 3B (140). Subsequent DNA methylation reactions are conducted by DNMT-1. DNMT-1o, a genetic variant of DNMT-1, maintains the parent-specific DNA methylation patterns of imprinted genes during the demethylation and remethylation process, between the one-cell cycle to the eight-cell cycle stage (141). The crucial role of DNMT-1 during early embryogenesis was confirmed in mice where the deletion of DNMT-1 in the gastrulation stage was found to impair embryo development (133).  A recent study of Belgian pregnant women found that global DNA methylation measured in maternal samples (i.e., white blood cells) significantly increased from 6.2% in the first (11-13 weeks of gestation) to 6.7% in the second (18-22 weeks of gestation), and decreased to 6.4% in  25  the third trimester (30-34 weeks of gestation) (142). Additionally, findings from the same study indicated that maternal dietary intake of methionine was positively associated with global DNA methylation (142). Furthermore, maternal dietary folate and total dietary choline intakes (i.e., the sum of free choline, glycerophosphocholine, PC, phosphocholine and sphingomyelin) estimated in the same group of Belgium pregnant women, were positively associated with DNA methylation levels of DNMT-1, measured in buccal cells of 3-mo infants (24). However, to date, there is no available information about the potential relationship between DNA methylation levels of DNMT-1 and DNMT-3 in the offspring and maternal methyl nutrient status at early pregnancy. As previously noted, pregnant women from mandatory folic acid fortified populations have significantly higher folate concentration compared to those from non-mandatory folic acid fortified populations (67,71). The potential excess supply of methyl donors before and in early pregnancy may promote DNA methylation alterations during fetal programming by modifying DNMT-1 and DNMT-3 methylation levels. The findings of cellular studies have indicated that the expression of DNMT-3A and DNMT-3B is regulated by nucleosome methylation (143,144). Thus, further studies exploring the relationship between maternal methyl nutrient status, as well as recommended folic acid use or voluntary multivitamin use, with DNA methylation levels of DNMT genes in mother-newborns dyads from mandatory folic-acid fortified populations are still needed.   1.2.4.1 Methyl nutrients, DNA methylation and obesity programming  B-vitamin imbalance may promote epigenetic changes that may promote the activation of obesity enhancing metabolic pathways. Evidence from animal studies has suggested that early-life exposure (i.e. pre- and post-conception) to methyl nutrient imbalance may modify the  26  phenotype in the offspring. Cooney et al. (145) found that compared to the control group, offspring of mice fed with increased amounts of methyl nutrients [i.e., choline (15 g), betaine (15 g), folic acid (15 mg), B-12 (1.5 mg) and methionine (7.5 g)] during pregnancy had a significantly different coat color and increased DNA methylation levels. These results were confirmed by Waterland et al. (20), who reported that offspring of mice that received a methyl nutrient-supplemented diet from 2 weeks before mating to lactation, showed higher DNA methylation levels of specific CpG sites, as well as a different coat color compared to the controls. Additionally, findings of the same study showed that the determination of DNA methylation levels in tissues derived from the three germ layers, i.e., endoderm (liver), mesoderm (kidney), and ectoderm (brain), of the mice indicated that DNA methylation is programmed in the embryogenesis stage (i.e., <8 weeks of gestation) (20). Although these epigenetic changes occurred in early embryogenesis, their influence on the phenotype, such as coat color, persisted during adulthood (20). Maternal folate-deficient diets, as well as B-12 and sulfur amino acid (mainly methionine) restricted diets in animals, during the pre-pregnancy and prenatal periods, have shown to impact body composition, likely promoting body fat gain and obesity in the offspring (146,147). Maternal dietary methyl nutrient restriction promotes hypo- or non-methylation of CpG sites in liver tissues of offspring sheep, suggesting that DNA methylation may be involved in the expression of body composition-related genes (146). Despite that the fetal programming theory was based on famine or under-nutrition conditions (148), studies in mice have found that high doses of maternal folic acid impact weight gain, fat composition, as well as body weight control-related hormones, such as adiponectin in the offspring receiving a post-weaning control diet (149,150).   27  1.2.4.2 Obesity related genes  1.2.4.2.1 Retinoid X-receptor \u03b1 gene  The gene retinoid X-receptor \u03b1 gene (RXRA) is a gene that has been suggested to be involved in obesity programming (151). It has been previously described that RXRA gene depends of peroxisome proliferator-activated receptors, which are related to insulin sensitivity and adipose tissue metabolism (152,153). In children, adiposity has been related to the DNA methylation levels of RXRAchr9:136355885(151). The Princess Anne Hospital study and the Southampton Women\u2019s Survey (SWS) \u2013 both large cohort studies conducted in the United Kingdom (UK), explored the association of RXRA chr9:136355885 methylation measured in cord blood and adiposity of 9- and 6-year old children, respectively (151). In both cohorts, RXRA chr9:136355885 DNA methylation levels were positively associated with fat mass and percentage fat in children. A recent study conducted with Belgian pregnant women found that maternal dietary folate intake in the third trimester was negatively associated with DNA methylation levels of CpG sites associated with RXRA, measured in the cord blood of the newborns (23), whereas the duration of folic acid supplementation before conception was positively associated with DNA methylation levels in most of the studied RXRA CpG sites (23). Similar results were found in the same study in infants at 6 months of age, where maternal betaine and folate intake before pregnancy was positively associated with RXRA DNA methylation levels measured in buccal epithelial cells (BECs) of the infants (24), which suggest that the early exposure to maternal methyl donors intake (i.e., betaine or folate intake) may be involved in obesity programming of the offspring.    28  1.2.4.2.2 Leptin gene  In addition to the RXRA gene, DNA methylation levels of CpG sites associated with the leptin gene, LEP, has been associated with obesity programming. Leptin is a protein produced mainly by adipose tissue, which was first identified in an obese-mouse model (154). However, it has also been indicated that leptin is produced in the hypothalamus, placenta and stomach. Leptin is one of the most important hormones participating in energy balance and satiety (155). Specifically, leptin stimulates neurons of the arcuate nucleus of the hypothalamus which results in an anorexigenic (i.e., feeding inhibition) effect driven by the melanocyte-stimulating hormone (\u03b1-MSH) and amphetamine-related transcript, which is produced from pro-opiomelanocortin (POMC), and inhibits the production of orexigenic molecules (i.e., feeding inducers) such as peptides neuropeptide Y and agouti-related protein (AGRP) in different areas of the hypothalamus (156). The \u03b1-MSH acts as an agonist of melanocortin-4 receptor which inhibits food intake, while AGRP acts as an antagonist (157). Thus, the leptin-mediated effect on satiety results from its action in anorexigenic and orexigenic pathways. Morash et al. (158) reported that LEP is regulated by dietary intake (i.e., nutrient availability) in the brain. Additionally, it has been described that leptin stimulates lipid oxidation in skeletal tissue, which may promote a negative energy balance (159). Pre-pregnancy dietary folate intake (not reported in the study) in Belgium women has been negatively associated with the DNA methylation levels of the LEP measured in BECs in the infant (24). Additionally, second-trimester maternal folate intake [mean 263.2 (standard error 92.3) \u00b5g] was negatively associated with LEP gene methylation measured in cord blood (24), which may suggest that maternal methyl nutrient intake, and potentially, maternal methyl nutrient status are involved in obesity programming.   29  1.2.4.3 Hypoxia inducible factor gene  The expression of the hypoxia-inducible factor (HIF) can be induced by pathological and environmental conditions, for example, cancer and oxygen saturation, respectively. Under adequate oxygen conditions, the expression of the gene HIF is inhibited by molecular mechanisms such as hydroxyl-methylation and ubiquitination. While under low oxygen supply (i.e., hypoxic conditions), the expression of the HIF gene is promoted (160). HIF is activated under hypoxic conditions due to its high sensitivity to oxygen concentrations which leads to vascular and metabolic adaptations (161). Increased adipocyte hypertrophy is one of the most common metabolic alterations associated with obesity. This process leads to an insufficient vascular and oxygen supply, promoting hypoxic conditions in adipose tissue. Additionally, obesity is an important risk factor of obstructive sleep apnea, which is characterized by intermittent episodes of hypoxia (162). Furthermore, results of an animal study indicated that HIF is expressed in the hypothalamus. HIF senses the presence of glucose in the brain and regulate satiety by the POMC pathway, independent of leptin action (163).  In humans, the relationship between HIF and obesity was first described in adults. Among the three isoforms of HIF, (HIF1\u03b1, HIF2\u03b1, and HIF3\u03b1), HIF3\u03b1 has been recognized as a gene potentially related to obesity. A recent large study that included three cohort studies, showed that DNA methylation levels at three specific CpG sites related to HIF3\u03b1 (i.e., cg22891070, cg27146050, and cg16672562) determined in blood were positively associated with body mass index (BMI) in adults. Furthermore, DNA methylation levels of CpG sites related to HIF3\u03b1 were negatively associated with HIF3\u03b1 expression in adipose tissue (164). Results from the Nurses\u2019 Health Study (women aged 30-55 years) and the Health Professionals Follow-Up Study (men aged 40-75 years) showed, for the first time, a significant interaction between DNA methylation  30  levels of the variant HIF3\u03b1 rs3826795 and dietary B-vitamin intake (165). Participants who had lower dietary riboflavin, B-12, and folate intake at the beginning of the study reported a smaller change in BMI after a 10-year follow-up. In contrast, participants with a higher dietary riboflavin, B-12, and folate intake had a greater increase in BMI over time (165).  Evidence of the relationship between maternal factors and DNA methylation levels of HIF in the offspring is limited. A recent study conducted by Mansell et al. (166) showed that gestational diabetes (GD) and weeks of gestation were positively associated with DNA methylation levels of the HIF-3A gene measured in cord blood, whereas preeclampsia and the rs3810298 HIF-3A variant were negatively associated. Whether maternal methyl nutrient concentrations are associated with DNA methylation levels of the HIF-3A gene has not been elucidated to date.  In summary, different genes have been identified as fetal-growth and obesity-related genes. Changes in DNA methylation levels of CpG sites related to these genes have been associated with changes of BMI and body fat in adults and children. IGF-2, LEP, and RXRA genes have shown variation in their DNA methylation levels in the offspring by maternal dietary methyl nutrient intake, whereas HIF-3\u03b1 DNA methylation levels have shown variations by dietary intake only in adults. Given that early pregnancy is considered a critical plasticity window for fetal programming, future pregnancy studies exploring the relationship between early-pregnancy methyl nutrient status, including multiple direct biomarkers of maternal methyl nutrient status, and DNA methylation levels of obesity-related genes measured in the offspring are crucial to understand the potential role of methyl nutrients on obesity programming. Additionally, the inclusion of pregnant women from folate-replete populations is important to  31  explore if the exposure to a potential excess of methyl nutrients (e.g., high folate concentration) in early pregnancy may play a role in obesity programming through DNA methylation.   1.2.5 Maternal methyl nutrients and related metabolites and their relationship with neonatal anthropometric outcomes  1.2.5.1 Indicators of neonatal anthropometric outcomes BW and BL are important predictors of mortality and morbidity in infants (167\u2013169). Evidence from epidemiological studies has shown a positive relationship between BW and BL, and BMI and risk of obesity later in life (170\u2013173). Additionally, BMI in infants at 2 months of age has been indicated as a significant predictor of adiposity at the age of 2 years (174). Findings from a meta-analysis conducted by Yu et al. (173) showed that BW >4000 g was associated with a higher risk of obesity in childhood and adulthood, compared to newborns with a BW <4000g. In addition, a large prospective study in which 4,376 newborns were followed up until 18 years of age, found that BL was positively associated with height at 16 years. Also, BL was positively associated with BMI in females, but not in men (172). Similar results were found in a large Brazilian prospective study (>5,200 participants) in which BL was positively associated with a higher BMI and adiposity at 11 years (175).  In addition to BW and BL, HC has been indicated as a key anthropometric outcome for the tracking of brain growth (176,177). Results of a retrospective study in which HC was measured by ultrasound between 16-24 weeks of gestation showed that having a small HC at the second trimester (i.e., <5th percentile) was associated with a higher risk of perinatal mortality and major malformations such as spina bifida, polycystic kidneys and aortic and mitral atresia (178).  32  Additionally, studies in children have reported that HC is an excellent predictor of brain volume in children between 1.7-6 years (177). Results of the Helsinki study, a prospective study conducted in Finland, indicated that early adiposity rebound, which is defined as the second rise in adiposity after the first year of life, is negatively associated with HC (179) and positively associated with a higher risk of obesity later in life (180,181). Together, these findings suggest that neonatal anthropometric measurements may be predictors for obesity later in life.   Recently, evidence from the INTERGROWTH 21th project (182), a multicenter study which included eight countries, and aimed to develop new standards and reference values for feta, newborn and infant growth, reported that not only BW, BL, and HC are critical anthropometric outcomes to predict obesity later in life, but that BW\/BL ratio (kg\/m) was the best predictor for neonatal fat mass (g), fat-free mass (g) and fat (%) measured by air displacement plethysmography. PI and neonatal BMI were also associated with neonatal fat-free mass and fat mass (183). Furthermore, the relationship between neonatal PI with anthropometric measurements later in life was explored by Pietil\u00e4inen et al.(172) in a large longitudinal study (n=4,376) conducted in Finland. The authors found that neonatal PI was positively associated with height, weight and BMI at 16 years of age.  In summary, neonatal anthropometric outcomes have been shown to predict obesity and adiposity later in life. Early stages of pregnancy are a critical period for fetal programming due to the high rate of de novo DNA methylation. Whether DNA methylation levels of obesity-related genes such as RXRA and HIF-3A in the offspring are associated with neonatal anthropometric outcomes is unknown. Thus, exploring the relationship between DNA methylation levels of obesity-related genes in the offspring and neonatal anthropometric outcomes is key to identify potential factors that may predispose for obesity later in life.  33  1.2.5.2 Betaine and neonatal anthropometric outcomes  Betaine has vital roles as an osmolyte in early embryo development (i.e., <8 weeks of gestation) and as the methyl donor of the \u201calternative\u201d remethylation pathway of homocysteine to methionine in liver and kidneys (47,48,103). Van Lee et al. (39) found that higher maternal betaine concentration measured between 26-28 weeks of gestation was associated with lower neonatal weight (\u03b2=-57.6 g; 95%CI -110, -25.3 g, P=0.03), smaller HC (\u03b2=-0.20 cm; 95%CI -0.38, -20.0 cm, P=0.03), and decreased mid-upper arm circumference (MUAC) (\u03b2=-0.16 cm; 95%CI -0.30, -0.03 cm, P=0.02), after adjustment for maternal plasma folate, free choline and B-12 concentrations (Table 1.1). Also, higher maternal betaine concentration was associated with a higher risk for SGA (39). Similar results were found by Du et al. (184) in a prospective study that included Chinese pregnant women in which higher maternal betaine concentration measured at late pregnancy was associated with a lower BW (\u03b2=-130 g; 95%CI -245, -15.9 g, P=0.03), after adjustment for plasma free choline, DMG, tHcy and methionine concentrations. In contrast, a study including Dutch pregnant women found no association between third-trimester betaine concentration and BW (35).  Potential mechanisms underlying the relationship between maternal betaine concentration and neonatal anthropometric outcomes may be related to the suggested role of betaine as a modulator of the expression of fetal growth-related factors. Findings of an animal study showed that rats supplemented with betaine (10 gr\/kg) during pregnancy had significantly reduced weaning weight, insulin growth factor-1 (IGF-1) expression and serum IGF-1 concentration in the first litter, while an opposite effect was seen in the second litter, in which betaine supplementation significantly increased weaning weight, IGF-1 expression and IGF-1 concentration (185). Another mechanism proposed is the potential regulatory effect of betaine in  34  the lipogenesis pathway. Betaine supplementation (1,000mg of betaine\/kg to the diet) significantly increased fatty acid synthase (FAS) expression, a key enzyme of the lipogenesis pathway, in adipose tissue of broilers (186). In contrast, results reported by Huang et al.(187) showed that the addition of 1250 mg of betaine to a pig\u2019s diet significantly decreased the expression of FAS and acetyl-CoA carboxylase in subcutaneous adipose tissue compared to pigs receiving the control diet. Despite previous studies and emerging evidence showing the inverse relationship between plasma betaine concentration and BMI and body fat in newborns and adults (39,188), the role of early-pregnancy betaine concentration during fetal-growth programming and neonatal anthropometric outcomes such as BW and BL is not fully understood. Findings from animal studies indicated that maternal betaine supplementation may modify methylation patterns in the promoter region of fetal growth and fatty acid metabolism-related genes (186,187); however the relationship of maternal betaine concentration and DNA methylation of fetal growth and obesity-related genes in the offspring needs to be confirmed in human studies (185). Additionally, whether the association of maternal betaine and neonatal and infant anthropometric outcomes depends on other maternal methyl nutrient concentrations such as maternal folate or total B-12 concentrations, needs to be elucidated.  1.2.5.3 Choline and neonatal anthropometric outcomes Choline is the precursor of betaine, the methyl nutrient donor of the alternative methylation pathway. Pregnancy studies exploring the relationship between maternal free choline concentration measured at late pregnancy (i.e., > 26 weeks of gestation) or at delivery have found no significant association between maternal free choline concentration and BW (35,189).  35  However, results of the Growing Up in Singapore Towards Healthy Outcomes (GUSTO) study indicated that maternal plasma free choline concentration at late-pregnancy was positively associated with neonatal BMI z-score, triceps and subscapular skinfolds, while findings of the UK Southampton Women\u2019s Survey showed that maternal serum free choline concentration measured at 11 weeks of pregnancy was positively associated only with subscapular skinfolds (190).  In animal studies, choline has been shown to play a key role in DNA methylation and fetal programming. Choline-deficient diets (i.e.,0g choline chloride) in rats resulted in DNA damage, as well as hypomethylation of global DNA and specific brain development-related genes (191). In addition, changes in maternal total choline intake in rats was shown to modify DNMT-1 expression and IGF-2 methylation in the offspring. In humans, two pregnancy studies have explored the relationship between maternal total dietary choline intake and DNA methylation of fetal-growth related genes in the offspring. A study conducted by Pauwels et al. (142) in Belgian pregnant women found that maternal second-trimester total dietary choline intake was negatively associated with DNA methylation levels of CpG sites related to DNMT-1 (\u03b2=-0.30; 95%CI -0.57,-0.03, P=0.03), while maternal third-trimester total dietary choline intake was positively associated with DNMT-1 DNA methylation levels (\u03b2=0.29; 95%CI 0.10,0.84, P=0.02) measured in the cord blood of newborns (23). In a sample of North American pregnant women, periconceptional total dietary choline intake was negatively associated with global DNA methylation in the cord blood of male newborns; however, this relationship was not significant in female newborns (192), which may suggest that the impact of maternal total dietary choline intake on DNA methylation is sex-dependent.   36  To summarize, our current knowledge on the relationship between maternal free choline concentration and neonatal anthropometric outcomes is limited to a few studies that only determined maternal free choline concentration at late pregnancy. In addition, evidence regarding the relationship between maternal total dietary choline intake and DNA methylation of fetal growth and obesity-related genes in the offspring is only available from one study in Belgium pregnant women. Given that de novo DNA methylation occurs at early stages of pregnancy and the potential role of choline on fetal growth programming (23,24), further studies exploring the relationship between early-pregnancy choline status (e.g., maternal free choline concentration), DNA methylation of fetal growth-related genes and neonatal anthropometric outcomes are still needed. 1.2.5.4 Folate and folic acid and neonatal anthropometric outcomes Folate is the key methyl donor nutrient in all cells and has an important role for methylation reactions and nucleotide synthesis, and subsequently in cellular differentiation and proliferation (41). Pregnancy studies exploring the relationship between maternal folate status and neonatal anthropometric outcomes have shown conflicting results. Current evidence from pregnancy studies conducted in UK (26), Indian (27), and Japanese pregnant women (25), have indicated that maternal RBC folate concentration, an indicator of long-term folate status, measured at early (i.e., <20 weeks of gestation) and late pregnancy (i.e., >20 weeks of gestation) is positively associated with BW z-score and BW, respectively. However, in Italian (30), Dutch (193), and Brazilian pregnant women (194), no relationship was found between RBC folate concentration at early and late pregnancy and BW.  37  In contrast to pregnancy studies that determined maternal RBC folate concentration, only one study has found a significant association between early-pregnancy plasma folate concentration, an indicator of short-term folate status and BW and BL. Bergen et al. (28) showed that low maternal folate concentration (defined as plasma folate concentration \u22649.1 nmol\/L) measured at 13 weeks of gestation was associated with a lower BW (\u03b2=-113.0 g; 95%CI -160, -66.3 g) and BL (\u03b2=-3.3 cm; 95%CI -5.8, -0.8 cm) compared to newborns of pregnant women with high maternal folate concentration (defined as plasma folate concentration \u226525.8 nmol\/L). Furthermore, pregnancy studies that measured maternal plasma folate or serum folate concentrations at late pregnancy (i.e., >20 weeks of gestation) reported no significant association between maternal folate concentration and neonatal anthropometric outcomes (25,30,32\u201334,36\u201338,194\u2013197).  Despite the positive results of the association between early-pregnancy RBC and plasma folate concentration and neonatal anthropometric outcomes, emerging evidence has suggested that not only low maternal folate, but also high maternal folate concentration, may not be beneficial for the offspring development. A study conducted by van Uitert et al. (198) in Dutch pregnant women reported that low maternal folate concentration, defined as RBC folate concentration <1513 nmol\/L measured at 8 weeks of gestation, was inversely associated with embryo growth measured before 13 weeks of gestation (198). Additionally, high RBC folate concentration, defined as RBC folate concentration >1813 nmol\/L, was also inversely associated with embryo growth. Furthermore, results of the same study showed that low and high folate concentrations were associated with lower cerebellar growth, compared to those offspring of pregnant women with RBC folate concentration between 1538\u20131813nmol\/L at early pregnancy (199). Findings of a study conducted in North American pregnant women showed a positive  38  association between early RBC folate concentration (mean 12 weeks of gestation) and BW, however, this relationship was only significant in pregnant women with RBC folate concentration in the second quartile, and not in those with higher RBC folate concentrations (i.e., third and fourth quartiles of RBC folate concentration) (22).  Potential explanations for the relationship between early pregnancy folate status with neonatal anthropometric outcomes and fetal growth may be related to the role of folate, as a methyl donor nutrient, in fetal growth programming by DNA methylation reactions. However, evidence regarding the relationship between early pregnancy folate concentration and DNA methylation of fetal growth-related genes is limited. A meta-analysis including two population-based birth cohort studies conducted in Northern Europe found that maternal plasma folate concentration (measured at < 22 weeks of gestation) was associated with the DNA methylation levels of more than 48 genes (specific CpG sites) assessed in cord blood (200). Furthermore, some of these genes were involved in the occurrence NTDs and brain development of the offspring during pregnancy, which suggest that maternal folate plays a key role in early stages of zygote and blastocyst differentiation. Additionally, one of the studies included in the meta-analysis, The Generation Study, found that maternal plasma folate concentration measured at early pregnancy (median 13 weeks of gestation) in Dutch pregnant women, was positively associated with fetal HC (201). However, in this study, maternal plasma folate concentration [mean 19.1 \u00b11 nmol\/L] was not significantly associated with fetal growth-related DNA methylation genes (202). Similar findings were reported by Haggarty et al. (203) in pregnant women living in the UK in which maternal RBC folate concentration at 19 weeks of gestation [geometric mean 456 (95%CI 442, 471) nmol\/L] was not associated with DNA methylation levels of IGF-2, a well-known gene involved in fetal growth, measured in cord blood (Table 1.2).  39  In contrast, results from the Newborn Epigenetics Study (NEST) study, a prospective cohort study with pregnant women in the United States (U.S.A) with a mean of RBC folate concentration ~494 nmol\/L measured at early pregnancy (mean 12 weeks of gestation) found that only low maternal RBC folate concentration (defined as second quartile folate concentration) was positively associated with the DNA methylation levels of the IGF-2 gene measured in cord blood (22), which may suggest a potential role of folate in fetal growth programming. Although, further studies in pregnant women are needed to replicate the results reported in European and North American pregnant women, and to explore the potential role of folate in fetal growth programming. According to international recommendations for the prevention of NTDs (204,205), women who are capable to conceive should consume a daily dose of 400\u00b5g folic acid until 12 weeks of gestation. In Canada, the SOGC recommends in their Clinical Practice Guidelines that women with a low risk of NTD and capable to conceive should consume \u201c400 \u00b5g folic acid for at least 2 to 3 months before conception throughout the pregnancy, and for 4 to 6 weeks postpartum or as long as breast-feeding continues\u201d (11). The Canadian Maternity Experiences Survey found that 58% of Canadian women took folic acid supplements three months before pregnancy (206), while 84% of pregnant women reported folic acid supplement use throughout pregnancy (207). In Alberta, the APrON study reported that 97%, 95% and 91% of women used folic acid supplements in the first, second and third trimester, respectively (68). In Toronto, the PREFORM study indicated that 90% of the participants consumed folic acid supplements at early pregnancy (i.e., 16 weeks of gestation) (16). Folic acid supplementation has led to high folate status in childbearing-aged and pregnant women (15,16,67). Furthermore, folic acid supplementation and mandatory folic acid fortification have led to the reduction of NTDs occurrences (63). Despite the  40  WHO (204) and the SOGC (11) recommendation of 400\u00b5g\/day of folic acid intake for women with a low risk of NTDs starting before conception, prenatal supplements containing 400-1000 \u00b5g of folic acid are readily available in Canada (13,208).  A recent animal study showed that maternal folic acid supplementation (20 mg\/kg diet) before and during pregnancy decreases the DNMT-3A expression in the brains of the offspring (94). Furthermore, a folic acid supplemented diet (i.e., 20 mg\/kg) before and during pregnancy in female mice was shown to significantly decrease embryo and placenta weight compared to those embryos of female mice that received a control diet (i.e., 2 mg\/kg) (94).   In humans, periconceptional folic acid intake (i.e., folic acid use from at least 4 weeks before conception until 8 weeks of gestations after conception) was positively associated with DNA methylation levels of the IGF-2 gene measured in whole blood of 17-month-old toddlers, while results of the same study showed that DNA methylation levels of the IGF-2 gene was negatively associated with BW (209). Findings of a UK pregnancy study also showed that folic acid supplementation beyond the first 12 weeks of gestation was positively associated with IGF-2 methylation measured in the cord blood of the newborn (203).  In contrast, a prospective study conducted in North American pregnant women reported that DNA methylation levels of the IGF-2 gene in newborns of pregnant women who took prenatal folic acid supplement (i.e., during pregnancy) were significantly lower, compared to newborns of pregnant women not using folic acid supplements in the second and third trimester (210). Results from a randomized placebo-controlled trial in Irish pregnant women showed that newborns from women who consumed folic acid supplements (400\u00b5g\/day) in all trimesters, had significantly lower cord blood DNA methylation levels (%) in IGF-2 compared to those of mothers who took folic acid supplements only the first trimester (211). Potential explanations for  41  the discrepancies found between the observational studies may be related to the use of different specimens for the determination of DNA methylation levels (i.e., whole blood vs. cord blood). The time point of specimen collection (i.e., toddlers vs. newborn), and the duration of the use of folic acid supplementation differed between the studies may also contribute to explain the different results between the studies. Furthermore, the replication of the results of the randomized controlled trial is needed to understand the effect of folic acid supplementation on DNA methylation levels of fetal growth-related genes in the offspring.  In summary, current evidence from a limited number of studies in apparently healthy pregnant women and their offspring, the majority of them conducted in non-mandatory folic acid fortified populations, indicated that early and late maternal RBC folate concentrations are positively associated with BW, while only one study reported a positive association between early-pregnancy plasma folate concentration and BW and BL. However, emerging evidence has indicated that increasing folate status is inversely related with embryo and brain growth at early stages of pregnancy. Given the well-known effect of folic acid fortification and extended use of folic acid supplementation during pregnancy on increasing folate status, future studies are needed to understand the relationship between maternal folate status and neonatal anthropometric outcomes under presumably high folate concentrations. Also, whether potentially high (i.e., RBC folate concentration >1360 nmol\/L) early and late pregnancy folate concentrations are associated with DNA methylation levels of fetal growth or obesity-related genes in the offspring is currently unknown. Thus, further pregnancy studies in folate-replete populations are needed to understand if DNA methylation is an underlying mechanism linking maternal folate status, fetal growth, obesity programming and neonatal anthropometric outcomes.  42  1.2.5.5 Homocysteine and neonatal anthropometric outcomes Folate, in the form of 5-methyl-THF serves as the methyl donor in the folate-B-12 dependent homocysteine remethylation pathway, which emphasizes the role of folate not only as a key nutrient for fetal growth and development, but also for preventing an increase in tHcy concentration (212). Given that homocysteine is a key and central metabolite of the remethylation and transsulfuration pathways (7,42), tHcy concentration is considered as a non-specific functional biomarker for low folate, B-12, B-6 and riboflavin status (213\u2013217). Furthermore, lifestyle factors such as increasing age (218), impaired renal function (219), and smoking (217) have shown a positive association with tHcy concentration in adults. In early pregnancy, elevated tHcy concentration (defined as tHcy concentration \u22658.3\u00b5mol\/L) measured at 13 weeks of gestation was associated with a reduced neonatal HC (\u03b2=-1.6 cm; 95%CI -3.1,-0.1 cm, P<0.05) and weight (\u03b2=-102 g; 95%CI -139,-6.65 g, P<0.05) in newborns of Dutch pregnant women (28). Similarly, in the U.S.A, McCullough et al. (29) found that tHcy concentration between 5.1-6.0 \u00b5mol\/L (i.e., the third quartile of the tHcy concentration) at early pregnancy (mean 12 weeks of gestation) was associated with a lower BW in male (\u03b2=-210 g; SE 102 g, P=0.04) but not female newborns. In Italian pregnant women, an inverse association was found between early-pregnancy tHcy concentrations and BW, in non-users and users of folic acid supplements (30). In Canada, a study conducted by Wu et al. (36) in pregnant women living in BC found no significant association between maternal tHcy concentration, measured at 16th weeks of gestation, and BW.  Additionally, several pregnancy studies have found an inverse association between late pregnancy maternal tHcy concentration and BW (25,30,32). A recent study led by Jiang et al. (31) found that maternal tHcy concentration at delivery was negatively correlated with BW (r=- 43  0.46, P=0.01), BL (r=-0.43, P=0.03), HC (r=-0.40, P=0.03) and neonatal BMI (r=-0.36, P=0.04). Similar results were found in a pregnancy study in Indian women, in which maternal tHcy concentration at 30\u00b12 week of gestation was negatively associated with BW (\u03b2=-0.13 g; 95%CI-0.21,-0.05 g, P=0.001), MUAC (\u03b2=-0.15;95 cm%CI-0.23,-0.07 cm, P<0.001), triceps (\u03b2=-0.12, 95%CI -0.20, -0.03, P=0.006) and subscapular skinfold thickness (\u03b2=-0.12, 95%CI -0.20,-0.04, P=0.006)(33).  Unlike the findings of the relationship between maternal folate status and newborn anthropometric outcomes, results of the majority of pregnancy studies indicated that tHcy concentration at any time point of pregnancy is negatively associated with neonatal anthropometric outcomes (25,28\u201330). To date, available evidence from pregnancy studies indicates no significant relationship between maternal tHcy concentration and DNA methylation levels of fetal growth and obesity-related genes in the offspring (29,202). tHcy concentration is an indicator that may reflect an impaired one-carbon metabolism. Further studies are needed to explore potential mechanisms underlying the inverse relationship between maternal tHcy concentration and neonatal anthropometric outcomes.  1.2.5.6 Vitamin B-12, B-6, riboflavin and neonatal anthropometric outcomes  Considering the role of B-12 in one-carbon metabolism and specifically as a cofactor for the enzyme methionine synthase (MS) that catalyzes the folate-B12 dependent homocysteine remethylation, recent studies have suggested that maternal B-12 status may also play a crucial role in fetal growth and programming. To date, conflicting results have been reported about the relationship between maternal B-12 status and neonatal anthropometric outcomes. A meta-analysis conducted by Rogne et al. (220) that included 18 studies in which 9 of the studies were  44  conducted in Europe, 6 in Asia, 1 study was conducted in North America, 1 in Africa, and 1 in Oceania, reported no significant linear association between maternal serum total B-12 concentration and BW. In contrast a pregnancy study conducted by Jiang et al. (31) found that maternal serum total B-12 measured at delivery was positively correlated with BW, HC, and BMI in newborns of Chinese pregnant women. However, this study included an equal proportion of fetal growth restriction (FGR), appropriate for gestational age (AGA) and large for gestational age (LGA) newborns. Thus, these findings cannot be extrapolated to other populations. In contrast, Frery et al. (195) showed an inverse association between maternal plasma total B-12 concentration at delivery and BW only in pregnant women who were smokers, while in Turkish pregnant women, no significant relationship was found between third-trimester total B-12 concentration and BW (37). Additionally, pregnancy studies conducted in Canadian (36), Asian (34) and Indian pregnant (221) women have found no significant relationship between early and late-pregnancy maternal total B-12 concentration and BW.  Data from animal studies showed that B-12 deficiency induces DNA hypomethylation and inadequate uracil incorporation in the DNA of rat colonic epithelial cells (222,223). The underlying mechanism between maternal B-12 and DNA methylation in the offspring is suggested to be due to the cofactor role of B-12 in the homocysteine remethylation pathway, whereby lower total B-12 concentration may lead to an insufficient supply of methyl groups from folate for DNA methylation. In humans, maternal third-trimester serum total B-12 concentration <148 pmol\/L was associated with a lower methylation of the IGF-2 promoter region in cord blood (224). In contrast, McKay et al. (225) found that total B-12 concentration in the first trimester was not associated with IGF-2 methylation levels measured in cord blood. Similar results were reported in Asian pregnant women (i.e., Chinese, Malay or Indian) where maternal  45  serum total B-12 concentration, measured at 26-28 weeks of gestation, was not associated with DNA methylation of fetal-growth related genes measured in cord blood (226). B-6 participates as a coenzyme in the form of PLP in several reactions of one-carbon metabolism. Results from the NEST study showed that maternal PLP concentration, the  biomarker of B-6 status, measured between 4.0\u201332.5 weeks of gestation (mean 12 weeks of gestation), was positively associated with early life weight gain, defined as the body weight gained in the first 36 months of age (29). Additionally, results of the same study indicated that maternal PLP concentration >7.48 nmol\/L was positively associated with MEG3 DNA methylation levels in cord blood, which is a cell-growth related gene (29).  Riboflavin functions as a coenzyme including for the formation of 5-methyl-THF that serves as methyl donor in the folate-B12 dependent homocysteine remethylation reaction. Dietary riboflavin intake measured in late pregnancy (i.e. 36 weeks of gestation) has been positively associated with BW (227), whereas late-pregnancy dietary riboflavin intake has been associated with BL (\u03b2=0.72 cm, P =0.002) and BW (\u03b2=149 g, P=0.001) in Dutch pregnant women (228). However, in another study of Dutch pregnant women, plasma riboflavin concentration measured at 6 and 10 weeks of gestation showed no significant relationship with BW (193). It has been previously reported that dietary riboflavin intake before pregnancy was positively correlated with the DNA methylation index of ZAC1 DMR, which is a gene involved in fetal growth and metabolism (117). However, whether circulating maternal riboflavin concentration is associated with DNA methylation of fetal growth-related genes needs to be elucidated in future prospective pregnancy studies.  To summarize, previous evidence suggested a link between individual maternal methyl nutrient concentrations and neonatal anthropometric outcomes. However, this evidence is limited  46  to pregnancy studies conducted in non-mandatory folic acid fortified countries and that have a lower prevalence of prenatal folic acid supplement use and a lower dose of folic acid in prenatal supplements, and which explored the relationship between individual maternal methyl nutrients (e.g., maternal folate concentration) and neonatal anthropometric outcomes. Also, methyl nutrient concentrations were determined at different trimesters of pregnancy (i.e., first trimester or at delivery). Given that de novo DNA methylation and fetal programming occur at early stages of pregnancy and the potential role of maternal methyl nutrients in fetal programming through DNA methylation, the relationship between early-pregnancy methyl nutrient concentrations and neonatal anthropometric outcomes need to be explored in future pregnancy studies including folate-replete populations. According to pregnancy studies conducted in Canada, more than 90% of pregnant women consume prenatal supplements containing folic acid supplements across trimesters (16,68) with a median daily intake of 1000\u00b5g of folic acid from supplements. The high compliance of folic acid supplementation has led to high folate status (i.e., RBC folate >1360 nmol\/L) in Canadian pregnant women (16,67), which represents an ideal opportunity to explore the relationship between early-pregnancy methyl nutrient concentrations and neonatal anthropometric outcomes in a folate-replete population.  Considering that all methyl nutrients participate in interrelated reactions of one-carbon metabolism which is critical for the provision of methyl groups for DNA methylation, as well as the lack of evidence of the potential role of maternal methyl nutrients in fetal growth and obesity programming through DNA methylation, the determination of the relationship of all maternal methyl nutrients and related-metabolites with DNA methylation of fetal growth and obesity-related genes in the offspring is still needed.    47  Table 1.1 Longitudinal studies reporting on the relationship between maternal methyl nutrient concentrations and neonatal anthropometric outcomes  1Data are presented as median (IQR) or mean \u00b1 SD unless otherwise noted. Folate and B-12 units were converted to [nmol\/L] and [pmol\/L], respectively. Conversion factor for folate concentration (1\u00b5g\/L=2.266 nmol\/L)(213). 2Fasting samples; 3RBC folate, serum folate, and B-12 were not reported; 4Non-fasting samples; AGA, appropriate for gestational age; BMI, body mass index; BW, birth weight; BL, birth length; FGR, fetal growth restriction; GUSTO, Growing Up in Singapore Towards Healthy Outcomes study; HC, head circumference; LGA, large for gestational age; PLP, pyridoxal 5\u2019-phosphate; RBC, red blood cell; SWS, UK Southampton Women\u2019s Survey; tHcy, total  homocysteine; total B12, total vitamin B-12; UK, United Kingdom; U.S.A, United States; wks, weeks of gestation.  Author\/ Country  (year) n Sample collection (wks) Maternal biomarker concentrations1 Neonatal outcomes Findings Du et al. (2019)(184)  China 115 Median 26.3 (range 22.7-33)2 Plasma free choline, < 35y: 9.1 (7.0,10.6); \u226535 y: 8.4 (7.8,10.4) \u00b5mol\/L Plasma betaine:< 35y: 15.1 (12.3,17.3); \u226535 y:14.5 (11.1,16.3) \u00b5mol\/L Plasma methionine:< 35y: 18.7 (15.7, 21.3); \u226535 y:18.8 (16.1,22.9) \u00b5mol\/L BW, BL Plasma betaine was negatively associated with BW van Lee et al. (2019)(190)  SWS: UK GUSTO: Singapore SWS, 985 GUSTO, 995 SWS, 11 GUSTO, 26-282 SWS, serum total choline: 6.03\u00b10.86 \u00b5mol\/L GUSTO, plasma free choline: 9.12\u00b11.6 \u00b5mol\/L BW, BL, BMI z-score, total body fat, abdominal circumference, Subscapular and triceps skinfolds GUSTO: Choline concentration was associated with higher BMI z-score, subscapular and triceps skinfold SWS: Choline concentration was associated with a higher subscapular skinfold at birth Cawley et al. (2019)(196)  Ireland 498 First prenatal  visit  Plasma tHcy: 7.1\u00b12.1 \u00b5mol\/L3  BW No relationship  Bergen et al. (2016)(28)   The Netherlands 5890 Median 14 (range 11-17)4 Plasma folate: 15.7 (6.2, 34.3) nmol\/L Plasma tHcy: 6.9 (4.9,10.5) \u00b5mol\/L Serum total B-12: 169.0 (83.0,351) pmol\/L Serum active B-12: 42.0 (20.0,83.0) pmol\/L BW, BL, HC Folate concentration \u22649.1 nmol\/L was associated with lower BW and BL  tHcy concentration (\u22658.3 \u00b5mol\/L) associated with lower BW and smaller HC   McCullough et al. (2016)(29)  U.S.A 496 Mean 12  (range 4-33) Plasma total B-12 range: 56.1,3665 ng\/L [41.5, 2704 pmol\/L] Plasma PLP range: 0,174 nmol\/L Plasma tHcy range: 0.3,1.54 \u00b5mol\/L BW tHcy concentration (>6\u00b5mol\/L) was associated with a lower BW in males van Lee et al. (2016)(39)  Singapore 955 26-282 Plasma betaine: 13.2\u00b12.7 \u00b5mol\/L Plasma free choline: mean 9.2 \u00b5mol\/L Plasma tHcy: mean 5.0 \u00b5mol\/L BW, BL, HC Fat mass, abdominal, and midupper arm circumferences Betaine was negatively associated with BW BL, HC and mid-upper arm circumference  Jiang et al. (2016)(31)  China 116 At delivery Serum B-12: FGR, 51.6\u00b139.2 ng\/L, [38.1\u00b128.9 pmol\/L] AGA, 333\u00b1195 ng\/L [238\u00b1144 pmol\/L], LGA, 299\u00b1202 ng\/L [221\u00b1149 pmol\/L]  Serum tHcy: FGR, 17.5\u00b128.4, AGA, 5.86\u00b16.17, LGA, 5.54\u00b16.21 \u00b5mol\/L BW, BL, HC, BMI, BW\/BL  Total B-12 was positively correlated with BW, HC and neonatal BMI  tHcy was negatively correlated with BW, BL, HC, and neonatal BMI Chen et al. (2015)(34)   Singapore 999 26-282 Plasma folate: 34.4 (24.5,44.6) nmol\/L Plasma total B-12: 209 (167,258) pmol\/L Plasma PLP: 61.8 (25.9,113) nmol\/L BW, BL No relationship  Gadgil et al. (2014)(197)  India 49 36 Plasma folate: 17.8 (15.1, 18.9) ng\/ml [40.3 (34.2, 42.8)] nmol\/L Plasma total B-12: 197 (139, 261) pg\/ml [145 (103, 193) pmol\/L] Plasma tHcy: 08.9 (07.1,10.4) \u00b5mol\/L BW, BL, HC, abdominal and chest circumferences Folate\/B-12 ratio was negatively correlated with BW, BL, HC and chest circumference Hoyo et al. (2014)(22)  U.S.A 496 All trimesters  Mean 12 RBC folate, Non-Hispanic Black: 195\u00b183.2 \u00b5g\/L [442\u00b1189 nmol\/L]; Non-hispanic white: 248\u00b166.6 \u00b5g\/L [562\u00b1151 nmol\/L]; Hispanic: 222\u00b164.1 \u00b5g\/L (503\u00b1145 nmol\/L); Others: 204\u00b183.6 \u00b5g\/L [462\u00b1189 nmol\/L] BW BW increased in newborns from women in the second RBC folate quartile compared to newborns of pregnant women with lower RBC folate concentration   48  Table 1.1. Longitudinal studies reporting on the relationship between maternal methyl nutrient concentrations and neonatal anthropometric outcomes (con)  1Data are presented as median (IQR) or mean\u00b1 SD unless otherwise noted. Folate and B-12 units were converted to [nmol\/L] and [pmol\/L], respectively. Conversion factor for folate concentration (1\u00b5g\/L=2.266 nmol\/L)(213). 2Fasting samples; 3median; 4Non-fasting. BW, birth weight; BL, birth length; HC, head circumference; MUAC, mid\u2013upper arm circumference; PLP, pyridoxal 5\u2019-phosphate; Parthenon, Parthenon study; PMNS, Pune Maternal Nutrition Study; RBC, red blood cell; tHcy, total homocysteine; total B-12, total vitamin B-12; wks, weeks of gestation.  Author  (year) n Sample collection (wks) Maternal biomarker concentrations1 Neonatal outcomes Findings Yajnik et al. (2014)(27)      India PMNS, 526 Parthenon, 515 PMNS: 29\u00b112 Parthenon:29\u00b122 PMNS; Plasma tHcy: 8.6 (6.7,10.8) \u00b5mol\/L RBC folate: 958 (734, 1261) nmol\/L Plasma total B-12: 122 (94, 164) pmol\/L Parthenon;  Plasma tHcy: 6.0 (5.1,7.1) \u00b5mol\/L Plasma folate: 34.4 (16.8, 51.2) nmol\/L Plasma total B-12:162 (123, 223) pmol\/L BW PMNS: RBC folate was positively associated with BW Parthenon: Higher tHcy concentration was negatively associated with BW Krishnaveni et al. (2014)(33)  India  654 30\u00b12 Plasma folate: 35.4 (17.9, 51.2) nmol\/L Plasma total B-12: 164 (123, 224) pmol\/L Plasma tHcy: 6.0 (5.1, 7.0) \u00b5mol\/L BW, BL, HC, MUAC, subscapular and triceps skinfold Higher maternal tHcy was associated with smaller BW, MUAC and triceps skinfold thickness Hogeveen et al. (2013) (35)  The Netherlands  366 30-34 Plasma free choline: 7.0 (6.1, 8.3) \u00b5mol\/L Plasma betaine: 11.0 (9.2, 12.9) \u00b5mol\/L  BW No relationship  Wu et al.  (2013)(36)   Canada 270 16th and 32th2  Plasma total B-12: 259 (199,351) and 206 (150,284) pmol\/L Plasma free choline:6.90 (5.96,8.10); and 9.40 (8.30,11.3) \u00b5mol\/L Plasma betaine:13.0 (10.6,15.8) and 13.1 (11.3,14.8) \u00b5mol\/L Plasma tHcy: 4.10 (3.50,4.80) and 4.80 (4.20,5.70) \u00b5mol\/L Plasma folate: 35.9 (33.3,38.1) and 35.8 (32.8,38.8) \u00b5mol\/L BW No relationship Halicioglu et al. (37)(2012)   Turkey  208 37th Serum total B-12: 163 pg\/ml3 [120 pmol\/L] Serum folate: 8.0 ng\/ml3 [18.1 nmol\/L] BW, BL, HC No relationship  Hogeveen et al. (2010)(32)   The Netherlands  366 30-34 Plasma folate: 9.1 (6.1,16.4) nmol\/L  Plasma total B-12: 179 (134,219) pmol\/L Plasma tHcy: 5.5 (4.5,67) \u00b5mol\/L BW tHcy concentration was negatively correlated with BW Nilsen et al. (2010) (38)  Norway 2934 Median 184 Not reported  BW, HC No relationship  Parazzini et al. (2010)(30)       Italy 244 8-12, 16 and 224  Non-folic acid users; RBC folate: 39 (19,76), 61 (36,109), 70 (49,109) nmol\/L Plasma folate: 11.2 (9.1,18.4), 15.2 (10.6,21.3), 18.4 (12.8,28.8) nmol\/L Plasma tHcy: 7.9 (5.0,11.2, 8.1 (5.5,11.2), 8.5 (5.6,12.0) \u00b5mol\/L Folic acid users; RBC folate: 52 (26,103), 67 (42,106), 85 (58,124) nmol\/L Plasma folate: 13.8 (9.2,20.3), 18.0 (11.7,21.7), 20.3 (14,30.1) nmol\/L Plasma tHcy: 6.9 (4.7,10.6), 7.3 (5.3,10.4), 7.4 (5.3,11.1) \u00b5mol\/L BW Higher tHcy concentration at 8-12 and 22 weeks of gestation was associated with lower BW in both groups    49  Table 1.1. Longitudinal studies reporting on the relationship between maternal methyl nutrient concentrations and neonatal anthropometric outcomes (con)  1Data are presented as median (IQR) or mean\u00b1 SD unless otherwise noted. Folate and B-12 units were converted to [nmol\/L] and [pmol\/L], respectively. Conversion factor for folate concentration (1\u00b5g\/L=2.266 nmol\/L)(213). 2Non-fasting sample; 3mean\u00b1standard error; 4geometric mean (95%CI); 5median (5th-95th percentiles). BW, birth weight; BL, birth length; HC, head circumference; PLP, pyridoxal 5\u2019-phosphate; RBC, red blood cell; tHcy, total homocysteine; total B-12, total vitamin B-12; UK, United Kingdom; wks, weeks of gestation.  Author  (year) n Sample collection (wks) Maternal biomarker concentrations1 Neonatal  Outcomes Findings Takimoto et al. (2007)(25)     Japan 94 7\u201314, 26\u201329 and 34\u2013362 Plasma tHcy: 5.1\u00b11.5, 5.0\u00b11.3, 5.9\u00b11.4 \u00b5mol\/L Serum total B-12: 405\u00b1146, 301\u00b196, 265\u00b195 pmol\/L Serum folate: 23.2\u00b19.7, 19.3\u00b123.8, 23.1\u00b169.3 nmol\/L RBC folate: 1317\u00b1824, 909\u00b1551, 813\u00b1475 nmol\/l BW, BL, HC Second-trimester RBC folate and tHcy concentrations were positively associated with HC Third trimester tHcy concentration was negatively, while RBC folate was positively associated with BW  Relton et al. (2005)(26)  UK  683 12\u00b16 RBC folate: 418\u00b1178 ng\/ml [947\u00b1403 nmol\/L] Plasma total B-12: 324 \u00b1132 pg\/ml [241\u00b197.4 pmol\/L] BW z-score RBC folate concentration was positively associated with BW Sram et al. (2005)(229)  Czech Republic  766 At delivery Plasma folate: 24.7\u00b116.3 nmol\/L3 BW Plasma folate was positively associated with BW (Prague-smoker group) Guerra-Shinohara et al. (2004)(194)   Brazil  119 At delivery Serum total B-12: 130 (122, 138) pmol\/L4 RBC folate: 643 (591, 701) nmol\/L4 Serum folate: 12.9 (12, 14) nmol\/L4 Serum tHcy: 6.5 (6.1, 6.9) \u00b5mol\/L4 Serum methionine: 16.9 (16.2,17.5) \u00b5mol\/L4 BW No relationship de Weerd et al (2003)(193)    The Netherlands  188 6 and 10 Plasma PLP: 47 (40,56), 48 (41,57) nmol\/L Plasma riboflavin: 270 (250,300), 280 (250,300) nmol\/L Plasma total B-12: 280 (230,330), 250 (180,310) pmol\/L Serum folate: 13 (10,17), 14 (11,19) nmol\/L RBC folate: 520 (440,650), 550 (440,710) nmol\/L BW No relationship  Buchman et al. (2001)(189)  U.S.A 25 At delivery Plasma free choline: 8.4\u00b13.1 nmol\/mL BW and BL  No relationship  Frery et al. (1992)(195)  France  188 At delivery   Plasma total B-12: 248 (100,539) pg\/mL5 [183 (73.8,398) pmol\/L] Plasma folate: not reported BW Total B-12 negatively correlated with BW in smokers    50  Table 1.2 Longitudinal studies reporting on the relationship between maternal methyl nutrient concentrations and DNA methylation of fetal growth and obesity related genes in the newborns  1Data are presented as median (IQR) or mean \u00b1 SD unless otherwise noted. Folate, B-12, and tHcy units were converted to [nmol\/L], [pmol\/L] and [\u00b5mol\/L], respectively. Conversion factor for folate concentration (1\u00b5g\/L=2.266 nmol\/L)(213).2range; 3geometric mean (95%CI) of log-transformed folate concentration;4median (90% range);5Fasting samples;6mean (range). CB, cord blood, DMR, Differentially methylated regions; EWAS, epigenome-wide association study; IGF-2, insulin growth factor-2; PCR, polymerase chain reaction; PLP, pyridoxal 5\u2019-phosphate; RBC, red blood cell; tHcy, total homocysteine; total B-12, total vitamin B-12; UK, United Kingdom; U.S.A, United States; wks, weeks of gestation.Author  (year) n Sample collection (wks) Maternal biomarker concentration1 Neonatal genes Tissue  Technique  Findings Lin et al. (2017)(226)  Singapore 987 26-28 Not reported EWAS CB Microarray assay No relationship  McCullough et al. (2016)(29)  U.S.A 496 12  (range 4-32.5) Plasma total B-12: 56.1,3665 ng\/L2 [41.4, 2704 pmol\/L] Plasma PLP: 0,174 nmol\/L2 Plasma tHcy: 0.3,1.54 \u00b5mol\/L2 IGF-2\/H19 DMR CB Pyrosequencing  PLP concentration >7.48 nM\/L was associated with higher MEG3 DNA methylation levels Hoyo et al.  (2014)(22)  U.S.A 528 All trimesters Mean 12 RBC folate, Non-Hispanic Black: 1945\u00b183.2 [4407\u00b1188 nmol\/L]; Non-hispanic white: 248\u00b166.6 [561\u00b1151 nmol\/L]; Hispanic: 222\u00b164.1 [502\u00b1145 nmol\/L]; Others: 204\u00b183.6 \u00b5g\/L [461\u00b1189 nmol\/L] IGF2\/H19  CB Pyrosequencing Low RBC folate concentration (i.e., second quartile) was positively associated with IGF-2 DNA methylation levels Haggarty et al. (2013)(203)  UK 913 19 RBC folate: 456 (442, 471) nmol\/L3 IGF-2 CB Pyrosequencing No relationship  Van Mil et al. (2014)(202)  The Netherlands 540 10-17 Plasma folate: 19.1\u00b19.0 nmol\/L Plasma tHcy: 7.0 (4.9,10.9) \u00b5mol\/L4  IGF2 DMR, H19 CB MassARRAY EpiTYPER Analyzer No relationship McKay et al. (2012)(225)  UK 197 11\u00b14 RBC folate: 379 (298,512) ng\/ml [859 (675-1160 nmol\/L] Serum total B-12: 283 (226,389) pg\/ml [209 (167,287 pmol\/L] IGF-2   CB Pyrosequencing No relationship  Ba et al. (2011)(224)    China 99 At delivery5 Serum folate:  2.29 (0,11.2) ng\/ml6 [5.19 (0, 25.4) nmol\/L] Serum total B-12: 175 (57,1120) pg\/ml6 [129 (42.1, 826 pmol\/L) IGF-2 (P2 and P3 promoter regions)  CB Real-time methylation specific PCR Serum total B-12 concentration was negatively associated with IGF-2 DNA (P2) methylation   51   1.3 Summary and rationale Methyl nutrients participate in interrelated reactions in one-carbon metabolism, which are crucial for cellular proliferation and DNA methylation reactions (7,41,50). Betaine functions as a methyl donor in the folate-B-12 independent remethylation of homocysteine to methionine (48). High maternal tHcy concentration is associated with adverse pregnancy outcomes such as preeclampsia (88). In Canada, a folate-replete population, B-12, and not folate, has been indicated as the major determinant of tHcy concentration (101). Lower total B-12 concentration has been reported in South Asian compared to European pregnant women in Vancouver, BC (230). The South Asian population is the largest ethnic minority in Canada (99). Thus, the understanding of the relationship between methyl nutrient and tHcy concentrations in pregnant women of a folate-replete population would provide important evidence of the role of other methyl nutrients (i.e., not only folate) in the provision of methyl groups for methylation reactions.  Despite methyl nutrients have interrelated roles in one-carbon metabolism (7,50), evidence from pregnancy studies  has revealed conflicting results about the relationship between individual methyl nutrient concentrations during pregnancy (e.g., maternal folate concentration) and neonatal anthropometric outcomes. BW is a prognostic indicator for obesity later in life (172,231), whereas HC and neonatal length are indicators of brain development and adiposity later in childhood, respectively. Additionally, neonatal PI, BMI and BW\/BL ratio have been associated with adiposity and changes in BMI later in life (183,232\u2013234). Given the interdependent role of methyl nutrients in one-carbon metabolism, a better understanding of the interrelationship of maternal methyl nutrient concentrations and its relationship with neonatal anthropometric outcomes and indicators is needed.  52  DNA methylation, histone modification and microRNA expression are well-known epigenetic mechanisms. DNA methylation is one of the most stable epigenetic mechanisms involved in fetal programming (132). Early-pregnancy folate and late maternal B-12 concentrations, respectively, have been associated with DNA methylation of fetal-growth related genes measured in cord blood (22,224). However, these studies have included a limited portfolio of circulating maternal methyl nutrient concentrations involved in methylation reactions, and none of these studies have explored the relationship of combined maternal methyl nutrient concentrations and DNA methylation levels in the offspring. Additionally, the measurement of DNA methylation levels in more homogeneous tissues (e.g., BECs or white blood cells), is recommended to partially avoid variations in DNA methylation levels due to cellular type. Given the occurrence of de novo DNA methylation at early stages of pregnancy, exploring the relationship between early-pregnancy methyl nutrient concentrations and DNA methylation levels of obesity and fetal-growth related genes measured in a homogeneous tissue (e.g., BECs) may contribute to the understanding of the role of methyl nutrients on obesity programming.  1.4 Research objectives and hypothesis The objectives of my dissertation are as follows:   Objective 1: To compare peripheral betaine concentration between European and South Asian pregnant women in their first and second trimester of pregnancy, and to determine the relationship between betaine and tHcy concentrations in early pregnancy (i.e., <20 weeks of gestation), in a group of women living in Metro Vancouver, BC, Canada.  53  a. Null hypothesis: Serum betaine concentration does not differ between pregnant women of South Asian and European descent in their first and second trimester of pregnancy, and serum betaine concentration is not associated with tHcy concentration in early pregnancy (i.e., <20 weeks of gestation).  b. Research hypothesis: Serum betaine concentration is higher in pregnant women of South Asian descent compared to pregnant women of European descent in their first and second trimester of pregnancy (P <0.05), and serum betaine concentration is associated with tHcy concentration in early pregnancy (i.e., <20 weeks of gestation) (P<0.05).   Objective 2: To explore the interrelationship of maternal methyl nutrient concentrations in the first and second trimester, and to determine the association of first- and second-trimester maternal methyl nutrient patterns with neonatal anthropometric outcomes among Canadian mother-newborn dyads.  a. Null hypothesis: First- and second-trimester maternal methyl nutrient patterns are not characterized by specific maternal methyl nutrient concentrations and are not associated with neonatal anthropometric outcomes among Canadian mother-newborn dyads.  b. Research hypothesis: First- and second-trimester maternal methyl nutrient patterns are characterized by specific maternal methyl nutrient concentrations and are positively associated with neonatal anthropometric outcomes among Canadian mother-newborn dyads (P <0.05).   54   Objective 3: To determine DNA methylation level (%) of CpG sites associated with IGF-2, LEP, leptin receptor (LEP-R), HIF3\u03b1, RXRA, DNMT-1, DNMT-3A, DNMT-3B genes in 3-month old infants of mothers who had high RBC folate concentration (>1360nmol\/L) in early pregnancy (i.e., <20 weeks of gestation), compared to infants of mothers with folate status in the reference range.  a. Null hypothesis: DNA methylation level (%) of CpG sites associated with IGF-2, LEP, LEP-R, HIF3\u03b1, RXRA , DNMT-1, DNMT-3A, DNMT-3B genes in 3-month old infants of pregnant women with high RBC folate concentration (>1360nmol\/L) in early pregnancy (i.e., <20 weeks of gestation) does not differ from those of infants from pregnant women with folate status in the reference range.  b. Research hypothesis: DNA methylation level (%) of CpG sites related to IGF-2, LEP, LEP-R, HIF3\u03b1, RXRA, DNMT-1, DNMT-3A, DNMT-3B genes in 3-month old infants of pregnant women with high RBC folate concentration (>1360nmol\/L) in early pregnancy (i.e., <20 weeks of gestation) are higher compared with those of 3-month-old infants of pregnant women with folate status in the reference range (P <0.05).  Objective 4: To determine the relationship between maternal methyl nutrient concentrations in early pregnancy (i.e., <20 weeks of gestation) and DNA methylation level (%) of CpG sites associated with IGF-2, LEP, LEP-R, HIF3\u03b1, RXRA , DNMT-1, DNMT-3A, DNMT-3B genes in 3-month-old infants.  55   a. Null hypothesis: DNA methylation level (%) of CpG sites associated with IGF-2, LEP, LEP-R, HIF3\u03b1, RXRA, DNMT-1, DNMT-3A, DNMT-3B genes in 3-month-old infants are not associated with maternal methyl nutrient patterns in early pregnancy (i.e., <20 gestational weeks).  b. Research hypothesis: DNA methylation level (%) of CpG sites related to IGF-2, LEP, LEP-R, HIF3\u03b1, RXRA, DNMT-1, DNMT-3A, DNMT-3B genes in 3-month-old infants are positively associated with maternal methyl nutrient patterns in early pregnancy (i.e., <20 gestational weeks) (P <0.05).      56  Chapter 2: MATERNAL SERUM BETAINE CONCENTRATION AND ITS RELATIONSHIP WITH TOTAL HOMOCYSTEINE CONCENTRATION IN EUROPEAN AND SOUTH ASIAN PREGNANT WOMEN Acknowledgement: A version of this chapter has been published. Maria F Mujica-Coopman, Amy Tan, Theresa H Schroder, Graham Sinclair, Hilary D Vallance and Yvonne Lamers. Serum betaine and dimethylglycine are higher in South Asian compared with European pregnant women in Canada, with betaine and total homocysteine inversely associated in early and mid-pregnancy, independent of ethnicity. Journal of Nutrition 2019; 149 (12):2145-2155. All inferences, opinions, and conclusions drawn in this chapter are those of the author and do not reflect the opinions or policies of Perinatal Services British Columbia.  2.1 Introduction Betaine, with the chemical structure of trimethylglycine, is a methyl donor nutrient with a metabolic role in the remethylation of homocysteine to methionine. This one-carbon reaction is critical for DNA methylation, a key epigenetic mechanism involved in fetal programming. Betaine can be obtained from a variety of dietary sources (e.g., wheat, beets, spinach) or by endogenous synthesis through choline oxidation, which is catalyzed by choline dehydrogenase (CHDH) and betaine aldehyde dehydrogenase (48,49). Homocysteine remethylation occurs in two pathways, one is by the action of the ubiquitous enzyme MS using 5-MTHF as the methyl group donor and B-12 as a coenzyme (235). In addition to B-12, the recycling of reduced folate forms depends on B-6 in its coenzyme form PLP and riboflavin. The second pathway uses betaine as the methyl group donor independent of folate and B-12, and is catalyzed by BHMT  57  with methionine and DMG as products (Figure 1.1). This pathway is responsible for 50% of homocysteine remethylation to methionine in mammalian liver cells (90). High plasma tHcy concentration during pregnancy is associated with adverse pregnancy outcomes such as preeclampsia and LBW (88,91,236). Circulating tHcy concentration is influenced by lifestyle, e.g., smoking (98), and nutritional status (237), and genetic variants (93). In Dutch men and post-menopausal women (50-75 y), folate was a two-fold stronger determinant of tHcy concentration at fasting state compared to plasma betaine and serum total B-12 concentrations (98). Plasma betaine, however, was the strongest determinant of the tHcy increase after a methionine loading test, compared to folate, B-12, and B-6 in healthy men and women aged 34-69 y (93). In folate-replete populations, such as the Canadian population (14), B-12 becomes a major determinant of tHcy concentration (101).  In pregnant women from non-mandatory folic acid fortified populations, plasma folate predicted tHcy across trimesters in women with low folate status (i.e., below median of 27.6 nmol\/L) (64). Similarly, a study conducted in the Caribbean with pregnant women of West African descent found that plasma folate was a significant predictor of tHcy only at 9 weeks of gestation and at delivery (65). Plasma betaine was associated with increased tHcy concentration after 20 weeks of gestation (64\u201366), with the strongest association at low folate status (64,66). Plasma betaine was a stronger predictor of plasma tHcy in the second and third trimester, compared to folate and B-12 (64). In contrast, in Canadian pregnant women with the majority being of European ethnicity, no association between betaine and tHcy concentrations was observed at early- and late-pregnancy (36), which may be related to the high folate status in Canadian pregnant women (16,67).   58  South Asians are the largest visible ethnic minority in Canada (5% of Canada\u2019s total population)  (99). Circulating betaine concentrations in South Asian populations in Canada or worldwide have not been described to date. Our group previously reported that South Asian pregnant women in BC, Canada, have significantly lower B-12 status compared to European pregnant women in early pregnancy (230). According to the Canadian Health Measures Survey, B-12 becomes the main determinant of tHcy in the folate-replete Canadian population (14,101). With betaine being a strong determinant of tHcy concentration, we hypothesize that betaine and B-12 are major determinants of tHcy in Canadian pregnant women, with betaine being a stronger determinant than B-12, especially in South Asian pregnant women shown to have lower B-12 status. Thus, the objective of this study was to assess betaine and DMG concentrations, and to determine the association of betaine with tHcy concentration in early pregnancy, among Canadian pregnant women of South Asian and European ethnicity.  2.2 Study design and methods This was a retrospective cohort study using bio-banked residual samples of 723 healthy pregnant women (19-44 years) who participated in the BC Prenatal Genetic Screening Program. In brief, the Prenatal Genetic Screening Program is a voluntary and free-of-charge program that offers a screening exam for Down\u2019s syndrome, Trisomy 18 and open NTDs during early fetal development. During each of the two blood collection visits, with one in the first trimester (between 8 and 13 weeks of gestation) and one in the second trimester (between 14 and 20 weeks of gestation), maternal blood samples along with information on maternal age, ethnicity, and weeks of gestation are collected. Around 50% of pregnant women in the province of BC, Canada, are estimated to participate in the program (238).   59  The primary goal of this retrospective cohort study was to determine the difference in serum total B-12 concentration between pregnant women of South Asian and European ethnicity, as described earlier (230). Briefly, maternal serum samples collected between November 2014 and May 2016 were retrieved for this study. Samples were excluded if women were current smokers, HIV positive, had a diagnosis of diabetes mellitus Type I or II, had in vitro fertilization or intracytoplasmic sperm injection treatment, had twins, were using intravenous or oral steroid medication, and had no ethnicity information. Additionally, pregnant women with an increased risk of chromosome disorders or structural anomalies according to the Prenatal Screening Guidelines for Down\u2019s syndrome, Trisomy 13, 18 and open NTDs (238) were not included in this study. The present study was reviewed and approved by The University of British Columbia Children\u2019s and Women\u2019s Research Ethics board, Vancouver, Canada (H15-00820).  2.2.1 Sample size The sample size calculation was based on the primary objective of this pregnancy cohort study that was to detect a difference of 40 pmol\/L in serum total B-12 concentration between women of European and South Asian ethnicity in both first and second trimesters of pregnancy, with a power of 0.8 and a confidence level (\u03b1) of 0.05 (230). In total, available first- and second-trimesters serum samples from up to 750 pregnant women were retrieved for the cohort study (European=350; South Asian=350).  Considering that the objectives of the research presented herein were: 1) to determine the relationship between betaine and homocysteine concentrations in South Asian pregnant women and, by comparison in European pregnant women, across the first and second trimester of pregnancy; and 2) to determine the relationship between maternal betaine, across the first and  60  second trimester of pregnancy, I conducted a power analysis. For power calculation, I considered a sample size of a total of 700 pregnant women. Using a generalized linear model, I would be able to detect an effect size of 0.15 in maternal tHcy concentration with a level of significance (\u03b1) of 0.01 and a power (\u03b2) of 0.9.  2.2.2 Maternal demographic, obstetric and neonatal information Maternal age, ethnicity (self-reported), and weeks of gestation at blood collection (calculated according to the crown-rump length ultrasound) were obtained from the Prenatal Genetic Screening Program. Pre-pregnancy weight and height, smoking status, parity (nulliparous-multiparous) and neonatal sex (male-female) were obtained from the British Columbia Perinatal Data Registry (BCPDR) (239). Briefly, the BCPDR \u201ccontains data abstracted from obstetrical and neonatal medical records on nearly 100% of births in the province of BC from over 60 acute care facilities as well as births occurring at home attended by BC registered midwives, including women who had pregnancies ending in a live or still birth of at least 20 weeks of gestation or 500 g BW. The BCPDR also collects data on maternal postpartum readmissions up to 42 days post-delivery and baby transfers and readmissions up to 28 days after birth\u201d (239). Pre-pregnancy BMI [weight in kg by height in m2] of the women was classified as underweight (BMI <18.5 kg\/m2), healthy weight (BMI \u226518.5 to 24.9 kg\/m2), overweight (BMI \u226525 to 29.9 kg\/m2) and obese (BMI \u226530 kg\/m2) according to the WHO criteria (240).    61  2.2.3  Biochemical analyses 2.2.3.1 Blood processing Maternal non-fasting blood samples were collected using serum separator tubes, following the clinical protocol. Blood samples were left at room temperature to allow for serum separation and, subsequently, were stored at 4\u00b0C, until centrifugation (1890 x g for 5 min at 4\u00b0C) within 24h of sample collection. Serum aliquots were stored at \u201380\u00b0C within 4 days. Details about aliquot preparation and thaw-freeze cycles of the samples analysed in this study have been previously reported by Schroder et al (230).   2.2.3.2 Biomarker quantitation Maternal serum betaine, DMG, methionine, and tHcy concentrations were determined by isotope-dilution liquid chromatography-tandem mass spectrometry (LC-MS\/MS) using a single platform (230). The inter-assay coefficient of variation (CV)% for the analytes were estimated based on in-house plasma control samples included in each of the in total 25 batches, and were \u226410% for betaine, DMG, tHcy, and methionine. In each batch, two external quality control samples for tHcy and methionine (Clinchek 23082 and IRIS Technologies International) were included; the mean (\u00b1SD) tHcy concentration was within the target analytical range for the low (8.0\u00b10.4 \u00b5mol\/L; target: 9.04; control range 7.23-10.8 \u00b5mol\/L) and high control samples (23.8\u00b11.0 \u00b5mol\/L; target: 25.9; control range 20.7-31.0 \u00b5mol\/L). The mean (\u00b1Standard deviation (SD)) methionine concentration was within the target analytical range for the low (30.1\u00b12.1 \u00b5mol\/L; target: 29.1; control range 23.3-34.9 \u00b5mol\/L) and high control samples (61.0\u00b15.6 \u00b5mol\/L; target: 67.0; control range 57.0-77.1 \u00b5mol\/L).   62  Serum PLP was quantified by isotope-dilution LC-MS\/MS using a method modified by Midttun et al. (241). In each batch, three external quality control samples for PLP concentrations (Clinchek 8873 and IRIS Technologies International) were included. The mean (\u00b1SD) PLP concentration were within the target analytical range for the low (67.5\u00b16.1 nmol\/L; target: 61.9; control range 49.4-74.5 nmol\/L), medium (109\u00b15.1 nmol\/L; target: 97.1; control range 77.1-117 nmol\/L) and high control samples (144\u00b18.6 nmol\/L; target: 138; control range 110-165 nmol\/L). The intra- and inter-assay CV% for PLP were estimated based on 14 batches. The intra-assay CV% was \u22645% for the low and medium concentration quality control samples, and \u226410% for the high concentration quality control samples. The inter-assay CV% was \u226410% for all control samples.  The quantitation of serum total folate and B-12 biomarker concentrations was previously described (230). In brief, serum total folate concentration was determined by the microbiological assay according to O\u2019Broin and Kelleher (242) and Molloy and Scott (243), using the chloramphenicol-resistant Lactobacillus rhamnosus (ATCC 27773) and 5-MTHF [(6S)-5-methyl-5,6,7,8-tetrahydro-pteroyl-L-glutamic acid, sodium salt; Merck Eprova] as calibrator. An in-house serum control sample with folate concentration of 42.8 nmol\/L (inter-assay CV%: 10.1%) and an external whole blood control sample (NIBSC 95\/528) with a folate content of 28.3 nmol\/L (i.e., 12.5 ng\/mL; target: 13ng\/mL) (inter-assay CV%: 8.8%) were included in each of the twenty-two batches. Serum total B-12 was quantified by a microparticle-enzyme immunoassay (Access 2 by Beckman Coulter) at the BC Children\u2019s Hospital Pathology Laboratory (Vancouver, BC, Canada). Results from the inter-assay (CV%) of four manufacturer control samples for serum total B12 (Bio-Rad, mean concentration of 93.6, 245, 335, 407 pmol\/L) analysed weekly over 4 months (n=18) ranged from 2.4% to 7.1%.  63   2.2.4 Statistical analyses Descriptive analyses. The normality distribution of the data was tested using the Shapiro-Wilk test. All continuous variables had a skewed distribution and data were, thus, expressed as the median and interquartile range (IQR). Maternal age, weeks of gestation and pre-pregnancy BMI were compared between ethnic groups using a Wilcoxon rank-sum test, whereas parity, pre-pregnancy BMI categories, and neonatal sex were compared using the Chi-square test. In each trimester of pregnancy, maternal biomarker concentrations were compared between ethnic groups using the Wilcoxon rank-sum test. Differences between the first- and second-trimester biomarker concentrations, in the total sample and each ethnic group, were tested using the Wilcoxon signed-rank test.  Longitudinal analysis. Generalized linear models (GLM) were used to assess 1) the change in betaine and DMG concentrations across weeks of gestation, and 2) the change in betaine and DMG concentrations between the first and second trimester according to first-trimester betaine and DMG concentrations, respectively. The longitudinal association between betaine concentration, as an independent variable, and tHcy concentration, as the dependent variable, was tested using four multi-level mixed effect models. Model 1 was adjusted for weeks of gestation, serum folate and total B-12 concentrations. Model 2 was additionally adjusted for PLP. Considering previous evidence that highlighted the potential influence of methionine concentration in the relationship between maternal betaine and tHcy concentrations (66), maternal serum methionine concentration was included in Model 3. In Model 4, maternal age, ethnicity, neonatal sex, pre-pregnancy BMI categories and parity were additionally included as confounding variables. All four models were adjusted for repeated measures (by including the  64  study ID as a random effect to account for the general pattern of change of biomarker concentrations over time). All \u03b2-coefficients and 95%CI values were multiplied by ten for interpretability purposes. Results were considered significant at P<0.05. All the statistical analyses were conducted using Stata version IC 15 (StataCorp). Graphs were made using R Studio version 3.5.1.  2.3 Results 2.3.1 Maternal characteristics Maternal and neonatal characteristics are presented in Table 2.1. The median (IQR) age of the women was 31 (28-34) y, and 50% were of South Asian (n=365) and 50% of European (n=358) ethnicity. The median (IQR) weeks of gestation was 11.4 (10.7-12.3) and 16.1 (15.7-17.1) at the first and second trimester blood collection visit, respectively. Maternal age, weeks of gestation, parity and neonatal sex did not differ by ethnicity. Most women were in the reference range for healthy weight (59.9%). Median pre-pregnancy BMI and the distribution of women by BMI categories did not differ between ethnic groups.  2.3.2 Betaine, DMG, and related metabolite concentrations across weeks of gestation and by ethnicity Biomarker concentrations are presented in Table 2.2. Median (IQR) betaine concentration decreased from 17.6 (13.7-22.6) \u00b5mol\/L to 12.9 (10.6-16.7) \u00b5mol\/L in European pregnant women and from 19.8 (16.3-25.0) \u00b5mol\/L to 16.1 (12.9-19.8) \u00b5mol\/L in South Asian pregnant women between the first and second trimester. Serum concentrations of DMG, total B-12, and tHcy also decreased in both ethnic groups, whereas serum folate increased in South Asian  65  pregnant women between trimesters. Per week of gestation, betaine concentration decreased by 0.99 \u00b5mol\/L in European and by 0.86 \u00b5mol\/L in South Asian pregnant women (both P<0.001) (Figure 2.1). Maternal DMG concentration decreased by 0.04 \u00b5mol\/L and 0.03 \u00b5mol\/L per week of gestation in European and South Asian pregnant women, respectively (Figure 2.2). The change in betaine concentration between first and second trimester was negatively associated with first-trimester betaine concentration (\u03b2=-0.62; 95%CI -0.68, -0.56), while the change in DMG concentration was positively associated with first-trimester DMG concentration (\u03b2=0.36; 95%CI 0.30, 0.42) (both P<0.001).   Table 2.1 Maternal and neonatal characteristics of apparently healthy European and South Asian pregnant women and their offspring1 1Data are presented as median (IQR) for continuous variables, and as prevalence (%) for categorical variables; 2Maternal age, weeks of gestation and pre-pregnancy BMI were compared between ethnic groups using a Wilcoxon rank-sum test. Parity, pre pregnancy BMI categories and neonatal sex were compared using a Chi-square test; 3 pre-pregnancy BMI categories; underweight (BMI <18.5 kg\/m2), healthy weight (BMI \u226518.5 to 24.9 kg\/m2), overweight (BMI \u226525 to 29.9 kg\/m2) and obese (BMI \u226530 kg\/m2) are expressed as median (IQR) and prevalence (%).     All n European  n South Asian  n P2 Age (years) 31 (28-34)  723 31 (28-35)   358 31 (28-34)   365 0.87 Weeks of gestation at first trimester  11.4 (10.7-12.3)   707 11.4 (10.7-12.4)  356 11.4 (10.6-12.3)  351 0.27 Weeks of gestation at second trimester  16.1 (15.7-17.1)  707 16.1 (15.6-17.0)  355 16.2 (15.7-17.3)  352 0.32 Parity, %       Nulliparous      Multiparous  51.1  48.9   358 343  53.9  46.1   188 161  48.3  51.7   170 182 0.14 Neonatal sex, %       Male      Female  50.5  49.5   357 350  49.4  50.6   174 178  51.5  48.5   183 172 0.32 Pre-pregnancy  BMI (kg\/m2), %  23.5 (21.2-26.7)  558 23.7 (21.3-27.5)  288 23.4 (20.9-26.2)  270 0.08 Underweight  17.7 (16.8-18.0) 3.94%  22 17.9 (17.7-18.2)  2.78%  8 17.4 (16.5-17.9)  5.19%  14 0.17 Healthy weight  21.9 (20.7-23.4) 59.9%  334 21.9 (20.7-23.3) 58.3%  168 21.9 (20.7-23.5) 61.5%  166 Overweight 26.8 (26.0-28.2) 23.7%  132 27.2 (26.0-28.3) 23.9%  69 26.5 (25.8-27.8)  23.3%  63 Obesity  33.2 (31.3-36.9) 12.5%  70 33.6 (31.6-36.9) 14.9%  43 32.4 (30.8-37.3) 10.0%  27  66               Figure 2.1 Maternal serum betaine concentration across weeks of gestation in European and South Asian pregnant women. Longitudinal linear regression coefficients (95%CIs) for weeks of gestation (P<0.001 unless otherwise stated) are as follows: European pregnant women (black line) \u03b2=-0.99; 95%CI -1.16, -0.82, P<0.001; South Asian pregnant women (grey line) \u03b2=-0.86; 95%CI -1.04, -0.68, P<0.001.                   Figure 2.2 Maternal serum DMG concentration across weeks of gestation in European and South Asian pregnant women Longitudinal linear regression coefficients (95%CIs) for weeks of gestation (P<0.001 unless otherwise stated) are as follows: European pregnant women (black line) \u03b2=-0.04; 95%CI -0.06, -0.03, P<0.001; and South Asian pregnant women (grey line) \u03b2=-0.03; 95%CI -0.05, -0.006, P=0.01.    67  Serum betaine and DMG concentrations were significantly higher in women of South Asian ethnicity compared to those of European ethnicity in the first and second trimester (Figure 2.3 and Figure 2.4). Serum total B-12 concentration was significantly lower in South Asian women, while methionine, PLP, folate, and tHcy concentrations did not differ between ethnicities (Table 2.2).              Figure 2.3 First and second-trimester betaine concentration in European and South Asian pregnant women.  Tukey box plots show median values (solid horizontal line), 50th percentile values (box outline; Q1 and Q3), the upper and lower adjacent values (whiskers; most extreme values within Q3+1.5*(Q3-Q1) and Q1-1.5*(Q3-Q1)), and outlier values (solid circles). Serum betaine concentration without a common letter differ significantly between European (first trimester n=306; second trimester n=286) and South Asian (first trimester n=282; second trimester n=291) pregnant women in each trimester, P<0.05 (Wilcoxon rank-sum test).         68                    Figure 2.4 First and second trimester DMG concentration in European and South Asian pregnant women. Tukey box plots show median values (solid horizontal line), 50th percentile values (box outline; Q1 and Q3), the upper and lower adjacent values  (whiskers; most extreme values within Q3+1.5*(Q3-Q1) and Q1-1.5*(Q3-Q1)), and outlier values (solid circles). Serum DMG concentration without a common letter differ significantly between European (first trimester n=306; second trimester n=286) and South Asian (first trimester n=286; second trimester n=292) pregnant women in each trimester, P<0.05 (Wilcoxon rank-sum test).   69      Table 2.2 Serum biomarker concentrations in the first and second trimester of pregnancy by ethnicity1 Serum biomarker All women European  South Asian    First trimester n Second trimester n P2 First trimester n Second trimester n P2 First trimester n Second trimester n P2 P3 P4 Betaine (\u00b5mol\/L) 18.8a (15.1-23.7) 589 14.5b (11.6-18.1) 577 <0.001 17.6a,d (13.7-22.6)  307 12.9b,f (10.6-16.7)  286 <0.001 19.8a,c (16.3-25.0)  282 16.1b,e (12.9-19.8)  291 <0.001 <0.001 <0.001 DMG (\u00b5mol\/L) 1.47a (1.19-1.90)  593 1.29b (1.04-1.68) 578 <0.001 1.38a,d (1.12-1.77)  307 1.19b,f (0.97-1.52)  286 <0.001 1.55a,c (1.30-1.96)  286 1.42b,e (1.14-1.81)  292 <0.001 <0.001 <0.001 Total B-12 (pmol\/L) 218a (157-296) 666 198b (146-270) 717 <0.001 247a,c (189-318)  336 224b,e (165-299)  355 <0.001 191a,d (137-255) 330 170b,f (126-242)  362 <0.001 <0.001 <0.001 Folate (nmol\/L) 66.8b (53.0-81.3) 656 69.6a (55.3-87.8) 675 0.031 67.9 (54.2-82.5)  330 68.7 (53.8-88.2)  332 0.148 66.5b (52.0-78.8)  326 70.9a (56.8-86.8)  343 0.006 0.301 0.564 tHcy  (\u00b5mol\/L) 5.01a (4.47-5.69) 610 4.37b (3.78-5.10)  578 <0.001 4.99a (4.37-5.65)  324 4.38b (3.67-5.10)  286 <0.001 5.03a (4.51-5.75)  286 4.36b (3.85-5.13)  292 <0.001 0.432 0.707 PLP (nmol\/L) 90.1a (50.7-176) 575 66.3b (43.1-121) 496 <0.001 95.4a (51.4-191)  287 66.9b (44.8-129)  254 <0.001 82.1a (50.3-164)  288 63.3b (41.5-116)  242 <0.001 0.184 0.381 Methionine (\u00b5mol\/L) 22.1a (19.3-25.4) 610 21.2b (18.4-24.7) 578 0.002 22.4 (19.4-25.6)  324 21.4 (18.3-25.2)  286 0.0529 21.7a (19.2-25.1)  286 21.1b (18.5-24.0)  292 0.012 0.314 0.445 1Data are presented as median (IQR). DMG, dimethylglycine; PLP, pyridoxal 5\u2019-phosphate; tHcy, total homocysteine. 2Biomarker concentrations without a common letter (a,b) differ significantly by trimester (P<0.05, Wilcoxon signed-rank test).  3First-trimester biomarker concentrations without a common letter (c,d) differ significantly by ethnicity (P<0.05, Wilcoxon rank-sum test). 4Second-trimester biomarker concentrations without a common letter (e,f) differ significantly by ethnicity (P<0.05, Wilcoxon rank-sum test).   70  2.3.3 Association between maternal betaine and total homocysteine concentrations in early pregnancy Serum tHcy concentration decreased by 0.19 \u00b5mol\/L for each ten-unit increase in serum betaine concentration across weeks of gestation after including B-12 and folate concentrations into the model (Table 2.3). We did not observe substantial changes in the relationship between betaine and tHcy concentrations after inclusion of methionine, PLP, maternal age, ethnicity, neonatal sex, parity, and pre-pregnancy BMI categories into the models. Also, serum tHcy concentration decreased by 0.03 \u00b5mol\/L for each ten-unit increase in serum total B-12 concentration across weeks of gestation, after adjusting for confounding factors.   71 Table 2.3 Association of maternal serum betaine and total homocysteine concentrations across weeks of gestation in 495 apparently healthy European and South Asian pregnant using longitudinal regression models1 Independent variables Model 1  \u03b2 (95%CI) P2 Model 2  \u03b2 (95%CI) P3 Model 3  \u03b2 (95%CI) P4 Model 4  \u03b2 (95%CI) P5 Serum betaine, \u00b5mol\/L -0.19 (-0.32, -0.06) 0.004 -0.19 (-0.32, -0.06) 0.005 -0.19 (-0.3, -0.63) 0.004 -0.21 (-0.34, -0.07) 0.002 Weeks of gestation -1.36 (-1.62, -1.10)   <0.001 -1.38 (-1.65, -1.11) <0.001 -1.37 (-1.64, -1.11) <0.001 -1.38 (-1.65, -1.11) <0.001 Serum total B-12, pmol\/L -0.03 (-0.04, -0.02) <0.001 -0.03 (-0.04, -0.02) <0.001 -0.03 (-0.04, -0.02) <0.001 -0.03 (-0.04, -0.02) <0.001 Serum folate, nmol\/L -0.01 (-0.04, 0.01) 0.33 -0.14 (-0.04, 0.01) 0.31 -0.01 (-0.04, 0.01) 0.34 -0.01 (-0.04, 0.01) 0.29 Serum PLP, nmol\/L   -0.003 (-0.01, 0.005) 0.45 -0.004 (-0.01, 0.004) 0.35 -0.003 (-0.01, 0.004) 0.38 Serum methionine, \u00b5mol\/L     0.13 (0.02, 0.28) 0.10 0.11 (-0.05, 0.26) 0.18 Maternal age       0.14 (-0.07, 0.35) 0.19 Ethnicity          South Asian       -1.72 (-3.58, 0.14) 0.07 Neonatal sex         Female       -0.68 (-2.43, 1.07) 0.45 Parity         Multiparous       -3.20 (-5.03, -1.38) 0.001 Pre-pregnancy BMI                  Healthy weight        Reference          Underweight        3.34 (-1.24, 7.92) 0.15         Overweight       -1.02 (-3.16, 1.12) 0.35         Obesity       -0.48 (-3.21, 2.25) 0.73 1Data presented as coefficients with 95%CI; coefficients and 95%CI values were multiplied by ten for interpretability purposes. PLP, pyridoxal 5\u2019-phosphate; BMI, body mass index; underweight (BMI <18.5 kg\/m2), healthy weight (BMI \u226518.5 to 24.9 kg\/m2), overweight (BMI \u226525 to 29.9 kg\/m2) and obese (BMI \u226530 kg\/m2);2Model 1 was adjusted for repeated measures, weeks of gestation, total B-12, and folate concentrations; 3Model 2 was additionally adjusted for PLP concentration; 4Model 3 was additionally adjusted for methionine concentration; 5Model 4 was additionally adjusted for maternal age, ethnicity, neonatal sex, parity, and pre-pregnancy BMI.  72  2.4 Discussion Betaine is the methyl donor nutrient required for the folate-B-12 independent remethylation of homocysteine to methionine (48). There is a higher requirement for methyl nutrients during rapid stages of growth including pregnancy. South Asian pregnant women are at risk for B-12 deficiency during pregnancy. To the best of my knowledge, this is the first study to determine betaine concentration in South Asian pregnant women, and the first to explore the association between maternal betaine and tHcy concentration in this ethnicity.  In this retrospective cohort study with samples of 723 pregnant women, serum betaine and DMG concentrations declined across weeks of gestation. Betaine concentration decreased from 18.8 \u00b5mol\/ to 14.5 \u00b5mol\/L between the first and second trimester, and DMG concentration from 1.47\u00b5mol\/L to 1.29 \u00b5mol\/L, respectively. Similarly, Velzing et al. (65) reported a decrease in plasma betaine concentration from 16.3 \u00b5mol\/L to 11.5 \u00b5mol\/L, and in DMG concentration from 1.96 \u00b5mol\/L to 1.54 \u00b5mol\/L, between 9 and 16 weeks of gestation, respectively, in 50 West African pregnant women. A significant decrease in betaine and DMG concentrations in early pregnancy (i.e., <20 weeks of gestation) was also reported in a prospective cohort study with 522 Spanish pregnant women (64). The observed decrease in these metabolites may be partially related to the preferential transport of betaine and DMG from the mother to the fetus. Betaine and DMG concentrations at delivery were significantly higher in umbilical cord blood (21.1 \u00b5mol\/L and 2.45 \u00b5mol\/L, respectively) compared to maternal blood (9.52 \u00b5mol\/L and 1.81 \u00b5mol\/L, respectively) in Irish pregnant women (84). Similarly, cord blood betaine and DMG concentrations were two-fold higher than maternal betaine and DMG concentrations in Spanish pregnant women (64). Hemodilution may be a contributing factor in the decline of betaine and DMG during pregnancy, however, maternal betaine concentration was not associated with   73 hematocrit concentration in West African pregnant women (65). Other pregnancy-related physiologic changes may lead to the decline, such as the increase in the glomerular filtration rate (244) and hormonal changes (245,246). Betaine and DMG are excreted by urine and changes in renal clearance may promote betaine excretion. The decreased osmolarity in kidney during pregnancy (247) may also promote a lower resorption of betaine by the sodium- and chloride-coupled betaine\/GABA transporter (248), thereby increasing the excretion of betaine. Hormonal changes such as the increase in estradiol can stimulate BHMT gene expression (245,246) and thereby the use of betaine as methyl donor, which may contribute to the decrease in maternal plasma betaine concentration during the first and second trimester of pregnancy. Additionally, it has been previously reported that the homocysteine remethylation rate is significantly higher in the third compared to the first trimester (249) likely due to an increased demand of cellular proliferation and differentiation that may also explain the decreasing betaine concentrations that we found between the first and second trimester. Thus, we hypothesize that the use of betaine in the remethylation pathway depends not only on the availability of betaine in circulation but also on methylation rates in each trimester of pregnancy.  South Asian pregnant women had significantly higher serum betaine and DMG concentrations compared to those of European ethnicity in this study. South Asians are the largest ethnic minority in Canada, and the second largest in Metro Vancouver, BC (99) The main dietary source of betaine are vegetables and roots such as spinach and beets, respectively. Given the retrospective study design, it was not possible to retrieve dietary intake data. However, in recent cross-sectional studies conducted in Metro Vancouver BC, it was observed an up to two-fold higher prevalence of vegetarianism in South Asian pregnant (100) and non-pregnant (250) young adult women, while the use of supplement was not significantly different between South Asian and other ethnicities included in these studies (100,250). South Asian adults in Vancouver were   74 reported to have increased their vegetable and fruit intake since their immigration to Canada (251). Also, higher betaine concentration might be related with a total dietary choline intake and subsequent choline oxidation, forming betaine. It has been shown that the administration of intravenous choline, in the form of choline chloride, in pigs significantly increased plasma, liver and kidney concentration of betaine (252). However, the intake of eggs,  the main dietary source of choline, was 50% lower in South Asian pregnant women compared to other ethnicities which may suggest that dietary choline intake is not immediately oxidized to betaine after consumption (100). Serum betaine concentration responds acutely to betaine intake in healthy adults (253). In rat liver tissue, betaine concentration and BHMT activity doubled within 3 days of a betaine-enriched versus a betaine-choline-deficient diet (i.e., 0mg\/kg) (254). A higher dietary betaine intake, and thereby increased BHMT activity, may lead to higher betaine and DMG concentrations, respectively. Additionally, an accumulation of serum betaine concentration may have also promoted higher serum DMG concentration, due to its inhibitory role on BHMT activity (255), as it was found in South Asian pregnant women. Current research is underway to investigate whether the differing metabolite concentrations, such as higher betaine, in South Asian women are related to dietary factors or genetic variants such as the common variant BHMT 742G>A (256). Although the BHMT 742G>A genetic variant has not been associated with differences in plasma concentrations of folate, free choline, betaine and tHcy through pregnancy and at delivery, pregnant women carrying the heterozygous BHMT GA or the homozygous BHMT AA genotypes had significantly lower DMG concentration in early pregnancy (<15 weeks of gestation) compared to those carrying the common BHMT GG genotype (97). Further pregnancy studies are needed to confirm the potential differences on BHMT by ethnicity and its influence on betaine and DMG concentration.    75  Maternal tHcy concentration was associated with serum betaine and total B-12 but not by folate concentrations across weeks of gestation in this cohort. In contrast, plasma betaine concentration was not associated with tHcy concentration in early pregnancy in Spanish or West African pregnant women (64,65), but was a strong predictor in late pregnancy (64\u201366) and under low methionine concentrations (66). In the Reus-Tarragona Birth Cohort Study, fasting plasma folate concentration was negatively associated with tHcy concentration across weeks of gestation  and at labour in women with low folate status, defined as having plasma folate concentration below the median of 27.6 nmol\/L (64), and in late pregnancy in women with high folate status. Betaine was associated with tHcy in late pregnancy and labour in women with low folate status, but only at labor in women with high folate status. Similarly, in a study conducted in 50 West-African pregnant women from a non-mandatory folic acid fortified population, fasting plasma folate concentration (geom. mean of 16.4 nmol\/L) was a strong predictor of maternal tHcy concentration at 9 weeks of gestation, while betaine concentration was the strongest predictor only after 20 weeks of gestation (65).  Potential explanations for the discrepancies found between the results of this study (i.e., a significant association between betaine and tHcy concentrations) and the lack of association between betaine and tHcy concentrations at early pregnancy reported by previous studies, might be related to an interaction between the folate and betaine dependent methylation pathways (257,258). Rats fed a diet deficient in choline (i.e., 0.006% choline), the main precursor of betaine, had 31% lower hepatic folate concentration (258), whereas the folate deficient diet caused hepatic choline depletion after 2 weeks (259). In healthy adults, who had fasting median folate concentration ~12 nmol\/L at baseline, folic acid supplementation (<800\u00b5g\/day) increased plasma betaine concentration by ~15% after 12 weeks (98), which may suggest a sparing effect of betaine as methyl donor and promoting the use of folate for homocysteine remethylation. In   76 contrast, offspring of mice receiving a high folic acid diet (20 mg folic acid\/kg diet) during pregnancy and lactation had 3-fold lower liver 5,10-methylenetetrahydrofolate reductase (MTHFR) protein levels (94). 5,10-MTHFR is the enzyme catalyzing the generation of 5-MTHF, the folate form required in homocysteine remethylation (260). In maternal liver of mice, messenger ribonucleic acid levels of MS, the enzyme catalyzing homocysteine remethylation using 5-MTHF, were decreased as a consequence of the high folic acid diet (94). In this study, a high non-fasting serum folate concentration of ~67 nmol\/L was reported in both trimesters, which reflects a high folate status and\/or folic acid intake, and may impair the folate-dependent methylation pathway, while diminishing the use of folate as methyl donor nutrient and promoting the use of betaine for the remethylation of homocysteine. High serum folate concentration reported in the current study may be explained by the high prevalence of prenatal supplement use in Canadian pregnant women and high dose of folic acid in commercially available products. Due to the retrospective design of this cohort study, we were unable to collect data on supplement use. Results from the Canadian Maternal Experiences Survey showed that 94% of pregnant women consume folic acid-containing prenatal supplements in the first three months of pregnancy, with >90% continuing until the end of pregnancy and\/or lactation (206). The long duration of supplement use and high folic acid content, most commonly 1mg\/d, throughout pregnancy (206), likely explains the high circulating folate concentrations in Canadian pregnant women (16).  This is the first study to assess betaine and DMG concentrations as well as the relationship between betaine and tHcy concentrations in South Asian pregnant women. Strengths of this study include the longitudinal approach, the large sample size, and the lack of consent bias. Due to the retrospective study design, the current study however is limited by the absence of maternal dietary intake information, and the lack of data on supplement use. Choline is the endogenous precursor of betaine; however, the biobanked specimen available was serum and given the   77 instability of free choline concentration in serum samples (261), we did not include maternal free choline data in our analyses. Also, the determination of tHcy concentration in serum samples requires coagulation time at room temperature. Plasma tHcy concentration increases by about 10% per hour at room temperature which may affect the interpretation of serum tHcy (262). However, tHcy concentration determined in the current study is comparable to previous pregnancy studies in Canadian pregnant women using plasma tHcy concentration (36), and also were lower than pregnancy studies using serum tHcy concentration in a non-mandatory folic acid fortified population (66). In summary, South Asian pregnant women had significantly higher betaine and DMG concentration compared to European pregnant women. For first time, this study is showing that maternal betaine concentration is an important predictor of tHcy concentration at early pregnancy in a folate-replete population. Results of this study emphasized the key role of betaine at pregnancy, and may be used to potentially decrease the risk of adverse pregnancy outcomes associated with elevated tHcy concentrations. Further pregnancy studies in folate-replete populations including clinical outcomes are still needed. Additionally, further studies are needed to explain the ethnic differences and should include dietary intake and supplement use data, as well as determine genetic variants related to homocysteine and betaine metabolism, e.g., for BHMT, to understand potential underlying mechanisms.   78 Chapter 3:  FIRST- AND SECOND-TRIMESTER MATERNAL METHYL NUTRIENT PATTERNS AND THEIR ASSOCIATION WITH NEONATAL ANTHROPOMETRIC OUTCOMES IN CANADIAN MOTHER-NEWBORN DYADS  Acknowledgments: All inferences, opinions, and conclusions drawn in this chapter are those of the author and do not reflect the opinions or policies of Perinatal Services British Columbia.  3.1 Introduction Maternal adequacy in folate, riboflavin, vitamins B-12 and B-6, methionine, choline, and betaine \u2013 also commonly defined as methyl nutrients \u2013 is crucial for optimal fetal growth and development given the nutrients\u2019 interdependent roles in one-carbon metabolism (41,42) (Figure 1.1). Folate, in the form of 5-MTHF with B-12 acting as a cofactor, and betaine serve as methyl donors in the remethylation of homocysteine to methionine (41,48,59). Methionine is the precursor of the universal methyl donor, S-adenosylmethionine, which is vital for DNA methylation and cell proliferation.  DNA methylation is one of the most stable epigenetic mechanisms involved in fetal programming. Demethylation and remethylation, also known as de novo DNA methylation, of embryonic DNA strands occur in the first ten weeks of gestation \u2013 also referred to as the plasticity window (130,132) \u2013 and are vital for early growth and development, as well as for potential anatomical, endocrine and metabolic adaptions which may impact health conditions later in life (22,139). BW, BL and HC are key indicators of fetal growth (177). In the U.S.A, the results of the NEST study showed a negative association between low maternal RBC folate   79 concentration in early pregnancy (i.e., first RBC folate quartile at ~12 weeks of gestation) and methylation levels of the IGF-2\/H19 gene in cord blood, a well-known gene involved in fetal growth, which may suggest a critical role of maternal methyl nutrient concentrations in fetal growth programming by DNA methylation modifications.   Despite that findings from animal and human studies have indicated that the utilization of methyl donors (i.e., folate and betaine) for remethylation reactions depend on the availability of methyl donors as well as on coenzyme concentrations (i.e., B-12 for the folate-dependent reaction) (64,94,263), evidence of the interrelationship of maternal methyl nutrients in early pregnancy in human studies is lacking. Additionally, pregnancy studies have only investigated the association of individual methyl nutrients with neonatal anthropometric outcomes, which have revealed conflicting results. For example, in European women, low maternal plasma folate concentration (defined as folate concentration \u22649.1 nmol\/L) at <15 weeks of gestation was negatively associated with BW and BL (28), while in Asian pregnant women, only late-pregnancy betaine, and not folate or B-12, concentration was negatively associated with BW, HC, BL, and subcutaneous adipose tissue in newborns (34,39).  Given the interdependent role of methyl nutrients in one-carbon metabolism, I sought, to my knowledge for the first time, to explore the interrelationship of maternal methyl nutrient and tHcy concentrations in the first and second trimester of pregnancy using a principal component analysis (PCA) approach. Also, in this study I explored the association between the derived maternal methyl nutrient pattern scores with neonatal anthropometric outcomes in Canadian mother-newborn pairs.    80 3.2 Study design and methods    This was a retrospective pregnancy cohort study for which bio-banked, non-fasting serum samples of 729 apparently healthy pregnant women, aged between 19-44 years who participated in the BC Prenatal Genetic Screening Program, Canada, between November 2014 and May 2016 were retrieved. Information about maternal samples collection, exclusion criteria, ethical approval and BC Prenatal Genetic Screening Program was provided in Chapter 2.2.  3.2.1 Maternal and neonatal information 3.2.1.1 Maternal data Information on maternal age, weeks of gestation at sample collection and self-reported ethnicity (European or South Asian) was obtained from the BC Prenatal Genetic Screening Program. Data on pre-pregnancy maternal weight (kg), height (cm), parity (nulliparous\/multiparous), delivery type (vaginal birth\/caesarean section), hypertensive disorder of pregnancy (HTN) (yes\/no) and GD (yes\/no) were obtained from the BCPDR, Canada (239). Briefly, the BCPDR \u201ccontains data abstracted from obstetrical and neonatal medical records on nearly 100% of births in the province of BC from over 60 acute care facilities as well as births occurring at home attended by BC registered midwives, including women who had pregnancies ending in a live or still birth of at least 20 weeks of gestation or 500g BW\u201d.  Neighborhood income was estimated using the median family income of each pregnant woman\u2019s residential Forward Sortation Area (FSA) reported in the Canadian Census 2016 (22). The FSA of each pregnant woman was retrieved from the BCPDR. According to Statistics Canada, the FSA is defined as \u201ca way to designate a geographical unit based on the first three characters in a Canadian postal code\u201d (264). The FSA-median income was adjusted for the   81 average family size in each FSA, i.e., FSA median household income\/\u221aaverage number of FSA household members. The median neighborhood income adjusted for the average family size for all FSAs in BC were divided into quintiles and each pregnant woman was allocated to the corresponding quintile (99). Pre-pregnancy BMI [weight (kg)\/height (m) 2] was calculated using the pre-pregnancy body weight, or body weight collected before 11 weeks of gestation, and height. Nutritional status was classified using BMI according to the WHO criteria (240).  3.2.1.2 Neonatal data Gestational age (GA) at birth (weeks), BW (kg), BL (cm), HC (cm), and neonatal sex were obtained from the BCPDR. Missing data on gestational age at birth, BW and sex from the BCPDR (n=8) was replaced by data available from the BC Newborn Screening Program, Canada. LBW was classified as BW <2,500 g (265). SGA, AGA and LGA were defined as BW <10th percentile, BW between 10th-90th percentile and BW >90th percentile, respectively, according to the Perinatal Services BC (PSBC) growth chart (266). The PSBC growth charts for BW are based on BC data registered by the Vital Statistics Agency between 1981-2000, whereas PSBC growth charts for BL and HC are based on BC data registered by the Vital Statistics Agency between 1995-2000. The Vital Statistics Agency registers all births that occur in BC. BW, BL and HC z-scores were derived from the PSBC growth charts (266). Furthermore, neonatal PI, BMI and BW\/BL ratio were included as proxy indicators of neonatal body composition (183), which were calculated as follows: PI was calculated as BW (kg) divided by BL (m3), whereas BW\/BL was calculated as BW (kg) by BL (m). Neonatal BMI was calculated as BW (kg) divided by BL (m2). In addition to the PSBC growth chart, BW z-scores using   82 Kramer\u2019s growth charts (267), which were developed based on Canadian population born between 1994-1996 (except for Ontario) was calculated.  3.2.2 Biochemical analyses Non-fasting serum samples were collected using serum separator tubes. Details about sample collection, processing, aliquot preparation and freeze-thaw cycles have been described earlier (230).  3.2.2.1  Biomarker quantitation Analytical methods used for serum tHcy, methionine, PLP, betaine, folate and total B-12 concentrations were described in Chapter 2.2.3. Riboflavin was quantified by isotope-dilution LC-MS\/MS using a method modified by Midttun et al. (241). The intra- and inter-assay CVs% for riboflavin were estimated based on 14 batches and were <10%. Maternal serum holotranscobalamin (holoTC) and total B-12 were quantified by automated immunoassays (Architect by Abbott Laboratories and Access 2 by Beckman Coulter, respectively) at the pathology laboratories at St. Paul\u2019s Hospital and BC Children\u2019s Hospital, respectively. The mean (SD) concentration for the holoTC control samples run once per batch for three batches of study samples was 46 (1.5) pmol\/l (CV: 3.3 %) for high and 6 (2.0) pmol\/l (CV: 12 %) for low control samples, respectively.   3.2.3 Statistical analyses Because of skewed distributions, descriptive statistics for biomarker and demographic variables are reported as median and IQR for continuous variables and as percentages (%) for categorical variables. A Wilcoxon signed-rank test was used to compare biomarker   83 concentrations between trimesters, while a Wilcoxon rank-sum test was used to compare biomarker concentration between pregnant women included in the PCA in the first and second trimester. Additionally, a Wilcoxon rank-sum test was also used to compare neonatal anthropometric outcomes by neonatal sex, maternal ethnicity, HTN, GD, and parity. A Kruskal-Wallis test and a Dunn-Bonferroni test as a post hoc test were used to compare neonatal anthropometric outcomes by neighborhood income quintile and pre-pregnancy BMI categories. Additionally, a chi-square test was used to compare birth status (preterm\/term), delivery type, and the prevalence of SGA, AGA, and LGA by neonatal sex and maternal ethnicity.   3.2.3.1 Principal components analysis  Considering the importance of methyl nutrients for the provision of methyl groups for DNA methylation at specific time windows of pregnancy, e.g., for de novo DNA methylation reactions through interrelated reactions of the one-carbon metabolism, uncorrelated maternal methyl nutrient principal components for first- and second-trimester biomarker concentrations were derived separately, by conducting two PCAs. The derivation of the first (PCA 1) and second (PCA 2) trimester maternal methyl nutrient principal components involved several steps which are detailed as follows. First, maternal biomarkers were selected to be included in the PCAs. The selection of the maternal biomarkers was based on their role in related methylation reactions of one-carbon metabolism: folate, betaine and methionine, serve as methyl group donors, whereas B-12 participates as a coenzyme for the folate-dependent remethylation reaction, and PLP participates as a coenzyme in the catabolism of homocysteine, an intermediate metabolite of the remethylation reaction, as well as in the provision of one carbon units for the formation of 5-MTHF in one-carbon metabolism. Also, riboflavin concentration was included due to its role as a   84 precursor of FMN and FAD, both coenzymes in one-carbon metabolism, while holoTC was included because this is the B-12 form taken up by cells, also referred to as \u2018active B-12\u2019. Given the role of homocysteine as an intermediate metabolite of methylation reactions in one-carbon metabolism, tHcy concentration was included in the PCAs. The PCAs were restricted to pregnant women with complete data for maternal biomarker concentrations, i.e., without missing observations. The sample size of complete data for first- and second-trimester biomarker concentrations was n=505 and n=393, respectively. The next step included the log-transformation of the maternal serum PLP, tHcy, total B-12, folate, betaine, methionine, holoTC, and riboflavin concentrations to improve the normality of the data, and the subsequent exploration of the correlation between the maternal biomarker concentrations at each trimester using a Pearson correlation test.  After this, the PCAs were conducted. The number of maternal methyl nutrient principal components derived equaled the number of variables included in each PCA (i.e., 8 maternal biomarker concentrations). The PCA enabled us to summarize 8 correlated maternal biomarker variables to uncorrelated maternal methyl nutrient principal components in each trimester, which were subjected to orthogonal rotation. After the derivation of the 8 maternal methyl nutrient principal components, the number of them to be retained was based on the examination of 1) Eigenvalues, which represents the variance of the data explained by each maternal methyl nutrient principal component using Kayser\u2019s rule (i.e., >1) (268), 2) the exploration of scree plots of the maternal methyl nutrient principal components for each trimester, which are the graphical representation of the percentages of variation that each maternal methyl nutrient principal component accounts for (Figure 3.1 and Figure 3.2) and 3) explained cumulative variance provided by the maternal methyl nutrient principal components (i.e., >50% of the variance of the biomarker data). Retained maternal principal methyl nutrient components, henceforth referred to   85 maternal methyl nutrient patterns, were labeled according to the factor loadings, also known as the contribution or proportion of each biomarker to the retained component (i.e., |factor loadings| \u22650.5 were considered to contribute significantly to each component) and overall interpretability. Finally, scores of the retained maternal methyl nutrients patterns were predicted for each pregnant woman.          Figure 3.1  Scree plot of eigenvalues of derived first-trimester principal components           Figure 3.2 Scree plot of eigenvalues of derived second-trimester principal components    86 Characteristics of mother-newborn pairs included in PCA 1 versus those included in PCA 2 were compared using Wilcoxon rank-sum test for continuous variables (i.e., maternal age, weeks of gestation, median of neighborhood income, GA at birth, BW, BL, HC, PI, neonatal BMI and BW\/BL) and chi-square test for categorical variables (i.e., pre-pregnancy BMI categories, parity, ethnicity, neonatal sex, BW categories, mode of delivery, and prevalence of GD, HTN, preterm birth, and LBW).  Linear regression models were used to explore the association between the predicted scores of each retained maternal methyl nutrient pattern in each trimester, considered as the exposure variable, and neonatal anthropometric variables as outcome variables. Specifically, the models included as outcome variables: BW (Model 1), BL (Model 2), HC (Model 3), PI (Model 4), BW\/BL (Model 5), and neonatal BMI (Model 6), BW z-score (Model 7), BL z-score (Model 8), and HC z-score (Model 9). For significant associations between retained maternal methyl nutrient patterns and neonatal anthropometric outcomes, additional regression models were used in each trimester to explore the association between log-transformed individual maternal methyl nutrient concentrations with the highest contribution (i.e., log-transformed individual methyl nutrient concentrations with |factor loadings| \u22650.5 to each pattern) as the exposure variable, and neonatal anthropometric outcomes.  Based on results from previous studies (183,269\u2013274), all models were adjusted, except Models 7-9, for maternal age, pre-pregnancy BMI, maternal height, GA at birth, and neighborhood income quintiles, parity, GD, HTN, maternal ethnicity and neonatal sex. Models 7-9 were not adjusted for GA at birth and neonatal sex, because the derivation of BW, BL and HC z-scores considers both variables. Because previous studies reported body composition and neonatal anthropometric outcomes to differ between preterm and term newborns (183), a   87 sensitivity analysis was conducted which restricted the regression analysis to term newborns, excluding preterm newborns.  In the first trimester analysis (n=505), missing data for maternal variables (i.e., weeks of gestation at blood collection (n=6), GD (n=16), HTN (n=16), parity (n=16), and neighborhood median household income per year (n=17)), and neonatal variables (i.e., sex (n=15), BW (n=15), BW z-score (n=15), BL (n=24), BL z-score (n=24), HC (n=23), HC z-score (n=23), BW\/BL (n=24), BMI (n=24), PI (n=24), GA at birth (n=15), preterm (n=15), and delivery type (n=16), represented <5% of the total sample size. In the second trimester analysis (n=393); maternal variables (i.e., GD (n=5), HTN (n=5), parity (n=5), and neighborhood median household income per year (n=6)), and neonatal variables (i.e., sex (n=6), BW (n=6), BW z-score (n=6), BL (n=13), BL z-score (n=13), HC (n=12), HC z-score (n=12), BW\/BL (n=13), BMI (n=13), PI (n=13), GA at birth (n=6), preterm (n=6) and delivery type (n=5)) represented <4% of the total sample size. Given that in the first and second-trimester analysis, missing data for the aforementioned variables represented <5% of the data, missing observations were omitted, and no missing data categories were created.  For pre-pregnancy BMI, 23% (n=115) and 22% (n=87) of data were missing for women included in the first- and second trimester PCAs, respectively. Thus, an additional BMI category for missing data was created, coded as 999, to account for the missing BMI data in the regression analysis. Data for maternal height were missing for 16% (n=81) and 15% (n=57) of women included in PCA 1 and 2, respectively. Missing data for height were coded as 999, and an indicator variable with two categories (missing height data=0; height data available=1) was created. Regression models were run including maternal height as a continuous variable, as well as the indicator variable. Results were considered significant at P<0.05. All the statistical   88 analyses were conducted using STATA version IC 15 (StataCorp). Graphs were made using R Studio version 3.5.1.   89 3.3 Results 3.3.1 Maternal characteristics  Median (IQR) maternal age was (28-34) y. Of all pregnant women, 49.5% of them were of European and 50.5% of South Asian ethnicity.  First- and second-trimester blood sample collection visits occurred at 11.4 (10.7-12.3) and 16.1 (15.7-17.1) weeks of gestation, respectively. Fifty-one percent of women were nulliparous and 49% multiparous. The women\u2019s pre-pregnancy BMI indicated that 3.9% (n=22), 59.9% (n=334), 23.7% (n=132) and 12.5% (n=70) were underweight, healthy weight, overweight and obese, respectively, before entering pregnancy or at <11 weeks of gestation. During pregnancy, 15.8% were diagnosed with GD and 4.14% with HTN. The prevalence of GD was significantly higher in South Asian pregnant women (23.3%) compared to European pregnant women (8.3 %) (P<0.001).  3.3.1 Neonatal characteristics Neonatal characteristics are presented in (Table 3.1). Most of the newborns were born at term and were AGA. Female newborns had a lower BW and BL, and had a smaller HC compared to males (P<0.05 for all). Newborns from South Asian pregnant women had lower BW, BW z-score, BL, BL z-score, and smaller HC and HC z-score (P<0.05), while newborns from mothers of European ethnicity had higher PI and BW\/BL ratio compared to newborns of South Asian women (P<0.001 for all). Additionally, the prevalence of SGA was higher in newborns of South Asian women, while LGA prevalence was higher in newborns of European women (P<0.001 for all). BW of newborns from pregnant women with GD [3239 (3024-3575) g] or HTN [3130 (3005-3620) g] was lower compared to those newborns of pregnant women without GD [3403 (3102-3703) g] or HTN [3386 (3100-3700) g] (both P<0.05). Newborns of pregnant women with GD had lower BMI [12.6 (11.9-13.5) vs. 12.8 (12.0-13.8) kg\/m3] and BW\/BL ratio [6.41 (6.02-7.0) vs. 6.60 (6.17-7.14) kg\/m] compared to newborns of non-GD pregnant women. Additionally,   90 within European or South Asian pregnant women, BW of newborns from GD-pregnant women did not significantly differ, compared to those newborns from non-GD-pregnant women (P>0.05).  Pregnant women with HTN had shorter newborns [50 (49-52) vs. 51 (50-53) cm] compared to those not reported to have HTN (P<0.05). Newborns from mothers with pre-pregnancy obesity had higher BW z-score, HC and HC z-score compared to newborns of women healthy weight or underweight before pregnancy (P<0.05). Newborns of pregnant women who lived in a neighborhood with a median neighborhood income in the lowest quintile had a lower BW, BW z-score, BL, BL z-score, BMI, BW\/BL ratio and PI compared to newborns of pregnant women who lived in a neighborhood with a median neighborhood income in the 4th or 5th quintiles (P<0.05).   3.3.2 Maternal methyl nutrient concentrations  The sample sizes for maternal methyl nutrient data are shown in Figure 3.3. Missing methyl nutrient data resulted from insufficient serum volume to quantify the metabolite. Serum PLP, betaine, methionine, riboflavin, total B-12, holoTC and tHcy concentrations significantly decreased from the first to the second trimester (P<0.05). In contrast, serum folate concentration increased (P<0.05) (Table 3.2).    91 Table 3.1 Newborn characteristics1 1Values are median (IQR) for continuous variables or prevalence (%) for categorical variables;P2 are from Wilcoxon rank-sum test for comparison of continuous variables and chi-square test for categorical variables by neonatal sex;  P3 are from Wilcoxon rank-sum test for comparison of continuous variables and chi-square test for categorical variables by maternal ethnicity. AGA, appropriate for gestational age; BL, birth length; BMI, body mass index; BW, birth weight; BW\/BL, weight-to-length ratio; GA, gestational age; HC, head circumference; LGA, large for gestational age; LBW, low birth weight; PI, ponderal index; SGA, small for gestational age.Neonatal characteristics n All n Female n Male n European  n South Asian  P2 P 3 Birth status, %     Term     Preterm 699  93.4  6.58 348  94.3 5.75 351  92.6 7.40 350  94.6 5.43 349  92.3 7.74 0.38 0.22 Delivery type, %     Vaginal birth     Caesarean section 701  61.3  38.7 343  64.1  35.9 349  57.6 42.4 349  66.8 33.2 352  56.0 44.0 0.08 0.003 GA at birth, weeks 699 39.0  (38.0-40.0) 348 39.0  (38.0-40.0) 351 39.0  (38.0-40.0) 350 39.0  (38.0-40.0) 349 39.0  (38.0-40.0) 0.69 0.0003 BW, g 700 3380  (3086-3691) 349 3322  (3070-3652) 351 3424  (3115-3735) 350 3540  (3240-3790) 350 3244  (3005-3525) 0.006 <0.001 LBW, % 20 2.86 10 2.87 10 2.85 <5 \u22641.43 18 5.14 0.99 <0.001 BW z-score 699 0.05  (-0.58-0.68) 348 0.09  (-0.54-0.66) 351 0.005  (-0.61-0.71) 350 0.32  (-0.31-0.96) 349 -0.17 (-0.77-0.41) 0.52 <0.001 BL, cm 688 51.0  (50.0-53.0) 341 51.0  (50.0-52.0) 347 52.0  (50.0-53.0) 346 52.0  (50.0-53.0) 342 51.0  (50.0-52.0) <0.001 <0.001 BL z-score 688 0.10  (-0.31-0.50) 341 0.04  (-0.27-0.54) 347 0.10  (-0.31-0.50) 346 0.15  (-0.27-0.61) 342 0.003 (-0.37-0.42) 0.80 0.003 HC, cm 689 35.0 (34.0-36.0) 342 35.0  (34.0-35.0) 347 35.0  (34.0-36.0) 346 35.0  (34.0-36.0) 343 35.0 (34.0-36.0) <0.001 <0.001 HC z-score 689 0.23  (-0.19-0.61) 342 0.23  (-0.06-0.54) 347       0.25  (-0.21-0.67) 346 0.25  (-0.15-0.71) 343 0.17  (-0.21-0.54) 0.92 0.04 PI, kg\/m3 688 24.9  (23.3-26.8) 341 25.1 (23.3-27.4) 347 24.8  (23.3-26.6) 346 25.4  (23.8-27.5) 342 24.5  (22.9-26.1) 0.26 <0.001 BW\/BL ratio, kg\/m 688 6.56  (6.14-7.09) 341 6.52  (6.06-7.04) 347 6.61  (6.19-7.18) 346 6.82  (6.36-7.25) 342 6.37  (5.96-6.85) 0.08 <0.001 BMI, kg\/m2 688 12.8  (11.9-13.7) 341 12.8  (11.9-13.8) 347 12.8  (12.1-13.7) 346 13.1 (12.3-14.0) 342 12.6  (11.7-13.3) 0.66 <0.001 SGA, % 699 5.29 348 5.46 351 5.13 350 3.43 349 7.16 0.94 <0.001 AGA, %  83.7  83.9   83.4  81.7  85.7   LGA, %  11.0  10.6   11.4   14.9  7.16     92 Figure 3.3. Flowchart showing serum samples analyzed and sequence of biomarkers determined for each trimester of pregnancy HoloTC, holotranscobalamin; tHcy, total homocysteine; PLP, pyridoxal 5\u2019-phosphate; NSQ, no sufficient quantity.  93 Table 3.2 First and second-trimester methyl nutrient and related metabolite concentrations of apparently healthy Canadian pregnant women in the total sample and in those women included in the first or second-trimester PCA1  All  PCA   Serum concentrations n First trimester n Second trimester P2 n First trimester  n Second trimester P3 Folate, nmol\/L 658 66.8 (53.2-81.3)  680 69.4 (55.3-87.6)  0.03 505 66.7 (51.9-81.4)  393 67.9 (53.4-86.8) 0.19 PLP, nmol\/L 577 90.1 (50.8-176.0) 499 66.4 (43.2-121.0) <0.001 505 93.4 (51.8-181)  393 66.4 (42.7-124) <0.001 Riboflavin, nmol\/L 579 22.0 (16.6-32.4) 499 21.4 (15.5-32.7)  0.02 505 22.0 (16.6-32.6)  393 21.2 (15.5-33.2) 0.11 Betaine, \u00b5mol\/L 591 18.8 (15.0-23.9)  580 14.5 (11.5-18.1)  <0.001 505 18.9 (15.0-24.1)  393 14.5 (11.4-18.3) <0.001 Methionine, nmol\/L 612 22.1 (19.3-25.4)  581 21.2 (18.4-24.7)  0.002 505 22.1 (19.2-25.4)  393 21.2 (18.4-25.0) 0.03 Total B-12, pmol\/L 668 218 (157-297)  723 199 (146-271)  <0.001 505 218 (159-297) 393 209 (151-279) 0.04 HoloTC, pmol\/L 656 83.6 (60.3-115.2) 634 77.9 (54.9-107.7) <0.001 505 84 (60.4-117) 393 81.0 (55.5-109) 0.07 tHcy, \u00b5mol\/l 612 5.0 (4.5-5.9)  581 4.4 (3.8-5.1)  <0.001 505 5.0 (4.5-5.7) 393 4.38 (3.76-5.03) <0.001 1Values are median (IQR). P2values are from Wilcoxon signed-rank test; P3 values are from Wilcoxon rank-sum test. B-12, vitamin B-12; HoloTC, holotranscobalamin; PCA, principal component analysis; PLP, pyridoxal-5\u2019-phosphate; tHcy, total homocysteine.       94 3.3.3 Maternal methyl nutrient patterns  After conducting the PCAs, 3 maternal methyl nutrient patterns were retained for each trimester of pregnancy (Table 3.3) (Figure 3.4 and Figure 3.5). First-trimester maternal methyl nutrient patterns explained 57% of the variation of the biomarker data, whereas second-trimester maternal methyl nutrient patterns explained 59% of the variation of the biomarker data. In the first trimester, maternal methyl nutrient patterns were labeled according the highest biomarker contributor of each retained component as follows 1) B-12 pattern (in this case, total B-12 and holoTC concentrations), 2) methyl donor\/riboflavin pattern (including folate, betaine as well as riboflavin concentrations), and 3) methionine pattern (i.e., methionine concentration). The B-12 pattern score tended to increase with increasing log-transformed total B-12 and holoTC, which were the highest contributors for the B-12 pattern. Similarly, the methyl donor\/riboflavin pattern score showed an increasing trend with increasing log-transformed folate, betaine and riboflavin that were the most important contributors for methyl donor\/riboflavin pattern. The methionine pattern score increased with increasing log-transformed methionine.  In the second trimester, maternal methyl nutrient patterns were characterized primarily by 1) B-12 pattern (in this case, total B-12, holoTC and tHcy concentrations), 2) B-6\/riboflavin pattern (i.e., PLP and riboflavin concentrations), and 3) methyl donor pattern (including betaine and methionine concentrations). Similar to the results of the first trimester, the B-12 pattern score showed an increasing trend as log total B-12 and log holoTC increased, and a decreasing trend with increasing log tHcy concentration. The B-6\/riboflavin pattern score showed also an increasing trend with increasing log-transformed B-6 and riboflavin, whereas the methyl donor pattern score tended to increase with increasing log-transformed betaine and methionine.    95 Maternal age, pre-pregnancy BMI, prevalence of GD and HTN, ethnicity, parity, neighborhood median income, prevalence of preterm birth, delivery type, GA at birth, neonatal sex, and neonatal anthropometric outcomes did not differ between mother-newborn pairs included in the first-trimester versus the second-trimester PCA (Table 3.4 and Table 3.5). Table 3.3. Loadings of methyl nutrients for the first and second-trimester maternal methyl nutrient patterns First trimester B-12 pattern1  Methyl donor\/riboflavin pattern2 Methionine pattern3 Variance explained 25.7% 16.7% 14.5% Folate -0.0096 0.6163 -0.1337 PLP 0.2333  0.1865 0.3594 Total B-12 0.5913 0.0052 0.0657 Riboflavin 0.0573 0.5411 0.2844 Methionine 0.0013 -0.0734 0.7624 tHcy -0.4697 0.0556 0.3833 HoloTC 0.6048 -0.0056 0.0384 Betaine -0.0790   0.5330 -0.1952 Second trimester B-12 pattern4  B6\/riboflavin pattern5 Methyl donor pattern6 Variance explained 25.1% 18.2% 15.5% Folate 0.0909 0.4323 -0.1669 PLP 0.1102 0.5140 0.1268 Total B-12 0.5772 0.0736 -0.0138 Riboflavin -0.0715 0.7129 0.0012 Methionine 0.0335 -0.0036 0.6953 tHcy -0.5107 0.1866 -0.0187 HoloTC 0.6150 0.0140 0.0001 Betaine -0.0344 0.0191 0.6871 1First trimester B-12 pattern refers to maternal total B-12 and holoTC loadings;2Methyl donor\/riboflavin pattern refers to maternal folate, riboflavin and betaine loadings;3Methionine patters refers to maternal methionine loadings; 4Second trimester B-12 pattern refers to maternal total B-12 and holoTC, and tHcy loadings;5B6-riboflavin pattern refers to maternal PLP and riboflavin loadings;6Methyl donor pattern refers to maternal methionine and betaine loadings. B-12, vitamin B-12; HoloTC, holotranscobalamin; PLP, pyridoxal-5\u2019-phosphate; tHcy, total homocysteine.              96                     Figure 3.4. Retained maternal methyl nutrient patterns in the first trimester  B-2, riboflavin; HoloTC, holotranscobalamin; PLP, pyridoxal 5\u2032 phosphate; tHcy, total homocysteine.                  Figure 3.5. Retained maternal methyl nutrient patterns in the second trimester  B-2, riboflavin; HoloTC, holotranscobalamin; PLP, pyridoxal 5\u2032 phosphate; tHcy, total homocysteine.     97 Table 3.4. Characteristics of apparently healthy Canadian pregnant women included in first or second trimester PCA1  First trimester PCA Second trimester PCA  Maternal characteristics n    n  P Age, y  505 31.0 (28.0-34.0) 393 31.0 (28.0-35.0) 0.45 Weeks of gestation  499 11.4 (10.7-12.3) 393 16.3 (15.7-17.1) <0.001 Pre-pregnancy BMI, kg\/m2 390 23.5 (21.2-26.9) 306 23.4 (21.2-26.4) 0.55 BMI categories,      0.94     Underweight, % (BMI) 14 3.59 (17.8; 16.8-18.0) 12 3.92 (17.7; 17.1-17.9)     Healthy weight, % (BMI) 231 59.2 (21.8; 20.6-23.4) 189 61.8 (21.9; 20.7-23.4)     Overweight, % (BMI) 95 24.4 (26.9; 26.0-28.3) 71 23.2 (26.7; 26.0-28.3)     Obesity, % (BMI) 50 12.8 (33.3; 31.4-36.4) 34 11.1 (33.1; 31.1-36.3) GD, % 72 14.7 63  16.2 0.54 HTN, % 22 4.50 18 4.60 0.92 Ethnicity, %     European      South Asian 505 263  242   52.1  47.9  393 199  194   50.6 49.4  0.67 Parity, %      Nulliparous     Multiparous 488 254  235   51.9  48.1  388 197  191   50.8  49.2 0.73 Median neighborhood income per year, CAD 488 51,217 (45,066-55,370) 387 51,439 (43,778-55,370) 0.99 1Values are medians (IQR) for continuous variables or prevalence (%) for categorical variables. P values are from Wilcoxon rank-sum test for comparison of continuous variables and chi-square test for categorical variables between pregnant women included in first and second trimester PCA analysis. Pre-pregnancy BMI categories; underweight (BMI <18.5 kg\/m2), healthy weight (BMI \u226518.5 to 24.9 kg\/m2), overweight (BMI \u226525 to 29.9 kg\/m2) and obese (BMI \u226530 kg\/m2). BMI, body mass index; GDM; gestational diabetes; HTN, hypertensive disorder of pregnancy; PCA, principal component analysis.                    98   Table 3.5 Characteristics of newborns of pregnant women included in the first or second trimester PCA 1  1Values are median (IQR) for continuous variables or prevalence (%) for categorical variables. P values are from Wilcoxon rank-sum test for comparison of continuous variables and chi-square test for categorical variables between newborns included in first and second trimester PCA analysis. AGA, appropriate for gestational age; BL, birth length; BMI, body mass index; BW, birth weight; BW\/BL, birth weight-to-length ratio; GA, gestational age; HC, head circumference; LGA, large for gestational age; LBW, low birth weight; PCA, principal component analysis; PI, ponderal index; SGA, small for gestational age.      First trimester PCA  Second trimester PCA   Neonatal characteristics  n  n  P  Ethnicity, %     European     South Asian   263 242  52.1 47.9  199 194  50.6 49.4  0.67 Neonatal sex, %     Female     Male  245 245  50.0 50.0  181 206  46.8 53.2  0.34 Birth status, %      Term     Preterm 490  92.7 7.4 387  93.3 6.72 0.71 Delivery type, %      Vaginal birth     Caesarean section 489  62.0 38.0 388  38.4 61.6 0.91 GA at birth, weeks 490 39.0 (38.0-40.0) 387 39.0 (38.0-40.0) 0.62 BW, g 490 3380 (3075-3682) 387 3390 (3111-3715) 0.26 LBW, % 14 2.86 11 2.84 0.99 BW, z-score  490 0.06 (-0.53-0.66) 387 0.16 (-0.52-0.71) 0.38 BL, cm 481 51.0 (50.0-53.0) 380 51.0 (50.0-53.0) 0.22 BL, z-score 481 0.05 (-0.32-0.49) 380 0.10 (-0.28-0.54) 0.31 HC, cm 482 35.0 (34.0-36.0) 381 35.0 (34.0-36.0) 0.11 HC z-score 482 0.23 (-0.21-0.60) 381 0.25 (-0.15-0.61) 0.29 PI, kg\/m3 481 25.0 (23.3-26.8) 380 25.0 (23.2-26.6) 0.99 BW\/BL, kg\/m 481 6.57 (6.12-7.08) 380 6.57 (6.19-7.14) 0.36 BMI, kg\/m2 481 12.8 (11.9-13.7) 380 12.9 (12.0-13.7) 0.69 SGA, % 27 5.51 17 4.40 0.54 AGA, % 415 84.7 325 83.9  LGA, % 48 9.80 45 11.6    99 3.3.4 Association between first-trimester maternal methyl nutrient pattern scores and neonatal anthropometric outcomes  The first-trimester maternal methyl nutrient pattern scores were not significantly associated with neonatal anthropometric outcomes in this sample of term and preterm newborns (Table 3.6). After the exclusion of preterm newborns (n=36), however, a 1-unit increase in score of the methyl donor\/riboflavin pattern, which mainly reflects increasing maternal betaine, riboflavin and folate concentrations, was associated with a decrease of 0.18 cm and 0.06 in BL and BL z-score, respectively (P<0.05 for all). Among the biomarkers with the highest contribution for the methyl donor\/riboflavin pattern (i.e., factor loadings \u2265 |0.5|), only maternal log-transformed betaine was negatively associated with BL [(\u03b2=-0.53; 95%CI -1.04, -0.02, P=0.04). Additionally, maternal log-transformed betaine was weakly associated with BL z-score (\u03b2=-0.17, 95%CI -0.33, -0.01; P=0.03)] after adjustment for confounding factors. No significant association was found between log folate or log riboflavin and BL or BL z-score, after controlling for confounding variables. Consistent with findings obtained using BW z-score calculated from PSBC growth charts, BW z-score calculated using Kramer\u2019s growth chart was not associated with maternal methyl nutrient patterns (P<0.05) before or after the exclusion of preterm newborns (Table 3.7)  3.3.5 Association between second-trimester maternal methyl nutrient pattern scores and neonatal anthropometric outcomes  In the second trimester, only the B-12 pattern score, but not methyl nutrient donor and B-6\/ riboflavin pattern scores, was associated with HC and HC z-score in the complete sample of newborns, including term and preterm newborns. The 1-unit increase in score of the B-12 pattern, which mainly reflects decreasing tHcy and increasing holoTC and B-12 concentrations, was   100 associated with a decrease of 0.13 cm in HC and 0.05 in HC z-score, after adjustment for confounding factors (P\u22640.05 for all) (Table 3.6) Among the biomarkers with a higher contribution to the B-12 pattern, only log total B-12 was negatively associated with HC (\u03b2=-0.43; 95%CI -0.77, -0.08, P=0.02) and HC z-score (\u03b2=-0.16; 95%CI -0.31, -0.02, P=0.03). The exclusion of mother-newborn pairs with preterm birth (n=26) did not significantly modify the association between B-12, methyl donor and riboflavin\/B-6 patterns and neonatal anthropometric outcomes. Additionally, BW z-score calculated using Kramer\u2019s growth chart was not associated with second-trimester maternal methyl nutrient patterns before or after the exclusion of preterm (Table 3.7)  101 Table 3.6 Associations of maternal methyl nutrient patterns in the first and second trimester of apparently healthy Canadian pregnant women with neonatal anthropometric outcomes P1 value is from linear regression model for the corresponding coefficient. Models adjusted for maternal age, gestational age at birth, pre-pregnancy BMI, maternal height, gestational diabetes, hypertensive disorder of pregnancy, neighborhood income quintiles, maternal ethnicity, neonatal sex and parity. P2 value is from linear regression model for the corresponding coefficient. Models adjusted for maternal age, pre-pregnancy BMI, maternal height, gestational diabetes, hypertensive disorder of pregnancy, neighborhood income quintiles, maternal ethnicity and parity.3B-12 pattern refers to maternal B-12 and holoTC loadings; 4Methyl donor\/riboflavin pattern refers to maternal folate, riboflavin and betaine loadings;5 Methionine pattern refers to maternal methionine loadings;6B-12 pattern refers to maternal B-12, holoTC and tHcy loadings;7B-6\/riboflavin pattern refers to maternal PLP and riboflavin loadings;8Methyl donor pattern refers to maternal betaine and methionine loadings.9Adjusted R2=0.25; 10Adjusted R2=0.07. B-6, vitamin B-6; B-12, vitamin B-12; BL, birth length; BMI, body mass index; BW, birth weight; BW\/BL, birth weight-to-length ratio; CI, confidence interval; HC, head circumference; PI, ponderal index. Median (IQR) neighborhood income, 1st quintile: $ 41,116 ($ 40,386-$ 42,062) CAD; 2nd quintile: $ 49,294 ($ 47,641-$ 50,586) CAD; 3rd quintile: $ 53,218 ($ 52,143-$ 54,713) CAD; 4th quintile: $ 57,107 ($ 56,456-$ 59,251) CAD; 5th quintile: $ 65,716 ($ 63,478-$ 68,554) CAD.            BW BW z-score BL BL z-score HC HC  z-score First Trimester \u03b2 (95%CI) P1 \u03b2 (95%CI) P2 \u03b2 (95%CI) P1 \u03b2 (95%CI) P2 \u03b2 (95%CI) P1 \u03b2 (95%CI) P2 B-12 pattern3  -22.7  (-47.5,1.98)  0.07 -0.05  (-0.11,0.005) 0.07  -0.12 (-0.26,0.01)  0.07 -0.04 (-0.08,0.005) 0.09  -0.03 (-0.12,0.07)  0.58 -0.002  (-0.04,0.04) 0.93 Methyl donor\/ riboflavin pattern4  -23.4 (-53.7,6.89)  0.13  -0.06  (-0.13,0.009)  0.09  -0.13  (-0.30,0.03)  0.11  -0.05 (-0.09,0.003)  0.07  0.05 (-0.06,0.17)  0.38  0.02 (-0.03,0.07)  0.34 Methionine pattern5 -9.26 (-23,4.16) 0.57 0.01 (-0.06,0.09) 0.71 0.13 (-0.05,0.31) 0.16 0.03 (-0.02,0.09) 0.23 0.02 (-0.10,0.14) 0.73 -0.0002 (-0.05,0.05) 0.99 Second Trimester \u03b2 (95%CI) P1 \u03b2 (95%CI) P2 \u03b2 (95%CI) P1 \u03b2 (95%CI) P2 \u03b2 (95%CI) P1 \u03b2 (95%CI) P2 B-12 pattern6 -17.8 (-47.1,11.5) 0.23 -0.04  (-0.11,0.02) 0.21 -0.13 (-0.28,0.03) 0.11 -0.04 (-0.08,0.01) 0.15 -0.13 (-0.24, -0.03) 0.019 -0.05  (-0.09, -0.009) 0.0210 B6\/ riboflavin pattern7 9.55 (-24.2,43.3) 0.58 0.02 (-0.06,0.09) 0.67 -0.09 (-0.27,0.09) 0.35 -0.03 (-0.09,0.02) 0.24 0.05 (-0.08,0.17) 0.46 0.01 (-0.04,0.06) 0.62 Methyl donor pattern8 8.13 (-28.9,45.1) 0.67 0.001 (-0.08,0.08) 0.99 -0.11 (-0.31,0.08) 0.26 -0.04  (-0.09,0.02) 0.20 0.04 (-0.09,0.18) 0.54 0.03 (-0.03,0.08) 0.34   102  Table 3.6 Associations of maternal methyl nutrient patterns in the first and second trimester of apparently healthy Canadian pregnant women with neonatal anthropometric outcomes (con) P1 value is from linear regression model for the corresponding coefficient. Models adjusted for maternal age, gestational age at birth, pre-pregnancy BMI, maternal height, gestational diabetes, hypertensive disorder of pregnancy, neighborhood income quintiles, maternal ethnicity, neonatal sex and parity. BMI, body mass index..2B-12 pattern refers to maternal B-12 and holoTC loadings; 3Methyl donor\/riboflavin pattern refers to maternal folate, riboflavin and betaine loadings;4 Methionine pattern refers to maternal methionine loadings;5B-12 refers to maternal B-12, holoTC and tHcy loadings;6B-6\/riboflavin refers to maternal PLP and riboflavin loadings;7Methyl donor refers to maternal betaine and methionine loadings. B-6, vitamin B-6; B-12, vitamin B-12; BL, birth length; BMI, body mass index; BW, birth weight; BW\/BL, weight-to-length ratio; CI, confidence interval; HC, head circumference; PI, ponderal index. Median (IQR) neighborhood income, 1st quintile: $ 41,116 ($ 40,386-$ 42,062) CAD; 2nd quintile: $ 49,294 ($ 47,641-$ 50,586) CAD; 3rd quintile: $ 53,218 ($ 52,143-$ 54,713) CAD; 4th quintile: $ 57,107 ($ 56,456-$ 59,251) CAD; 5th quintile: $ 65,716 ($ 63,478-$ 68,554) CAD.            BW\/BL BMI PI First Trimester \u03b2 (95%CI) P1 \u03b2 (95%CI) P1 \u03b2 (95%CI) P1 B-12 pattern2  -0.03  (-0.07,0.01)  0.19  -0.03 (-0.12,0.06)  0.49  -0.009 (-0.20,0.18)  0.92 Methyl donor\/riboflavin pattern3  -0.03  (-0.08,0.03)  0.31  -0.02 (-0.12,0.09)  0.71  0.02 (-0.22,0.26)  0.86 Methionine pattern4 0.006 (-0.05,0.06) 0.82 -0.01 (-0.12,0.09) 0.83 -0.08 (-0.33,0.18) 0.56 Second Trimester \u03b2 (95%CI) P1 \u03b2 (95%CI) P1 \u03b2 (95%CI) P1 B-12 pattern5 -0.02 (-0.08,0.03) 0.34 -0.02  (-0.12,0.07) 0.67 0.01 (-0.20,0.23) 0.90 B6\/riboflavin pattern6 0.03  (-0.03,0.09) 0.27 0.09 (-0.02,0.20) 0.12 0.22 (-0.03,0.47) 0.08 Methyl donor pattern7 0.03 (-0.04,0.09) 0.39 0.07 (-0.05,0.20) 0.25 0.19 (-0.08,0.46) 0.18   103   Table 3.7 Associations of maternal methyl nutrient patterns in the first and second trimester of apparently healthy Canadian pregnant women with neonatal anthropometric outcomes using Kramer\u2019s growth charts  P1 value is from linear regression model for the corresponding coefficient. 1z-score calculated using Kramer\u2019s growth chart; 2z-scores calculated using Kramer\u2019s growth chart including only full-term newborns. Models adjusted for maternal age, pre-pregnancy BMI, maternal height, gestational diabetes, hypertensive disorder of pregnancy, neighborhood income quintiles, maternal ethnicity and parity. 3B-12 pattern refers to maternal total B-12 and holoTC loadings, 4Methyl donor\/riboflavin pattern refers to maternal folate, betaine, and riboflavin loadings; 5Methionine pattern refers to maternal methionine loadings; 6B-12 pattern refers to maternal total B-12, holoTC and Hcy loadings; 7B6\/riboflavin pattern refers to maternal PLP and riboflavin loadings; 8Methyl donor pattern refers to maternal methionine and betaine loadings; BW, birth weight; CI, confidence interval; Median (IQR) neighborhood income, 1st quintile: $ 41,116 ($ 40,386-$ 42,062) CAD; 2nd quintile: $ 49,294 ($ 47,641-$ 50,586) CAD; 3rd quintile: $ 53,218 ($ 52,143-$ 54,713) CAD; 4th quintile: $ 57,107 ($ 56,456-$ 59,251) CAD; 5th quintile: $ 65,716 ($ 63,478-$ 68,554) CAD.         BW z-score  BW z-score First trimester \u03b2 (95%CI) P1 \u03b2 (95%CI) P2 B-12 pattern3 -0.05 (-0.11,0.003) 0.07 -0.05 (-0.11,0.009) 0.10 Methyl donor\/riboflavin pattern4 -0.06 (-0.13,0.007) 0.08 -0.07 (-0.14,0.003) 0.06 Methionine pattern5 0.02 (-0.06,0.09) 0.70 0.02 (-0.06,0.09) 0.65 Second trimester \u03b2 (95%CI) P1 \u03b2 (95%CI) P2 B-12 pattern6 -0.04 (-0.11,0.02) 0.20 -0.03 (-0.09,0.04) 0.40 B6\/riboflavin pattern7 0.02 (-0.06,0.09) 0.67 0.03 (-0.05,0.11) 0.49 Methyl donor pattern8 -0.002 (-0.09,0.08) 0.95 0.003 (-0.08,0.09) 0.95   104 3.4 Discussion To the best of my knowledge, this is the first study to describe first and second-trimester maternal methyl nutrient patterns using a PCA approach, and to explore the association between maternal methyl nutrient patterns and neonatal anthropometric outcomes. Maternal methyl nutrient concentrations. First- and second-trimester serum betaine concentrations (18.8 and 14.5 \u00b5mol\/L) in this sample of 723 pregnant women were similar to fasting plasma betaine concentrations reported in Spanish (~21 and ~15 \u00b5mol\/L at <12 and 15 weeks of gestation, respectively) (64,275) and Canadian pregnant women (13 \u00b5mol\/L at 16 weeks of gestation) (36). First- and second-trimester serum holoTC concentrations (83.6 and 77.9 pmol\/L) were also comparable to those reported in Spanish pregnant women (86.0 and 72.4 pmol\/L at <12 and 15 weeks of gestation, respectively) (64,275). Serum total B-12 concentration was similar to findings of other pregnancy cohorts in Canada, especially when comparing total B-12 concentration of European pregnant women in this study (247 and 224 pmol\/L in first and second trimester, respectively) with cohorts that mostly included European women in Vancouver (259 pmol\/L at 16 weeks of gestation) (36) and Toronto (219 pmol\/L at 12-16 weeks of gestation) (72).  Serum tHcy concentrations in both trimesters (5.0 and 4.4 \u00b5mol\/L) were also comparable to plasma tHcy concentration reported for Vancouver pregnant women at 16 weeks of gestation (4.1 \u00b5mol\/L) (36). In contrast, serum PLP concentration in this current study (90.1 and 69.4 nmol\/L) was lower compared to pregnant women from Toronto (107 nmol\/L at 12-16 weeks of gestation) (71), while serum folate concentration in both trimesters (~68 nmol\/L) was higher compared to those reported in previous pregnancy studies conducted in European and Asian countries (28,34,39,65). Serum riboflavin concentration (~22 nmol\/L) was almost two-fold   105 higher than that in non-fasting plasma samples at 18 gestational weeks (11.8 \u00b1 0.26 nmol\/L) of pregnant women in the Norwegian Mother and Child Cohort Study (MoBa) (276). Samples analyzed in the current study were collected under non-fasting conditions, which may partially explain the discrepancies found in biomarker concentrations. Recent dietary intake of folate and riboflavin rich foods or supplements acutely raise circulating plasma concentrations of these metabolites (16,277\u2013279). Information on supplement use was not available in this retrospective study. Previous pregnancy cohort studies in Canada reported about 90% of pregnant women to consume a folic acid containing multivitamin-multimineral prenatal supplement in the first and second trimesters (68,71). A major contributor to the higher serum folate concentration in the current study is likely the fact that most prenatal supplements contain folic acid doses of \u22651000\u00b5g (206) as well as the extended use of folic acid containing supplements throughout pregnancy until post-partum (11) in Canada. Results of the MoBa study showed 36% of women consumed folic acid supplements in the first 4 months of pregnancy, whereas 31% consumed multivitamin-minerals supplements (280). The higher prevalence of prenatal supplement use and the mandatory flour fortification with riboflavin in Canada, may also explain the higher serum riboflavin concentration observed in this study. Additionally, differences in analytical methods used to determine PLP concentration (i.e., LC-MS\/MS and nonradioactive enzymatic assay) may also explain the differences that we found between PLP concentration of the current study and PLP concentration reported in the PREFORM study (281).  First and second-trimester maternal methyl nutrient patterns. Of the maternal methyl nutrient patterns, the B-12 patterns explained the highest variation of the first (26.7%) and second trimester (25.1%) biomarker data, with log total B-12 and log holoTC as the main positive contributors to the first-trimester B-12 pattern. The derivation of the B-12 pattern is likely explained by the stronger correlation between first-trimester log B-12 and holoTC (r=0.70,   106 P<0.001), compared to correlations among other methyl nutrients included in PCA 1. Similar results between the B-12 biomarkers was reported in first-trimester Spanish pregnant women (r=0.62, P<0.001) (38). Despite that the loading of log tHcy to the B-12 pattern was slightly below the cut-off used to characterize the pattern (i.e., |0.5|), this biomarker still had a meaningful and negative loading to the B-12 pattern (factor loading: -0.47), and it was correlated with total B-12 (r=-0.30) and holoTC (r=-0.35) (P<0.001 for both), which is expected because serum tHcy is a functional B-12 indicator, and a specific biomarker of B-12 status in folate-replete populations such as the Canadian population (47). Similarly, in the second trimester, log B-12 and log holoTC were positive contributors, while log tHcy was a negative contributor to the B-12 pattern. Also, these results are explained by the stronger correlation of the B-12 biomarkers (total B-12 and holoTC: r=0.68; tHcy and B-12: r=-0.30; tHcy and holoTC: r=-0.39) (P<0.001 for all). The negative contribution of log tHcy to the pattern is in line with the inverse relationship of B-12 with tHcy (47). Given that the derivation of the patterns is partly based on the correlation of the data, we speculate that under folate-replete conditions such as in the Canadian population, B-12, becomes a critical nutrient for the remethylation of Hcy to methionine, which results in the strongest correlation between B-12 biomarkers (i.e., B-12, holoTC and tHcy), and subsequently, the derivation of the B-12 pattern in both trimesters. After the B-12 pattern, the second highest variation of the methyl nutrient concentrations was explained by the first-trimester methyl-donor\/riboflavin pattern with log folate, betaine and riboflavin being the main contributors, and the second-trimester B-6\/riboflavin pattern with log riboflavin and B-6 as the main contributors. The positive correlation between first-trimester log folate and riboflavin (r=0.25, P<0.001) likely explained the derivation of methyl-donor\/riboflavin pattern, which may be related to the coenzyme role of riboflavin in the form of FAD for the   107 formation of 5-MTHF (48). Riboflavin, in the form of FMN, functions also as a coenzyme in the formation of the B-6 coenzyme form PLP by the action of pyridoxine-phosphate oxidase (49), which may explain the positive correlation between second-trimester log PLP and riboflavin (r=0.35, P<0.001), and the subsequent derivation the B-6\/riboflavin pattern. Lastly, first-trimester methionine pattern and the second-trimester methyl donor pattern explained 15% and 16% of the maternal methyl nutrients variation, respectively. Only log methionine was a positive contributor for the methionine pattern, which can be due to the lack of a strong correlation between log methionine and other methyl nutrients, and the combined regulation of methionine concentration by nutritional (e.g., folate, dietary protein intake) and hormonal factors (i.e., insulin concentration) (1,50). In the second trimester, not only log methionine but also log betaine were positive contributors to the methyl donor pattern, which is likely explained by the significant correlation between these methyl nutrients (r=0.23, P=0.001). We speculate that a higher remethylation rate in the second compared to the first trimester (51), in addition to an impaired use of folate as a methyl donor for the remethylation of homocysteine to methionine, due to a high folic acid intake which acts as folate antagonist (10), promote the use of betaine as a methyl donor, resulting in a stronger correlation between log betaine and methionine.   First-trimester maternal methyl donor\/riboflavin pattern and neonatal anthropometric outcomes. First-trimester methyl donor\/riboflavin pattern was negatively associated with BL and BL z-score, only in full-term newborns. We speculate that differences in median (IQR) folate concentrations (i.e., mothers of term babies: ~67 (51-81) nmol\/L, mothers of preterm babies: ~79 (60-89) nmol\/L) may modify the pattern score, and subsequently, the association with BL and BL z-score. Among the main contributors to this pattern, only log betaine was negatively associated with BL and BL z-score. Our results are similar to those reported in the GUSTO study, in which a 1-unit increase in plasma betaine concentration at late-pregnancy was significantly associated   108 with a decrease of 0.29 cm in BL (39). These findings  can be partially explained by the previously reported critical role of betaine, as a methyl donor, in fetal programming and de novo DNA methylation (47,102,263). Results from animal studies have indicated that maternal supplementation with betaine decreased the expression of genes involved in adipose tissue metabolism (186), increased serum growth-hormone concentration (282) and regulated the expression of IGF-1 and IGF-2 genes in the offspring, which may affect fetal growth, and subsequently, anthropometric outcomes of the offspring (185). However, the relationship between maternal betaine concentration and DNA methylation levels of fetal growth-related genes needs to be confirmed in human studies.   Second-trimester maternal B-12 pattern and neonatal anthropometric outcomes.  In the second trimester, an inverse association between the B-12 pattern scores with HC and HC z-score was found. Among the main contributors to this pattern, only log total B-12 was negatively associated with HC and HC-z-score. We hypothesized that the relationship between maternal B-12 concentration and HC and HC z-score became significant at the second trimester due to the occurrence of the \u201contogenesis window\u201d of brain development. The formation of the frontal and occipital sulci, and the neural cellular migration peak occurs mainly between 8-24 weeks of gestation(283,284). Given that DNA methylation is key for cellular and organ differentiation (e.g., neural cell), we speculate that B-12 may become critical for the provision of methyl groups for DNA methylation reactions in this folate-replete population. Despite that available evidence from pregnancy studies have reported no significant linear associations between maternal B-12 and neonatal anthropometric outcomes (28,220), results of a pregnancy study in China showed that maternal serum total B-12 concentration at delivery was negatively associated with DNA   109 methylation levels of IGF-2 in cord blood, which highlight the potential role of second and third trimester B-12 in fetal growth and brain programming (224). Strengths and limitations. To my knowledge, this is the first observational study in a large sample size of pregnant women that explored the interrelationship of maternal methyl nutrient concentrations by using PCA. PCA allowed us to derive non-correlated maternal methyl nutrient patterns, which prevented collinearity between maternal methyl nutrient concentrations in the regression models. PSBC growth charts were used to calculate BW, BL, and HC-z-scores, which are based on the BC population and portray neonatal anthropometric outcomes in the context of the BC health care system. Limitations of the current study include the equal proportion of apparently healthy South Asian and European pregnant women, which limits the generalization of our results to other populations. Additionally, this study lacks other maternal methyl nutrients in the PCAs such as free choline concentration. Due to the instability of maternal free choline in serum samples (261), data for maternal free choline concentration was not included in the PCA analysis.  In conclusion, this study shows that first and second-trimester maternal B-12 patterns explained the highest variation of the maternal methyl nutrient concentrations in both trimesters in 700 pregnant women residing in BC. Second-trimester maternal B-12 pattern was negatively associated with HC and HC z-score which may suggest a critical role of B-12 on fetal brain programming. However, further pregnancy studies exploring the interrelationship of maternal methyl nutrients using PCA are needed to assess the reproducibility of the derivation of the maternal methyl nutrients patterns and their relationship with neonatal anthropometric outcomes in other populations.   110 Chapter 4: COMPARISON BETWEEN DNA METHYLATION LEVELS IN CPG SITES OF FETAL GROWTH AND OBESITY-RELATED GENES IN 3-MONTH OLD INFANTS OF MOTHERS WITH RBC FOLATE CONCENTRATION WITHIN AND ABOVE THE REFERENCE RANGE IN EARLY PREGNANCY   4.1 Introduction Folate is a methyl donor nutrient that has a critical role in nucleotide synthesis and methylation reactions, such as DNA methylation, through one-carbon metabolism (7). Findings from two randomized controlled trials conducted in pregnant women with high risk for NTDs and low risk for NTDs showed that supplementation with 4000\u00b5g\/day (8) and 0.8\u00b5g\/day (9), respectively, of folic acid, the synthetic form of folate, before and during the first trimester of pregnancy reduced the incidence of NTDs by >60%. Currently, the SOGC recommends a daily dose of 400\u00b5g of folic acid, starting at least 3 months before pregnant to lactation, for low risk women to prevent the occurrence of NTD (11). However, most of the available prenatal supplements on the Canadian market contain 1000\u00b5g of folic acid (12,13). According to the Canadian Maternity Experiences Survey, 58% of Canadian women took multivitamin supplements containing folic acid during the three months before conception, and 90% took folic acid supplements at some point during the first trimester (206). Prospective cohort studies conducted in Canadian pregnant women (i.e., the PREFORM and APrON studies) have reported that >90% of the participants consume folic acid containing prenatal supplements by the end of the first trimester (16,68). Findings of from the PREFORM and the APrON studies, both   111 Canadian pregnancy studies, indicated that >50% of pregnant women had high RBC folate concentration (>1360 nmol\/L) in early and late pregnancy (16,67).  Early pregnancy is considered a critical window for fetal growth programming mainly due to the occurrence of de novo DNA methylation in the offspring (18). Evidence from previous pregnancy studies have shown that maternal factors such as pre-pregnancy BMI may influence fetal growth and obesity programming of the offspring through modifications in fetal DNA methylation levels (285,286). However, evidence of the relationship between maternal folate status and DNA methylation levels of fetal growth or obesity related genes in the offspring is limited and with conflicting results. Results of a previous study conducted in North American pregnant women (n=496 mother-newborn dyads) with mean RBC folate concentration of ~494 nmol\/L, indicated that DNA methylation levels of IGF-2 measured in cord blood, were higher in offspring of pregnant women with low maternal RBC folate concentration (i.e., second quartile) at ~12 weeks of gestation, compared to offspring of pregnant women with RBC folate concentration in the third and fourth quartiles (22). In contrast, findings from two pregnancy studies conducted in the UK, where women had RBC folate concentration <500 nmol\/L, showed no significant association between maternal RBC folate concentration at early pregnancy (i.e., <20 weeks of gestation) and DNA methylation levels of IGF-2 measured in cord blood (203,225). Whether early pregnancy folate status is associated with infant DNA methylation levels of CpG sites associated with genes involved in fetal growth and obesity programming in folate-replete populations is currently unknown. Given the high maternal folate status reported in Canadian pregnant women, the current study sought to compare DNA methylation levels of CpG sites associated with fetal growth and obesity programming between infants of pregnant women with high folate status (i.e., \u22651360 nmol\/L) and folate status in the reference range (i.e., <1360 nmol\/) measured before 22 weeks of gestation, in Canadian mother-infant dyads.   112 4.2 Study design and methods The current study is nested within the APrON study. The APrON study is a prospective cohort study of pregnant women and their children with follow-up visits until 3 years postpartum, which was conducted in the central and southern regions of Alberta, Canada. Details of the APrON study have been previously described (287). Briefly, during the study, 2192 pregnant women, living in Edmonton, Calgary, or nearby areas, were enrolled between May 2009 and July 2012 if they were >16 years, at <27 weeks of gestation, were able to speak in English, and who had planned to not move out of the province during the time of the study. A study visit was conducted in the first trimester in women recruited between 1-13 weeks of gestation, while for those recruited after, a study visit was conducted at the second (14-26 weeks of gestation) and third (27-42 weeks of gestation) trimester. All study visits were conducted by trained research assistants. Non-fasting maternal blood samples were collected at each study visit. Blood or cheek swab samples were collected from infants at 3 months of age.  The APrON study was approved by the University of Calgary Health Research Ethics Board and the University of Alberta Health Research Ethics Biomedical Panel (REB14-1702, Pro00002954). The present study was reviewed and approved by The University of British Columbia \/ Children\u2019s and Women\u2019s Health Centre of British Columbia Research Ethics Board, Vancouver, Canada (H17-03141). Inclusion criteria The present study included mother-infant dyads that have available maternal samples collected at <22 weeks of gestation, and information of DNA methylation levels measured in infants at 3 months of age, from women who reported to not smoke (i.e., non-smokers or smoked <100 cigarettes in her life) or consume alcohol (i.e., never consumed alcohol) or drug users (i.e., never used recreational drugs) or had type 1 or type 2 diabetes.    113 4.2.1 Maternal lifestyle, demographic and anthropometric information Data about maternal age (years), ethnicity [self-reported; Black (i.e., African, African North American), Caucasian\/white (i.e., English, French, German, Greek, Irish, Polish, Russian, Scottish, Ukrainian) Chinese, Filipino, Japanese, Korean, Latin American, Native\/Aboriginal Peoples of North America (i.e., First Nations, North American Indian, Metis, Inuit) South Asian, South East Asian, Arab, West Asian], education (less than high school, completed high school diploma, completed university undergraduate degree diploma, completed post-graduate degree, completed technical, vocational school, or trade, business\/community college), total family income per year (<$20, 0000, $20,000-$39,999, $40,000-$69,999, $70,000-$99,999, \u2265$100,000), marital status (single, common-law, married, widowed, divorced and separated), parity (nulliparous\/multiparous), GD, type 1 and type 2 diabetes, primary hypertension, gestational hypertension, preeclampsia, smoking history, drug use, and alcohol consumption was collected at the first study visit (287). Additionally, maternal weight (kg) and height (cm) were measured at each study visit by trained research assistants. Weight was measured using a professional scale with 200 kg capacity and variation of 0.01 kg Healthometer Professional 752KL, Pelstar LLC, Bridgeview, IL, USA. Height (m) was measured using a digital stadiometer (HM200P Portstad Portable Stadiometer, Charder, Seattle, WA, USA) with a variation of 0.1 cm.   4.2.2 Maternal supplement intake A supplement intake questionnaire (SIQ) developed by the APrON study team was used to estimate the daily supplement intake of vitamins and minerals (e.g., prenatal multi-vitamins) at each trimester, as described in details by Gomez et al (68). Briefly, the SIQ contained three sections. The first section included a list of 20 commonly used multivitamins\/minerals; the second section contained a list of 23 single-nutrient supplements, and the third section contained   114 a list of 43 herbal products, homeopathic remedies, probiotic, traditional medicines and animal-derived products.   4.2.3 Infant demographic and anthropometric information BW, BL, sex, birth status (term\/preterm), and gestational age at birth were obtained from medical records. Infant age (weeks) and infant weight (kg) information was collected by the APrON study\u2019s research team at the ~12-week post-partum visit. Infant weight was measured by calculating the difference in body weight between the mother holding her infant (undressed except for a dry diaper) and the mother alone using a scale Healthometer Professional 752KL, Pelstar LLC, IL, U.S.A with a precision of 0.01 kg (288). LBW was classified as BW <2,500 g (265). Infant length (cm) was obtained from medical records from a pediatrician appointment one week before the study visit, whereas infant age was provided by the mother.   4.2.4 Infant DNA methylation information Infant DNA methylation data were available for 246 infants. Briefly, a blood (n=109) or cheek swab (n=137) sample was collected from infants ~12 weeks post-partum. Blood samples were collected by a certified phlebotomist using either a butterfly needle or a 25G3\/4 standard infant needle and EDTA tubes. Blood samples were processed within 6 hours after collection and were centrifuged at 3000 r\/min for 15 min to separate the plasma, buffy coat (leukocytes and platelets), and erythrocytes. The buffy coat was separated, and processed leukocytes were stored in microcentrifuge tubes at -80\u00b0C for DNA extraction. For those infants who were not able to provide a blood sample, BEC samples were collected. BECs were collected by rolling a sterile cytology brush up and down 10 times along each cheek of the infant (duplicate). Both BECs samples were dried for 10\u201315 min in a drinking glass at room temperature and were placed and   115 sealed in a labeled plastic bag and then stored at -80\u00b0C until DNA extraction. DNA was extracted from leukocytes and BECs by cellular lysis and then it was purified using Gentra Buccal DNA purification kits. Approximately 750 ng of genomic DNA was bisulfite converted using the EZ DNA Methylation Kit (Zymo Research, Irvine, CA). After, \u223c160 ng of bisulfite converted DNA was processed using the 450K array, according to manufacturer\u2019s instructions (Illumina) (287) at the Kobor Laboratory at the University of British Columbia. The selection of CpG sites to include in the current study was based on the highest variability of DNA methylation levels at each CpG site associated with LEP, LEP-R, HIF3\u03b1 , IGF-2, RXRA, DNMT-1, DNMT-3A and DNMT-3B genes according to the information provided in the BEcon platform (289). The detailed list of CpG sites is provided in Table 4.1. Table 4.1 CpG sites associated with fetal growth and obesity-related genes included in this study LEP  LEP receptor HIF3\u03b1 IGF-2 RXRA DNMT-1 DNMT-3A DNMT-3B cg00666422 cg03050981 cg01552731 cg00570518 cg02127980 cg07627628 cg00277048 cg00300969 cg00840332 cg11228758 cg07684068 cg02045936 cg13487983 cg21063296 cg00912598 cg09835408 cg03084214 cg11807188 cg12068280 cg02807948 cg13510651 cg21892967 cg05544807 cg13788819 cg07464571 cg23688719 cg14153927 cg07096953 cg13595495 cg01347596  cg09986894 cg14224313 cg12782180  cg16672562 cg05384664 cg13677043  cg10749994 cg17475857 cg13381984  cg20667364 cg11717189 cg13689699  cg11354105 cg17482740 cg25435800  cg22891070 cg12528452 cg13847322  cg15150970 cg21235334 cg19594666  cg23548163 cg13165070 cg13931640  cg15843262 cg22052056 cg26814075  cg25196389 cg26401390 cg13941235  cg19489797 cg22605822   cg25460031  cg14051662  cg20303441 cg24403338   cg26749414  cg14121282  cg20669908    cg27146050  cg14236758  cg26544247      cg14472716  cg27369452      cg14570632  cg00856404      cg21201924       116 4.2.5 Biochemical analyses 4.2.5.1 Blood processing  Non-fasting maternal samples were collected by a certified phlebotomist. Blood samples were drawn using vacutainer tubes with EDTA or serum separator gel and processed immediately after collection. RBC samples were aliquoted and stored at \u221280 \u00b0C (67,287).  4.2.5.2 Biomarker quantification  RBC folate concentration was determined in a sub-sample of pregnant women (n=154) using an on-capture method. Details of the determination of RBC folate concentration have been described previously (67,287). Briefly, a hemolysate (1:10) was prepared directly after blood sampling. The hemolysate was prepared using 0.1 mL of blood (with EDTA) and 0.9 mL ascorbic acid (1%). Then, the hemolysate was incubated at 37\uf0b0C for 30\u00b15 minutes and underwent a 1:2 dilution with AxSYM Folate RBC Protein Diluent. Samples were analysed using the AXSYM\u00ae analyzer (Abbott, Mississauga, ON, Canada) according to the manufacturer\u2019s instructions. RBC folate concentration was classified into two groups; 1) RBC folate concentration \u22641360 nmol\/L, and 2) RBC folate concentration >1360 nmol\/L, the latter indicated the cut-off for high folate status in the adult population (290).   4.2.6 Statistical analysis Descriptive analysis The normality of the data was tested using a Shapiro-Wilk test. Maternal age is normally distributed and is presented as mean (SD). All other continuous variables had skewed distributions and, thus, are presented as median (IQR). Categorical maternal and infant variables   117 are presented as prevalence (%). Differences in the percentage of pregnant women with RBC folate concentration \u22641360 nmol\/L and those with RBC folate concentration >1360 nmol\/L between tissues (i.e., BECs and buffy coat) were compared using a chi-square test. An independent Student\u2019s t-test was used to compare maternal age between pregnant women with RBC folate concentration \u22641360 nmol\/L and those with RBC folate concentration >1360 nmol\/L. A Wilcoxon rank-sum test was used to compare weeks of gestation at blood collection, gestational age at birth, BW, BL, infant age, infant weight and infant length between infants of pregnant women with RBC folate concentration \u22641360 nmol\/L and those with RBC folate concentration  >1360 nmol\/L, while parity, birth status (term\/preterm), infant sex, and LBW were compared using a chi-square test. Because the sample size for at least one cell was <5, pre-pregnancy BMI categories (i.e., underweight, overweight, healthy weight and obesity) and maternal ethnicity between pregnant women with RBC folate concentration \u22641360 nmol\/L and pregnant women with RBC folate >1360 nmol\/L were compared using a Fisher\u2019s exact test.  Furthermore, an independent Student\u2019s t-test was used to compare maternal age between the total sample of pregnant women (n=246) and the subset of pregnant women with available maternal RBC folate concentration (n=154). A Wilcoxon rank-sum test was used to compare weeks of gestation, gestational age at birth, BW, BL, infant age, infant weight and infant length between the total sample of pregnant women and the subset of pregnant women with available maternal RBC folate concentration. Additionally, a chi-square test was used to compare pre-pregnancy BMI categories (i.e., underweight, healthy weight, overweight, and obesity), parity, maternal ethnicity, infant sex, birth status (preterm\/term) and LBW.     118 DNA methylation analysis A Wilcoxon-rank sum test was used to compare DNA methylation levels (i.e., \u03b2-values) of each CpG site measured in infants of pregnant women with RBC folate concentration \u22641360 nmol\/L and infants of pregnant women with RBC folate concentration >1360 nmol\/L, separately by tissue type (291). The \u03b2-value is calculated as a ratio between the intensities of methylated probes divided by the sum of unmethylated + methylated probes (\u03b2=\ud835\udc40\ud835\udc52\ud835\udc61\u210e\ud835\udc66\ud835\udc59\ud835\udc4e\ud835\udc61\ud835\udc52\ud835\udc51 \ud835\udc56\ud835\udc5b\ud835\udc61\ud835\udc52\ud835\udc5b\ud835\udc60\ud835\udc56\ud835\udc61\ud835\udc66 \ud835\udc5d\ud835\udc5f\ud835\udc5c\ud835\udc4f\ud835\udc52\ud835\udc60\ud835\udc48\ud835\udc5b\ud835\udc5a\ud835\udc52\ud835\udc61\u210e\ud835\udc66\ud835\udc59\ud835\udc4e\ud835\udc61\ud835\udc52\ud835\udc51 \ud835\udc56\ud835\udc5b\ud835\udc61\ud835\udc52\ud835\udc5b\ud835\udc60\ud835\udc56\ud835\udc61\ud835\udc66 \ud835\udc5d\ud835\udc5f\ud835\udc5c\ud835\udc4f\ud835\udc52\ud835\udc60+\ud835\udc40\ud835\udc52\ud835\udc61\u210e\ud835\udc66\ud835\udc59\ud835\udc4e\ud835\udc61\ud835\udc52\ud835\udc51 \ud835\udc56\ud835\udc5b\ud835\udc61\ud835\udc52\ud835\udc5b\ud835\udc60\ud835\udc56\ud835\udc61\ud835\udc66 \ud835\udc5d\ud835\udc5f\ud835\udc5c\ud835\udc4f\ud835\udc52\ud835\udc60) (292). Results were corrected for multiple comparisons using the Benjamini-Hochberg test. All the statistical analyses were conducted using Stata version IC 15 (StataCorp).  119 4.3 Results 4.3.1 Maternal characteristics  Maternal characteristics are presented in Table 4.2. For the sample of 246 women mean \u00b1 SD maternal age was 30.9 \u00b13.8 y. In the present study, 78% identified as Caucasian, and 70% had a completed university degree or postgraduate level degree, while only 2% had an uncompleted high school diploma. Furthermore, >70% of pregnant women had a total annual family income \u2265 $ 70,000, whereas only 2% had a total annual family income <$ 20,000. Most of pregnant women (87.5%) were married, 9.8% were living with their partners, and only 2.5% were single. Regarding the BMI classification, 38% of pregnant women had overweight or obesity, whereas 59% had a healthy weight. Only 1.2% (n=3) and 0.8% (n=2) had GD and preeclampsia, respectively, while no pregnant women had primary hypertension, gestational hypertension or type 1 or type 2 diabetes. In total, 97.5% of pregnant women consumed a daily folic acid supplement in early pregnancy. The median daily folic acid dose used was 1000\u00b5g (IQR: 862-1000\u00b5g; minimum-maximum 14.3-6000 \u00b5g), and 23.8% consumed >1000\u00b5g folic acid\/day. Of the total sample of pregnant women (n=246), 154 had available RBC folate concentration data. Median (IQR) RBC folate concentration was 1334 (1111-1668) nmol\/L. No significant differences in maternal and infant characteristics were found comparing the sample size of 246 pregnant women and those with available RBC folate data.  Of pregnant women with available RBC folate concentration, 51.3% (n=79) had RBC folate concentration \u22641360 nmol\/L with a median (IQR) RBC concentration of 1116 (996-1231) nmol\/L, and 48.7% (n=75) had RBC folate concentration >1360 nmol\/L with a median (IQR) RBC concentration of 1702 (1516-1887) nmol\/L. Weeks of gestation were significantly higher in pregnant women with RBC folate concentration >1360 nmol\/L compared to those with RBC   120 folate concentration \u22641360 nmol\/L. No further significant differences were found in maternal characteristics comparing pregnant women with RBC folate concentration \u22641360 nmol\/L and those with RBC folate concentration >1360 nmol\/L (Table 4.2). Table 4.2 Maternal characteristics of the total sample of pregnant women and those with RBC folate concentration \u22641360 nmol\/L and RBC folate concentration >1360 nmol 1Values are median (IQR) or mean (SD) for continuous variables and prevalence (%) for categorical variables. P values are from Wilcoxon rank-sum test or independent Student\u2019s t-test for comparison of continuous variables and chi-square test or Fisher\u2019s exact test for categorical variables between pregnant women with RBC folate \u22641360 nmol\/L and those with RBC folate >1360 nmol\/L. Pre-pregnancy BMI categories; underweight (BMI <18.5 kg\/m2), healthy weight (BMI \u226518.5 to 24.9 kg\/m2), overweight (BMI \u226525 to 29.9 kg\/m2) and obese (BMI \u226530 kg\/m2). BMI, body mass index; RBC, red blood cells.      4.3.1 Infant characteristics Most of the infants included in the current study were full-term newborns (94%) (Table 4.3), and only 5.7% of them were classified as LBW. Of the total sample of infants, 52% of were male (n=128) and 48% were female (n=118). No significant differences in infant characteristics  Maternal characteristics n All n RBC folate  \u22641360 nmol\/L n RBC folate  >1360 nmol\/L P Maternal age, y 238 30.9\u00b13.8 75 31.0\u00b14.2 75 31.3\u00b13.5 0.61 Weeks of gestation at blood sample collection 246 15.4 (13.6-17.4) 79 15.0  (12.6-16.6) 75 15.7  (13.4-17.4) 0.03 Pre-pregnancy BMI, % (kg\/m2) Underweight  Healthy weight  Overweight  Obesity 241  7  144  52  38 23.7 (21.5-26.8)  2.90  (18.3;17.8-18.4) 59.8  (22.0;20.9-23.5) 21.6  (26.7;25.6-28.0) 15.8 (34.0;31.9-37.2) 76  2        50  15  9 23.0 (21.6-25.4)  2.63  (18.1;17.8-18.5) 65.8  (22.0; 21.2-23.1) 19.7  (26.4; 25.2-27.7) 11.8  (36.1; 33.2-38.1) 74  0  39  18  17 24.8 (22.5-29.4)  - 52.7 (22.6; 20.8-24.2) 24.3  (26.6; 25.9-28.3) 23.0  (33.1; 31.9-36.2)   0.11 Parity, % Nulliparous Multiparous 240 91 149  37.9 62.1 77 30 47  39.0 61.0 75 31 44  41.3 58.7 0.77 Ethnicity, % European Chinese Latin American South East Asian Other 238 190 8 9 8 23  80 3.4 3.8 3.4 9.4 75 57 3 6 2 7  76.0 4.0 8.0 2.7 9.3 75 69 3 1 1 1  92.0 4.0 1.3 1.3 1.3 0.24   121 were found comparing infants of the total sample size of pregnant women with infants of pregnant women with available RBC folate concentration data. Also, the characteristics of infants of pregnant women with RBC folate concentration \u22641360 nmol\/L did not significantly differ from infants of pregnant women with RBC folate concentration >1360 nmol\/L.  Table 4.3 Infant characteristics of the total sample of infants and those of pregnant women with RBC folate concentration \u22641360 nmol\/L and RBC folate concentration >1360 nmol  1Values are median (IQR) or mean (SD) for continuous variables and prevalence (%) for categorical variables. P values are from Wilcoxon rank-sum test or independent Student\u2019s t-test for comparison of continuous variables and chi-square test for categorical variables between infants of pregnant women with RBC folate concentration \u22641360 nmol\/L and those with RBC folate concentration >1360 nmol\/L. BW, birth weight; BL, birth length; LBW, low birth weight; RBC, red blood cell. Infant characteristics n All n RBC folate \u22641360 nmol\/L n RBC folate >1360 nmol\/L P         Birth status, % Term Preterm 246 231 15  93.9 6.10 79 76 3  96.2 3.80 75 67 8  89.3 10.7 0.09 Gestational age at birth, wks  246 39.7 (38.4-40.1) 79 39.7 (39.0-40.0) 75 39.9 (38.3-40.4) 0.88 BW, g 246 3380 (3060-3731) 79 3400 (3070-3710) 75 3390 (3140-3840) 0.56 LBW, % 14 5.69 6 7.59 4 5.33 0.57 BL, cm 243 51.1 (49.5-53.3) 79 51.0 (50.0-53.3) 74 51.8 (49.5-53.3) 0.62 Infant age, wks  209 12.0 (11-13) 70 12.0 (12.0-13.0) 65 12.0 (11.0-13.0) 0.16 Infant weight, g 236 6095 (5600-6600) 77 6110 (5620-6570) 74 5925 (5450-6550) 0.18 Infant length, cm 181 59.0 (57.7-61.0) 59 59.5 (58.0-61.0) 59 59.0 (56.0-61.0) 0.30   122  4.3.2 Comparison of DNA methylation levels of CpG sites associated with fetal growth and obesity-related and DNA methyltransferase genes in offspring by maternal RBC folate concentration  The proportion of pregnant women with folate status in the reference range and high folate status did not significantly differ by tissue sample (i.e., buffy coat or BECs) (P> 0.05). DNA methylation levels in specific CpG sites associated with DNMT-3A\/3B, HIF-3\u03b1, LEP, LEP-R and RXRA, sites, measured in BECs, and CpG sites associated with genes, measured in the buffy coat, differed comparing folate status in pregnant women (Table 4.4). However, these results were no longer significant after Benjamini-Hochberg correction.     123 Table 4.4 Comparison of DNA methylation levels of CpG sites associated with fetal growth and obesity-related genes in infants of mothers with RBC folate \u22641360 nmol\/L and RBC folate >1360 nmol\/L at early pregnancy1 CpG sites BECs Buffy coat    P2    P3 RBC folate  \u22641360 nmol\/L (n=46) RBC folate  >1360 nmol\/L (n=37) RBC folate  \u22641360 nmol\/L (n=33) RBC folate  >1360 nmol\/L (n=38) DNMT-1       cg07627628 0.28 (0.25-0.30) 0.29 (0.25-0.33) 0.85 (0.83-0.86) 0.84 (0.83-0.86) 0.33 0.86 cg21063296 0.04 (0.04-0.05) 0.05 (0.04-0.05) 0.46 (0.45-0.47) 0.46 (0.45-0.47) 0.90 0.84 cg21892967 0.19 (0.17-0.19) 0.19 (0.17-0.20) 0.18 (0.18-0.19) 0.19 (0.18-0.20) 0.67 0.34 cg01347596 0.81 (0.76-0.84) 0.82 (0.79-0.84) 0.94 (0.94-0.95) 0.94 (0.94-0.95) 0.40 0.36 DNMT-3A       cg00277048 0.96 (0.95-0.97)  0.96 (0.95-0.96)  0.96 (0.96-0.97) 0.96 (0.96-0.97) 0.85 0.51 cg00856404 0.75 (0.74-0.76) 0.77 (0.74-0.77) 0.30 (0.29-0.31) 0.30 (0.29-0.30) 0.03 0.11 cg00912598 0.06 (0.06-0.07)  0.06 (0.06-0.07) 0.06 (0.06-0.07) 0.06 (0.06-0.07) 0.58 0.95 cg05544807 0.23 (0.22-0.24) 0.23 (0.22-0.25) 0.72 (0.70-0.73) 0.71 (0.69-0.72) 0.96 0.05 cg09986894 0.93 (0.92-0.94) 0.93 (0.93-0.93) 0.94 (0.94-0.95) 0.95 (0.94-0.95) 0.99 0.36 cg10749994 0.09 (0.08-0.11) 0.09 (0.08-0.11) 0.03 (0.02-0.03) 0.02 (0.02-0.03) 0.99 0.34 cg11354105 0.16 (0.15-0.18) 0.17 (0.16-0.19) 0.81 (0.80-0.82) 0.81 (0.80-0.82) 0.11 0.84 cg15150970 0.51 (0.49-0.52) 0.51 (0.50-0.52) 0.20 (0.19-0.21) 0.20 (0.19-0.21) 0.51 0.64 cg15843262 0.46 (0.42-0.49) 0.46 (0.43-0.48) 0.02 (0.02-0.02) 0.02 (0.02-0.03) 0.99 0.09 cg19489797 0.07 (0.06-0.08) 0.06 (0.06-0.08) 0.07 (0.06-0.07) 0.07 (0.06-0.08) 0.70 0.96 cg20303441 0.42 (0.39-0.43) 0.40 (0.38-0.43) 0.01 (0.008-0.01) 0.01 (0.01-0.02) 0.45 0.03 cg20669908 0.80 (0.79-0.81) 0.80 (0.79-0.81) 0.33 (0.33-0.34) 0.33 (0.32-0.34) 0.22 0.51 cg26544247 0.43 (0.40-0.44) 0.43 (0.40-0.45) 0.04 (0.04-0.05) 0.05 (0.04-0.05) 0.67 0.11 cg27369452 0.03 (0.03-0.04) 0.04 (0.03-0.04) 0.04 (0.04-0.05) 0.04 (0.04-0.05) 0.16 0.78 DNMT-3B       cg00300969 0.88 (0.87-0.89) 0.88 (0.87-0.89) 0.58 (0.56-0.60) 0.58 (0.56-060) 0.79 0.64 cg09835408 0.76 (0.74-0.77) 0.75 (0.74-0.76) 0.42 (0.39-0.45) 0.41 (0.40-0.44) 0.04 0.91 cg13788819 0.88 (0.86-0.89)  0.88 (0.86-0.89) 0.31 (0.30-0.33) 0.32 (0.30-0.34) 0.97 0.35 cg14224313 0.89 (0.88-0.90) 0.89 (0.88-0.90) 0.92 (0.92-0.93) 0.92 (0.92-0.93) 0.63 0.58 cg17475857 0.86 (0.84-0.88) 0.86 (0.84-0.87) 0.58 (0.56-0.60) 0.58 (0.57-0.60) 0.65 0.56 cg17482740 0.47 (0.47-0.49) 0.47 (0.46-0.49) 0.59 (0.58-0.60) 0.59 (0.57-0.60) 0.45 0.52 cg21235334 0.09 (0.08-0.10) 0.08 (0.07-0.09) 0.24 (0.22-0.25) 0.23 (0.21-0.25) 0.89 0.92 cg22052056 0.95 (0.94-0.95) 0.95 (0.94-0.95) 0.82 (0.81-0.83) 0.82 (0.80-0.83) 0.29 0.78 cg22605822 0.86 (0.85-0.86) 0.86 (0.85-0.87) 0.87 (0.87-0.88) 0.88 (0.87-0.88) 0.04 0.49 cg24403338 0.82 (0.81-0.83) 0.82 (0.81-0.83) 0.67 (0.64-0.69) 0.66 (.065-0.68) 0.71 0.95 IGF-2       cg00570518 0.40 (0.39-0.41) 0.41 (0.39-0.42) 0.59 (0.57-0.60) 0.58 (0.57-0.59) 0.07 0.30 cg02045936 0.84 (0.83-0.86) 0.84 (0.83-0.86) 0.91 (0.91-0.92) 0.91 (0.91-0.92) 0.85 0.64 cg02807948 0.30 (0.29-0.32) 0.30 (0.29-0.31) 0.51 (0.50-0.54) 0.52 (0.51-0.54) 0.93 0.73 cg05384664 0.44 (0.43-0.45) 0.45 (0.43-0.46) 0.62 (0.60-0.63) 0.62 (0.60-0.64) 0.15 0.84 cg07096953 0.11 (0.10-0.12) 0.11 (0.10-0.13) 0.34 (0.32-0.38) 0.35 (0.32-0.38) 0.93 0.94 cg11717189 0.32 (0.30-0.34) 0.31 (0.30-0.35) 0.38 (0.35-0.41) 0.39 (0.37-0.42) 0.80 0.26 cg12528452 0.50 (0.46-0.53) 0.50 (0.46-0.54) 0.61 (0.57-0.63) 0.62 (0.58-0.64) 0.88 0.57 cg13165070 0.33 (0.32-0.35) 0.33 (0.31-0.35) 0.36 (0.34-0.38) 0.35 (0.33-0.37) 0.21 0.08 cg26401390 0.84 (0.83-0.86) 0.84 (0.83-0.86) 0.91 (0.91-0.92) 0.91 (0.91-0.92) 0.84 0.60 LEP       cg00666422 0.41 (0.38-0.44) 0.40 (0.36-0.42) 0.47 (0.46-0.49) 0.46 (0.44-0.49) 0.24 0.28 cg00840332 0.04 (0.04-0.05) 0.04 (0.03-0.05) 0.16 (0.14-0.18) 0.14 (0.13-0.16) 0.62 0.03 cg03084214 0.79 (0.77-0.81) 0.79 (0.76-0.81) 0.90 (0.89-0.91) 0.90 (0.89-0.91) 0.91 0.13 cg07464571 0.10 (0.08-0.12) 0.09 (0.08-0.11) 0.37 (0.35-0.38) 0.35 (0.34-0.37) 0.51 0.08 cg13381984 0.16 (0.14-0.18) 0.15 (0.13-0.18) 0.36 (0.34-0.37) 0.34 (0.33-0.37) 0.93 0.23 cg19594666 0.09 (0.06-0.10) 0.08 (0.07-0.11) 0.40 (0.37-0.43) 0.38 (0.35-0.41) 0.64 0.03 cg26814075 cg12782180 0.14 (0.11-0.16) 0.11 (0.08-0.13) 0.14 (0.12-0.17) 0.10 (0.08-0.12) 0.40 (0.37-0.42) 0.40 (0.38-0.42) 0.38 (0.36-0.40) 0.39 (0.37-0.41) 0.48 0.07 0.10 0.44 cg25435800 0.07 (0.05-0.09) 0.07 (0.06-0.09) 0.94 (0.93-0.95) 0.94 (0.93-0.95) 0.48 0.65 1Values are median (IQR) for beta values. P2 values are from Wilcoxon rank-sum test between DNA methylation levels of CpG sites measured in BECs of infants of pregnant women with RBC folate concentration \u22641360 nmol\/L and those with RBC folate concentration >1360 nmol\/L. P3values are from Wilcoxon rank-sum test between DNA methylation levels of CpG sites measured in buffy coat of infants of pregnant women with RBC folate concentration \u22641360 nmol\/L and those with RBC folate concentration >1360 nmol\/L. BECs, buccal epithelial cells; RBC, red blood cell.    124 Table 4.4 Comparison of DNA methylation levels of CpG sites associated with fetal growth and obesity-related genes in infants of mothers with RBC folate \u22641360 nmol\/L and RBC folate >1360 nmol\/L at early pregnancy (con)1 CpG sites BECs Buffy coat    P2    P3 RBC folate \u22641360 nmol\/L (n=46) RBC folate >1360 nmol\/L (n=37) RBC folate \u22641360 nmol\/L (n=33) RBC folate >1360 nmol\/L (n=38) HIF3\u03b1       cg14153927 0.40 (0.39-0.41) 0.40 (0.38-0.42) 0.28 (0.27-0.29) 0.28 (0.27-0.29) 0.62 0.10 cg16672562 0.10 (0.09-0.12) 0.11 (0.10-0.13) 0.97 (0.96-0.98) 0.97 (0.96-0.97) 0.34 0.67 cg20667364 0.11 (0.10-0.13) 0.13 (0.12-0.15) 0.27 (0.24-0.33) 0.27 (0.24-0.33) 0.008 0.59 cg22891070 0.23 (0.22-0.26) 0.24 (0.24-0.26) 0.91 (0.89-0.92) 0.90 (0.90-0.92) 0.85 0.63 cg23548163 0.21 (0.19-0.22) 0.22 (0.20-0.23) 0.55 (0.50-0.62) 0.55 (0.49-0.62) 0.10 0.82 cg25196389 0.57 (0.53-0.58) 0.56 (0.55-0.59) 0.83 (0.80-0.85) 0.83 (0.80-0.84) 0.65 0.48 cg25460031 0.87 (0.86-0.89) 0.86 (0.87-0.89) 0.92 (0.92-0.93) 0.93 (0.92-0.94) 0.35 0.61 cg26749414 0.39 (0.36-0.42) 0.41 (0.39-0.43) 0.78 (0.71-0.84) 0.76 (0.71-0.81) 0.05 0.91 cg27146050 0.17 (0.16-0.19) 0.18 (0.15-0.19) 0.66 (0.64-0.68) 0.66 (0.64-0.67) 0.65 0.50 cg01552731 0.64 (0.60-0.67) 0.66 (0.63-0.68) 0.89 (0.87-0.90) 0.90 (0.83-0.91) 0.09 0.64 cg07684068 0.84 (0.81-0.86) 0.84 (0.82-0.86) 0.93 (0.91-0.94) 0.93 (0.91-0.94) 0.52 0.79 cg12068280 0.37 (0.36-0.39) 0.37 (0.36-0.40) 0.69 (0.68-0.71) 0.69 (0.68-0.70) 0.95 0.53 RXRA       cg13941235 0.06 (0.05-0.07) 0.07 (0.06-0.08) 0.73 (0.70-0.75) 0.72 (0.69-0.75) 0.06 0.50 cg14051662 0.18 (0.17-0.19) 0.18 (0.17-0.19) 0.92 (0.91-0.92) 0.91 (0.91-0.92) 0.81 0.53 cg14121282 0.05 (0.04-0.05) 0.05 (0.04-0.05) 0.22 (0.21-0.23) 0.24 (0.22-0.26) 0.68 0.008 cg14236758 0.13 (0.12-0.14) 0.13 (0.11-0.14) 0.68 (0.65-0.72) 0.68 (0.63-0.71) 0.69 0.94 cg14472716 0.29 (0.28-0.31) 0.29 (0.28-0.30) 0.39 (0.37-0.42) 0.39 (0.38-0.42) 0.50 0.53 cg14570632 0.66 (0.61-0.72) 0.65 (0.60-0.72) 0.97 (0.96-0.98) 0.97 (0.96-0.97) 0.87 0.70 cg21201924 0.29 (0.27-0.31) 0.28 (0.25-0.31) 0.78 (0.77-0.80) 0.78 (0.77-0.79) 0.61 0.71 cg02127980 0.07 (0.06-0.07) 0.06 (0.05-0.07) 0.56 (0.54-0.60) 0.56 (0.54-0.59) 0.18 0.95 cg13487983 0.13 (0.13-0.14) 0.13 (0.12-0.14) 0.31 (0.30-0.32) 0.32 (0.31-0.33) 0.49 0.29 cg13510651 0.72 (0.70-0.73) 0.72 (0.70-0.73) 0.90 (0.89-0.90) 0.90 (0.90-0.91) 0.87 0.53 cg13595495 0.18 (0.17-0.19) 0.18 (0.17-0.19) 0.94 (0.94-0.95) 0.95 (0.94-0.95) 0.97 0.85 cg13677043 0.08 (0.07-0.10) 0.09 (0.07-0.10) 0.92 (0.91-0.92) 0.92 (0.91-0.92) 0.65 0.46 cg13689699 0.40 (0.38-0.43)  0.42 (0.37-0.44) 0.94 (0.94-0.95) 0.94 (0.94-0.95) 0.99 0.44 cg13847322 0.24 (0.22-0.25) 0.23 (0.22-0.25) 0.64 (0.62-0.66) 0.63 (0.61-0.65) 0.93 0.67 cg13931640 0.98 (0.97-0.98) 0.97 (0.96-0.98) 0.62 (0.60-0.64) 0.62 (0.59-0.64) 0.58 0.88 LEP-R       cg03050981 0.98 (0.97-0.99) 0.98 (0.97-0.99) 0.96 (0.95-0.97) 0.97 (0.96-0.97) 0.37 0.61 cg11228758 0.71 (0.69-0.73) 0.70 (0.69-0.72) 0.92 (0.91-0.92) 0.92 (0.91-0.92) 0.45 0.28 cg11807188 0.54 (0.51-0.56) 0.53 (0.52-0.55) 0.89 (0.89-0.90) 0.88 (0.88-0.89) 0.71 0.003 cg23688719 0.34 (0.32-0.35) 0.33 (0.31-0.36) 0.96 (0.95-0.96) 0.96 (0.95-0.96) 0.45 0.80 1Values are median (IQR) for beta values. P2 values are from Wilcoxon rank-sum test between DNA methylation levels of CpG sites measured in BECs of infants of pregnant women with RBC folate concentration \u22641360 nmol\/L and those with RBC folate concentration >1360 nmol\/L. P3values are from Wilcoxon rank-sum test between DNA methylation levels of CpG sites measured in buffy coat of infants of pregnant women with RBC folate concentration \u22641360 nmol\/L and those with RBC folate concentration >1360 nmol\/L.BECs, buccal epithelial cells; RBC, red blood cell.          125 4.4 Discussion Folate is the most important methyl donor for remethylation reactions. Findings from the current study indicate there are no significant differences in DNA methylation levels of CpG sites associated with LEP, LEP-R, RXRA, HIF3\u03b1, DNMT-1, IGF-2, DNMT-3A and DNMT-3B genes measured in infants of mothers with high folate concentration compared to those with folate concentration in the current reference range. Maternal RBC folate concentration. The median (IQR) maternal RBC folate concentration in the current study was 1334 (1111-1688) nmol\/L which was higher than maternal RBC folate concentration previously reported in European pregnancy studies. Cavalle-Busquets et al. (293) showed that the geometric mean (95%CI) of RBC folate concentration was 901 (859-945) nmol\/L at 12 weeks of gestation in Spanish pregnant women. Similarly, a study conducted by O\u2019Malley et al.(294) in Irish pregnant women showed that first-trimester RBC folate concentration was 1137.5 ( SD 443.0) nmol\/L (294), while in the UK, a geometric mean of 456 nmol\/L for RBC folate concentration was reported in pregnant women at 19 weeks of gestation (203). In contrast, the PREFORM study (16), a pregnancy study in Canadian pregnant women living in Toronto, showed that maternal RBC folate concentration measured at 12 weeks of gestation was higher (~2400 nmol\/L) compared to the results of the current study. Discrepancies in folate concentration between the pregnancy studies may be partially explained by differences in folic acid supplement use at early pregnancy (i.e., < 20 weeks of gestation). In the current study, 98% of pregnant women in the first trimester (i.e., <14 weeks of gestation) consumed folic acid supplement with at least 14\u00b5g\/day, 42% using a daily dose of folic acid of 1000\u00b5g, and 24% using a folic acid dose of greater than 1000\u00b5g\/day. In comparison, 94%, 98%, and 83% of Spanish (64,293), Irish (294) and UK pregnant women (203), respectively, consumed a daily dose of 400\u00b5g of folic acid in their first trimester of pregnancy (i.e., <14 weeks of gestation),   126 which likely explains the higher RBC folate concentration in the current study. Additionally, 83% of pregnant women that participated in the PREFROM study consumed a daily dose of folic acid \u22651000\u00b5g, which may result in higher RBC folate concentration, compared to the results of the present study (16). Furthermore, differences in the analytical methods used to measure folate concentration may also contribute to the discrepancies found among the aforementioned pregnancy studies. In the pregnancy studies conducted in Spain, Ireland, and the UK, RBC folate concentration was measured by microbiological assay, while in the present study RBC folate concentration was measured by the ion-capture method (67), and in the PREFORM study, it was measured by folate-binding protein with chemiluminescence detection. It has been previously suggested that the latter method may overestimate RBC folate concentration resulting in higher RBC folate concentration as it was reported in the PREFORM study, compared to RBC folate measured by microbiological assay (295).  Differences in DNA methylation levels of CpG sites associated with fetal growth and obesity-related genes by maternal folate status. In contrast to my hypothesis, DNA methylation levels of CpG sites associated with IGF-2, LEP, RXRA, HIF-3\u03b1, and LEP-R genes did not differ between infants of pregnant women with RBC folate concentration \u22641360 nmol\/L, and those of pregnant women with RBC folate concentration >1360 nmol\/L. Similar to our findings, Hoyo et al. (22) reported that higher maternal RBC folate concentration (i.e., third and fourth quartiles) at early pregnancy was not associated with DNA methylation levels of CpG sites associated with IGF-2 gene measured in cord blood.  To the best of my knowledge, there is no available evidence in the relationship between maternal RBC folate concentration and DNA methylation levels of CpG sites associated with LEP, LEP-R, RXRA, and HIF-3A in the offspring. Offspring of Belgium pregnant women that   127 used folic acid supplementation 6 months before conception, had higher DNA methylation levels in CpG sites associated with LEP, measured in cord blood, compared to offspring of women that did not take folic acid supplements (23). Folic acid supplementation is the most important positive predictor of RBC folate concentration in Canadian childbearing-aged women (15). Additionally, findings of the same study showed that offspring of Belgium pregnant women that used folic acid supplementation across all trimesters had higher DNA methylation levels in CpG sites associated with RXRA, compared to offspring of pregnant women that stopped consumption in the second trimester (23). Overall, previous studies suggest that maternal folate status and folic acid supplementation before and during pregnancy may have a role in fetal growth and obesity programming, however, the current evidence is limited.  Comparison between studies is challenging and it should be made with caution. It has been previously reported that DNA methylation levels measured in blood vary by age. For example, findings from a recent study indicated that DNA methylation levels of specific CpG sites (i.e., CpG 11) associated with the  LEP gene increased by 22% from birth to 12 months (296). Furthermore, genetic variants of obesity-related genes such as LEP have been associated with DNA methylation of CpG sites associated with the LEP gene (296). Results of a study conducted by Mansell et al. (166) showed that DNA methylation levels measured in cord blood significantly differed by genotypic variants of HIF  (i.e., rs112087991 TT\/CT\/CC). Additionally, differences between methods to measure DNA methylation (i.e., microarray vs. pyrosequencing) may also contribute to the differences found between studies (297). For example, pyrosequencing detects small differences of DNA methylation levels (i.e., <5%) (297), which may also contribute to the discrepancies between studies.  Given these and other factors may influence DNA methylation levels of CpG sites associated with obesity-related genes, further prospective pregnancy studies measuring DNA   128 methylation levels in the same specimen (e.g., BECs) between studies, as well as, the determination of genetic variant information, are needed to replicate the results, and understand the potential role of maternal folate concentration at early pregnancy in obesity programming.  Findings of the present study also showed that DNA methylation levels of CpG sites associated with DNMT-1, DNMT-3A and DNMT-3B genes did not differ comparing offspring of pregnant women with RBC folate concentration \u22641360 nmol\/L and those of pregnant women with RBC folate concentration >1360 nmol\/L. Results from animal studies reported that male mice that received a postnatal methyl donor-supplemented diet (i.e., folic acid, betaine, choline chloride, methionine and B-12) during 3 weeks after weaning, increased the expression of DNMT-3A by 6% in brain tissue (298). In contrast, Bahous et al. (94) showed that the offspring of dams fed with a folic acid supplemented diet during pregnancy showed a significantly lower expression of the DNMT-3A gene in the cortex and hippocampus, which, we speculate, may potentially be mediated by modifications in DNA methylation. However, further animal studies are needed to replicate and confirm whether the association between methyl nutrient intake and DNMT-3A expression is underlaid by modifications in DNA methylation levels of CpG sites associated with DNMT-3A.   Potential mechanisms underlying the lack of differences in DNA methylation levels in CpG sites associated with fetal growth and obesity related genes We hypothesize that several and complex factors may explain the lack of significant differences. First, findings of the current study indicate that most of pregnant women consumed folic acid supplements and 24% used a supplemental folic acid dose >1,000\u00b5g\/day. High folic acid consumption, and the subsequent production of DHF (299), has been shown to lead to a pseudo-MTHFR deficiency, characterized by a decrease in MTHFR expression and protein levels, which we speculate, may lead to a decreased availability of 5-MTHF for the remethylation   129 of homocysteine to methionine and further decreased synthesis of SAM, the universal methyl donor for DNA methylation (300). Second, the median RBC folate concentration was ~1300 nmol\/L, which may promote an adequate synthesis of SAM for the provision of methyl groups for methylation reactions; however, a potential negative feedback may occur under RBC folate concentration >1360 nmol\/L that results in no changes in DNA methylation levels (200). The provision of methyl groups through one-carbon metabolism is regulated. High SAM concentration has an inhibitory effect on the enzyme MTHFR which catalyzes the formation of 5-MTHF for homocysteine remethylation to methionine. Inhibition of MTHFR may thereby decrease the synthesis of methionine, and subsequently, of SAM (7), that may partially also explain the lack of differences in DNA methylation levels between pregnant women with RBC folate \u22641360 nmol\/L and those with RBC folate >1360 nmol\/L. Further pregnancy studies including a larger sample size of pregnant women with early-pregnancy folate status data and DNA methylation levels of CpG sites associated with obesity-related genes measured only in one type of specimen (i.e., BECs) collected from their offspring are needed to confirm the reproducibility of the current findings. Strengths and limitations. The strengths of the current study include the prospective design as well as the availability of early-pregnancy RBC samples collected at a critical developmental window for fetal programming. Furthermore, the determination of DNA methylation levels in cord blood has been disputed due to the heterogeneity of the specimen (i.e., varied cell composition) which may challenge the interpretation of DNA methylation. The determination of DNA methylation levels in a more homogeneous specimen such as BECs or buffy coat (i.e., leucocytes) is recommended, which were used in this study. However, the current study also has limitations. The present study included apparently healthy pregnant women with high educational levels and high household income, which limits the generalizability of the   130 findings. The determination of DNA methylation levels in two tissues (i.e., BECs and leucocytes) resulted in a smaller sample size as results cannot be combined. The small sample size may have not allowed detect differences in DNA methylation levels between pregnant women with RBC folate \u22641360 nmol\/L and those with RBC folate >1360 nmol\/L. Additionally, the current study did not include the determination of genetic variants, which may influence DNA methylation levels of the genes examined.  In conclusion, early-pregnancy RBC folate concentration was not significantly associated with DNA methylation levels of CpG sites associated with DNA methyltransferase coding genes and fetal growth and obesity related genes, in these 154 mother-infant dyads. Further pregnancy studies with a larger sample size, data on early pregnancy folate status, and data on genetic variants of interrogated genes in folate-replete and non-folate-replete populations are needed to expand the findings of the current investigation.     131 Chapter 5: RELATIONSHIP BETWEEN MATERNAL METHYL NUTRIENT PATTERNS IN EARLY PREGNANCY AND INFANT DNA METHYLATION LEVELS OF FETAL GROWTH, OBESITY-RELATED GENES AND METHYLTRANSFERASE CODING GENES. 5.1 Introduction Maternal nutrition is critical for adequate fetal growth and development and possibly even health in later life. Methyl nutrients such as folate, betaine, choline, B-6, B-12 and riboflavin have an interrelated role the one-carbon metabolism. Vitamins B-6, in the form of PLP, riboflavin, in the form of FMN and FAD, and B-12 participate as coenzymes for the synthesis of SAM (7,50,59), the universal methyl donor, whereas folate and betaine donate methyl groups for SAM formation (48), which is critical for the provision of methyl donors for methylation reactions such as DNA methylation. DNA methylation is a well-recognized stable epigenetic mechanism, which plays a key role in the regulation of gene expression (139). In early pregnancy, de novo DNA methylation occurs, which is characterized by the removal of the methyl groups in embryonic DNA, and the subsequent re-methylation of the DNA strand (18,138,139). This physiological process makes early pregnancy a critical plasticity window for fetal programing and development (20). Current evidence from animal and human studies support the notion that modifications of DNA methylation levels of fetal growth and obesity-related genes in the offspring may predispose the offspring to obesity later in life (151,301). DNA methylation levels of LEP and HIF3\u03b1 genes, which participate in appetite control and hypoxic cellular conditions such as obesity, have shown an inverse and positive association, respectively, with BMI in both children (302) and adults (164). DNA methylation levels of IGF-2 and RXRA genes, both genes involved in fetal growth   132 and cellular proliferation, showed an inverse and positive association with subcutaneous adipose tissue in young adults (303) and fat mass in children, respectively (151).  Despite findings from previous pregnancy studies that indicated pre-pregnancy dietary folate and betaine intake was associated with DNA methylation levels of IGF-2 and LEP genes measured in cord blood (23) and BECs of the offspring (24), evidence of pregnancy studies exploring the relationship between individual maternal methyl nutrient concentrations and DNA methylation levels of IGF-2 in the offspring has shown conflicting results. Haggarty et al. (203) found no relationship between maternal RBC folate concentration measured at 19 weeks of gestation in British pregnant women (n=913) and DNA methylation levels of IGF-2 in cord blood. Whereas, third-trimester serum total B-12 concentration measured in 99 Chinese pregnant women (224), was negatively associated with DNA methylation levels of IGF-2 measured in cord blood, which may suggest a potential role of maternal B-12 status in concentrations in obesity programming. However, the aforementioned studies included only a single maternal biomarker (e.g., maternal RBC folate concentration) as the exposure variable and without adjustment for other circulating methyl nutrients, which, from a metabolic perspective, is challenging to interpret due to the interrelated role of these methyl nutrients in one-carbon metabolism and in the provision of methyl groups. Furthermore, the use of cord blood for the determination of DNA methylation levels has been debated mainly due to the presence of different types of cells. Given the crucial role of de novo DNA methylation in fetal programming and the interrelated functions of methyl nutrients in one-carbon metabolism and thus potentially also DNA methylation, we aimed to determine the relationship between maternal methyl nutrient patterns in early pregnancy, derived using PCA, and DNA methylation levels of CpG sites associated with LEP, HIF3\u03b1, RXRA, IGF-2, LEP-R, and DNA methyltransferase genes (i.e., DNMT-1 and DNMT-3) in their infants.    133  5.2 Study design and participants Information on the study design, exclusion criteria, ethical approval, maternal, and infant characteristics and infant DNA methylation data is provided in Chapter 4.2. Briefly, the present study is nested within the APrON study, and includes maternal lifestyle and anthropometric information, in addition to infant DNA methylation and anthropometric data collected as part of the APrON study. The APrON study is a prospective cohort study that included pregnant women and their children, which were followed until 3 years postpartum, and was conducted in Alberta, Canada.  5.2.1 Biochemical analyses 5.2.1.1 Blood processing  As detailed in Chapter 4.2.5.1, non-fasting maternal blood samples were collected using vacutainer tubes with EDTA or serum separator gel by a certified phlebotomist and processed immediately after collection. Serum, plasma and buffy coat were aliquoted and stored at \u221280\u00b0C (67,287).  5.2.1.2 Biomarker quantification  Plasma free choline, betaine, methionine, and tHcy concentrations were determined by LC-MS\/MS using a single platform (304). We used in-house plasma control samples in each of the in total 4 batches to estimate the inter-assay CV% for the analytes. The inter-assays CVs were \u226410% for free choline, \u226410% for betaine, \u22645% for tHcy, and \u22647% for methionine concentrations. Two external quality control samples for tHcy and methionine concentrations (ClinChek 23082,   134 RECIPE Chemicals and Instruments; IRIS Technologies International) were included in each batch; the mean (\u00b1SD) tHcy concentration was within the target analytical range for the low (8.7\u00b10.5\u00b5mol\/L; target: 9.04; control range 7.23-10.8 \u00b5mol\/L) and high control samples (24.9\u00b12.3 \u00b5mol\/L; target: 25.9; control range 20.7-31.0 \u00b5mol\/L). The mean (\u00b1SD) methionine concentration was within the target analytical range for the low (28.4\u00b10.4 \u00b5mol\/L; target: 29.1; control range 23.3-34.9 \u00b5mol\/L) and high control samples (61.9\u00b13.5 \u00b5mol\/L; target: 67.0; control range 57.0-77.1 \u00b5mol\/L). Plasma PLP, riboflavin, FMN and FAD concentrations were determined by isotope-dilution LC-MS\/MS using a method modified by Midttun et al. (23). In each batch, three external quality control samples for PLP concentration (ClinChek 8873, RECIPE Chemicals and Instruments; IRIS Technologies International) were included. The mean (\u00b1SD) PLP concentration was within the target analytical range for the low (62.7\u00b15.0 nmol\/L; target: 61.9; control range 49.4-74.5 nmol\/L), medium (100\u00b11.9 nmol\/L; target: 97.1; control range 77.1-117 nmol\/L) and high (128\u00b15.0 nmol\/L; target: 138; control range 110-165 nmol\/L) control samples. The intra- and inter-assay CV% for PLP concentration were estimated based on 3 batches. The intra-assay CV% was \u22647% for the low concentration quality control samples, and \u22645% for the medium and high concentration quality control samples. The inter-assay CV% was \u22645% for all control samples. For riboflavin, FMN and FAD in-house plasma control samples were included in each of the three batches to estimate the inter-assay CV% of these metabolites, and were \u226413% for riboflavin, \u226410% for FAD and \u226413% for FMN concentrations. Serum total B-12 was quantified by a microparticle-enzyme immunoassay (Access 2 by Beckman Coulter) at the BC Children\u2019s Hospital Pathology Laboratory (Vancouver, BC, Canada). RBC folate concentration was determined using a by ion-capture method. Details of the determination of RBC folate concentration were described in Chapter 4.2.5.2.   135 5.2.2 Statistical analysis A Shapiro-Wilk test was used to test the normality of the data. Maternal age is normally distributed and is presented as mean (\u00b1SD). Because of skewed distributions, weeks of gestation at maternal blood collection, gestational age at birth, pre-pregnancy BMI, maternal biomarker concentrations, BW, BL, infant age, infant weight, and infant length are presented as median (IQR). Categorical variables are presented as percentages (%). Student\u2019s t-test was used to compare maternal age, while a Wilcoxon rank-sum test was used to compare weeks of gestation and maternal biomarker concentrations, and chi-square test was used to compare parity, between tissue type for DNA methylation data (i.e., BECs or buffy coat) and between pregnant women included in the PCA and the total sample of pregnant women. Given the expected values for at least one cell were <5, pre-pregnancy BMI categories (i.e., underweight, healthy weight, overweight and obesity) and maternal ethnicity categories were compared between tissue type, and between PCA sample and the total sample using Fisher\u2019s exact test.   For infant characteristics, a Wilcoxon rank-sum test was used to compare gestational age at birth, BW, BL, infant age, infant weight and infant length by tissue, sex and by infant of pregnant women included in the PCA with the total sample of pregnant women. A chi-square test was used to compare birth status (preterm\/term) and LBW by tissue, sex and by infants of pregnant women included in the PCA with the total sample of pregnant women.  5.2.2.1 Principal component analysis Given the interrelated role of methyl nutrients in one-carbon metabolism, and the critical role that they play in the provision of methyl groups for DNA methylation, especially for de novo DNA methylation at early stages of development, PCA was used to derive uncorrelated maternal   136 methyl nutrient principal components at early pregnancy.  The derivation of maternal methyl nutrient patterns was conducted through several steps. Based on their role as methyl donors (i.e., folate, betaine, choline), coenzymes (i.e., total B-12, PLP, FMN and FAD) or intermediate metabolites of one-carbon metabolism (i.e., methionine, tHcy), the aforementioned biomarkers were included in the PCA. PCA was restricted to pregnant women with complete data for maternal biomarker concentrations, i.e., without missing observations. Of the total sample size (n=246), 92 pregnant women had missing RBC folate data. Of the remaining sample (n=154) 12 had missing total-B12 data and 3 had missing PLP\/FMN\/FAD data. In total, 139 pregnant women with complete biomarker data were included in the PCA (n=139). After the selection of maternal biomarkers to include in the PCA, maternal biomarker concentrations were log-transformed to improve the normality of the data, and the correlation between the maternal biomarker concentrations was explored, using a Pearson correlation test.  The PCA was conducted. The number of maternal methyl nutrient principal components derived indicated the number of variables included in the PCA (i.e., 9 maternal biomarker concentrations). The maternal methyl nutrient principal components were subjected to orthogonal rotation. Of the total derived maternal methyl nutrient components (n=9), retained maternal methyl nutrient components were selected using the following criteria; 1) the examination of Eigenvalues, which represents the variance of the data explained by each maternal methyl nutrient principal component using Kayser\u2019s rule (i.e., >1) (233), 2) the exploration of scree plots of the maternal methyl nutrient principal components, which are the graphical representation of the percentages of variation that each maternal methyl nutrient principal components accounts for (Figure 5.1), and 3) the explained cumulative variance provided by the maternal methyl nutrient principal components (i.e., >50% of the variance of the biomarker data).    137 Retained maternal methyl nutrient principal components, henceforth referred to as maternal methyl nutrient patterns, were labeled according to the factor loadings, which indicates the contribution of each biomarker of the retained component (i.e., |factor loadings| \u22650.45 were considered to contribute significantly to each component) and overall interpretability. Finally, scores of the retained maternal methyl nutrient patterns were predicted for each pregnant woman.            Figure 5.1 Scree plot of eigenvalues of derived early-pregnancy principal components  Longitudinal analysis. Linear regression models were used to explore the association between the predicted scores of each retained maternal methyl nutrient pattern at early pregnancy, as the exposure variable, and 1) DNA methylation levels of each of the 77 CpG sites (i.e. M values) (292) associated with IGF-2, RXRA, HIF3\u03b1, DNMT-1, DNMT-3, LEP and LEP receptor genes measured in offspring buffy coat, as the outcome variables; and 2) DNA methylation levels of each of the 77 CpG sites associated with IGF-2, RXRA, HIF3\u03b1, DNMT-1, DNMT-3, LEP and LEP-R genes measured in offspring BECs, as the outcome variables. Given previous evidence has shown that infant age, sex, maternal ethnicity (Caucasian\/Non-Caucasian)   138 and pre-pregnancy BMI may influence DNA methylation levels of CpG sites (166,296,305,306), all models were adjusted for these variables. Benjamini-Hochberg procedure was used to control for false-positive rate. Because missing data for maternal variables [i.e., maternal age (n=4), pre-pregnancy BMI (n=1), parity (n=2), ethnicity (n=4), type 1 diabetes (n=2), type 2 diabetes (n=2), gestational diabetes (n=2), primary hypertension (n=2), gestational hypertension (n=2), preeclampsia (n=2)] represented <5% of the database, no missing categories were created. Results were considered significant at P<0.05. Descriptive analyses and PCA were conducted using Stata version IC 15 (StataCorp). Longitudinal regression analyses were conducted using R studio version 3.5.1.   5.3 Results 5.3.1 Maternal characteristics Maternal characteristics are presented in Table 5.1. Mean (SD) age of pregnant women at their first study visit was 30.9 (3.8) y, and 78% reported being Caucasian. Only 2% of women had an incomplete high school diploma, whereas 9% had a high school diploma, 19% had a technical degree, 48% had completed university and 22% had a post-graduate level degree. Also, >50% of women had a total family income\/year \u2265$100,000, while less than 10% had a total family income\/year <$40,000. Most of the women had a healthy pre-pregnancy BMI, while 16% of women had a BMI that classified them as obese. None of the participants had primary hypertension, gestational hypertension, type 1 or type 2 diabetes. Only 1.2% (n=3) of the participants had GD, whereas 0.8% (n=2) developed preeclampsia during pregnancy. Maternal characteristics and methyl nutrient concentrations did not differ between pregnant women who had infants in which DNA methylation levels were measured in BECs compared to those in which DNA methylation levels were measured in buffy coat (Table 5.1 and Table 5.2).   139 Table 5.1 Maternal characteristics of all women with infant DNA methylation data by tissue and those included in the PCA1 Maternal characteristics                 All women                                               Specimen DNA methylation Women included in the PCA n  n Buffy coat n BECs P2 n PCA P3 Maternal age, y 238 30.9\u00b13.8 107 31.1\u00b14.2 131 30.8\u00b13.5  0.69 135 31.2\u00b13.9 0.64 Weeks of gestation at blood sample collection 246 15.4 (13.6-17.4) 109 15.4 (13.1-17.4) 137 15.3 (13.7-17.1) 0.08 139 15.1 (12.7-16.7) 0.74 Pre-pregnancy BMI, % (kg\/m2) Underweight Healthy weight Overweight Obesity 241  7 144 52 38 23.8 (21.7-26.9)  2.90 (18.3;17.8-18.4) 59.8 (22.0;20.9-23.5) 21.6 (26.7;25.6-28.0) 15.8 (34.0;31.9-37.2) 109  2 64 26 17 23.7 (21.5-23.4)  1.83 (17.9;17.8-18) 58.7 (21.8;20.7 22.9) 23.9 (26.8;25.9-28.0) 15.6 (35.2;31.9-37.0) 132                 5 80 26 21 23.7 (21.6-26.7)  3.8 (18.3;18.3-18.4) 50.6 (22.7;21.0-23.7) 19.7 (26.4;25.6-28.1) 15.9 (33.2;32.1-37.2) 0.88 135  2 80 33 20 24.1 (21.8-26.8)  2.5 (18.1; 17.8-18.5) 56.1 (20.1; 21.1-23.5) 22.3 (26.5; 25.5-28.0) 19.1 (32.9; 31.6-36.1)  0.75 Parity, % Nulliparous Multiparous 240 91 149  37.9 62.1 107 40 67  37.4 62.6 133 51 82  38.4 61.7  0.98 137 53 84  38.7 61.3 0.88 Ethnicity, % European Chinese Latin American South East Asian Other 238 190 8 9 8 23  80 3.4 3.8 3.4 9.7 107 92 2 3 1 9  86.0 1.9 2.8 0.93 8.4  131 98 6 6 7 14  74.8 4.58 4.58 5.34 10.7 0.97 135 114 4 6 3 8  84.4 2.94 4.44 2.22 6 0.17 1Values are mean (SD) or median (IQR) for continuous variables or prevalence (%) for categorical variables. P2 values are from Wilcoxon rank-sum test or independent Student\u2019s t-test for comparison of continuous variables and chi-square test or Fisher\u2019s exact test for categorical variables between pregnant women who had infant DNA methylation data measured in buffy coat or BECs. P3 values are from Wilcoxon rank-sum test or independent Student\u2019s t-test, and chi-square test for categorical variables between the total sample and those pregnant women that were included in the PCA. or Fisher\u2019s exact test for categorical variables between pregnant women with RBC folate \u22641360 nmol\/L and those with RBC folate >1360 nmol\/L. Pre-pregnancy BMI categories; underweight (BMI <18.5 kg\/m2), healthy weight (BMI \u226518.5 to 24.9 kg\/m2), overweight (BMI \u226525 to 29.9 kg\/m2) and obese (BMI \u226530 kg\/m2). BMI, body mass index; BECs, buccal epithelial cells, PCA, principal component analysis; Wks, weeks.       140  5.3.2 Maternal methyl nutrient patterns No significant differences were found in maternal characteristics or maternal biomarker concentrations between pregnant women included in the PCA compared to the total sample of pregnant women (Table 5.2). After conducting the PCA, 5 maternal methyl nutrient patterns were retained which explained 72.3% of the variation of maternal methyl nutrient data (Table 5.3). Maternal methyl nutrient patterns were labeled according to loading of each maternal methyl nutrient to the pattern: 1) B-12\/methyl donor pattern (including total B-12, betaine and free choline concentrations), 2) Homocysteine pattern (tHcy concentration), 3) Riboflavin pattern (including FAD and FMN concentrations), 4) Folate pattern (RBC folate concentration) and 5) PLP\/methionine pattern (including PLP and methionine concentrations). Maternal B-12\/methyl donor pattern score tended to increase with increasing log-B-12, log- free choline and log-betaine concentrations, the main contributors to the pattern. Similarly, maternal homocysteine and riboflavin pattern scores tended to increase with increasing log-tHcy, and log-FMN and log-FAD concentrations, respectively. Maternal folate pattern scores also increased with increasing log-RBC folate, and the PLP\/methionine pattern scores increased with increasing log-PLP and log-methionine concentrations.    141 Table 5.2 Maternal biomarker concentrations of all women with infant DNA methylation data by tissue and those included in the PCA1 1Values are median (IQR). P2 values are from Wilcoxon rank-sum test for comparison of continuous variables between pregnant women who had infant DNA methylation data measured in buffy or BECs. P3 values are from Wilcoxon rank-sum test for comparison between the total sample and those pregnant women that were included in the PCA. BECs, buccal epithelial cells; FAD, flavin adenine dinucleotide; FMN, flavin mononucleotide; PCA, principal component analysis; PLP, pyridoxal-5\u2019-phosphate; tHcy, total homocysteine.   Table 5.3 Loadings of methyl nutrients for early-pregnancy methyl nutrient patterns                1B-12\/methyl donor pattern refers to maternal total B-12, free choline and betaine loadings;2Homocysteine pattern refers to maternal tHcy loading; 3Riboflavin pattern refers to maternal FAD and FMN loadings; 4Folate pattern refers to maternal RBC folate loading; 5PLP\/methionine pattern refers to maternal PLP and methionine loadings. FAD,flavin adenine dinucleotide; FMN, flavin mononucleotide; PLP, pyridoxal-5\u2019-phosphate; RBC, red blood cell; tHcy, total homocysteine.    Biomarker concentrations                 All women                                               Specimen DNA methylation Women included in the PCA n  n Buffy coat  n BECs n P2 PCA P3 Plasma betaine, \u00b5mol\/L 238 18.1 (14.5-22.8) 107 18.0 (15.3-22.9) 131 18.1 (14.3-22.3) 139 0.73 18.4 (14.3-24.1) 0.98 Plasma free choline, \u00b5mol\/L 238 11.3 (9.35-13.5) 107 11.5 (9.45-13.5) 131 11.2 (9.27-13.4) 139 0.40 11.4 (9.45-13.6) 0.99 Plasma tHcy, \u00b5mol\/L 238 4.47 (3.72-5.43) 107 4.19 (3.40-5.37) 131 4.63 (4.05-5.51) 139 0.07 4.52 (3.76-5.46) 0.90 Plasma methionine, \u00b5mol\/L 238 20.6 (17.5-24.2) 107 20.0 (17.1-23.4) 131 21.0 (17.8-25.1) 139 0.26 20.5 (17.0-24.6) 0.55 Serum total B-12, pmol\/L 223 246 (184-322) 98 247 (190-322) 125 240 (169-315) 139 0.36 255 (188-338) 0.74 Plasma PLP, nmol\/L 239 72.8 (46.8-128) 107 64.4 (45.8-119) 132 74.5 (49.2-134) 139 0.33 74.0 (51.8-133) 0.64 Plasma FMN, nmol\/L 239 18.8 (16.1-21.6) 107 18.7 (15.8-21.1) 132 18.8 (16.2-22.1) 139 0.24 19.3 (16.2-22.1) 0.60 Plasma FAD, nmol\/L  238 78.0 (65.8-93.1) 107 77.7 (64.4-92.9) 131 78.6 (66.8-93.7) 139 0.55 76.0 (64.3-93.1) 0.49 RBC folate, nmol\/L 154 1334 (1111-1688) 71 1427 (1130-1688) 83 1306 (1095-1762) 139 0.53 1337 (1111-1702) 0.95 Maternal methyl nutrient patterns B-12\/methyl donor pattern1 Homocysteine pattern2 Riboflavin pattern3 Folate pattern4 PLP\/methionine  pattern5 Variance explained 17.4% 14.6% 14.4% 13.3% 12.6% Total B-12 0.4700 -0.3251 -0.0397 0.4030 0.0870 FAD -0.2485 0.3355 0.5816 0.2860 0.0276 FMN 0.1091 -0.2086 0.7845 -0.1554 0.0101 tHcy 0.0352 0.7500 -0.1370 -0.0670 0.0743 PLP 0.1364 0.1819 0.0311 0.2284 0.7060 RBC folate -0.084 -0.0432 -0.0632 0.7539 0.0050 Betaine 0.6717 0.0424 0.0091 -0.1709 0.0758 Free choline 0.4618 0.3426 0.1448 0.0329 -0.231 Methionine 0.1189 0.1393 -0.0012 0.2763 -0.6546   142  5.3.1 Infant characteristics Neonatal and infant characteristics are presented in Table 5.4. Of the total sample, 48% of infants were female (n=118) and 52% were male (n=128). Most of the infants were born at term. Females were lighter [3353 (2970-3545) g vs. 3494 (3143-3833) g] and shorter [52.0 (50.0-53.3) cm vs. 50.8 (49-53.3)] cm than male newborns (P<0.05 for both). No significant differences were found comparing neonatal and infant characteristics by tissue type in the total sample of pregnant women (P>0.05). Furthermore, no significant differences were found comparing infant characteristics of infants of pregnant women that were included in the PCA compared to the total sample of pregnant women.    5.3.1 Association of maternal methyl nutrient patterns and DNA methylation levels of fetal growth and obesity-related genes in the offspring Derived maternal methyl nutrient patterns were not significantly associated with DNA methylation levels of CpG sites associated with IGF-2, RXRA, HIF3\u03b1, DNMT-1, DNMT-3, LEP and LEP-R, measured in BECs or buffy coat, before or after the Benjamini-Hochberg test adjustment (P>0.05). Additionally, maternal pre-pregnancy BMI, maternal ethnicity, infant sex and, infant age were not significantly associated with DNA methylation levels of CpG sites associated with IGF-2, RXRA, HIF3\u03b1, DNMT-1, DNMT-3, LEP, and LEP-R.   143   Table 5.4 Infants characteristics of offspring with DNA methylation data by tissue and those included in the PCA1            Infant characteristics                 All infants                                               Specimen DNA methylation Women included in the PCA n  n Buffy coat n BECs P2 n PCA P3 Birth status, %        Term Preterm 246 231 15  93.9 6.10  109 106 3  97.3 2.75 137 125 12  91.2 8.76 0.05 139 129 10  92.8 7.19 0.61 Gestational age at birth, wks  246 39.7  (38.4-40.1) 109 39.7  (38.9-40.1) 137  39.7  (38.0-40.1) 0.45 139 39.9  (38.9-40.3) 0.27 BW, g 246 3380  (3060-3731) 109 3410  (3120-3731) 137 3343  (3020-3710) 0.33 139 3400  (3100-3785) 0.39 LBW, % 14 5.69 4 3.67 10 6.30 0.22 8 5.79 0.97 BL, cm 243 51.1  (49.5-53.3) 106 51.0 (49.5-53.3) 137 51.5  (50.0-53.3) 0.68 139 51.5  (50-53.3) 0.64 Infant age, wks  209  12 (11-13) 96 12 (12-13)  113 12 (11-13)  0.97 123 12 (11-13) 0.92 Infant weight, g 236 6095 (5600-6600) 108 6100  (5645-6745) 128 6000  (5500-6590) 0.35 136 6045  (5520-6555) 0.73 Infant length, cm 181 59.1 (57.1-61.0) 85 60.0 (58.0-61.0) 96 58.7 (57.0-61) 0.08 103 59.0 (57.0-70.0) 0.59 1Values are median (IQR) for continuous variables or prevalence (%) for categorical variables. P2 values are from Wilcoxon rank-sum test for comparison of continuous variables and chi-square test for comparison of categorical variables between tissues. P3 values are from Wilcoxon rank-sum test for comparison of continuous variables and chi-square test for comparison of categorical variables between the infants of pregnant women of the total sample and infants of pregnant women that were included in the PCA. BECs, buccal epithelial cells; BW, birth weight; BL, birth length; LBW, low birth weight; PCA, principal component analysis; wks, weeks          144  5.4 Discussion DNA methylation is the most stable epigenetic mechanism, and it has been suggested that it might be involved in fetal growth and obesity programming. Methyl nutrients such as folate, riboflavin, vitamins B-6 and B-12, choline, and betaine participate in interrelated pathways of one-carbon metabolism, which critical for the provision of methyl groups for DNA methylation. The findings of the present study showed no significant relationship between maternal methyl nutrients patterns at early pregnancy and DNA methylation levels of CpG sites associated with fetal growth and obesity-related genes (i.e., IGF-2, LEP, LEP-R, RXRA, and HIF3\u03b1 genes) and methyltransferase coding genes in the offspring.  Maternal methyl nutrient concentrations. Early-pregnancy betaine (median 18.1 nmol\/L), tHcy (median 4.47 \u00b5mol\/L), total B-12 (median 246 pmol\/L) and methionine (median 20.6 \u00b5mol\/L) concentrations reported in the current study are similar to maternal betaine, tHcy, total B-12 and methionine concentrations reported by Wu et al. (36) (betaine: median 13.0 \u00b5mol\/L; Hcy: median 4.10 \u00b5mol\/L; total B-12: median 259 pmol\/L; methionine: median 20.6 \u00b5mol\/L) in pregnant women living in Vancouver, BC, at 16 weeks of gestation, and in Spanish pregnant women (betaine: geometric mean ~18 \u00b5mol\/L between 12-15 weeks of gestation) (64). However, maternal non-fasting free choline concentration was almost two-fold higher (11.3 (9.35-13.5) \u00b5mol\/L) compared to early fasting maternal free choline concentration reported by Wu et al. (36) (median: 6.9 \u00b5mol\/L at 16 weeks of gestation). Differences in maternal free choline concentration might be partially explained by non-fasting conditions in the samples collected in the present study which may increase free choline concentration compared to those collected under fasting conditions (36,261,307). However, Visentin et al. (308) reported that Canadian pregnant who were not asked to fast had a free choline concentration of 7.2 \u00b5mol\/L at <16 weeks   145 of gestation. Additionally, maternal free choline concentration increases across weeks of gestation, showing a more pronounced physiological increase after 15-16 weeks of gestation (64,65), mainly due to de novo choline synthesis by the PEMT pathway, and the subsequent generation of free choline (87). In the present study, 40% of pregnant women were at >16 weeks of gestation during sample collection. Therefore, different time points of maternal sample collection (i.e., <22 weeks of gestation vs. <16 weeks of gestation) may also explain the differences found with the maternal free choline concentration reported in the PREFORM study (308).  Furthermore, maternal RBC folate concentration (median: 1334 nmol\/L) was lower compared to results reported in Canadian pregnant women who participated in the PREFORM study (mean RBC folate 2417 nmol\/L) at 16 weeks of gestation. The difference may be related to the folic acid dose in prenatal supplements consumed. In the PREFORM study, 93% of the participants consumed folic acid supplementation at early pregnancy with 83% of them using a dose \u22651000\u00b5g\/d (16). While, in the AprON study, 97% of the participants consumed folic acid supplements in the first trimester, however, only 22% used a dose >1000\u00b5g\/day (68). Also, PLP concentration (median: 73 nmol\/L) was lower in the current study compared to the PREFORM study (geometric mean: 107 nmol\/L; 95% CI 98, 116 nmol\/L). We speculate that differences in analytical methods used to determine PLP concentration (non-radioactive enzymatic assay vs. LC-MS-\/MS) may partially explain the differences (71).   To my knowledge, only one previous study conducted in the Netherlands has reported FMN and FAD concentrations during pregnancy. The study was an intervention trial that included women who had a previous NTD affected pregnancy and were planning to become pregnant. Findings of the study showed that FMN (~36 nmol\/L at 6 and 9 weeks of gestation) and FAD (~250 nmol\/L at 6 and 9 weeks of gestation) concentrations were at least two-fold higher   146 than the FMN (median:18.8 nmol\/L) and FAD (median: 78 nmol\/L) in the present study (309). We speculate that differences on FMN and FAD concentrations between studies may be partially related to the analytical methods used in each study (HPLC vs. LC-MS\/MS), however further pregnancy studies are needed to understand potential pregnancy-related metabolic adaptations that may influence FAD and FMN concentrations. Maternal methyl nutrient patterns. Maternal B-12\/methyl donor pattern explained the highest variation of the biomarker data (17.4%), followed by homocysteine (14.6%), riboflavin (14.4%), folate (13.3%) and PLP\/methionine (12.6%) patterns, which explained similar variation of the biomarker data. The derivation of the maternal B-12\/methyl donor pattern is likely attributed to the significant and stronger correlation between log-betaine and log-free choline (r=0.45, P<0.001) and log-betaine and log-total B-12 (r=0.23, P=0.007), compared to the correlation among other methyl nutrients included in the PCA. A positive correlation between maternal free choline and betaine was previously reported by Wu et al. (36) in Canadian pregnant women (\u03c1=0.60, P<0.01). The strong correlation can be related to the role of choline as the precursor of betaine due to the action of the enzyme CHDH (49). Furthermore, choline supplementation (i.e., choline chloride), the precursor of betaine, resulted in higher serum holoTC and holoTC\/total B-12 ratio in North American pregnant women (310), which may contribute to explain the significant correlation between plasma betaine, of which choline is precursor, and B-12 found in the current study.  Given the interrelationship between the folate-B-12-dependent and the alternative methylation pathways, the provision of choline may decrease the demand for the folate-B-12 dependent pathways, which may result in higher B-12 biomarker concentrations. Despite that a higher proportion of choline is used for PC during pregnancy, results of a randomized controlled trial in pregnant women showed that the provision of supplemented choline diets (i.e., 380 mg of   147 total dietary choline plus 100 mg of choline chloride or 550 mg choline chloride) enhanced the production of betaine by the choline oxidation pathway (87). Results of a prospective pregnancy study conducted in BC indicated that third-trimester plasma betaine concentration was lower in those pregnant women with deficient (<148 pmol\/L) or marginal (148-221 pmol\/L) B-12 status, compared to those classified with a sufficient B-12 (>221 pmol\/L) status (36), which may indicate an interaction between betaine and B-12, especially in folate-replete populations.  Considering that homocysteine is a central intermediary metabolite in the one-carbon metabolism and is remethylated through either the folate-B-12 dependent or the alternative-betaine dependent pathways, tHcy concentration may be influenced by total B-12, folate, betaine, PLP, and riboflavin related coenzymes. Therefore, the derivation of the maternal tHcy pattern with the only significant loading of log-tHcy (factor loading> \u01c00.45\u01c0) was unexpected. It has been previously reported that total B-12 concentration is significantly associated with tHcy concentration in Canadian population (101). Additionally, as reported in Chapter 2, maternal betaine concentration was negatively associated with tHcy concentration across the first and second trimester. We speculate that the potential small sample size of the current study may not have allowed the detection of a stronger correlation of log-tHcy with the other methyl nutrients included in the PCA. The derivation of the riboflavin pattern is likely explained by the strong correlation of log-FMN and log-FAD (r=0.26, P=0.001) which is expected because both metabolites are coenzyme forms and potential biomarkers of riboflavin status.  The derivation of the folate pattern, in which only log-RBC folate had a significant loading to the pattern (factor loading> \u01c00.45\u01c0), and the subsequent lack of a strong correlation with the other methyl nutrients and related metabolites included in the PCA, such as tHcy concentration is partially expected in pregnant women from folate-replete populations. For example, in pregnant women living in Vancouver, plasma folate was not associated with the tHcy   148 concentration at early or late pregnancy (36). However, the lack of correlation between RBC folate and the other biomarkers such as total B-12, free choline, betaine, FMN, and FAD needs to be confirmed in further pregnancy studies. Among the retained patterns, the PLP\/methionine pattern explained the lowest variation of biomarker data. PLP has a ubiquitous role as a coenzyme in the provision of one-carbon groups for the formation of 5-MTHF, the methyl donor form of folate for the remethylation of homocysteine to methionine, which may partially explain the contributions of PLP and methionine to the pattern. Overall, the derivation of early-pregnancy methyl patterns was driven mainly by maternal total-B12, betaine, and homocysteine concentrations, which is similar to the results reported in Chapter 2.3. However, the explained biomarker variation for each pattern tended to be equal (i.e., ~17-14%), which may be related to the small sample size, and the lack of strong correlations among biomarker concentrations potentially due to \u2018replete methyl nutrient conditions\u2019.  Association between maternal methyl nutrient patterns and DNA methylation levels of fetal growth and obesity-related genes in the offspring. The findings of this study indicated that there is no significant association between early pregnancy methyl nutrient patternsand DNA methylation levels of CpG sites associated with the IGF-2, RXRA, HIF3\u03b1, LEP, LEP-R, DNMT-1 and DNMT-3 genes in the offspring. To my knowledge, evidence on the association between early-pregnancy methyl nutrient status and DNA methylation levels of CpG sites associated with fetal growth and obesity-related genes in the offspring is limited. In the U.S.A, Hoyo et al. (22) found that in a sample of 496 pregnant women, low maternal RBC folate concentration (i.e., second quartile, folate concentration not reported) was associated with increased DNA methylation compared to 3rd and 4th quartiles.  Different factors may be involved in the discrepancies between the results of the current study and the findings reported by Hoyo et al. (22) First, Hoyo et al. (22) explored the association   149 between an individual maternal biomarker concentration (RBC folate concentration) and DNA methylation levels of IGF-2 in the offspring, while in the current study, I explored the relationship of combined  early-pregnancy methyl nutrient concentrations and DNA methylation levels of IGF-2 in the offspring. Second, Hoyo et al. (22) measured DNA methylation levels in cord blood, while in our study DNA methylation levels were measured in infant buffy coat (i.e., leucocytes) and BECs. It has been previously reported that DNA methylation levels vary by cell type (311). For example, DNA methylation levels of IGF-2 were consistently lower in BECs compared to DNA methylation levels of IGF-2 measured in cord blood (311) which may attenuate the potential relationship between maternal methyl nutrient patterns and DNA methylation levels of IGF-2, measured in BECs, in the offspring.  The methods used to measure DNA methylation in the aforementioned studies were pyrosequencing or mass spectrometer, whereas in the present study DNA methylation was measured using a microarray assay (Illumina 450k). Pyrosequencing measures specific genes of interest (target genes), while the microarray assay examines DMRs, covering more than 450,000 CpG sites of the genome. It has been reported that pyrosequencing can detect small differences of DNA methylation levels (<5%) compared to other technologies used (297), including the Illumina 450K, which may partially explain the lack of association in the present study, and the significant association reported in North American pregnant women (22). The current study included a relatively small sample size (n <100) which may have not allowed the detection of potential associations with small effect sizes between maternal methyl nutrient patterns and DNA methylation levels of CpG sites associated with IGF-2, LEP, LEP-R, RXRA, HIF3\u03b1, DNMT-1 and DNMT-3 genes in the offspring.  Strengths and limitations. Strengths of the present study include a comprehensive portfolio of maternal methyl nutrient concentrations at early pregnancy, and DNA methylation   150 levels of CpG sites associated with fetal growth and obesity-related genes in the offspring which allowed to test the association of maternal methyl nutrient status and DNA methylation levels in the offspring in a critical developmental window of pregnancy. Additionally, as it was described in Chapter 3.4, the statistical approach used (i.e., PCA) allowed us to explore the interrelationship between maternal methyl nutrients by the derivation of non-correlated maternal methyl nutrient patterns, which prevented potential multicollinearity among the maternal biomarkers included in the regression models. However, this prospective cohort study was limited by its small sample size as well as by the determination of DNA methylation levels in two different tissues (i.e., BECs and buffy coat), which potentially affected the capacity to detect small effect sizes in the association between maternal methyl nutrient patters and infant DNA methylation levels. Furthermore, the small sample size did not allow the inclusion of weeks of gestation at collection of maternal blood samples in the regression model. We prioritized the inclusion of maternal ethnicity, infant age and pre-pregnancy BMI because previous evidence has shown that DNA methylation levels vary according to ethnic background and age (312). Also, pre-pregnancy BMI has been associated with obesity and overweight in the offspring, which may be related to DNA methylation (312). Of the total maternal samples that we included in the present study, 30% of them were collected at the first trimester (i.e., 14 weeks of gestation) and 70% were collected between 14 and <22 weeks of gestation. We acknowledge that maternal methyl nutrient concentrations change across weeks of gestation during early pregnancy, for example, maternal betaine (i.e., ~0.5 \u00b5mol\/L\/week at < 20 weeks of gestation) (65), total B-12 (i.e., ~9 pmol\/L\/week at <20 weeks of gestation) (65) and PLP concentrations (i.e., ~2 nmol\/L\/week between 12-16 weeks of gestation and delivery) (71). Given that most of the samples included in this study were collected in the second trimester of pregnancy, it is unlikely that the inclusion of weeks of gestation in the regression model would have modified our findings. Additionally, as shown in   151 Chapter 33.3, the derivation of the maternal methyl nutrient patterns was characterized by maternal B-12 status in both trimesters. Therefore, we speculate that the potential derivation of maternal methyl nutrient patterns at the first and second trimester separately would likely provide similar retained patterns in both trimesters.  Additionally, mother-infants dyads included in this study were apparently healthy, predominantly of Caucasian ethnicity, with a high household income and maternal education level, limiting the generalizability of our results. Also, the current study did not include other factors such as genetic variants, which were suggested as important contributors to the variation of DNA methylation levels of CpG sites associated with obesity-related genes (e.g., LEP) (313,314).  In conclusion, the current study showed that early-pregnancy B-12\/methyl donor pattern explained the highest variation of the biomarker data. Additionally, no significant associations between early pregnancy methyl nutrient patterns and DNA methylation levels of CpG sites associated with IGF-2, LEP, LEP-R, RXRA, HIF3\u03b1, DNMT-1 and DNMT-3 genes in the offspring were found. Further pregnancy studies including larger sample size or pooled data including other pregnancy cohort studies are needed to explore the relationship between maternal methyl nutrients at early pregnancy and DNA methylation of CpG sites associated with fetal growth and obesity-related genes in the offspring.     152 Chapter 6: GENERAL DISCUSSION, CONCLUSIONS AND FUTURE DIRECTIONS 6.1 General discussion of the key findings In Chapter 6, a summary of the key findings of my Ph.D. research work is presented, followed by overall limitations of the present research. Lastly, future research directions and the novelty, significance and contribution of my work to the current knowledge of the relationship between maternal methyl nutrient concentrations and fetal growth and obesity programming, and neonatal anthropometric measurements are presented.   6.1.1 Summary of my Ph.D. research work Folate, betaine, choline, methionine, riboflavin and vitamin B-12 and B-6, - also known as methyl nutrients, have a combined role in the one-carbon metabolism, which is key for nucleotide synthesis, cellular proliferation and the provision of methyl groups for methylation reactions, such as DNA methylation. DNA methylation is a well-recognized stable epigenetic mechanism that is involved in fetal growth and obesity programming (Chapter 1.2). My Ph.D. research is focused on exploring the interrelationship of maternal methyl nutrient concentrations in early pregnancy, and its relationship with infant DNA methylation levels of CpG sites associated with fetal growth and obesity-related genes, and neonatal anthropometric outcomes in Canadian mother-newborn dyads.  1. The first objective of this research was to compare circulating betaine concentration between European and South Asian pregnant women in their first and second trimester of pregnancy, and to determine the relationship between betaine and tHcy concentrations in early pregnancy (i.e., <20 weeks of gestation), in a group of women living in Metro   153 Vancouver, BC, Canada. For this objective, we hypothesized that serum betaine concentration is higher in pregnant women of South Asian descent compared to pregnant women of European descent in their first and second trimester of pregnancy (P <0.05), and that serum betaine concentration is associated with tHcy concentration in early pregnancy (P<0.05). This hypothesis was supported by results presented in Chapter 2. We described that maternal betaine and DMG concentrations (\u00b5mol\/L) in South Asian [first trimester, betaine 19.8 (16.3-25.0), DMG 1.55 (1.30-1.96); second trimester, betaine 16.1 (12.9-19.8), DMG 1.42 (1.14-1.81)] were significantly higher compared to European [first trimester, betaine 17.6 (13.7-22.6), DMG 1.38 (1.12-1.77); second trimester, betaine 12.9 (10.6-16.7), DMG 1.19 (0.97-1.52)] pregnant women living in Metro Vancouver. Furthermore, in this sample of Canadian pregnant women, early-pregnancy serum betaine concentration was negatively associated with tHcy concentration (\u03b2= -0.21; 95%CI -0.34, -0.07, P=0.002) across weeks of gestation after adjusting for serum folate, total B-12, methionine, PLP, maternal age, ethnicity, neonatal sex, parity, and pre-pregnancy BMI categories. Additionally, serum total B-12 concentration was negatively associated with serum tHcy concentration (\u03b2=-0.031; 95%CI -0.04, -0.02, P<0.001), after adjustment for confounding factors.  2. The second objective of this research was to explore the interrelationship of maternal methyl nutrient concentrations in the first- and second trimester, and to determine the association of first- and second-trimester methyl nutrient patterns with neonatal anthropometric outcomes among Canadian mother-newborn dyads. We hypothesized that first- and second-trimester maternal methyl nutrient patterns were characterized by specific maternal methyl nutrient concentrations and are positively associated with   154 neonatal anthropometric outcomes among Canadian mother-newborn dyads (P<0.05). The results presented in Chapter 3 partly supported this hypothesis. First and second trimester maternal methyl nutrient patterns were characterized by B-12 status (i.e., serum total B-12 and holoTC concentrations). First and second trimester maternal B-12 status explained the highest variation of the maternal biomarker data (i.e., 26% and 25%, respectively) in a subset of pregnant women of European and South Asian ethnicity living in BC, Canada, using trimester-specific PCA. In contrast to our hypothesis, second-trimester maternal B-12 pattern was negatively associated with HC (\u03b2=-0.13 cm; 95%CI -0.24, -0.03) and HC-z-score (\u03b2=-0.05; 95%CI -0.09, -0.009), after adjustment for confounding factors (P\u22640.05 for all).  3. The third objective of this thesis was to determine DNA methylation level (%) of CpG sites associated with IGF-2, LEP, LEP-R, HIF3\u03b1, RXRA, DNMT-1, DNMT-3A, DNMT-3B genes in 3-month old infants of mothers who had high RBC folate concentration (>1360 nmol\/L) in early pregnancy (i.e., <20 weeks of gestation), compared to infants of mothers who had RBC folate concentration \u22641360 nmol\/L. For this objective, we hypothesized DNA methylation level (%) of CpG sites related to IGF-2, LEP, LEP-R, HIF3\u03b1, RXRA, DNMT-1, DNMT-3A and DNMT-3B genes in 3-month old infants of pregnant women with high RBC folate concentration (>1360 nmol\/L) in early pregnancy are higher compared to those of 3-month-old infants of pregnant women who had RBC folate concentration \u22641360 nmol\/L (P <0.05). The results presented in Chapter 4 did not support this hypothesis. Median (IQR) RBC folate concentration was 1334 (1111-1668) nmol\/L in a subset of pregnant women (n=154) that participated in the APrON study, conducted in Alberta, Canada. Of the total   155 subset of pregnant women, 51% (n=79) had folate concentration \u22641360 nmol\/L [median (IQR) RBC folate concentration of 1116 (996-1231) nmol\/L], whereas 49% (n=75) had high folate status defined as RBC folate concentration >1360 nmol\/L [median (IQR) RBC folate concentration of 1702 (1516-1887) nmol\/L]. DNA methylation levels of DNMT-3B, LEP, LEP-R DNMT-3A, HIF-3\u03b1, RXRA, IGF-2, and DNMT-1 in the infants did not significantly differ by maternal folate status.  4. The fourth objective was to investigate the relationship between maternal methyl nutrient patterns in early pregnancy (i.e., <20 weeks of gestation) and DNA methylation level (%) of CpG sites associated with DNMT-3B, LEP, LEP-R DNMT-3A, HIF-3\u03b1, RXRA, IGF-2, and DNMT-1 genes in 3-month-old infants. We hypothesized that DNA methylation level (%) of CpG sites related to DNMT-3B, LEP, LEP-R DNMT-3A, HIF-3\u03b1, RXRA, IGF-2, and DNMT-1 genes in 3-month-old infants are positively associated with maternal methyl nutrient patterns in early pregnancy (P<0.05). The results presented in Chapter 5 did not support this hypothesis. Maternal methyl nutrient patterns were characterized by B-12 and methyl donor concentrations (i.e., serum total B-12, betaine and free choline). The maternal B-12\/methyl donor pattern explained the highest variation of the variation of biomarker concentrations at early pregnancy. No significant association was found between early-pregnancy maternal methyl nutrient patterns and infant DNA methylation of CpG sites associated with LEP, RXRA, HIF-3\u03b1, IGF-2, LEP-R, DNMT-1, DNTM-3A and DNTM-3B genes.  In conclusion, the findings of my Ph.D. research emphasize the role of betaine, in addition to B-12, for homocysteine remethylation in pregnant women from a folate-replete population. The derivation of maternal methyl nutrients patterns at early   156 pregnancy was mainly driven by total B-12, holoTC, tHcy, betaine and riboflavin related metabolite concentrations which indicate that other methyl nutrients such as B-12 and not folate, may have a stronger relationship with the other methyl nutrients in folate-replete pregnant women. I only found a significant relationship between early-pregnancy methyl nutrient patterns and HC, as well as, no significant relationship between early-pregnancy methyl nutrient patterns or maternal folate status and DNA methylation levels of CpG sites associated with the investigated genes.   6.2 Discussion of key findings 6.2.1 Maternal betaine concentration and its relationship with tHcy concentration in a folate-replete population Folate and betaine have been indicated as methyl donors of the B-12 dependent and alternative remethylation pathways, respectively, in one-carbon metabolism in which both nutrients provide a methyl group for the remethylation of homocysteine to methionine. Methionine is then converted to SAM, the universal methyl donor, which is critical for methylation reactions such as DNA methylation (Chapter 1.2). In my Ph.D. research, I found that early-pregnancy betaine concentration, but not folate, was negatively associated with tHcy concentration (\u03b2=-0.21; 95%CI -0.34, -0.07 \u00b5mol\/L) across early pregnancy (i.e., <20 weeks of gestation) in folate replete Canadian pregnant women. Additionally, total B-12 but not folate, was also negatively associated with tHcy concentration (\u03b2=-0.031; 95%CI -0.040, -0.022) (Chapter 2.3).  In the last years, betaine has become a nutrient of interest mainly due to its protective role as an intracellular osmolyte, as well as its role as a methyl donor in the alternative remethylation pathway in the liver and kidneys (47,48,103). An interaction between the folate-B12 dependent   157 and alternative remethylation pathways has been described in human studies (93,95,98). In Dutch men [plasma folate concentration: 11.8 (7.8\u201317.7) nmol\/L] and postmenopausal women [plasma folate concentration: 13.1 (8.7\u201319.3) nmol\/L], betaine concentration was a negative predictor of tHcy concentration (standardized \u03b2=-0.20) (98). Furthermore, results of a study in Spanish pregnant women (64) showed that betaine became a negative predictor of tHcy concentration at 24-27 weeks of gestation but only in women with low folate status (plasma folate concentration: geometric mean 7.2 nmol\/L), which may indicate lower intracellular availability of folate, with subsequent use of betaine as a methyl donor. Meanwhile, a daily dose of folic acid supplementation (50, 100, 200, 400, 600, or 800 \u03bcg) showed a dose-dependent effect, increasing betaine concentration which suggested a sparing effect of folate as a methyl donor (98). Although folate has been indicated as the most important methyl donor in pregnancy studies conducted in non-folate-replete populations (64\u201366), in folate-replete populations, such as the Canadian population (67,207), betaine, in addition to B-12 (101), are critical for the remethylation of homocysteine to methionine. As discussed in Chapter 2.4, findings from pregnancy studies conducted in Canada indicated that most of the pregnant women (>90%) use folic acid supplements before pregnancy and across all trimesters, with most supplements containing more than two-fold (i.e., 1000 \u00b5g of folic acid) of the recommended dose for women with a low risk of NTDs (i.e., 400 \u00b5g of folic acid), which has led to high folate status (i.e., RBC folate concentration >1360 nmol\/L). Evidence from a recent animal study showed that high doses of folic acid supplementation (i.e., 20 mg of folic acid\/kg) before and during pregnancy, and lactation caused a pseudo-MTHFR deficiency in the offspring, characterized by a decreased enzymatic activity and expression of MTHFR  in the liver (94). Additionally, offspring of dams that received folic acid supplementation, had significantly lower expression of the DNMT-3A gene in the brain, compared to offspring of dams that received the control diet (94). While no   158 effect was seen in the expression of the DNMT-1 gene. These findings suggest that under folate-replete conditions not only a pseudo-intracellular folate deficiency may occur with the subsequent use of betaine as a methyl donor, but also, potential critical adverse consequences in de novo DNA methylation. De novo DNA methylation, which occurs at early stages of pregnancy, is vital for cellular differentiation and offspring growth. Findings of the present study indicate that betaine is negatively associated with tHcy concentration in early pregnancy (i.e., <20 weeks of gestation), compared to previous pregnancy studies that showed an inverse association only after 20 weeks of gestation (64\u201366). These findings suggest that in pregnant women from folate-replete populations and characterized by extended use of high dose folic acid supplements from early pregnancy to lactation, the use of betaine, as methyl donor, is promoted earlier during pregnancy which may be crucial for fetal growth programming. A recent study in mice indicated that the offspring of female mice who received betaine supplementation (1% betaine anhydrous) showed a significantly lower expression of IGF-2 gene, which is a key gene involved in fetal growth and placental development (315), and may suggest a key role of betaine in fetal growth programming. Additionally, betaine has also shown a regulatory role in the expression of enzymes involved in lipid metabolism and body composition in pigs (187), which we speculate, might be critical for obesity  programming at the early stages of pregnancy.   In humans, results of the GUSTO study (65) showed that late-pregnancy betaine concentration was negatively associated with a lower neonatal total body fat and newborn size. Similarly, I reported in Chapter 3.3.4, that first-trimester betaine concentration of those pregnant women included in first-trimester PCA was negatively associated with BL and BL z-score only in full-term babies, which may suggest that early-pregnancy betaine concentration has not only a key role for the remethylation of homocysteine to methionine in this folate-replete population,   159 but also may have a crucial role in fetal growth and body composition programming through DNA methylation.  Given the lack of national data in Canadian pregnant women, the present study contributes to the knowledge of maternal methyl nutrient concentrations in a sample of South Asian and European pregnant women in early pregnancy, which is known as a critical plasticity window for fetal development. Additionally, this research provides key information about the role of betaine in the provision of methyl groups in the remethylation pathway of one-carbon metabolism, which are further use for DNA methylation and potentially for fetal programming.   6.2.2 Early-pregnancy methyl nutrient patterns are characterized by betaine and B-12 and riboflavin related metabolites in Canadian pregnant women  Early stages of pregnancy are a critical developmental and plasticity window in which the fate of pluripotent stem cells and offspring organs is programmed. One of the mechanisms suggested to be involved in fetal programming is DNA methylation. In this context, the interrelationship of maternal methyl nutrients is crucial in providing methyl groups as well as for the synthesis of DNA, and subsequently, for cellular proliferation (Figure 1.2).  In the present research, early-pregnancy methyl nutrient patterns were derived in two subsets of pregnant women living in BC and Alberta, Canada, using a PCA approach. Overall, they were comparable as is detailed in in Chapter 3.3.3 and Chapter 5.3.2. In both subsets of pregnant women, maternal B-12 patterns explained the highest variation of the maternal biomarker data. Additionally, maternal B-12 biomarker and betaine concentrations were the most important drivers of the maternal methyl nutrient patterns. The derivation of the early pregnancy methyl nutrient patterns indicated that in these two subsets of Canadian pregnant women from a folate-replete population, direct biomarkers of B-12 status (i.e., holoTC, total B-12) explained the   160 highest variation of the maternal biomarker data. As discussed in Chapter 3.3, a strong correlation seen between total B-12 and holoTC, likely explains the subsequent derivation of the B-12 pattern in pregnant women of BC, while in pregnant women of the APrON study, B-12 also showed a significant contribution to the B-12\/methyl donor pattern. Furthermore, tHcy concentration, a functional biomarker of B-12 status, was a significant contributor to the B-12 pattern only in BC pregnant women, and it showed a significant and inverse relationship with total B-12. Regardless of the lack of a significant relationship between total B-12 and tHcy concentration in the subset of pregnant women from the APrON study, and the subsequent lack of contribution of tHcy to the B-12\/methyl donor pattern, we speculate that this is likely due to the lower sample size of the subset of APrON participants included in the PCA. The correlation of tHcy and B-12 concentration in the total sample size (n=253) was weak, but significant (r=-0.17, P=0.02) compared to the results obtained in the subset of pregnant women included in the PCA (r=-0.12, P=0.03).   Furthermore, 3 patterns were retained in pregnant women of BC, while 5 patterns were retained in the subset of participants of the APrON study. Differences in the amount of retained patterns may be explained by the different biomarkers included in each PCA, which we speculate may contribute to a partial redistribution of the relationship between the biomarkers and the subsequent retention of more patterns in the subset of pregnant women of the APrON study. For example, first trimester serum folate showed a significant correlation with riboflavin (r=0.25, P<0.001) in BC pregnant women, whereas RBC folate concentration was not significantly correlated with any of the biomarkers determined in pregnant women of the APrON study, which may contribute to the differences in the retained patterns.  Additionally, the subset of pregnant women included in each PCA (i.e., BC pregnant women: n=505 at first trimester and n=392 at the second trimester; APrON study n=139: <22   161 weeks of gestation) was different. For example, 50% of BC pregnant women were European and 50% were South Asian, while 80% of the pregnant women of the APrON study were Caucasian and 6% were South Asian. In the BC study, South Asian pregnant women had significantly lower total B-12 and higher betaine concentration compared to European pregnant women, which may have also influenced the strength of the relationship among the biomarkers and the subsequent derivation of maternal methyl nutrients patterns.  Furthermore, differences in the time point of maternal blood samples collection between BC pregnant women and the subset of the participants of the APrON study may also explained the derivation of different maternal methyl nutrients patterns. Two PCAs were conducted in the BC pregnancy study; the first PCA was conducted using biomarker concentrations measured in samples collected at ~11 weeks of gestation. The second PCA was conducted using biomarker concentrations measured in collected at ~16 weeks of gestation. While in the subset of participants of the APrON study only, one PCA was conducted using biomarker concentrations of maternal samples collected ~15 weeks of gestation. It has been previously reported that maternal methyl nutrients may decrease (i.e., PLP concentration ~2 pmol\/L per week) (71) or increase (i.e., free choline concentration ~0.2 \u00b5mol\/L per week) (65) across weeks of gestation due to pregnancy-related factors, such as hemodilution or increased estrogen concentration.  Given the folate-replete status of Canadian pregnant women, as well as the extended use of high dose of folic acid (16,67), we speculate that B-12, in addition to betaine, may become a critical nutrient for the remethylation of homocysteine to methionine. High doses of folic acid intake have shown to inhibit the expression of MTHFR and MS genes, and MTHFR and MS are both critical enzymes for the folate-B-12 dependent methylation pathway (94). In contrast, evidence from an in vitro study indicated that B-12 modulates the translation of MS (316,317). The inhibition of the expression of MTHFR and MS, and the subsequent decreased activity of   162 MTHFR and MS enzymes may cause pseudo-folate deficiency conditions. In this context, B-12 may become a key nutrient attenuating the inhibitory effect of folic acid on MS activity and promoting the remethylation of homocysteine to methionine.  In addition to the role of B-12 in the remethylation pathway of one-carbon metabolism, evidence from animal studies has indicated that B-12 has an important role at early stages of pregnancy. A study conducted by Kumar et al. (318) showed that offspring of dams that received a B-12 deficient diet (i.e., without B-12) had a decreased lean body mass (%) and fat-free mass (%), and had higher fasting plasma glucose and insulin concentrations at 12 months. Additionally, findings of the same study indicated that the B-12 deficient diet led to alterations in DNA methylation levels of genes involved in fatty acid metabolism and energy expenditure in the offspring (319). Offspring of dams that received a B-12 deficient diet had higher DNA methylation levels of the Adipor2 gene that encodes adiponectin receptor 2, and it was expressed in adipose tissue and liver (319). Adiponectin receptor 2 is involved in glucose and insulin metabolism (320). These results suggest that B-12 may have an important role not only in the remethylation of homocysteine to methionine but also in fetal programming through DNA methylation reactions. A potential role of B-12 in early stages of pregnancy has also been suggested in human studies. In Canada, low maternal B-12 concentration was associated with a higher risk of NTDs in the era of folic acid fortification (321). Neural tube closure is part of the primary neurulation process that occurs in the first 4 weeks of gestation (322). Neural tube closure involves cellular differentiation (i.e., neural stem cells) and migration; therefore, these findings suggest that B-12 may potentially play a crucial role, in addition to folate, in the cellular fate and early formation of the nervous system in the offspring (321).   In addition to the direct and functional biomarkers of B-12 status, findings of the current study indicate that betaine and riboflavin related metabolites had a significant contribution to   163 maternal methyl donor patterns (Chapters 3.3.3 and 5.3.2). We speculate that under pseudo-folate deficiency conditions, the use of the alternative remethylation pathway may be promoted (48). It has been previously suggested that folate deficiency induces choline depletion, which is the precursor of betaine, in the liver (259). Limited evidence is available regarding the potential role of betaine in early stages of pregnancy. Findings of animal studies have shown that both, betaine and folate, are important for the provision of methyl groups during embryogenesis stage (263). Additionally, specific betaine transporters have been identified in mice embryos which may indicate that betaine is needed for early stages of development and growth (102,103). Furthermore, in humans, late-pregnancy betaine concentration was inversely associated with neonatal anthropometric outcomes such as BW, BL and adiposity measurements (39). However, whether betaine is involved in the programming of fetal growth and neonatal anthropometric outcomes needs to be explored in future pregnancy studies.  In addition to betaine and B-12, plasma riboflavin was an important contributor to the patterns which is likely explained by the role of riboflavin in the form of FAD as a coenzyme for the formation of 5-MTHF (7), the form of folate involved in homocysteine remethylation. In cells, riboflavin is the precursor of FMN which is a coenzyme for the formation of PLP. PLP functions as a coenzyme in the reaction providing one-carbon groups, e.g., the reaction catalyzed by the enzyme cytosolic serine hydroxymethyltransferase (SHMT1), and in the irreversible catabolism to cysteine in the transsulfuration pathway (79).  Findings from studies conducted in healthy adults indicated that plasma riboflavin concentration showed a negative association with tHcy concentration in adults carrying the CT\/TT (323) and the CC\/TT genotype of the MTHFR C677T polymorphism (324), which may suggest an important role of riboflavin in the remethylation pathway of homocysteine to methionine. It has been reported that individuals carrying the TT genotype have 30% MTHFR   164 activity (325). Results of the Framingham Offspring Cohort study in adults showed that the inverse relationship between plasma riboflavin and tHcy concentration was restricted to those with low folate status (defined as plasma folate <12.5 nmol\/L) and TT genotype (326). We speculate that under pseudo-folate deficiency due to impaired activity of MTHFR (i.e., under potentially excessive folic acid consumption), riboflavin becomes an important nutrient for the remethylation of homocysteine to methionine, which may potentially explain the contribution of riboflavin to the derived patterns in this sub-sample of Canadian pregnant women of the APrON study. In addition, findings from animal studies have indicated that riboflavin deficiency in dams was associated with impaired embryo growth and cardiac development in the offspring (110), which may also suggest that riboflavin participates in early stages of development and cellular proliferation and differentiation. However, the relationship between maternal riboflavin status and offspring development needs to be confirmed in human studies.  Taken together, the derivation of maternal methyl nutrient patterns in early pregnancy indicated that under folate-replete conditions, B-12, betaine and riboflavin and related metabolites are the main drivers of maternal methyl nutrient patters in folate-replete Canadian pregnant women. These findings suggest that in pregnant women with high folate status and\/or folic acid intake, B-12, betaine and riboflavin may have a predominant role in the interrelationship of methyl nutrients in early pregnancy, which is vital for provision of methyl groups for DNA methylation, and potentially in fetal growth programming.        165 6.2.3 Early-pregnancy methyl nutrient patterns were associated with HC, while DNA methylation levels of fetal growth and obesity-related genes in the infants were not associated with maternal methyl nutrient patterns or maternal folate status at early pregnancy Given the interrelated role of the methyl nutrients in one-carbon metabolism for the provision of methyl groups for DNA methylation, we hypothesized that early-pregnancy methyl nutrient patterns are associated with neonatal anthropometric outcomes. Additionally, we speculated that this relationship is mediated by DNA methylation of CpG sites associated with fetal growth and obesity-related genes in infants. However, the findings of the current research did not support our hypothesis. To my knowledge, there is no available evidence on the relationship between maternal methyl nutrient patterns and neonatal anthropometric outcomes. Previous pregnancy studies that have explored the relationship between individual early-pregnancy methyl nutrient concentration and neonatal anthropometric outcomes have shown conflicting results. A pregnancy study conducted in Dutch pregnant women showed that low maternal folate concentration (i.e., folate concentration \u22649.1 nmol\/L) measured at 14 weeks of gestation was associated with a lower BW (28), while studies including Italian and Canadian pregnant women have reported that early-pregnancy concentrations of folate, total B-12, free choline, and betaine were not significantly associated with BW (30,36).  Although maternal methyl nutrient patterns were not associated with neonatal anthropometric outcomes, which is similar to the findings of previous pregnancy studies that explored the relationship of individual maternal methyl nutrients and neonatal anthropometric outcomes, the derivation of maternal methyl nutrient patterns provides useful information about the interrelationship of maternal methyl nutrients, taking into consideration the contribution of all maternal methyl nutrients. As described in Chapter 1.2.1, methyl nutrients participate in   166 interdependent reactions of the one-carbon metabolism, thus, restricting the analysis to individual maternal methyl nutrient concentration (i.e., including only maternal folate concentration as exposure variable) may limit the understanding of the role of methyl nutrients in early growth and development.  Consistent with the results of the association between maternal methyl nutrient patterns and neonatal anthropometric outcomes, my findings showed no significant associations between maternal methyl nutrient patters and DNA methylation of CpG sites associated with fetal growth and obesity-related genes measured in the infants. Additionally, findings of the present study showed no significant differences in DNA methylation levels of CpG sites associated with the interrogated genes comparing RBC folate \u22641360 nmol\/L and high maternal folate status (i.e., RBC folate >1360 nmol\/L). Evidence of the relationship between individual methyl nutrient concentrations at early pregnancy and DNA methylation levels of CpG sites associated with fetal growth genes is limited. Results of a prospective pregnancy study conducted in North American pregnant women (22) indicated that low RBC folate concentration (i.e., second quartile), measured at ~12 weeks of gestation, was positively associated with DNA methylation levels of CpG sites associated with IGF-2, determined in cord blood. In contrast, McKay et al. (225) found no significant association between maternal RBC folate concentration measured at ~11 weeks of gestation and DNA methylation levels of CpG sites associated with IGF-2 in cord blood.  Given the differences between methods for the determination of DNA methylation levels (i.e., microarray assay vs. pyrosequencing) and tissue and cell types used (i.e., cord blood vs. infant blood and BECs), comparisons between studies is challenging. It has been previously reported that pyrosequencing is able to detect smaller differences on DNA methylation levels (<5%), compared to other methods (297). In addition, a study conducted by Murphy et al. (311)   167 showed that DNA methylation levels of IGF-2 DMR measured in 9 pairs of matched cord blood and BECs samples collected at birth, were slightly lower in BECs compared to DNA methylation levels measured in cord blood, which may contribute to partially explain the discrepancies found. Furthermore, in the current study, we included the measurement of single CpG sites, while, the prospective study conducted by Hoyo et al. (22) in North American pregnant women, measured DNA methylation levels in IGF-2 DMR. DMR consists of a group of CpG sites associated with IGF-2. We speculate that differences in the results of the relationship between RBC folate concentration and DNA methylation of CpG sites associated with IGF-2 may be also related to differences in DNA methylation levels obtained from measuring DMR or single CpG sites.  Differences in RBC folate concentration between the studies may also contribute to explain the discrepancies found between the studies. In the current study, the median RBC folate concentration was ~1300 nmol\/L, while in the study conducted by Hoyo et al. (22) the mean RBC folate concentration was ~494 nmol\/L. We speculate that IGF-2 DMR may have a more sensitive response to \u2018low folate status\u2019 compared to \u2019high folate status\u2019. However, further prospective pregnancy cohort studies including pregnant women from populations with different folate status, as well as the measurement of DNA methylation levels in CpGs and DMRs associated with IGF-2 measured in the offspring, are needed to understand the potential differences in DNA methylation levels comparing individual CpG sites and DMRs.   6.3 Overall strengths and limitations Specific strengths and limitations of each study i.e., the BC retrospective pregnancy cohort study and the prospective APrON study, are discussed in each chapter. This section summarizes the overall strengths and limitations of the current Ph.D. research work. The strengths of this research included the longitudinal design of pregnancy studies with available   168 maternal blood samples collected at early pregnancy which allowed us to explore the relationship of early-pregnancy methyl nutrients patterns with neonatal anthropometric outcomes and DNA methylation levels of CpG sites associated with fetal growth and obesity-related genes in mother-offspring dyads in a critical developmental window of pregnancy. Additionally, the present research includes the determination of a comprehensive portfolio of maternal methyl nutrients and related metabolite concentrations. Furthermore, the statistical approach used (i.e., PCA) allowed us to explore the interrelationship between maternal methyl nutrients by the derivation of non-correlated maternal methyl nutrient patterns, which prevented potential multicollinearity among the maternal biomarkers included in the regression models.  We acknowledge some limitations. The blood samples retrieved for the BC retrospective pregnancy cohort study were from pregnant women of the BC Prenatal Genetic Screening Program, while pregnant women of the APrON study were recruited by convenience sampling, limiting the representativeness of the study. Given that the primary objective of the BC retrospective pregnancy cohort study was to compare serum total B-12 concentration between South Asian and European pregnant women in the first and second trimester, the research described in Chapters 2.2 and 3.2 included biospecimen and data from an equal distribution of European and South Asian pregnant women, which also limits the generalizability of my findings. Furthermore, according to their estimated income, pregnant women included in both studies had a higher income or they reside in higher income neighbourhoods compared to the average national individual income provided by Statistics of Canada in 2017 (327). Additionally, most pregnant women included in the APrON study as described in Chapter 4 and 5 had high education level and were non-smokers. Thus, it is possible that pregnant women included in the current research had better health care access, a greater willingness to participate in a study about nutrition, more knowledge of nutrition, and were more health conscious which may influence   169 their consumption of folic acid, multivitamin or mineral supplements, and subsequently affecting their folate status (i.e., higher RBC folate concentration), compared to pregnant women of the general population. However, despite the limitations regarding the generalizability of the findings, the present research provides valuable information on maternal characteristics and B-vitamin status of Canadian pregnant women of which there is no available national data. Thus, longitudinal pregnancy studies become an important source of information of health and nutritional status of this specific life stage and provides a key opportunity to study maternal health and fetal development at early stages of life. Additionally, the current research provides maternal and neonatal data on pregnant women of South Asian ethnicity, the largest ethnic minority in Canada and BC and thus, this research represented an opportunity to study this specific population group.  The inclusion of different biomarkers resulted in the derivation of different maternal methyl nutrient patterns; therefore, the comparison of maternal methyl patterns needs to be done with caution. Because of the blood collection protocol for the biobanked maternal serum samples retrieved for the retrospective pregnancy cohort study (Chapters 2 and 3), data on serum free choline concentration were not included in the data analysis. Choline participates in one-carbon metabolism, e.g., as a precursor for betaine, the methyl donor of the alternative homocysteine remethylation pathway. Although, betaine and total B-12 have been identified as significant predictors of tHcy, the inclusion of free choline could have offered a more comprehensive understanding of the interrelationship of methyl nutrients and the derivation of maternal methyl nutrient patterns in the BC study. A recent study conducted by van Lee et al. (190) which included maternal free choline, total choline and neonatal anthropometric data from two pregnancy cohort studies (i.e., GUSTO and SWS studies) showed that early-pregnancy total   170 choline and free choline concentrations were positively associated with total body fat (SWS study) and neonatal BMI (GUSTO study), respectively.  The authors indicated that potential mechanisms underlying the relationship between maternal choline concentration (i.e., free choline and total choline concentrations) with neonatal anthropometric outcomes and body fat are related to epigenetic mechanisms, such as DNA methylation. Specifically, the authors suggested that maternal choline may have a key role as a methyl donor in the control of gene expression of body composition-related genes through DNA methylation in the offspring (190). Additionally, it suggested that acetylcholine, a neurotransmitter synthesized from choline, may be involved in the regulation of glucose and lipids in adipose tissue. Thus, including maternal free choline in the derivation of maternal patterns could have also provided a better understanding of the relationship between maternal methyl nutrient patterns and neonatal anthropometric outcomes. Furthermore, because of the lack of availability of specimen (e.g., RBC samples in the BC study), it was not possible to derive maternal methyl nutrient patterns using the same maternal biomarker portfolio in the two subsets of pregnant women included in my Ph.D. research, which would have allowed for an accurate comparison of the patterns and an assessment of reproducibility. The determination of DNA methylation was conducted using two different types of tissues (i.e., BECs and buffy coat), therefore comparison and interpretation of the DNA methylation levels of the targeted CpG sites included in my Ph.D. research work should be made with caution. Despite that several patterns of DNA methylation are shared by different types of cells, it has been widely reported that DNA methylation may differ by cell type or tissue (311). Tissue and cell types have been indicated as the most important sources of variation in DNA methylation levels (313). For example, DNA methylation levels have greater variability for genome wide and specific CpG sites when measured in BECs compared to peripheral blood   171 mononuclear cells (328). Given the lack of an independent subset of mother-infants dyads data with available infant DNA methylation data and maternal folate status, I was unable to replicate the results of Chapters 4 and 5. Furthermore, the age of the offspring at the time of sample collection for DNA methylation analyses needs to be taken under consideration. DNA methylation levels of CpG sites associated with the LEP gene have been shown to significantly differ between newborns (i.e., measured in cord blood) and 1-year-old infants (i.e., measured in whole blood) (296). Therefore, discrepancies between our findings and results of previous studies may also be related to differences in the time point of sample collection for DNA methylation analyses.  6.4 Future research directions 6.4.1 Periconceptional methyl nutrient status in women and DNA methylation levels of fetal growth and obesity-related genes in the offspring Most of the current studies have explored the relationship between maternal methyl nutrient concentrations measured after 10 weeks of gestation and DNA methylation levels in the offspring. De novo DNA methylation occurs at the very early stage of pregnancy (i.e., <10 weeks of gestation), a period when most women are unaware of being pregnant. The determination of periconceptional methyl nutrient concentrations (i.e., one month before conception to two months after conception) would provide an ideal opportunity to explore the relationship of early maternal methyl nutrients concentrations, and DNA methylation levels of fetal growth and obesity-related genes, in this critical plasticity window.  Therefore, a prospective study including the determination of methyl nutrient concentrations before conception, and the measurement of DNA methylation levels of obesity-related genes in the offspring is needed to explore whether pre-conceptional maternal methyl   172 nutrient status is involved in obesity growth programming. Furthermore, maternal genetic variants have shown to contribute to the variation of DNA methylation levels of genes included in the present research (i.e., LEP gene) (296). Thus, the measurement of genetic variants of obesity-related genes in the offspring would provide a better understanding of maternal factors potentially involved in DNA methylation variation.  Additionally, evidence from pregnancy studies in Canada has shown that >90% of pregnant women use folic acid supplements in the first trimester of pregnancy; however, results of the Canadian Maternity Experience Survey indicated that only 60% of women consumed folic acid supplements before pregnancy (206). Critical milestones in the embryonic development (e.g., closure of the neural tube or the differentiation of germ layers) occur before 28 days of gestation. This is a gestational period in which most women are unaware of being pregnant. Furthermore, findings of a recent study in Canadian pregnant women indicated that women had a lower knowledge related to the DOHaD theory, compared to their knowledge of pregnancy guidelines (329). These findings can be considered as a lack of knowledge translation about the DOHaD-related evidence. However, these findings can also be considered as an opportunity to conduct studies that contribute novel evidence and knowledge of the critical role of preconceptional and early-pregnancy nutrition and health for later stages of life, and to develop strategies to improve the DOHaD-related knowledge translation.   6.4.2 Obesity programming and its consequences later in life DNA methylation is a stable and critical epigenetic mechanism involved in fetal programming. However, whether these modifications in DNA methylation levels of fetal growth or obesity-related genes in the offspring result in modifications in the expression of the target molecules (i.e., serum IGF-2 or insulin concentrations) is not fully understood. Prospective   173 cohort studies including the measurement of DNA methylation levels of CpG sites associated with fetal growth and obesity related-genes (i.e., LEP, HIF-3A, RXRA) at birth or early childhood (i.e., infants), the determination of biochemical parameters potentially involved in obesity such as insulin or leptin concentrations, and the occurrence of obesity during adolescence are needed to understand the metabolic pathway behind the relationship of DNA methylation levels of obesity-related genes and the risk or occurrence of obesity later in life.  Furthermore, evidence from observational studies that followed newborns until early adolescence indicated that the trajectory of weight gain as well as the timing of the adiposity rebound are important predictors of obesity later in life (330). Thus, exploring the relationship between DNA methylation levels of fetal growth and obesity-related genes in the offspring and weight gain in childhood and adiposity rebound is key to understand the potential role of DNA methylation levels in the offspring in obesity programming and its consequences later in life.  6.4.3 Paternal methyl nutrient concentrations and DNA methylation in the offspring Maternal nutrition and health are fundamental for adequate fetal growth and development. The majority of pregnancy studies to date have focused on studying the association or effect of maternal factors on offspring outcomes. In addition to maternal nutrition, evidence from animal and human studies have reported that paternal lifestyle factors such as obesity and paternal dietary intake before conception are also involved in fetal programming (331\u2013333). Recently, Sahara et al. (334) indicated that the offspring of male mice who received a methyl donor deficient diet showed a decrease in anxiety\u2010like behavior which was mediated by modifications of DNA methylation levels of cognitive-related genes. While in humans, paternal dietary betaine and methionine intakes were positively associated with global DNA methylation levels and DNA   174 methylation levels of IGF-2 in the newborn, respectively (333). Therefore, future studies including the determination of paternal methyl nutrient concentrations before conception and its association with DNA methylation levels of fetal growth and obesity-related genes in the offspring are needed to understand the potential paternal role in fetal growth and obesity programming.   6.4.4 Derivation of maternal methyl nutrient patterns in non-folate-replete populations The derivation of maternal methyl nutrient patterns is highly influenced by biomarker concentrations and the relationship among them. Despite the central role of folate as methyl donor, folate was not a significant driver of maternal methyl nutrient patterns in the present research. Pregnant women included in this research were from a folate-replete population, and with a primarily high SES. Thus, maternal methyl nutrient patterns derived in this research may not be replicable in other populations with different folate status. Whether the interrelationship of maternal methyl nutrients and the subsequent characteristics of the derived maternal methyl patterns are different in non-folate-replete populations is unknown. Therefore, further prospective studies conducted across all trimesters including pregnant women from other populations are needed to explore potential differences in the relationship between maternal methyl nutrients, as well as to identify other drivers of maternal methyl nutrient patterns under non-folate-replete conditions. Given that critical methylation reactions occur during pregnancy and that the relationship of methyl nutrients varies according to methyl nutrient status (i.e., low or high folate status), the understanding of the relationship of methyl donors and homocysteine, as a central metabolite of the remethylation pathway, is needed in non-folate-replete populations.    175 6.5 Significance and contribution of research The current research work contributed three new findings about the interrelationship of methyl nutrient concentrations during early pregnancy and its relationship with DNA methylation of fetal growth and obesity-related genes and neonatal anthropometric outcomes. Findings of the present study showed that maternal betaine concentration was negatively associated with tHcy concentration in early pregnancy, while folate concentration did not show a significant association with tHcy in this sample of Canadian pregnant women from a folate-replete population. Given the higher daily folic acid intake in pregnant women (335) (i.e., 1000\u00b5g) compared to the recommended dose of 400\u00b5g\/day for NTD prevention, these results are particularly important because they might provide a novel perspective about the homocysteine remethylation reaction of one carbon metabolism, in particular the key role, of betaine as a methyl donor of the alternative remethylation pathway under folate-replete conditions. Second, the derivation of maternal methyl nutrient patterns in the present study allowed to explore the interrelationship of maternal methyl nutrient concentrations in early pregnancy. Maternal methyl nutrient patterns were mainly driven by the relationship of direct and functional biomarkers of B-12 of status (i.e., total B-12 and holoTC and tHcy concentrations). Additionally, betaine and riboflavin related metabolites (i.e., free riboflavin, FMN and FAD) showed significant contributions to the patterns. These results suggest that in folate-replete pregnant women, total B-12, betaine, and riboflavin concentrations may play critical roles in homocysteine remethylation, for the formation of methionine, the precursor of SAM. These findings are particularly relevant because they may suggest that not only folate, as it has been reported in non-folate-replete populations, but also betaine and B-12 may become critical for the provision of methyl groups for DNA methylation reactions that are central to de novo DNA methylation and fetal programming.    176 Third, early-pregnancy methyl nutrient patterns showed only a significant association with HC at birth, while no significant association was seen between maternal methyl nutrient patterns or maternal folate status, and DNA methylation levels of CpG sites associated with LEP, LEP-R, IGF-2, RXRA, HIF-3\u03b1, DNMT-1, DNMT-3A and DNMT-3B in infants. These findings suggest that in apparently healthy mother-offspring dyads, the interrelationship of maternal methyl nutrient concentrations and maternal folate status alone during early pregnancy may not be significantly related to neonatal anthropometric outcomes and DNA methylation levels of CpG sites associated with fetal growth and obesity-related genes.     177 References 1.  Harding R, Bocking AD. Fetal growth and development. Cambridge. New York; 2001.  2.  Abu-Saad K, Fraser D. Maternal nutrition and birth outcomes. Epidemiol Rev. 2010;32:5\u201325.  3.  Wu G, Bazer FW, Cudd T, Meininger CJ, Spencer TE. Maternal nutrition and fetal development. J Nutr. 2004;134(1):2169\u201372.  4.  Bergen NE, Jaddoe VWV, Timmermans S, Hofman A, Lindemans J, Russcher H, et al. Homocysteine and folate concentrations in early pregnancy and the risk of adverse pregnancy outcomes\u202f: the Generation R Study. BJOG. 2012;739\u201351.  5.  Furness DLF, Yasin N, Dekker GA, Steven D, Roberts CT. Maternal red blood cell folate concentration at 10 \u2013 12 weeks gestation and pregnancy outcome. J Matern Fetal Neonatal Med. 2012;7058.  6.  Serrano NC, Quintero-Lesmes DC, Becerra-Bayona S, Guio E, Beltran M, Paez MC, et al. Association of pre-eclampsia risk with maternal levels of folate, homocysteine and vitamin B12 in Colombia: A case-control study. PLoS One. 2018;13(12):1\u201315.  7.  Fox JT, Stover PJ. Chapter 1 Folate-Mediated One-Carbon Metabolism. Vitam Horm. 2008;79:1\u201344.  8.  MRC Vitamin Study Research Group. Prevention of neural tube defects: Results of the Medical Research Council Vitamin Study. Lancet. 1991;338(8760):131\u20137.  9.  Czeizel A, Dudas I. Prevention of the first occurrence of neural tube defects by periconceptional vitamin supplementation. N Engl J Med. 1992;327(26):1832\u20131835.  10.  Institute of Medicine (US) Standing Committee on the Scientific Evaluation of Dietary Reference Intakes and its Panel on Folate, Other B Vitamins  and C. Dietary Reference   178 Intakes for Thiamin, Riboflavin, Niacin, Vitamin B6, Folate, Vitamin B12, Pantothenic Acid, Biotin, and Choline. National A. Washington, D.C; 1998.  11.  Wilson D, Audibert F, Brock J, Carroll J, Cartier L, Gagnon A, et al. Pre-conception folic acid and multivitamin supplementation for the primary and secondary prevention of neural tube defects and other folic acid-sensitive congenital anomalies. J Obstet Gynaecol Canada. 2015;37(6):534\u201349.  12.  Health Canada. Licensed Natural Health Products Database (LNHPD). [Internet]. Ottawa, Canada: Government of Canada; 2015, [cited 2020 Aug 15]. http:\/\/www.hc-sc.gc.ca\/dhp-mps\/prodnatur\/applications\/licen-prod\/lnhpd-bdpsnh-eng.php.  13.  Health Canada. Prenatal nutrition guidelines for health professionals: folate[Internet]. Ottawa, Canada: Government of Canada; 2009. p. 1\u20138.. Report No.: H164\u2013109\/2\u20132009E-PDF. Available from: http:\/\/www.hc-sc.gc.ca\/fn-an\/alt_formats\/hpfb-dgpsa\/pdf\/pubs\/fo.  14.  Colapinto CK, O\u2019Connor DL, Tremblay MS. Folate status of the population in the Canadian Health Measures Survey. Can Med Assoc J. 2011;183(2):E100-06.  15.  Colapinto CK, O\u2019Connor DL, Dubois L, Tremblay MS. Folic acid supplement use is the most significant predictor of folate concentrations in Canadian women of childbearing age. Appl Physiol Nutr Metab. 2012;37(2):284\u201392.  16.  Plumptre L, Masih SP, Ly A, Aufreiter S, Sohn K, Croxford R, et al. High concentrations of folate and unmetabolized folic acid in a cohort of pregnant Canadian women and umbilical cord blood. Am J Clin Nutr. 2015;102(4):848\u201357.  17.  Lamers Y, MacFarlane AJ, O\u2019Connor DL, Fontaine-Bisson B. Periconceptional intake of folic acid among low-risk women in Canada: Summary of a workshop aiming to align prenatal folic acid supplement composition with current expert guidelines. Am J Clin Nutr. 2018;108(6):1357\u201368.    179 18.  Okano M, Bell DW, Haber DA, Li E. DNA methyltransferases Dnmt3a and Dnmt3b are essential for de novo methylation and mammalian development. Cell. 1999;99(3):247\u201357.  19.  Dunlevy LPE, Burren KA, Mills K, Chitty LS, Copp AJ, Greene NDE. Integrity of the methylation cycle is essential for mammalian neural tube closure. Birth Defects Res Part A - Clin Mol Teratol. 2006;76(7):544\u201352.  20.  Waterland RA, Jirtle RL. Transposable elements\u202f: targets for early nutritional fffects on epigenetic gene regulation. Mol Cell Biol. 2003;23(15):5293\u2013300.  21.  Dominguez-Salas P, Moore SE, Baker MS, Bergen AW, Cox SE, Dyer RA, et al. Maternal nutrition at conception modulates DNA methylation of human metastable epialleles. Nat Commun. 2014;5:3746.  22.  Hoyo C, Daltveit AK, Iversen E, Benjamin-Neelon SE, Fuemmeler B, Schildkraut J, et al. Erythrocyte folate concentrations, CpG methylation at genomically imprinted domains, and birth weight in a multiethnic newborn cohort. Epigenetics. 2014;9(8):1120\u201330.  23.  Pauwels S, Ghosh M, Duca RC, Bekaert B, Freson K, Huybrechts I, et al. Dietary and supplemental maternal methyl-group donor intake and cord blood DNA methylation. Epigenetics. 2017;12(1):1\u201310.  24.  Pauwels S, Ghosh M, Duca RC, Bekaert B, Freson K, Huybrechts I, et al. Maternal intake of methyl-group donors affects DNA methylation of metabolic genes in infants. Clin Epigenetics. 2017;9(1):16.  25.  Takimoto H, Mito N, Umegaki K, Ishiwaki A, Kusama K, Abe S, et al. Relationship between dietary folate intakes, maternal plasma total homocysteine and B-vitamins during pregnancy and fetal growth in Japan. Eur J Nutr. 2007;46(5):300\u20136.  26.  Relton CL, Pearce MS, Parker L. The influence of erythrocyte folate and serum vitamin B 12 status on birth weight. Br J Nutr. 2005;93(5):593\u20139.    180 27.  Yajnik CS, Chandak GR, Joglekar C, Katre P, Bhat DS, Singh SN, et al. Maternal homocysteine in pregnancy and offspring birthweight: Epidemiological associations and Mendelian randomization analysis. Int J Epidemiol. 2014;43(5):1487\u201397.  28.  Bergen NE, Schalekamp-Timmermans S, Jaddoe VWV, Hofman A, Lindemans J, Russcher H, et al. Maternal and neonatal markers of the homocysteine pathway and fetal growth: The Generation R Study. Paediatr Perinat Epidemiol. 2016;30(4):386\u201396.  29.  McCullough LE, Miller EE, Mendez MA, Murtha AP, Murphy SK, Hoyo C. Maternal B vitamins: effects on offspring weight and DNA methylation at genomically imprinted domains. Clin Epigenetics. 2016;8(1):8.  30.  Parazzini F, Chiaffarino F, Ricci E, Improta L. Homocysteine, red cell, and plasma folate concentrations and birth weight in Italian women\u202f: results from a prospective study. J Matern Fetal Neonatal Med. 2011;24(3):427\u201331.  31.  Jiang HL, Cao LQ, Chen HY. Blood folic acid, vitamin B12, and homocysteine levels in pregnant women with fetal growth restriction. Genet Mol Res. 2016;15(4).  32.  Hogeveen M, Blom HJ, Heijden EH Van Der, Semmekrot BA, Sporken JM, Ueland PM, et al. Maternal homocysteine and related B vitamins as risk factors for low birthweight. Am J Obs Gynecol. 2010;202(6):572.e1-6.  33.  Krishnaveni G V, Veena SR, Karat SC, Yajnik CS, Fall CH. Association between maternal folate concentrations during pregnancy and insulin resistance in Indian children. Diabetologia. 2014;57(1):110\u201321.  34.  Chen LW, Lim AL, Colega M, Tint MT, Aris IM, Tan C, et al. Maternal folate status, but not that of vitamins B-12 or B-6 , is associated with gestational age and preterm birth risk in a multiethnic Asian population. J Nutr. 2015;145(1):113\u201320.  35.  Hogeveen M, den Heijer M, Semmekrot BA, Sporken JM, Ueland PM, Blom HJ.   181 Umbilical choline and related methylamines betaine and dimethylglycine in relation to birth weight. Pediatr Res. 2013;73(6):783\u20137.  36.  Wu BTF, Innis SM, Mulder KA, Dyer RA, King DJ. Low plasma vitamin B-12 is associated with a lower pregnancy-associated rise in plasma free choline in Canadian pregnant women and lower postnatal growth rates in their male infants. Am J Clin Nutr. 2013;98(5):1209\u201317.  37.  Halicioglu O, Sutcuoglu S, Koc F, Ozturk C, Albudak E, Colak A, et al. Vitamin B 12 and folate statuses are associated with diet in pregnant women, but not with anthropometric measurements in term newborns. J Matern Neonatal Med. 2012;25(9):1618\u201321.  38.  Nilsen RM, Vollset SE, Monsen AL, Ulvik A, Haugen M, Meltzer MH, et al. Infant birth size is not associated with maternal intake and status of folate during the second trimester in Norwegian pregnant women. J Nutr. 2010;140(3):572\u20139.  39.  Van Lee L, Tint MT, Aris IM, Quah PL, Fortier M V, Lee YS, et al. Prospective associations of maternal betaine status with offspring weight and body composition at birth: The Growing Up in Singapore Towards healthy Outcomes (GUSTO) cohort study. Am J Clin Nutr. 2016;104(5):1327\u201333.  40.  Scott JM. Folate and vitamin B 12 . Proc Nutr Soc. 1999;58(2):441\u20138.  41.  Bailey LB, Gregory JF. Folate metabolism and requeriments. J Nutr. 1999;129:779\u201382.  42.  Stipanuk MH. Sulfur amino acid metabolism: Pathways for production and removal of homocysteine and cysteine. Annu Rev Nutr. 2004;24(1):539\u201377.  43.  Zhao M, Lamers Y, Ralat M, Coats B, Chi Y, Muller K, et al. Marginal vitamin B-6 deficiency decreases plasma (n-3) and (n-6) PUFA concentrations in healthy men and women. J Nutr. 2012;142(10):1791\u20137.  44.  Zeisel SH, Blusztajn JK. Choline and human nutrition. Annu Rev Nutr. 1994;14:269\u201396.    182 45.  Guo-Ross SX, Clark S, Montoya DAC, Jones KH, Obernier J, Shetty AK, et al. Prenatal choline supplementation protects against postnatal neurotoxicity. J Neurosci. 2002;22(1):1\u20136.  46.  Zhang D, Jing H, Dou C, Zhang L, Wu X, Wu Q, et al. Supplement of betaine into embryo culture medium can rescue injury effect of ethanol on mouse embryo development. Sci Rep. 2018;8(1):1\u201313.  47.  Lee MB, Kooistra M, Zhang B, Slow S, Fortier AL, Garrow TA, et al. Betaine homocysteine methyltransferase is active in the mouse blastocyst and promotes inner cell mass development. J Biol Chem. 2012;287(39):33094\u2013103.  48.  Lever M, Slow S. The clinical significance of betaine, an osmolyte with a key role in methyl group metabolism. Clin Biochem. 2010;43(9):732\u201344.  49.  Craig SA. Betaine in human nutrition. Am J Clin Nutr. 2004;80(3):539\u201349.  50.  Stover PJ. One-carbon metabolism-genome interactions in folate-associated pathologies. J Nutr. 2009;139(12):2402\u20135.  51.  Inoue-Choi M, Nelson HH, Robien K, Arning E, Bottiglieri T, Koh WP, et al. Plasma S-adenosylmethionine, DNMT polymorphisms, and peripheral blood LINE-1 methylation among healthy Chinese adults in Singapore. BMC Cancer. 2013;13(389):doi: 10.1186\/1471-2407-13-389.  52.  Stipanuk MH, Dominy JE, Lee J, Coloso RM. Mammalian cysteine metabolism: new insights into regulation of cysteine metabolism. J Nutr. 2006;136(Suppl 6):1652S\u20131659S.  53.  Shin YS, Chan C, Vidal AJ, Brody T, Stokstad EL. Subcellular localization of \u03b3-glutamyl carboxypeptidase and of folates. Biochim Biophys Acta. 1976;444(3):794\u2013801.  54.  Bianchetti R, Lucchini G, Crosti P, Tortora P. Dependence of mitochondrial protein synthesis initiation on formylation of the iniator methionyl-tRNA. J Biol Chem.   183 1977;252(8):2519\u201323.  55.  Nijhout HF, Gregory JF, Fitzpatrick C, Cho E, Lamers Y, Ulrich CM, et al. A mathematical model gives insights into the effects of vitamin B-6 deficiency on 1-carbon and glutathione metabolism. J Nutr. 2009;139(4):784\u201391.  56.  Reo N, Adinehzadeh M, Foy BD. Kinetic analyses of liver phosphatidylcholine and phosphatidylethanolamine biosynthesis using 13C NMR spectroscopy. Biochim Biophys Acta. 2002;1580(2\u20133):171\u201388.  57.  Wittwer AJ, Wagner C. Identification of folate binding protein of mitochondria as dimethylglycine dehydrogenase. Proc Natl Acad Sci U S A. 1980;77(8):4484\u20138.  58.  Friso S, Udali S, De Santis D, Choi SW. One-carbon metabolism and epigenetics. Mol Aspects Med. 2017;54:28\u201336.  59.  Shane B. Folate, vitamin B12 and vitamin B6. 2nd ed. Stipanuk MH, editor. New York; 2006. 693\u2013732 p.  60.  Centers for Disease Control (CDC). Recommendations for use of folic acid to reduce number of spina bifida cases and other neural tube defects. MMWR. 1992;41:1\u20138.  61.  Morin VI, Mondor M, Wilson RD. Knowledge on periconceptional use of folic acid in women of British Columbia. Fetal Diagn Ther. 2001;16(2):111\u20135.  62.  Finer LB, Henshaw SK. Disparities in rates of unintended pregnancy in the United States, 1994 and 2001. Perspect Sex Reprod Health. 2006;38(2):90\u20136.  63.  De Wals P, Tairou F, Van Allen MI, Uh SH, Lowry RB, Sibbald B, et al. Reduction in neural-tube defects after folic acid fortification in Canada. N Engl J Med. 2007;357(2):135\u201342.  64.  Fern\u00e0ndez-Roig S, Cavall\u00e9-Busquets P, Fernandez-Ballart JD, Ballesteros M, Berrocal-Zaragoza MI, Salat-Batlle J, et al. Low folate status enhances pregnancy changes in   184 plasma betaine and dimethylglycine concentrations and the association between betaine and homocysteine. Am J Clin Nutr. 2013;97(6):1252\u20139.  65.  Velzing-Aarts F V, Holm PI, Fokkema MR, Van Der Dijs FP, Ueland PM, Muskiet FA. Plasma choline and betaine and their relation to plasma homocysteine in normal pregnancy. Am J Clin Nutr. 2005;81(6):1383\u20139.  66.  Wallace JM, Bonham MP, Strain JJ, Duffy EM, Robson PJ, Ward M, et al. Homocysteine concentration, related B vitamins, and betaine in pregnant women recruited to the Seychelles Child Development Study. Am J Clin Nutr. 2008;87(2):391\u20137.  67.  Fayyaz F, Wang F, Jacobs RL, O\u2019Connor DL, Bell RC, Field CJ. Folate, vitamin B12, and vitamin B6 status of a group of high socioeconomic status women in the Alberta Pregnancy Outcomes and Nutrition (APrON) cohort. Appl Physiol Nutr Metab. 2014;39(12):1402\u20138.  68.  G\u00f3mez MF, Field CJ, Olstad DL, Loehr S, Ramage S, Mccargar LJ, et al. Use of micronutrient supplements among pregnant women in Alberta: Results from the Alberta Pregnancy Outcomes and Nutrition (APrON) cohort. Matern Child Nutr. 2015;11(4):497\u2013510.  69.  Cikot RJ, Steegers-Theunissen RP, Thomas CM, de Boo TM, Merkus HM, Steegers EA. Longitudinal vitamin and homocysteine levels in normal pregnancy. Br J Nutr. 2007;85(1):49\u201358.  70.  Mujica-Coopman MF, Franco-Sena AB, Farias DR, Vaz JS, Brito A, Kac G, et al. Vitamin B-6 status in unsupplemented pregnant women is associated positively with serum docosahexaenoic acid and inversely with the n-6-to-n-3 fatty acid ratio. J Nutr. 2017;147(2):170\u20138.  71.  Plumptre L, Masih SP, Sohn KJ, Kim D, Visentin CE, Ly A, et al. Suboptimal maternal   185 and cord plasma pyridoxal 5\u2032 phosphate concentrations are uncommon in a cohort of Canadian pregnant women and newborn infants. Matern Child Nutr. 2018;14(1):e12467.  72.  Visentin CE, Masih SP, Plumptre L, Schroder TH, Sohn KJ, Ly A, et al. Low serum vitamin B-12 concentrations are prevalent in a cohort of pregnant Canadian women. J Nutr. 2016;146(5):1035\u201342.  73.  Murphy MM, Molloy AM, Ueland PM, Fernandez-Ballart JD, Schneede J, Arija V, et al. Longitudinal study of the effect of pregnancy on maternal and fetal cobalamin status in healthy women and their offspring. J Nutr. 2007;137(8):1863\u20137.  74.  Bruinse HW, van den Berg H. Changes of some vitamin levels during and after normal pregnancy. Eur J Obstet Gynecol Reprod Biol. 1995;61(1):31\u20137.  75.  Ek J. Plasma and red cell folate values in newborn infants and their mothers in relation to gestational age. J Pediatr. 1980;97(2):288\u201392.  76.  Hall MH, Pirani BBK, Campbell D. The cause of the fall in serum folate in normal pregnancy. Br J Obs Gynaecol. 1976;83(2):132\u20136.  77.  McPartlin J, Halligan A, Scott J, Darling M, Weir D. Accelerated folate breakdown in pregnancy. Lancet. 1993;341(8838):148\u20139.  78.  Higgins JR, Quinlivan EP, McPartlin J, Scott JM, Weir DG, Darling MRN. The relationship between increased folate catabolism and the increased requirement for folate in pregnancy. Br J Obstet Gynaecol. 2000;107(9):1149\u201354.  79.  Ueland PM, Ulvik A, Rios-Avila L, Midttun \u00d8, Gregory JF. Direct and functional biomarkers of vitamin B6 status. Annu Rev Nutr. 2015;35(1):33\u201370.  80.  Barnard HC, de Kock JJ, Vermaak WJ, Potgieter GM. A new perspective in the assessment of vitamin B-6 nutritional status during pregnancy in humans. J Nutr. 1987;117(7):1303\u20136.    186 81.  Lamers Y, Coats B, Ralat M, Quinlivan E, Stacpoole P, Gregory JF 3rd. Moderate vitamin B-6 restriction does not alter postprandial methionine cycle rates of remethylation, transmethylation, and total transsulfuration but increases the fractional synthesis rate of cystathionine in healthy young men and women. J Nutr. 2011;141(5):835\u201342.  82.  Ho CL, Quay TA, Devlin AM, Lamers Y. Prevalence and predictors of low vitamin B6 status in healthy young adult women in Metro Vancouver. Nutrients. 2016;8(9):1\u201311.  83.  Tucker KL, Selhub J, Wilson PW, Rosenberg IH. Human and clinical nutrition dietary intake pattern relates to plasma folate and homocysteine concentrations in the Framingham heart study. J Nutr. 1996;3025\u201331.  84.  Molloy AM, Mills JL, Cox C, Daly SF, Conley M, Brody L, et al. Choline and homocysteine interrelations in umbilical cord and maternal plasma at delivery. Am J Clin Nutr. 2005;82(4):836\u201342.  85.  Fischer LM, Da Costa KA, Kwock L, Galanko J, Zeisel SH. Dietary choline requirements of women: Effects of estrogen and genetic variation. Am J Clin Nutr. 2010;92(5):1113\u20139.  86.  Polese B, Gridelet V, Araklioti E, Martens H, Perrier d\u2019Hauterive S, Geenen V. The endocrine milieu and CD4 T-lymphocyte polarization during pregnancy. Front Endocrinol (Lausanne). 2014;5:1\u201311.  87.  Yan J, Jiang X, West AA, Perry CA, Malysheva O V, Brenna JT, et al. Pregnancy alters choline dynamics\u202f: results of a randomized trial using stable isotope methodology in pregnant and nonpregnant women. Am J Clin Nutr. 2013;98(6):1459\u201367.  88.  Kim MW, Hong S-C, Choi JS, Han J-Y, Oh M-J, Kim HJ, et al. Homocysteine, folate and pregnancy outcomes. J Obs Gynaecol. 2012;32:520\u20134.  89.  Hogeveen M, Blom HJ, Den Heijer M. Maternal homocysteine and small-for-gestational-age offspring: Systematic review and meta-analysis. Am J Clin Nutr. 2012;95(1):130\u20136.    187 90.  Vollset SE, Refsum H, Irgens LM, Emblem BM, Tverdal A, Gjessing HK, et al. Plasma total homocysteine , pregnancy complications , and adverse pregnancy outcomes\u202f: the Hordaland homocysteine study. Am J Clin Nutr. 2000;71(4):962\u20138.  91.  Kobus-Bianchini K, Bourckhardt GF, Ammar D, Nazari EM, Muller YMR. Homocysteine-induced changes in cell proliferation and differentiation in the chick embryo spinal cord: implications for mechanisms of neural tube defects (NTD). Reprod Toxicol. 2017;69:167\u201373.  92.  Sayyah-Melli M, Ghorbanihaghjo A, Alizadeh M, Kazemi-Shishvan M, Ghojazadeh M, Bidadi S. The effect of high dose folic acid throughout pregnancy on homocysteine (Hcy) concentration and pre-eclampsia: A Randomized Clinical Trial. PLoS One. 2016;11(5):e0154400.  93.  Holm PI, Ueland PM, Vollset SE, Midttun \u00d8, Blom HJ, Keijzer MBAJ, et al. Betaine and folate status as cooperative determinants of plasma homocysteine in humans. Arterioscler Thromb Vasc Biol. 2005;25(2):379\u201385.  94.  Bahous RH, Jadavji NM, Deng L, Cos\u00edn-Tom\u00e1s M, Lu J, Malysheva O, et al. High dietary folate in pregnant mice leads to pseudo-MTHFR deficiency and altered methyl metabolism, with embryonic growth delay and short-term memory impairment in offspring. Hum Mol Genet. 2017;26(5):888\u2013900.  95.  Holm PI, Bleie \u00d8, Ueland PM, Lien EA, Refsum H, Nordrehaug JE, et al. Betaine as a determinant of postmethionine load total plasma homocysteine before and after B-vitamin supplementation. Arterioscler Thromb Vasc Biol. 2004;24(2):301\u20137.  96.  Chmurzynska A, Malinowska AM. Homocysteine homeostasis in the rat is maintained by compensatory changes in cystathionine \u03b2-synthase, betaine-homocysteine methyltransferase, and phosphatidylethanolamine n-methyltransferase gene transcription   188 occurring in response to maternal protein an. Nutr Res. 2011;31(7):572\u20138.  97.  Colomina JM, Cavalle-Busquets P, Fernandez-Roig S, Sole-Navais P, Fernandez-Ballart JD, Ballesteros M, et al. Maternal folate status and the BHMT c.716g>A polymorphism affect the betaine dimethylglycine pathway during pregnancy. Nutrients. 2016;8(10):E621.  98.  Melse-Boonstra A, Holm PI, Ueland PM, Olthof M, Clarke R, Verhoef P. Betaine concentration as a determinant of fasting total homocysteine concentrations and the effect of folic acid supplementation on betaine concentrations. Am J Clin Nutr. 2005;81(6):1378\u201382.  99.  Statistics Canada. 2019. Immigration and ethnocultural diversity: Key results from the 2016 Census. Statistics Canada catalogue no. 11-001-X.Ottawa, Ontario. Last updated October, 25 2017. [Internet]. [cited 2019 Jul 2]. Available from: https:\/\/www150.statcan.gc.ca\/n1\/daily-quotidien\/171025\/dq171025b-eng.htm 100.  Jeruszka-Bielak M, Isman C, Schroder TH, Li W, Green TJ, Lamers Y. South Asian ethnicity is related to the highest risk of vitamin B12 deficiency in pregnant Canadian women. Nutrients. 2017;9(4):1\u201313.  101.  MacFarlane AJ, Greene-Finestone LS, Shi Y. Vitamin B-12 and homocysteine status in a folate-replete population: results from the Canadian Health Measures Survey. Am J Clin Nutr. 2011;94(4):1079\u201387.  102.  Anas MK, Lee MB, Zhou C, Hammer MA, Slow S, Karmouch J, et al. SIT1 is a betaine\/proline transporter that is activated in mouse eggs after fertilization and functions until the 2-cell stage. Development. 2008;135(24):4123\u201330.  103.  Anas MK, Hammer MA, Lever M, Staton JA, Baltz J. The organic osmolytes betaine and proline are transported by a shared system in early preimplantation mouse embryos. J Cell Physiol. 2007;210(1):210\u201377.    189 104.  Zhao R, Russell RG, Wang Y, Liu L, Gao F, Kneitz B, et al. 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. 2001;276(13):10224\u20138.  105.  Craciunescu CN, Brown EC, Zeisel SH, Albright CD, Mar M-H, Nadeau MR. Folic acid deficiency during late gestation decreases progenitor cell proliferation and increases apoptosis in fetal mouse brain. J Nutr. 2004;134(1):162\u20136.  106.  Rosario FJ, Nathanielsz PW, Powell TL, Jansson T. Maternal folate deficiency causes inhibition of mTOR signaling, down-regulation of placental amino acid transporters and fetal growth restriction in mice. Sci Rep. 2017;7(1):1\u201315.  107.  Lambrot R, Xu C, Saint-Phar S, Chountalos G, Cohen T, Paquet M, et al. Low paternal dietary folate alters the mouse sperm epigenome and is associated with negative pregnancy outcomes. Nat Commun. 2013;4(2889):doi: 10.1038\/ncomms3889.  108.  Daly LE, Kirke PN, Molloy A, Weir DG, Scott JM. Folate levels and neural tube defects: implications for prevention. JAMA. 1995;274(21):1698\u2013702.  109.  Craciunescu CN, Albright CD, Mar M, Song J, Zeisel SH. Choline availability during embryonic development alters progenitor cell mitosis in developing mouse hippocampus. J Nutr. 2003;133(11):3614\u20138.  110.  Chan J, Deng L, Mikael L, Yan J, Pickell L, Wu Q, et al. Low dietary choline and low dietary riboflavin during pregnancy influence reproductive outcomes and heart development in mice. Am J Clin Nutr. 2010;91(4):1035\u201343.  111.  Kwan ST (Cecilia), King JH, Yan J, Jiang X, Wei E, Fomin VG, et al. Maternal choline supplementation during murine pregnancy modulates placental markers of inflammation, apoptosis and vascularization in a fetal sex-dependent manner. Placenta. 2017;53(2017):57\u201365.    190 112.  Kwan ST (Cecilia), King JH, Yan J, Wang Z, Jiang X, Hutzler JS, et al. Maternal choline supplementation modulates placental nutrient transport and metabolism in late gestation of mouse pregnancy. J Nutr. 2017;147(11):2083\u201392.  113.  Shaw GM, Finnell R, Blom H, Carmichael S, Vollset S, Yang W, et al. Choline and risk of neural tube defects in a folate-fortified population. Epidemiology. 2009;20(5):714\u20139.  114.  Smedts HPM, Rakhshandehroo M, Verkleij-Hagoort AC, De Vries JHM, Ottenkamp J, Steegers EAP, et al. Maternal intake of fat, riboflavin and nicotinamide and the risk of having offspring with congenital heart defects. Eur J Nutr. 2008;47(7):357\u201365.  115.  Wallace DC. Mitochondrial DNA mutations in diseases of energy metabolism. J Bioenerg Biomembr. 1994;26(3):241\u201350.  116.  Larsson NG, Olfors A. Mitochondrial myopathies. Acta Physiol Scand. 2001;171(3):385\u201393.  117.  Azzi S, Sas TCJ, Koudou Y, Le Bouc Y, Souberbielle JC, Dargent-Molina P, et al. Degree of methylation of ZAC1 (PLAGL1) is associated with prenatal and post-natal growth in healthy infants of the EDEN mother child cohort. Epigenetics. 2013;9(3):338\u201345.  118.  Varrault A, Gueydan C, Delalbre A, Bellmann A, Houssami S, Aknin C, et al. Zac1 regulates an imprinted gene network critically involved in the control of embryonic growth. Dev Cell. 2006;11(5):711\u201322.  119.  Valente T, Junyent F, Auladell C. Zac1 is expressed in progenitor\/stem cells of the neuroectoderm and mesoderm during embryogenesis: Differential phenotype of the zac1-expressing cells during development. Dev Dyn. 2005;233(2):667\u201379.  120.  Menezo Y, Khatchadourian C, Gharib A, Hamidi J, Greenland T, Sarda N. Regulation of S-adenosyl methionine synthesis in the mouse embryo. Life Sci. 1989;44(21):1601\u20139.  121.  Smith BT, Mussell JC, Fleming PA, Barth JL, Spyropoulos DD, Cooley MA, et al.   191 Targeted disruption of cubilin reveals essential developmental roles in the structure and function of endoderm and in somite formation. BMC Dev Biol. 2006;6:30.  122.  Mani C, Kochhar P, Ravikumar G, Dwarkanath P, Sheela CN, George S, et al. Placental expression of ENG, VEGF, and FLT: Gender-specific associations with maternal vitamin B12 status. Eur J Clin Nutr. 2020;74(1):176\u201382.  123.  Aboelenain M, Balboula A, Kawahara M, El-Monem Montaser A, Zaabel S, Kim S, et al. Pyridoxine supplementation during oocyte maturation improves the development and quality of bovine preimplantation embryos. Theriogenology. 2017;91:127\u201333.  124.  Hendrina DB, Harding J. The developmental origins of adult disease (Barker) hypothesis. Aust New Zeal J Obstet Gynaecol. 2006;46(1):4\u201314.  125.  Barker DJ. The intrauterine origins of cardiovascular and obstructive lung disease in adult life. The Marc Daniels Lecture 1990. J R Coll Physicians Lond. 1991;25(2):129\u201333.  126.  Barker DJP, Osmond C. Infant mortality, childhood nutrition, and ischaemic heart disease in England and Wales. Lancet. 1986;327(8489):1077\u201381.  127.  Veenendaal MVE, Painter RC, De Rooij SR, Bossuyt PMM, Van Der Post JAM, Gluckman PD, et al. Transgenerational effects of prenatal exposure to the 1944-45 Dutch famine. BJOG. 2013;120(5):548\u201354.  128.  Painter RC, Osmond C, Gluckman P, Hanson M, Phillips DIW, Roseboom TJ. Transgenerational effects of prenatal exposure to the Dutch famine on neonatal adiposity and health in later life. BJOG. 2008;115(10):1243\u20139.  129.  Gomez-Verjan JC, Barrera-V\u00e1zquez OS, Garc\u00eda-Vel\u00e1zquez L, Samper-Ternent R, Arroyo P. Epigenetic variations due to nutritional status in early-life and its later impact on aging and disease. Clin Genet. 2020;1\u20139.  130.  Li E. Chromatin modification and epigenetic reprogramming in mammalian development.   192 Nat Rev Genet. 2002;3(9):662\u201373.  131.  Heijmans BT, Tobi EW, Stein AD, Putter H, Blauw GJ, Susser ES, et al. Persistent epigenetic differences associated with prenatal exposure to famine in humans. Proc Natl Acad Sci U S A. 2008;105(44):17046\u20139.  132.  Chen Z, Riggs AD. DNA methylation and demethylation in mammals. J Biol Chem. 2011;286(21):18347\u201353.  133.  Messerschmidt DM, Knowles BB, Solter D. DNA methylation dynamics during epigenetic reprogramming in the germline and preimplantation embryos. Genes Dev. 2014;28(8):812\u201328.  134.  Shen L, Inoue A, He J, Liu Y, Lu F, Zhang Y. Tet3 and DNA replication mediate demethylation of both the maternal and paternal genomes in mouse zygotes. Cell Stem Cell. 2014;15(4):459\u201371.  135.  Guo F, Li X, Liang D, Li T, Zhu P, Guo H, et al. Active and passive demethylation of male and female pronuclear DNA in the mammalian zygote. Cell Stem Cell. 2014;15(4):447\u201359.  136.  Wu X, Zhang Y. TET-mediated active DNA demethylation: mechanism, function and beyond. Nat Rev Genet. 2017;18(9):517\u201334.  137.  Kohli RM, Zhang Y. TET enzymes, TDG and the dynamics of DNA demethylation. Nature. 2013;502(7472):472\u20139.  138.  Nakamura T, Arai Y, Umehara H, Masuhara M, Kimura T, Taniguchi H, et al. PGC7\/Stella protects against DNA demethylation in early embryogenesis. Nat Cell Biol. 2007;9(1):64\u201371.  139.  Senner CE. The role of DNA methylation in mammalian development. Reprod Biomed Online. 2011;22(6):529\u201335.    193 140.  Bourc\u2019his D, Xu G, Lin C, Bollman B, Bestor TH. Dnmt3L and the establishment of maternal genomic imprints. Science (80- ). 2001;294(5551):2536\u20139.  141.  Howell CY, Bestor TH, Ding F, Latham KE, Mertineit C, Trasler JM, et al. Genomic imprinting disrupted by a maternal effect mutation in the Dnmt1 gene. Cell. 2001;104(6):829\u201338.  142.  Pauwels S, Duca R, Devlieger R, Freson K, Straetmans D, Van Herck E, et al. Maternal methyl-group donor intake and global DNA (Hydroxy)methylation before and during pregnancy. Nutrients. 2016;8(8):474.  143.  Sharma S, de Carvalho DD, Jeong S, Jones PA, Liang G. Nucleosomes containing methylated DNA stabilize DNA methyltransferases 3A\/3B and ensure faithful epigenetic inheritance. PLoS Genet. 2011;7(2).  144.  Novakovic B, Wong NC, Sibson M, Ng HK, Morley R, Manuelpillai U, et al. DNA methylation-mediated down-regulation of DNA methyltransferase-1 (DNMT1) is coincident with, but not essential for, global hypomethylation in human placenta. J Biol Chem. 2010;285(13):9583\u201393.  145.  Cooney CA, Dave AA, Wolff GL. Maternal methyl supplements in mice affect epigenetic variation and DNA methylation of offspring. J Nutr. 2002;132(8):2393S\u20132400S.  146.  Sinclair KD, Allegrucci C, Singh R, Gardner DS, Sebastian S, Bispham J, et al. DNA methylation, insulin resistance, and blood pressure in offspring determined by maternal periconceptional B vitamin and methionine status. Proc Natl Acad Sci U S A. 2007;104(49):19351\u20136.  147.  Kumar K, Lalitha A, Pavithra D, Padmavathi I, Ganeshan M, Rao K, et al. Maternal dietary folate and\/or vitamin B12 restrictions alter body composition (adiposity) and lipid metabolism in Wistar rat offspring. J Nutr Biochem. 2013;24(1):25\u201331.    194 148.  Tobi EW, Lumey LH, Talens RP, Kremer D, Putter H, Stein AD, et al. DNA methylation differences after exposure to prenatal famine are common and timing- and sex-specific. Hum Mol Genet. 2009;18(21):4046\u201353.  149.  Aleliunas R, Aljaadi A, Laher I, Glier M, Green T, Murphy M, et al. Folic acid supplementation of female mice, with or without vitamin B-12, before and during pregnancy and lactation programs adiposity and vascular health in adult male offspring. J Nutr. 2016;146(4):688\u201396.  150.  Keating E, Correia-Branco A, Araujo JR, Meireles M, Fernandes R, Guardao L, et al. Excess perigestational folic acid exposure induces metabolic dysfunction in post-natal life. J Endocrinol. 2015;224(3):245\u201359.  151.  Godfrey KM, Sheppard A, Gluckman PD, Lillycrop KA, Burdge GC, McLean C, et al. Epigenetic gene promoter methylation at birth is associated with child\u2019s later adiposity. Diabetes. 2011;60(5):1528\u201334.  152.  Sugden M, Holness M. Role of nuclear receptors in the modulation of insulin secretion in lipid-induced insulin resistance. Biochem Soc Trans. 2008;36(5):891\u2013900.  153.  Alvarez R, Checa M, Brun S, Vi\u00f1as O, Mampel T, Iglesias R, et al. Both retinoic-acid-receptor- and retinoid-X-receptor-dependent signalling pathways mediate the induction of the brown-adipose-tissue-uncoupling-protein-1 gene by retinoids. Biochem J. 2000;345(1):91\u20137.  154.  Wasim M, Awan FR, Najam SS, Khan AR, Khan HN. Role of leptin deficiency, inefficiency, and leptin receptors in obesity. Biochem Genet. 2016;54(5):565\u201372.  155.  Farr OM, Gavrieli A, Mantzoros CS. Leptin applications in 2015: what have we learned about leptin and obesity? Curr Opin Endocrinol Diabetes Obes. 2015;22(5):353\u20139.  156.  Bouret SG, Draper SJ, Simerly RB. Trophic action of leptin on hypothalamic neurons that   195 regulate feeding. Science (80- ). 2004;304(5667):108\u201310.  157.  Jequier E. Leptin Signaling, Adiposity, and Energy Balance. Ann N Y Acad Sci. 2006;967(1):379\u201388.  158.  Morash B, Li A, Murphy P, Wilkinson M, Ehud U. Leptin gene expression in the brain and pituitary gland. Endocrinology. 1999;140(12):5995\u20138.  159.  Steinberg G, Parolin ML, Heigenhauser G, Dyck DJ. Leptin increases FA oxidation in lean but not obese human skeletal muscle: evidence of peripheral leptin resistance. Am J Physiol Endocrinol Metab. 2002;283(1):E187-192.  160.  Gordan JD, Simon MC. Hypoxia-inducible factors: central regulators of the tumor phenotype. Curr Opin Genet Dev. 2007;17(1):71\u20137.  161.  Pfeiffer S, Kr\u00fcger J, Maierhofer A, B\u00f6ttcher Y, Kl\u00f6ting N, El Hajj N, et al. Hypoxia-inducible factor 3A gene expression and methylation in adipose tissue is related to adipose tissue dysfunction. Sci Rep. 2016;6(27969):doi: 10.1038\/srep27969.  162.  Tuomilehto H, Sepp\u00e4 J, Uusitupa M. Obesity and obstructive sleep apnea - Clinical significance of weight loss. Sleep Med Rev. 2013;17(5):321\u20139.  163.  Zhang H, Zhang G, Gonzalez FJ, Park S min, Cai D. Hypoxia-inducible factor directs POMC gene to mediate hypothalamic glucose sensing and energy balance regulation. PLoS Biol. 2011;9(7):1\u201316.  164.  Dick KJ, Nelson CP, Tsaprouni L, Sandling JK, A\u00efssi D, Wahl S, et al. DNA methylation and body-mass index: A genome-wide analysis. Lancet. 2014;383(9933):1990\u20138.  165.  Huang T, Zheng Y, Qi Q, Xu M, Ley SH, Li Y, et al. DNA Methylation variants at HIF3A locus, B-vitamin intake, and long-term weight change: gene-diet interactions in two U.S. cohorts. Diabetes. 2015;64(9):3146\u201354.  166.  Mansell T, Ponsonby AL, Januar V, Novakovic B, Collier F, Burgner D, et al. Early-life   196 determinants of hypoxia-inducible factor 3A gene (HIF3A) methylation: A birth cohort study. Clin Epigenetics. 2019;11(1):1\u201312.  167.  Wilcox AJ, Russell IT. Birthweight and perinatal mortality: I. On the frequency distribution of birthweight. Int J Epidemiol. 1983;12(3):314\u20138.  168.  Morris SS, Victora CG, Barros FC, Halpern R, Menezes AMB, C\u00e9sar JA, et al. Length and ponderal index at birth: Associations with mortality, hospitalizations, development and post natal growth in Brazilian infants. Int J Epidemiol. 1998;27(2):242\u20137.  169.  Kajantie E, Osmond C, Barker DJP, Fors\u00e9n T, Phillips DIW, Eriksson JG. Size at birth as a predictor of mortality in adulthood: A follow-up of 350 000 person-years. Int J Epidemiol. 2005;34(3):655\u201363.  170.  Parsons TJ, Power C, Manor O. Fetal and early life growth and body mass index from birth to early adulthood in 1958 British cohort: Longitudinal study. BMJ. 2001;323(7325):1331\u20135.  171.  S\u00f8rensen HT, Sabroe S, Rothman KJ, Gillman M, Fischer P, S\u00f8rensen TI. Relation between weight and length at birth and body mass index in young adulthood: cohort study. BMJ. 1997;315(7116):1137.  172.  Pietilainen KH, Kaprio J, Rasanen M, Winter T, Rissanen A, Rose R. Tracking of body size from birth to late adolescence: Contributions of birth length, birth weight, duration of gestation, parents\u2019 body size, and twinship. Am J Epidemiol. 2001;154(1):21\u20139.  173.  Yu ZB, Han SP, Zhu GZ, Zhu C, Wang XJ, Cao XG, et al. Birth weight and subsequent risk of obesity: A systematic review and meta-analysis. Obes Rev. 2011;12(7):525\u201342.  174.  Roy SM, Spivack JG, Faith MS, Chesi A, Mitchell JA, Kelly A, et al. Infant BMI or weight-for-length and obesity risk in early childhood. Pediatrics. 2016;137(5):e20153492.  175.  Ara\u00fajo CL, Hallal PC, Nader GA, Neutzling MB, deF\u00e1tima Vieira M, Menezes AMB, et   197 al. Effect of birth size and proportionality on BMI and skinfold thickness in early adolescence: Prospective birth cohort study. Eur J Clin Nutr. 2009;63(5):634\u20139.  176.  Maunu J, Parkkola R, Rikalainen H, Lehtonen L, Haataja L, Lapinleimu H. Brain and ventricles in very low birth weight infants at term: A comparison among head circumference, ultrasound, and magnetic resonance imaging. Pediatrics. 2009;123(2):617\u201326.  177.  Bartholomeusz H, Courchesne E, Karns C. Relationship between head circumference and brain volume in healthy normal toddlers, children, and adults. Neuropediatrics. 2002;33(5):239\u201341.  178.  Deloison B, Chalouhi GE, Bernard JP, Ville Y, Salomon LJ. Outcomes of fetuses with small head circumference on second-trimester ultrasonography. Prenat Diagn. 2012;32(9):869\u201374.  179.  Eriksson JG, Kajantie E, Lampl M, Osmond C, Barker DJP. Small head circumference at birth and early age at adiposity rebound. Acta Physiol. 2014;210(1):154\u201360.  180.  Cole TJ. Children grow and horses race: Is the adiposity rebound a critical period for later obesity? BMC Pediatr. 2004;4(6).  181.  Whitaker RC, Pepe MS, Wright JA, Seidel KD, Dietz WH. Early adiposity rebound and the risk of adult obesity. Pediatrics. 1998;101(3):E5.  182.  Villar J, Ismail LC, Victora CG, Ohuma EO, Bertino E, Altman DG, et al. International standards for newborn weight, length, and head circumference by gestational age and sex: The Newborn Cross-Sectional Study of the INTERGROWTH-21st Project. Lancet. 2014;384(9946):857\u201368.  183.  Villar J, Puglia FA, Fenton TR, Cheikh Ismail L, Staines-Urias E, Giuliani F, et al. Body composition at birth and its relationship with neonatal anthropometric ratios: The newborn   198 body composition study of the INTERGROWTH-21 st project. Pediatr Res. 2017;82(2):305\u201316.  184.  Du YF, Wei Y, Yang J, Cheng ZY, Zuo XF, Wu TC, et al. Maternal betaine status, but not that of choline or methionine, is inversely associated with infant birth weight. Br J Nutr. 2019;121(11):1279\u201386.  185.  Zhao N, Yang S, Hu Y, Dong H, Zhao R. Maternal betaine supplementation in rats induces intergenerational changes in hepatic IGF-1 expression and DNA methylation. Mol Nutr Food Res. 2017;61(8):DOI 10.1002\/mnfr.201600940.  186.  Xing J, Kang L, Jiang Y. Effect of dietary betaine supplementation on lipogenesis gene expression and CpG methylation of lipoprotein lipase gene in broilers. Mol Biol Rep. 2011;38(3):1975\u201381.  187.  Huang QC, Xu ZR, Han XY, Li WF. Effect of dietary betaine supplementation on lipogenic enzyme activities and fatty acid synthase mRNA expression in finishing pigs. Anim Feed Sci Technol. 2008;140(3\u20134):365\u201375.  188.  Yan J, Winter LB, Burns-Whitmore B, Vermeylen F, Caudill MA. Plasma choline metabolites associate with metabolic stress among young overweight men in a genotype-specific manner. Nutr Diabetes. 2012;2(10):e49.  189.  Buchman AL, Sohel M, Moukarzel A, Bryant D, Schanler R, Awal M, et al. Plasma choline in normal newborns, infants, toddlers, and in very-low-birth-weight neonates requiring total parenteral nutrition. Nutrition. 2001;17(1):18\u201321.  190.  Van Lee L, Crozier SR, Aris IM, Tint MT, Sadananthan SA, Michael N, et al. Prospective associations of maternal choline status with offspring body composition in the first 5 years of life in two large mother-offspring cohorts: The Southampton Women\u2019s Survey cohort and the Growing up in Singapore towards healthy Outcomes cohort. Int J Epidemiol.   199 2019;48(2):433\u201344.  191.  Niculescu MD, Craciunescu CN, Zeisel SH. Dietary choline deficiency alters global and gene-specific DNA methylation in the developing hippocampus of mouse fetal brains. FASEB J. 2006;20(1):43\u20139.  192.  Boeke CE, Baccarelli A, Kleinman KP, Burris HH, Litonjua AA, Rifas-Shiman SL, et al. Gestational intake of methyl donors and global LINE-1 DNA methylation in maternal and cord blood: Prospective results from a folate-replete population. Epigenetics. 2012;7(3):253\u201360.  193.  de Weerd S, Steegers-Theunissen RP, de Boo TM, Thomas CMG, Steegers EA. Maternal periconceptional biochemical and hematological parameters, vitamin profiles and pregnancy outcome. Eur J Clin Nutr. 2003;57(9):1128\u201334.  194.  Guerra-Shinohara EM, Morita OE, Peres S, Pagliusi RA, Neto LFS, Irazusta SP, et al. Low ratio of S-adenosylmethionine to S-adenosylhomocysteine is associated with vitamin deficiency in Brazilian pregnant women and newborns. Am J Clin Nutr. 2004;80(5):1312\u201321.  195.  Frery N, Huel G, Leroy M, Moreau T, Savard R, Blot P, et al. Vitamin B12 among parturients and their newborns and its relationship with birthweight. Eur J Obstet Gynecol Reprod Biol. 1992;45(3):155\u201363.  196.  Cawley S, O\u2019Malley E, Kennedy R, Reynolds C, Molloy A, Turner M. The relationship between maternal plasma homocysteine in early pregnancy and birth weight. J Matern Neonatal Med. 2020;33(18):3045\u20139.  197.  Gadgil M, Joshi K, Pandit A, Otiv S, Joshi R, Brenna JT, et al. Imbalance of folic acid and vitamin B12 is associated with birth outcome: An Indian pregnant women study. Eur J Clin Nutr. 2014;68(6):726\u20139.    200 198.  Van Uitert EM, Van Ginkel S, Willemsen SP, Lindemans J, Koning AHJ, Eilers PHC, et al. An optimal periconception maternal folate status for embryonic size: The Rotterdam Predict study. BJOG. 2014;121(7):821\u20139.  199.  Koning I V, Groenenberg IAL, Gotink AW, Willemsen SP, Gijtenbeek M, Dudink J, et al. Periconception maternal folate status and human embryonic cerebellum growth trajectories: The Rotterdam predict study. PLoS One. 2015;10(10):1\u201313.  200.  Joubert BR, den Dekker HT, Felix JF, Bohlin J, Ligthart S, Beckett E, et al. Maternal plasma folate impacts differential DNA methylation in an epigenome-wide meta-analysis of newborns. Nat Commun. 2016;7(10577):doi: 10.1038\/ncomms10577.  201.  Steenweg-de Graaff J, Roza SJ, Walstra AN, El Marroun H, Steegers EAP, Jaddoe VWV, et al. Associations of maternal folic acid supplementation and folate concentrations during pregnancy with foetal and child head growth: the Generation R Study. Eur J Nutr. 2017;56(1):65\u201375.  202.  Van Mil NH, Bouwl-Both MI, Stolk L, Verbiest MMPJ, Hofman A, Jaddoe VWV, et al. Determinants of maternal pregnancy one-carbon metabolism and newborn human DNA methylation profiles. Reproduction. 2014;148(6):581\u201392.  203.  Haggarty P, Hoad G, Campbell DM, Horgan GW, Piyathilake C, McNeill G. Folate in pregnancy and imprinted gene and repeat element methylation in the offspring. Am J Clin Nutr. 2013;97(1):94\u20139.  204.  World Health Organization. Periconceptional folic acid supplementation to prevent neural tube defects [Internet]. Geneva, Switzerland: WHO\u2014e-Library of Evidence for Nutrition Actions (eLENA). 2017 [cited 2020 May 13]. Available from: http:\/\/www.who.int\/elena\/titles\/folate_periconceptional\/en\/ 205.  Centers for Disease Control (CDC). Use of folic acid for prevention of spina bifida and   201 other neural tube defects\u20141983\u20131991. MMWR Morb Mortal Wkly Rep. 1991;40:513\u201316.  206.  Chalmers B, Dzakpasu S, Heaman M, Kaczorowski J. The Canadian Maternity Experiences Survey: An Overview of Findings. J Obstet Gynaecol Canada. 2009;31(4):303.  207.  Public Health Agency of Canada. Folic acid use among women in Canada. Results from the Canadian Maternity Experiences Survey 2006-2007. :1\u20132.  208.  Health Canada. Licensed Natural Health Products Database (LNHPD). [Internet]. Ottawa, Canada: Government of Canada; 2015, [cited 2018 Jun 19]. http:\/\/www.hc-sc.gc.ca\/dhp-mps\/prodnatur\/applications\/licen-prod\/lnhpd-bdpsnh-eng.php.  209.  Steegers-Theunissen RP, Obermann-Borst SA, Kremer D, Lindemans J, Siebel C, Steegers EA, et al. Periconceptional maternal folic acid use of 400 \u03bcg per day is related to increased methylation of the IGF2 gene in the very young child. PLoS One. 2009;4(11):1\u20135.  210.  Hoyo C, Murtha AP, Schildkraut JM, Jirtle R, Demark-Wahnefried W, Forman MR, et al. Methylation variation at IGF2 differentially methylated regions and maternal folic acid use before and during pregnancy. Epigenetics. 2011;6(7):928\u201336.  211.  Caffrey A, Irwin RE, McNulty H, Strain JJ, Lees-Murdock DJ, McNulty BA, et al. Gene-specific DNA methylation in newborns in response to folic acid supplementation during the second and third trimesters of pregnancy: Epigenetic analysis from a randomized controlled trial. Am J Clin Nutr. 2018;107(4):566\u201375.  212.  Van Oort FVA, Melse-Boonstra A, Brouwer IA, Clarke R, West CE, Katan MB, et al. Folic acid and reduction of plasma homocysteine concentrations in older adults: A dose-response study. Am J Clin Nutr. 2003;77(5):1318\u201323.  213.  Centers for Disease Control and Prevention (CDC). Second national report on biochemical indicators of diet and nutrition in the U.S. population. Atlanta, GA; 2012.    202 214.  Green R, Miller JW. Vitamin B12 deficiency is the dominant nutritional cause of hyperhomocysteinemia in a folic acid-fortified population. Clin Chem Lab Med. 2005;43(10):1048\u201351.  215.  McKinley MC, McNulty H, McPartlin J, Strain JJ, Pentieva K, Ward M, et al. Low-dose vitamin B-6 effectively lowers fasting plasma homocysteine in healthy elderly persons who are folate and riboflavin replete. Am J Clin Nutr. 2001;73(4):759\u201364.  216.  McKinley MC, McNulty H, McPartlin J, Strain JJ, Scott JM. Effect of riboflavin supplementation on plasma homocysteine in elderly people with low riboflavin status. Eur J Clin Nutr. 2002;56(9):850\u20136.  217.  Jacques PF, Bostom AG, Wilson PWF, Rich S, Rosenberg IH, Selhub J. Determinants of plasma total homocysteine concentration in the Framingham Offspring cohort. Am J Clin Nutr. 2001;73(3):613\u201321.  218.  Jacques PF, Rosenberg IH, Rogers G, Selhub J, Bowman BA, Gunter EW, et al. Serum total homocysteine concentrations in adolescent and adult Americans: Results from the third National Health and Nutrition Examination Survey. Am J Clin Nutr. 1999;69(3):482\u20139.  219.  Bostom A, Culleton B. Hyperhomocysteinemia in Chronic Renal Disease. J Am Soc Nephrol. 1999;10:891\u2013900.  220.  Rogne T, Tielemans MJ, Chong MF, Yajnik CS, Krishnaveni G V, Poston L, et al. Systematic reviews and meta- and pooled analyses associations of maternal vitamin B12 concentration in pregnancy with the risks of preterm birth and low birth weight\u202f: A systematic review and meta-analysis of individual participant data. Am J Epidemiol. 2018;185(3):212\u201323.  221.  Yajnik C, Deshpande S, Panchanadikar A, Naik S, Deshpande J, Coyaji K, et al. Maternal   203 total homocysteine concentration and neonatal size in India. Asia Pac J Clin Nutr. 2005;14(2):179\u201381.  222.  Choi SW, Friso S, Ghandour H, Bagley PJ, Selhub J, Mason JB. Vitamin B-12 deficiency induces anomalies of base substitution and methylation in the DNA of rat colonic epithelium. J Nutr. 2004;134(4):750\u20135.  223.  Kulkarni A, Dangat K, Kale A, Sable P, Chavan-Gautam P, Joshi S. Effects of altered maternal folic acid, vitamin B12 and docosahexaenoic acid on placental global DNA methylation patterns in Wistar rats. PLoS One. 2011;6(3):e17706.  224.  Ba Y, Yu H, Fudong L, Geng X, Cairong Z. Relationship of folate, vitamin B12 and methylation of insulin-like growth factor-II maternal and cord blood. Eur J Clin Nutr. 2011;65(4):480\u20135.  225.  McKay J, Groom A, Potter C, Coneyworth L, Ford D, Mathers J, et al. Genetic and non-genetic influences during pregnancy on infant global and site specific DNA methylation: Role for folate gene variants and vitamin B 12. PLoS One. 2012;7(3):1\u20139.  226.  Lin X, Lim IY, Wu Y, Teh AL, Chen L, Aris IM, et al. Developmental pathways to adiposity begin before birth and are influenced by genotype, prenatal environment and epigenome. BMC Med. 2017;15(1):1\u201318.  227.  Haste FM, Brooke OG, Anderson HR, Bland JM. The effect of nutritional intake on outcome of pregnancy in smokers and non-smokers. Br J Nutr. 1991;65(3):347\u201354.  228.  Badart-Smook A, Houwelingen A, Al M, Kester A, Hornstra G. Fetal growth is associated positively with maternal intake of riboflavin and negatively with maternal intake of linolein acid. J Am Diet Assoc. 1997;97(8):867\u201370.  229.  Sram RJ, Binkova B, Lnenickova Z, Solansky I, Dejmek J. The impact of plasma folate levels of mothers and newborns on intrauterine growth retardation and birth weight. Mutat   204 Res. 2005;591(1\u20132):302\u201310.  230.  Schroder TH, Sinclair G, Mattman A, Jung B, Barr SI, Vallace HD, et al. Pregnant women of South Asian ethnicity in Canada have substantially lower vitamin B12 status compared with pregnant women of European ethnicity. Br J Nutr. 2017;118(6):454\u201362.  231.  Evensen E, Emaus N, Kokkvoll A, Wilsgaard T, Furberg AS, Skeie G. The relation between birthweight, childhood body mass index, and overweight and obesity in late adolescence: A longitudinal cohort study from Norway, the Troms\u00f8 Study, Fit Futures. BMJ Open. 2017;7(6):e015576.  232.  Eriksson J, Forsen T, Tuomilehto J, Osmond C, Barker D. Size at birth, childhood growth and obesity in adult life. Int J Obes. 2001;25(5):735\u201340.  233.  Winter JD, Taylor Y, Mowrer L, Winter KM, Dulin MF. BMI at birth and overweight at age four. Obes Res Clin Pract. 2017;11(2):151\u20137.  234.  Winter JD, Langenberg P, Krugman SD. Newborn adiposity by body mass index predicts childhood overweight. Clin Pediatr (Phila). 2010;49(9):866\u201370.  235.  Finkelstein JD, Martin JJ. Methionine Metabolism in Mammals. J Biol Chem. 1984;259(15):9508\u201313.  236.  Van Wersch JWJ, Janssens Y, Zandvoort JA. Folic acid, vitamin B12, and homocysteine in smoking and non-smoking pregnant women. Eur J Obstet Gynecol Reprod Biol. 2002;103(1):18\u201321.  237.  Nelen WL, Blom HJ, Thomas CM, Steegers EA, Boers GH, Eskes TK. Methylenetetrahydrofolate reductase polymorphism affects the change in homocysteine and folate concentrations resulting from low dose folic acid supplementation in women with unexplained recurrent miscarriages. J Nutr. 1998;128(8):1336\u201341.  238.  Perinatal Services BC (2018) BC Prenatal Genetic Screening Program [Internet]. [cited   205 2018 Oct 4]. Available from: http:\/\/www.perinatalservicesbc.ca\/our-services\/screeningprograms\/prenatal-genetic-screening-program 239.  Perinatal Services BC. British Columbia Perinatal Data Registry. Years Provided: (2015 to 2016). Resource Type: Data Extract. Data Provided on (2018).  240.  World Health Organization. Obesity and overweight [Internet]. [cited 2019 Jan 5]. Available from: http:\/\/www.who.int\/mediacentre\/factsheets\/fs311\/en\/ 241.  Midttun O, Hustad S, Solheim E, Schneede J, Ueland PM. Multianalyte quantification of vitamin B6 and B2 species in the nanomolar range in human plasma by liquid chromatography-tandem mass spectrometry. Clin Chem. 2005;51(7):1206\u201316.  242.  O\u2019Broin S, Kelleher B. Microbiological assay on microtitre plates of folate in serum and red cells. J Clin Pathol. 1992;45(4):344\u20137.  243.  Molloy AM, Scott JM. Microbiological assay for serum, plasma, and red cell folate using cryopreserved, microtiter plate method. Methods Enzym. 1997;281:43\u201353.  244.  Cheung KL, Lafayette RA. Renal physiology of pregnancy. Adv Chronic Kidney Dis. 2013;20(3):209\u201314.  245.  Finkelstein JD, Kyle WE, Harris BJ. Methionine metabolism in mammals. Regulation of homocysteine methyltransferases in rat tissue. Arch Biochem Biophys. 1971;146(1):84\u201392.  246.  Park EI, Garrow TA. Interaction between dietary methionine and methyl donor intake on rat liver betaine-homocysteine methyltransferase gene expression and organization of the human gene. J Biol Chem. 1999;274(12):7816\u201324.  247.  Davison JM, Shiells EA, Philips PR, Undheimer MD. Serial evaluation of vasopressin release and thirst in human pregnancy. J Clin Invest. 1988;81:798\u2013806.  248.  Miyakawa H, Rim S, Handler J, Kwon M. Identification of the second tonicity-responsive   206 enhancer for the betaine transporter (BGT1) gene. Biochim Biophys Acta. 1999;1446:359\u201364.  249.  Dasarathy J, Gruca LL, Bennett C, Parimi PS, Duenas C, Marczewski S, et al. Methionine metabolism in human pregnancy. Am J Clin Nutr. 2010;91(2):357\u201365.  250.  Quay T, Schroder TH, Jeruszka-Bielak M, Li W, Devlin AM, Barr SI, et al. High prevalence of suboptimal vitamin B 12 status in young adult women of South Asian and European ethnicity. Appl Physiol Nutr Metab. 2015;40(12):1279\u201386.  251.  Lesser IA, Gasevic D, Lear SA. The association between acculturation and dietary patterns of South Asian immigrants. PLoS One. 2014;9(2):1\u20136.  252.  Haubrich D, Wang P, Wedeking P. Distribution and metabolism of intravenously administered choline[methyl- 3-H] and synthesis in vivo of acetylcholine in various tissues of guinea pigs. J Pharmacol Exp Ther. 1975;193(1):246\u201355.  253.  Schwab U, T\u00f6rr\u00f6nen A, Meririnne E, Saarinen M, Alfthan G, Aro A, et al. Orally administered betaine has an acute and dose-dependent effect on serum betaine and plasma homocysteine concentrations in healthy humans. J Nutr. 2006;136(1):34\u20138.  254.  Finkelstein JD, Harris BJ, Martin JJ, Kyle WE, Harris BJ, Kyle WE. Regulation of hepatic betaine-homocysteine methyltransferase by dietary betaine. J Nutr. 1983;113(3):519\u201321.  255.  Skiba WE, Taylor MP, Wells MS, Mangum JH, Awad WM. Human hepatic methionine biosynthesis. Purification and characterization of betaine:homocysteine S-methyltransferase. J Biol Chem. 1982;257(24):14944\u20138.  256.  Morin I, Platt R, Weisberg I, Sabbaghian N, Wu Q, Garrow TA, et al. Common variant in betaine-homocysteine methyltransferase (BHMT) and risk for spina bifida. Am J Med Genet. 2003;119A(2):172\u20136.  257.  Selhub J, Pomfret EA, Zeisel SH. Effects of choline deficiency and methotrexate treatment   207 upon liver folate content and distribution. Cancer Res. 1991;51(1):16\u201321.  258.  Varela-Moreiras G, Selhub J, daCosta KA, Zeisel SH. Effect of chronic choline deficiency in rats on liver folate content and distribution. J Nutr Biochem. 1992;3(10):519\u201322.  259.  Kim YI, Miller W, Costa DA, Nadeau M, Smith D, Selhub J, et al. Severe folate deficiency causes secondary depletion of choline and phosphocholine in rat liver. J Nutr. 1994;124(11):2197\u2013203.  260.  Selhub J. Homocysteine metabolism. Annu Rev Nutr. 1999;19:217\u201346.  261.  Holm PI, Ueland PM, Kvalheim G, Lien EA. Determination of choline, betaine, and dimethylglycine in plasma by a high-throughput method based on normal-phase chromatography-tandem mass spectrometry. Clin Chem. 2003;49(2):286\u201394.  262.  Rasmussen K, Moller J. Total homocysteine measurement in clinical practice. Ann Clin Biochem. 2000;37(5):627\u201348.  263.  Zhang B, Denomme MM, White CR, Leung KY, Lee MB, Greene NDE, et al. Both the folate cycle and betaine-homocysteine methyltransferase contribute methyl groups for DNA methylation in mouse blastocysts. FASEB J. 2015;29(3):1069\u201379.  264.  Statistics Canada.2019. 2006 Census Dictionary. Statistics Canada. Catalogue no 98-316-X2016001. Ottawa, Ontario. [Internet]. [cited 2019 Jan 30]. Available from: https:\/\/www150.statcan.gc.ca\/n1\/en\/catalogue\/98-316-X2016001 265.  World Health Organization (WHO). Global nutrition targets 2025: low birth weight policy brief. (WHO\/NMH\/NHD\/14.5). Geneva: World Health Organization;2014.  266.  Perinatal Services BC (2019) Birth Growth Charts [Internet]. [cited 2019 Jul 2]. Available from: http:\/\/www.perinatalservicesbc.ca\/health-professionals\/professional-resources\/health-promo\/birth-growth-charts 267.  Kramer MS, Platt RW, Wen Wu S, Joseph K., Allen A, Abrahamowicz M, et al. A new   208 and improved population-based Canadian reference for birth weight for gestational age. Pediatrics. 2001;108(2):E35.  268.  Kaiser H. The application of electronic computers to factor analysis. Educ Psychol Meas. 1960;20:141\u201351.  269.  S\u00f8rensen TIA, Ajslev TA, \u00c4ngquist L, Morgen CS, Ciuchi IG, Smith GD. Comparison of associations of maternal peri-pregnancy and paternal anthropometrics with child anthropometrics from birth through age 7 y assessed in the Danish National Birth Cohort. Am J Clin Nutr. 2016;104(2):389\u201396.  270.  Papazian T, Tayeh GA, Sibai D, Hout H, Melki I, Khabbaz LR. Impact of maternal body mass index and gestational weight gain on neonatal outcomes among healthy Middle-Eastern females. PLoS One. 2017;12(7):1\u201313.  271.  P\u00f6lzlberger E, Hartmann B, Hafner E, St\u00fcmpflein I, Kirchengast S. Maternal height and pre-pregnancy weight status are associated with fetal growth patterns and newborn size. J Biosoc Sci. 2017;49(3):392\u2013407.  272.  Fraser AM, Brockert JE, Ward RH. Association of young maternal age with adverse reproductive outcomes. N Engl J Med. 1995;332(17):1113\u20138.  273.  Strobino DM, Ensminger ME, Kim YJ, Nanda J. Mechanisms for maternal age differences in birth weight. Am J Epidemiol. 1995;142(5):504\u201314.  274.  Van Eijsden M, Smits LJM, Van Der Wal MF, Bonsel GJ. Association between short interpregnancy intervals and term birth weight: The role of folate depletion. Am J Clin Nutr. 2008;88(1):147\u201353.  275.  Sol\u00e9-Navais P, Salat-Batlle J, Cavall\u00e9-Busquets P, Fernandez-Ballart J, Ueland PM, Ballesteros M, et al. Early pregnancy folate-cobalamin interactions and their effects on cobalamin status and hematologic variables throughout pregnancy. Am J Clin Nutr.   209 2018;107(2):173\u201382.  276.  Nilsen R, Bj\u00f8rke-Monsen A, Midttun O, Nyg\u00e5rd O, Pedersen E, Ulvik A, et al. Maternal tryptophan and kynurenine pathway metabolites and risk of preeclampsia. Obs Gynecol. 2012;119(6):1243\u201350.  277.  Hansen CM, Shultz TD, Kwak H, Memon HS, Leklem JE. Assessment of vitamin B-6 status in young women consuming a controlled diet containing four levels of vitamin B-6 provides an Estimated Average Requirement and Recommended Dietary Allowance. J Nutr. 2001;131(6):1777\u201386.  278.  Berti C, Fekete K, Dullemeijer C, Trovato M, Souverein O, Cavelaars A, et al. Folate intake and markers of folate status in women of reproductive age, pregnant and lactating women: A meta-analysis. J Nutr Metab. 2012;2012:470656.  279.  Guo C, Wei J, Gao W, Pu L, Tang Z, Li L. Plasma riboflavin is a useful marker for studying riboflavin requirement in Chinese male adults. Nutr Res. 2016;36(6):534\u201340.  280.  Haugen M, Brants\u00e6ter AL, Alexander J, Meltzer HM. Dietary supplements contribute substantially to the total nutrient intake in pregnant Norwegian women. Ann Nutr Metab. 2008;52(4):272\u201380.  281.  Han Q, Hoffman RM. Nonradioactive enzymatic assay for plasma and serum vitamin B6. Nat Protoc. 2008;3(12):1815\u20139.  282.  Huang QC, Xu ZR, Han XY, Li WF. Effect of betaine on growth hormone pulsatile secretion and serum metabolites in finishing pigs. J Anim Physiol Anim Nutr (Berl). 2007;91(3\u20134):85\u201390.  283.  Stiles J, Jernigan TL. The basics of brain development. Neuropsychol Rev. 2010;20(4):327\u201348.  284.  Keunen K, Counsell SJ, Benders MJNL. The emergence of functional architecture during   210 early brain development. Neuroimage. 2017;160:2\u201314.  285.  Sharp GC, Lawlor DA, Richmond RC, Fraser A, Simpkin A, Suderman M, et al. Maternal pre-pregnancy BMI and gestational weight gain, offspring DNA methylation and later offspring adiposity: Findings from the Avon Longitudinal Study of Parents and Children. Int J Epidemiol. 2015;44(4):1288\u2013304.  286.  Liu X, Chen Q, Tsai H-J, Wang G, Hong X, Zhou Y, et al. Maternal preconception body mass index and offspring cord blood DNA Methylation: Exploration of early life origins of disease. Env Mol Mutagen. 2014;55(3):223\u201330.  287.  Kaplan BJ, Giesbrecht GF, Leung BMY, Field CJ, Dewey D, Bell RC, et al. The Alberta Pregnancy Outcomes and Nutrition (APrON) cohort study: Rationale and methods. Matern Child Nutr. 2014;10(1):44\u201360.  288.  World Health Organization (2008). Training Course on Child Growth Assessment. Geneva: WHO.  289.  Edgar RD, Jones MJ, Meaney MJ, Turecki G, Kobor MS. BECon: A tool for interpreting DNA methylation findings from blood in the context of brain. Transl Psychiatry. 2017;7(8):e1187.  290.  Pfeiffer CM, Sternberg MR, Fazili Z, Lacher DA, Zhang M, Johnson CL, et al. Folate status and concentrations of serum folate forms in the US population: National Health and Nutrition Examination Survey 2011\u20132. Br J Nutr. 2016;113(12):1965\u201377.  291.  Dedeurwaerder S, Defrance M, Bizet M, Calonne E, Bontempi G, Fuks F. A comprehensive overview of Infinium Human Methylation450 data processing. Brief Bioinform. 2013;15(6):929\u201341.  292.  Du P, Zhang X, Huang CC, Jafari N, Kibbe WA, Hou L, et al. Comparison of Beta-value and M-value methods for quantifying methylation levels by microarray analysis. BMC   211 Bioinformatics. 2010;11.  293.  Cavall\u00e9-Busquets P, Ingl\u00e8s-Puig M, Fernandez-Ballart JD, Haro-Barcelo J, Rojas-Gomez A, Ramos-Rodriguez A, et al. Moderately elevated first trimester fasting plasma total homocysteine is associated with increased probability of miscarriage. The Reus-Tarragona Birth Cohort Study. Biochemie. 2020;6\u201311.  294.  O\u2019Malley EG, Cawley S, Kennedy RAK, Reynolds CME, Molloy A, Turner MJ. Maternal anaemia and folate intake in early pregnancy. J Public Heal. 2018;40(3):e296\u2013302.  295.  Shane B, Tamura T, Stokstad EL. Folate assay: A comparison of radioassay and microbiological methods. Clin Chim Acta. 1980;100:13\u20139.  296.  Mansell T, Ponsonby AL, Collier F, Burgner D, Vuillermin P, Lange K, et al. Genetic variation, intrauterine growth, and adverse pregnancy conditions predict leptin gene DNA methylation in blood at birth and 12 months of age. Int J Obes. 2020;44(1):45\u201356.  297.  Kurdyukov S, Bullock M. DNA methylation analysis: Choosing the right method. Biology (Basel). 2016;5(1):1\u201321.  298.  McKee S, Zhang S, Chen L, Rabinowitz J, Reyes T. Perinatal high fat diet and early life methyl donor supplementation alter one carbon metabolism and DNA methylation in the brain. J Neurochem. 2018;145(5):362\u201373.  299.  Matthews RG, Daubner SC. Modulation of methylenetetrahydrofolate reductase activity by S-adenosylmethionine and by dihydrofolate and its polyglutamate analogues. Adv Enzyme Regul. 1982;20:123\u201331.  300.  Christensen KE, Mikael LG, Leung KY, L\u00e9vesque N, Deng L, Wu Q, et al. High folic acid consumption leads to pseudo-MTHFR deficiency, altered lipid metabolism, and liver injury in mice. Am J Clin Nutr. 2015;101(3):646\u201358.  301.  Waterland RA, Travisano M, Tahiliani KG, Rached MT, Mirza S. Methyl donor   212 supplementation prevents transgenerational amplification of obesity. Int J Obes. 2008;32(9):1373\u20139.  302.  Dunstan J, Bressler JP, Moran TH, Pollak JS, Hirsch AG, Bailey-Davis L, et al. Associations of LEP, CRH, ICAM-1, and LINE-1 methylation, measured in saliva, with waist circumference, body mass index, and percent body fat in mid-childhood. Clin Epigenetics. 2017;9(1):1\u20138.  303.  Huang R-C, Galati JC, Burrows S, Beilin LJ, Li X, Pennell CE, et al. DNA methylation of the IGF2\/H19 imprinting control region and adiposity distribution in young adults. Clin Epigenetics. 2012;4(1):1\u201311.  304.  Human Nutrition and Vitamin Metabolism Research Group [Internet]. [cited 2020 May 19]. Available from: https:\/\/vitamins.landfood.ubc.ca\/lab-services\/ 305.  Lesseur C, Armstrong DA, Paquette AG, Koestler DC, Padbury JF, Marsit CJ. Tissue-specific Leptin promoter DNA methylation is associated with maternal and infant perinatal factors. Mol Cell Endocrinol. 2013;381(1\u20132):160\u20137.  306.  Richmond RC, Sharp GC, Ward ME, Fraser A, Lyttleton O, McArdle WL, et al. DNA methylation and BMI: Investigating identified methylation sites at HIF3A in a causal framework. Diabetes. 2016;65(5):1231\u201344.  307.  Wiedeman AM, Dyer RA, Green TJ, Xu Z, Barr SI, Innis SM, et al. Variations in plasma choline and metabolite concentrations in healthy adults. Clin Biochem. 2018;60:77\u201383.  308.  Visentin CE, Masih S, Plumptre L, Malysheva O, Nielsen DE, Sohn K-JJ, et al. Maternal choline status, but not fetal genotype, influences cord plasma choline metabolite concentrations. J Nutr. 2015;145(7):1491\u20137.  309.  Mooij PNM, Steegers-Theunissen RPM, Thomas CMG, Doesburg WH, Eskes TKAB. Periconceptional vitamin profiles are not suitable for identifying women at risk for neural   213 tube defects. J Nutr. 1993;123(2):197\u2013203.  310.  King JH, Kwan ST, Bae S, Klatt KC, Yan J, Malysheva O V., et al. Maternal choline supplementation alters vitamin B-12 status in human and murine pregnancy. J Nutr Biochem. 2019;72:108210.  311.  Murphy SK, Huang Z, Hoyo C. Differentially methylated regions of imprinted genes in prenatal, perinatal and postnatal human tissues. PLoS One. 2012;7(7):e40924.  312.  Jones MJ, Moore SR, Kobor MS. Principles and challenges of applying epigenetic epidemiology to psychology. Annu Rev Psychol. 2018;69(1):459\u201385.  313.  Islam SA, Goodman SJ, MacIsaac JL, Obradovi\u0107 J, Barr RG, Boyce WT, et al. Integration of DNA methylation patterns and genetic variation in human pediatric tissues help inform EWAS design and interpretation. Epigenetics and Chromatin. 2019;12(1):1\u201318.  314.  Sherwood WB, Bion V, Lockett GA, Ziyab AH, Soto-Ram\u00edrez N, Mukherjee N, et al. Duration of breastfeeding is associated with leptin (LEP) DNA methylation profiles and BMI in 10-year-old children. Clin Epigenetics. 2019;11(1):1\u201310.  315.  Joselit Y, Nanobashvili K, Jack-Roberts C, Greenwald E, Malysheva O V, Caudill MA, et al. Maternal betaine supplementation affects fetal growth and lipid metabolism of high-fat fed mice in a temporal-specific manner. Nutr Diabetes. 2018;8(1):0\u201310.  316.  Oltean S, Banerjee R. A B12-responsive internal ribosome entry site (IRES) element in human methionine synthase. J Biol Chem. 2005;280(38):32662\u20138.  317.  Oltean S, Banerjee R. Nutritional modulation of gene expression and homocysteine utilization by vitamin B12. J Biol Chem. 2003;278(23):20778\u201384.  318.  Kumar KA, Lalitha A, Reddy U, Chandak GR, Sengupta S, Raghunath M. Chronic maternal vitamin B12 restriction induced changes in body composition & glucose metabolism in the Wistar rat offspring are partly correctable by rehabilitation. PLoS One.   214 2014;9(11).  319.  Tanwar VS, Ghosh S, Sati S, Ghose S, Kaur L, Kumar KA, et al. Maternal vitamin B12 deficiency in rats alters DNA methylation in metabolically important genes in their offspring. Mol Cell Biochem. 2020;468(1\u20132):83\u201396.  320.  Liu Y, Michael MD, Kash S, Bensch WR, Monia BP, Murray SF, et al. Deficiency of adiponectin receptor 2 reduces diet-induced insulin resistance but promotes type 2 diabetes. Endocrinology. 2007;148(2):683\u201392.  321.  Thompson MD, Cole DEC, Ray JG. Vitamin B-12 and neural tube defects: The Canadian experience. Am J Clin Nutr. 2009;89(2):697\u2013701.  322.  Jnah A, Trembath A. Fetal and neonatal physiology for the advanced practice nurse. Jnah A, Trembath A, editors. New York: Springer US; 2019.  323.  Hustad S, McKinley MC, McNulty H, Schneede J, Strain JJ, Scott JM, et al. Riboflavin, flavin mononucleotide, and flavin adenine dinucleotide in human plasma and erythrocytes at baseline and after low-dose riboflavin supplementation. Clin Chem. 2002;48(9):1571\u20137.  324.  Moat SJ, Ashfield-Watt PAL, Powers HJ, Newcombe RG, McDowell IFW. Effect of riboflavin status on the homocysteine-lowering effect of folate in relation to the MTHFR (C677T) genotype. Clin Chem. 2003;49(2):295\u2013302.  325.  Frosst P, Blom HJ, Milos R, Goyette P, Sheppard C, Matthews R, et al. A candidate genetic risk factor for vascular disease: a common mutation in methylenetetrahydrofolate reductase. Nat Genet. 1995;10:198\u2013200.  326.  Jacques PF, Kalmbach R, Bagley PJ, Russo GT, Rogers G, Wilson PWF, et al. The relationship between riboflavin and plasma total homocysteine in the Framingham offspring cohort is influenced by folate status and the C677T transition in the methylenetetrahydrofolate reductase gene. J Nutr. 2002;132(2):283\u20138.    215 327.  Canada S. Table 11-10-0239-01 Income of individuals by age group, sex and income source, Canada, provinces and selected census metropolitan areas. 2017.  328.  Jiang R, Jones MJ, Chen E, Neumann SM, Fraser HB, Miller GE, et al. Discordance of DNA methylation variance between two accessible human tissues. Sci Rep. 2015;5:8257.  329.  McKerracher L, Moffat T, Barker M, McConnell M, Atkinson SA, Murray-Davis B, et al. Knowledge about the Developmental Origins of Health and Disease is independently associated with variation in diet quality during pregnancy. Matern Child Nutr. 2020;16(2):1\u201313.  330.  Leunissen RWJ, Kerkhof GF, Stijnen T, Hokken-Koelega A. Timing and tempo of first-year rapid growth in relation to cardiovascular and metabolic risk profile in early adulthood. JAMA. 2009;301(21):2234\u201342.  331.  Soubry A, Murphy SK, Wang F, Huang Z, Vidal AC, Fuemmeler BF, et al. Newborns of obese parents have altered DNA methylation patterns at imprinted genes. Int J Obes. 2015;39(4):650\u20137.  332.  Soubry A, Schildkraut JM, Murtha A, Wang F, Huang Z, Bernal A, et al. Paternal obesity is associated with IGF2 hypomethylation in newborns: Results from a Newborn Epigenetics Study (NEST) cohort. BMC Med. 2013;11(1):29.  333.  Pauwels S, Truijen I, Ghosh M, Duca RC, Langie SAS, Bekaert B, et al. The effect of paternal methyl-group donor intake on offspring DNA methylation and birth weight. J Dev Orig Health Dis. 2017;8(3):311\u201321.  334.  Sahara Y, Matsuzawa D, Ishii D, Fuchida T, Goto T, Sutoh C, et al. Paternal methyl donor deficient diets during development affect male offspring behavior and memory-related gene expression in mice. Dev Psychobiol. 2019;61(1):17\u201328.  335.  Health Canada. Prenatal nutrition guidelines for health professionals: folate[Internet].   216 Ottawa, Canada: Government of Canada; 2009. p. 1\u20138. Report No.: H164\u2013109\/2\u20132009E-PDF. Available from: http:\/\/www.hc-sc.gc.ca\/fn-an\/alt_formats\/hpfb-dgpsa\/pdf\/pubs\/fo.      217 Appendices A 1. Maternal and neonatal information obtained from Prenatal Genetic Screening Program and British Columbia Perinatal Data Registry  1 Lastmenstrual period  Maternal variables Description Source of information Age - Prenatal Genetic Screening Program Ethnicity Self-reported: European or South Asian Prenatal Genetic Screening Program Weeks of gestation at blood collection (weeks) Crown-rump length by ultrasound Prenatal Genetic Screening Program Pre-pregnancy weight (kg) Defined as weight before pregnancy or \u2264 11 weeks British Columbia Perinatal Data Registry Pre-pregnancy height (m) Defined as height before pregnancy British Columbia Perinatal Data Registry Parity  Number of pregnancies that reached 20 weeks gestation or 500 grams birth weight (nulliparous\/multiparous) British Columbia Perinatal Data Registry Delivery type Vaginal birth\/caesarean section British Columbia Perinatal Data Registry Hypertensive disorder of pregnancy  Yes\/No British Columbia Perinatal Data Registry Gestational diabetes  Yes\/No British Columbia Perinatal Data Registry Neighborhood income Median family income of each pregnant woman\u2019s residential Forward Sortation Area (FSA) Canadian Census 2016 Neonatal variables   Gestational age (GA) at birth (weeks) Gestational age at birth, in completed weeks \u2013 calculated by algorithm incorporating LNMP1, first ultrasound, newborn examination, and maternal chart British Columbia Perinatal Data Registry Birth weight (kg) - British Columbia Perinatal Data Registry Birth length (cm) - British Columbia Perinatal Data Registry Head circumference (cm) - British Columbia Perinatal Data Registry Neonatal sex  Biological sex of the newborn British Columbia Perinatal Data Registry ","attrs":{"lang":"en","ns":"http:\/\/www.w3.org\/2009\/08\/skos-reference\/skos.html#note","classmap":"oc:AnnotationContainer"},"iri":"http:\/\/www.w3.org\/2009\/08\/skos-reference\/skos.html#note","explain":"Simple Knowledge Organisation System; Notes are used to provide information relating to SKOS concepts. There is no restriction on the nature of this information, e.g., it could be plain text, hypertext, or an image; it could be a definition, information about the scope of a concept, editorial information, or any other type of information."}],"Genre":[{"label":"Genre","value":"Thesis\/Dissertation","attrs":{"lang":"en","ns":"http:\/\/www.europeana.eu\/schemas\/edm\/hasType","classmap":"dpla:SourceResource","property":"edm:hasType"},"iri":"http:\/\/www.europeana.eu\/schemas\/edm\/hasType","explain":"A Europeana Data Model Property; This property relates a resource with the concepts it belongs to in a suitable type system such as MIME or any thesaurus that captures categories of objects in a given field. It does NOT capture aboutness"}],"GraduationDate":[{"label":"Graduation Date","value":"2020-11","attrs":{"lang":"en","ns":"http:\/\/vivoweb.org\/ontology\/core#dateIssued","classmap":"vivo:DateTimeValue","property":"vivo:dateIssued"},"iri":"http:\/\/vivoweb.org\/ontology\/core#dateIssued","explain":"VIVO-ISF Ontology V1.6 Property; Date Optional Time Value, DateTime+Timezone Preferred "}],"IsShownAt":[{"label":"DOI","value":"10.14288\/1.0394083","attrs":{"lang":"en","ns":"http:\/\/www.europeana.eu\/schemas\/edm\/isShownAt","classmap":"edm:WebResource","property":"edm:isShownAt"},"iri":"http:\/\/www.europeana.eu\/schemas\/edm\/isShownAt","explain":"A Europeana Data Model Property; An unambiguous URL reference to the digital object on the provider\u2019s website in its full information context."}],"Language":[{"label":"Language","value":"eng","attrs":{"lang":"en","ns":"http:\/\/purl.org\/dc\/terms\/language","classmap":"dpla:SourceResource","property":"dcterms:language"},"iri":"http:\/\/purl.org\/dc\/terms\/language","explain":"A Dublin Core Terms Property; A language of the resource.; Recommended best practice is to use a controlled vocabulary such as RFC 4646 [RFC4646]."}],"Program":[{"label":"Program (Theses)","value":"Human Nutrition","attrs":{"lang":"en","ns":"https:\/\/open.library.ubc.ca\/terms#degreeDiscipline","classmap":"oc:ThesisDescription","property":"oc:degreeDiscipline"},"iri":"https:\/\/open.library.ubc.ca\/terms#degreeDiscipline","explain":"UBC Open Collections Metadata Components; Local Field; Indicates the program for which the degree was granted."}],"Provider":[{"label":"Provider","value":"Vancouver : University of British Columbia Library","attrs":{"lang":"en","ns":"http:\/\/www.europeana.eu\/schemas\/edm\/provider","classmap":"ore:Aggregation","property":"edm:provider"},"iri":"http:\/\/www.europeana.eu\/schemas\/edm\/provider","explain":"A Europeana Data Model Property; The name or identifier of the organization who delivers data directly to an aggregation service (e.g. Europeana)"}],"Publisher":[{"label":"Publisher","value":"University of British Columbia","attrs":{"lang":"en","ns":"http:\/\/purl.org\/dc\/terms\/publisher","classmap":"dpla:SourceResource","property":"dcterms:publisher"},"iri":"http:\/\/purl.org\/dc\/terms\/publisher","explain":"A Dublin Core Terms Property; An entity responsible for making the resource available.; Examples of a Publisher include a person, an organization, or a service."}],"Rights":[{"label":"Rights","value":"Attribution-NonCommercial-NoDerivatives 4.0 International","attrs":{"lang":"*","ns":"http:\/\/purl.org\/dc\/terms\/rights","classmap":"edm:WebResource","property":"dcterms:rights"},"iri":"http:\/\/purl.org\/dc\/terms\/rights","explain":"A Dublin Core Terms Property; Information about rights held in and over the resource.; Typically, rights information includes a statement about various property rights associated with the resource, including intellectual property rights."}],"RightsURI":[{"label":"Rights URI","value":"http:\/\/creativecommons.org\/licenses\/by-nc-nd\/4.0\/","attrs":{"lang":"*","ns":"https:\/\/open.library.ubc.ca\/terms#rightsURI","classmap":"oc:PublicationDescription","property":"oc:rightsURI"},"iri":"https:\/\/open.library.ubc.ca\/terms#rightsURI","explain":"UBC Open Collections Metadata Components; Local Field; Indicates the Creative Commons license url."}],"ScholarlyLevel":[{"label":"Scholarly Level","value":"Graduate","attrs":{"lang":"en","ns":"https:\/\/open.library.ubc.ca\/terms#scholarLevel","classmap":"oc:PublicationDescription","property":"oc:scholarLevel"},"iri":"https:\/\/open.library.ubc.ca\/terms#scholarLevel","explain":"UBC Open Collections Metadata Components; Local Field; Identifies the scholarly level of the author(s)\/creator(s)."}],"Title":[{"label":"Title ","value":"Maternal methyl nutrients, obesity programming, and neonatal anthropometric outcomes","attrs":{"lang":"en","ns":"http:\/\/purl.org\/dc\/terms\/title","classmap":"dpla:SourceResource","property":"dcterms:title"},"iri":"http:\/\/purl.org\/dc\/terms\/title","explain":"A Dublin Core Terms Property; The name given to the resource."}],"Type":[{"label":"Type","value":"Text","attrs":{"lang":"en","ns":"http:\/\/purl.org\/dc\/terms\/type","classmap":"dpla:SourceResource","property":"dcterms:type"},"iri":"http:\/\/purl.org\/dc\/terms\/type","explain":"A Dublin Core Terms Property; The nature or genre of the resource.; Recommended best practice is to use a controlled vocabulary such as the DCMI Type Vocabulary [DCMITYPE]. To describe the file format, physical medium, or dimensions of the resource, use the Format element."}],"URI":[{"label":"URI","value":"http:\/\/hdl.handle.net\/2429\/75762","attrs":{"lang":"en","ns":"https:\/\/open.library.ubc.ca\/terms#identifierURI","classmap":"oc:PublicationDescription","property":"oc:identifierURI"},"iri":"https:\/\/open.library.ubc.ca\/terms#identifierURI","explain":"UBC Open Collections Metadata Components; Local Field; Indicates the handle for item record."}],"SortDate":[{"label":"Sort Date","value":"2020-12-31 AD","attrs":{"lang":"en","ns":"http:\/\/purl.org\/dc\/terms\/date","classmap":"oc:InternalResource","property":"dcterms:date"},"iri":"http:\/\/purl.org\/dc\/terms\/date","explain":"A Dublin Core Elements Property; A point or period of time associated with an event in the lifecycle of the resource.; Date may be used to express temporal information at any level of granularity. Recommended best practice is to use an encoding scheme, such as the W3CDTF profile of ISO 8601 [W3CDTF].; A point or period of time associated with an event in the lifecycle of the resource.; Date may be used to express temporal information at any level of granularity. Recommended best practice is to use an encoding scheme, such as the W3CDTF profile of ISO 8601 [W3CDTF]."}]}