MATERNAL DEPRESSION DURING PREGNANCY, METHYL NUTRIENT METABOLISM, AND SEROTONIN TRANSPORTER by Julia Jia Wei B.Sc., The University of British Columbia, 2009 A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF MASTER OF SCIENCE in The Faculty of Graduate Studies (Experimental Medicine) THE UNIVERSITY OF BRITISH COLUMBIA (Vancouver) June 2011 © Julia Jia Wei, 2011 Abstract Depression occurs in 15% of pregnant women and 1/3 are taking selective serotonin reuptake inhibitors (SSRIs) as antidepressants. The neurotransmitter, serotonin, plays a critical role in modulating stress responses and early brain development. Serotonin transporter (SLC6A4) regulates extracellular serotonin levels, and an insertion/deletion variant in the promoter (5HTTLPR) is associated with depression. Maternal mood and SSRIs may program newborns’ behaviour later in life. The underlying molecular mechanism for developmental programming may involve DNA methylation, which requires methyl nutrients as enzymatic cofactors. While low methyl nutrient status (folate and vitamin B12) and a genetic variant in methylenetetrahydrofolate reductase (MTHFR C677T) have been associated with depression in adults, the role of methyl nutrient metabolism in depression during pregnancy remains unclear. Furthermore, little is known about the combined roles of methyl nutrient status and depression in the epigenetic regulation of SLC6A4. The experiments in this thesis explored whether disturbances in methyl nutrient metabolism and depressed mood during the 3rd trimester of pregnancy affect SLC6A4 methylation and expression in mothers and their newborns. Maternal folate status was positively associated with maternal SLC6A4 methylation at CpGs 1, 4, and 8 (P<0.05). Maternal 3rd trimester mood was inversely associated with SLC6A4 CpG 10 methylation in both mothers and newborns (P<0.05). Methylation at SLC6A4 CpG 8 was lower in newborns with mothers with the MTHFR 677TT genotype, and methylation at CpGs 6 and 10 were lower in newborns with the MTHFR 677TT genotype (P<0.05). Maternal SLC6A4 mRNA level was positively associated with mean maternal methylation and methylation at CpGs 5, 7, 8, and 10 (P<0.05). Yet, newborn SLC6A4 mRNA level was negatively associated with newborn methylation at CpGs 4 and 7 ii (P<0.05). Homozygosity for the minor allele for MTHFR C677T and 5-HTTLPR insertion/deletion variants in mothers were not associated with maternal mood (P>0.05). These results provide evidence to suggest that maternal mood during pregnancy and methyl nutrient metabolism may program SLC6A4 gene expression through DNA methylation in both mothers and their newborns. Mood during pregnancy and disturbances in methyl nutrient metabolism could set up life-long health consequences in the development of the offspring. iii Preface Statement of co-authorship: I composed this thesis in its entirety, with direction and input from Dr Angela Devlin and Dr. Tim Oberlander. This thesis was revised by Dr. Angela Devlin, Dr. Tim Oberlander and Dr. Timothy Green. This is a retrospective study using archived blood and serum samples collected by members of the Oberlander Laboratory. Members of the Oberlander Laboratory were also responsible for administering mood questionnaires. I was responsible for executing monocyte isolation and all nucleic acid extractions. In addition, I was responsible for conducting the DNA methylation analyses, gene expression analyses, and genotyping assays. I prepared all serum samples for Janette King and Roger Dyer (Metabolomics Core Lab of the UBC Nutrition Research Program, directed by Dr. Sheila Innis), who performed the methyl nutrient analyses. I conducted statistical analyses for all results. A manuscript will be prepared for future publication based on results from Chapters 3-5. The manuscript will include relevant sections of this thesis, including introduction, methods, results and discussions found in Chapters 1-6. Statement of research ethics approval: This thesis was conducted under ethical approval from the University of British Columbia Research Ethics Board, and the Children’s and Women’s Health Centre of British Columbia Research Review Committee (Certificate number: H05-70629). iv Table of Contents ABSTRACT....................................................................................................................................... ii PREFACE ........................................................................................................................................ iv TABLE OF CONTENTS ..................................................................................................................... v LIST OF TABLES ........................................................................................................................... viii LIST OF FIGURES ........................................................................................................................... ix LIST OF ABBREVIATIONS................................................................................................................ x ACKNOWLEDGEMENTS................................................................................................................. xii CHAPTER I: INTRODUCTION ....................................................................................................... 1 1.1 Project Overview .............................................................................................................. 1 1.2 Depression and the Brain ................................................................................................. 5 1.2.1 Depression................................................................................................................. 5 1.2.2 Serotonergic System ................................................................................................. 5 1.2.3 Depression During Pregnancy .................................................................................. 7 1.3 Developmental Programming .......................................................................................... 8 1.3.1 Overview of Developmental Programming .............................................................. 8 1.3.2 Human Epidemiological Evidence ........................................................................... 9 1.3.3 Developmental Programming in Animal Models ................................................... 10 1.4 Epigenetics ..................................................................................................................... 11 1.4.1 Epigenetics Defined ................................................................................................ 11 1.4.2 Epigenetics as a Potential Programming Mechanism ............................................. 11 1.5 Methyl Nutrient Metabolism .......................................................................................... 13 1.5.1 Folate....................................................................................................................... 13 1.5.2 Vitamin B12 ............................................................................................................. 15 1.5.3 Methyl Nutrient Metabolism................................................................................... 15 1.5.4 Methylenetetrahydrofolate Reductase Polymorphism ............................................ 17 1.5.5 Deficiencies in Folate and Vitamin B12 .................................................................. 17 1.5.6 Methyl Nutrients and DNA Methylation ................................................................ 18 1.6 Rationale and Hypothesis ............................................................................................... 19 v CHAPTER 2: MATERIALS AND METHODS ................................................................................. 22 2.1 Study Population ............................................................................................................ 22 2.2 Maternal Mood Assessments ......................................................................................... 22 2.3 Subject Samples ............................................................................................................. 23 2.4 Nucleic Acid Extraction ................................................................................................. 23 2.4.1 DNA Isolation from Monocytes ............................................................................. 24 2.4.2 RNA Isolation from Whole Blood .......................................................................... 24 2.4.3 DNA Isolation from Whole Blood .......................................................................... 25 2.4.4 Assessment of Nucleic Acid Quality and Quantity ................................................ 25 2.5 DNA Methylation Assay ................................................................................................ 25 2.5.1 Target Gene ............................................................................................................. 25 2.5.2 Bisulphite Pyrosequencing...................................................................................... 26 2.6 Gene Expression Assay .................................................................................................. 30 2.7 Genotyping Assays ......................................................................................................... 30 2.8 Methyl Nutrient Quantification ...................................................................................... 33 2.9 Exclusions ...................................................................................................................... 33 2.10 Statistical Analyses ........................................................................................................ 33 CHAPTER 3: MOTHERS ............................................................................................................. 35 3.1 Maternal Methyl Nutrient Metabolism and Mood ......................................................... 36 3.2 Maternal Methyl Nutrient Metabolism, Mood, and Methylation................................... 40 3.3 Maternal Methyl Nutrient Metabolism, Methylation, and Gene Expression ................. 45 3.4 5HTTLPR ....................................................................................................................... 48 CHAPTER 4: MATERNAL INFLUENCES ON NEWBORNS............................................................ 49 4.1 Maternal Methyl Nutrient Metabolism, Maternal Mood, and Newborn Methylation ... 50 4.2 Maternal Methyl Nutrient Metabolism and Newborn Gene Expression ........................ 55 4.3 Maternal Mood and Newborn Gene Expression ............................................................ 55 vi CHAPTER 5: NEWBORNS ........................................................................................................... 56 5.1 Maternal and Newborn Methyl Nutrients ...................................................................... 57 5.2 Newborn Methyl Nutrient Metabolism and DNA Methylation ..................................... 60 5.3 Newborn Methyl Nutrient Metabolism, DNA Methylation, and Gene Expression ....... 62 5.4 5HTTLPR ....................................................................................................................... 65 CHAPTER 6: GENERAL DISCUSSION ......................................................................................... 66 6.1 Discussion of Results ..................................................................................................... 66 6.1.1 Methyl Nutrient Status ............................................................................................ 66 6.1.2 Discussion of First Aim – Mothers ......................................................................... 70 6.1.3 Discussion of Second Aim – Maternal Influences on Newborns ........................... 76 6.1.4 Discussion of Third Aim – Newborns .................................................................... 78 6.2 Limitations, Strengths, and Future Directions ............................................................... 80 6.2.1 Limitations .............................................................................................................. 80 6.2.2 Strengths ................................................................................................................. 83 6.2.3 Future Directions .................................................................................................... 83 6.3 Concluding Remarks ...................................................................................................... 84 BIBLIOGRAPHY ............................................................................................................................. 87 vii List of Tables Table 2.1 Primers for bisulphite pyrosequencing......................................................................... 29 Table 2.2 Primers for genotyping. ................................................................................................ 32 Table 3.1 Maternal MTHFR C677T genotype and demographic data of the pregnant women. .. 38 Table 3.2 Relationship between maternal folate levels and maternal SLC6A4 methylation (adjusted for EPDS scores). .......................................................................................................... 41 Table 3.3 Relationship between maternal folate levels and maternal SLC6A4 methylation (adjusted for HAM-A scores). ...................................................................................................... 42 Table 3.4 Relationship between maternal 3rd trimester EPDS scores and maternal SLC6A4 methylation. .................................................................................................................................. 43 Table 3.5 Relationship between maternal 3rd trimester HAM-A scores and maternal SLC6A4 methylation. .................................................................................................................................. 44 Table 3.6 Estimated difference in maternal SLC6A4 gene expression according to maternal SLC6A4 methylation sites (unadjusted). ....................................................................................... 46 Table 3.7 Estimated difference in maternal SLC6A4 gene expression according to maternal SLC6A4 methylation sites (adjusted). ........................................................................................... 47 Table 4.1 Maternal MTHFR C677T genotype and clinical data of their newborns. .................... 51 Table 4.2 Relationship between maternal MTHFR C677T genotype and newborn SLC6A4 methylation (adjusted for EPDS scores). ...................................................................................... 52 Table 4.3 Relationship between maternal MTHFR C677T genotype and newborn SLC6A4 methylation (adjusted for HAM-A scores). .................................................................................. 53 Table 4.4 Relationship between prenatal exposure to maternal 3rd trimester mood and newborn SLC6A4 methylation. .................................................................................................................... 54 Table 5.1 Methyl nutrient status in mothers and newborns. ........................................................ 58 Table 5.2 Relationship between newborn MTHFR C677T genotype and newborn SLC6A4 methylation. .................................................................................................................................. 61 Table 5.3 Relationship between newborn SLC6A4 methylation and newborn SLC6A4 expression (unadjusted)................................................................................................................................... 63 Table 5.4 Relationship between newborn SLC6A4 methylation and newborn SLC6A4 expression (adjusted)....................................................................................................................................... 64 viii List of Figures Figure 1.1 Methyl nutrient metabolism.......................................................................................... 3 Figure 1.2 Schematic representation of thesis project. .................................................................. 4 Figure 2.1 Schematic representation of the SLC6A4 promoter region analyzed for methylation status. ............................................................................................................................................ 28 Figure 3.1 Specific aim 1. ............................................................................................................ 35 Figure 3.2 Relationship between early 3rd trimester EPDS scores and HAM-D scores. ............ 39 Figure 3.3 Relationship between maternal vitamin B12 and holoTC status. ................................ 39 Figure 4.1 Specific aim 2. ............................................................................................................ 49 Figure 5.1 Specific aim 3. ............................................................................................................ 56 Figure 5.2 Relationships between maternal and newborn methyl nutrient levels. ...................... 59 Figure 5.3 Relationship between newborn vitamin B12 and holoTC levels. ................................ 59 ix List of Abbreviations 5,10-MTHF 5,10-methylenetetrahydrofolate 5-HT Serotonin 5-HTT Serotonin transporter 5-HTTLPR Serotonin transporter linked polymorphic region 5-MTHF 5-methyltetrahydrofolate AdoHcy S-Adenosylhomocysteine AdoMet S-Adenosylmethionine Bdnf Gene for brain derived neurotrophic factor bp Base-pair cDNA Complementary DNA DHF Dihydrofolate DHFR Dihydrofolate reductase DMR Differentially methylated region DNA Deoxyribonucleic acid DNMT DNA methyltransferase dTMP Deoxythymidine monophosphate EDTA Ethylenediaminetetraacetic acid EPDS Edinburgh postnatal depression scale GCP II Glutamate carboxypeptidase II GLM General linear model GR Glucocorticoid receptor H3K9 Histone 3 lysine 9 HAM-A Hamilton rating scale for anxiety HAM-D Hamilton rating scale for depression Hcy Homocysteine HDAC Histone deacetylase x HoloTC Holotranscobalamin HPA Hypothalamic-pituitary-adrenal IF Intrinsic factor Igf2 or IGF2 Gene for insulin-like growth factor 2 LG/ABN Licking/grooming-arched back nursing LP Low protein LSD Least significant difference MAT Methionine adenosyltransferase MeCP2 Methyl CpG binding protein 2 MS Methionine synthase MTHFD Methylenetetrahydrofolate dehydrogenase MTHFR Methylenetetrahydrofolate reductase NR3C1 or Nr3c1 Gene for glucocorticoid receptor NTD Neural tube defect PND Postnatal day RDA Recommended dietary allowance RNA Ribonucleic acid RQ Relative quantifiation SAHH S-Adenosylhomocysteine hydrolase SD Standard deviation SEM Standard error of the mean SHMT Serine hydroxymethyltransferase SLC6A4 Gene for serotonin transporter SSRI Selective serotonin reuptake inhibitor THF Tetrahydrofolate TSS Transcription start site TYMS Thymidylate synthase xi Acknowledgements First and foremost, I would like to express my sincere gratitude to my thesis cosupervisors Dr. Angela Devlin and Dr. Tim Oberlander for providing me with the unique opportunity to explore this remarkable research project. I am indebted to my supervisors for not only their exceptional scientific guidance, but also for their continuing support and patience. I would like to thank my supervisory committee member, Dr. Timothy Green for his advice, insight, and tremendous help in statistics. I am grateful for the help that Ursula Brain has provided me. She has been a key player in making sure that I stay organized in maintaining my databases. Further, I would like to thank all my lab mates at the Child and Family Institute (CFRI) and friends. To Tiffany Ngai, Melissa Glier, Dian Sulistyoningrum, Eugene Wang, Mihai Cirstea, Rika Aleliunas, Baki Cvijetinovic, Ashish Sharma, Tina Li, Edgar Chan-Wong, and Roger Jen – a heartfelt appreciation goes to them for teaching me molecular biology techniques, providing me with invaluable advice in trouble-shooting experiments and writing the thesis, offering me emotional support, and making my graduate studies experience enjoyable. Moreover, I would like to recognize Janette King and Roger Dyer for their technical assistance in methyl nutrient studies. The completion of this thesis was made possible by studentships from the Interdisciplinary Women’s Reproductive Health (IWRH) Research Training Program, the Canadian Institutes of Health Research, NeuroDevNet (NDN), and CFRI. I would also like to acknowledge IWRH, NDN, CFRI and UBC for providing me with travel funding. Finally, I would like to extend my deepest gratitude to my family for instilling in me the value of education and coaching me in so many aspects of life. I’m extremely fortunate to have their love and support. xii 1 CHAPTER I: Introduction 1.1 Project Overview Up to 15% of women experience depression during pregnancy, and one in three are taking selective serotonin reuptake inhibitors (SSRIs) as antidepressants (1). The neurotransmitter, serotonin (5-HT) plays a critical role in modulating stress responses and early brain development. Depressive maternal mood and SSRI usage alter 5-HT levels during development and may program the newborn’s responses to environmental shifts later in life (24). The molecular mechanisms underlying developmental programming remain poorly understood, but may involve epigenetic processes such as DNA methylation. DNA methylation requires several methyl nutrients (Figure 1.1), such as folate and vitamin B12, as enzymatic cofactors in the production of S-adenosylmethionine (AdoMet), the key methyl donor in the body. Methyl nutrient metabolism and early postnatal exposure to adverse maternal care giving are independently associated with gene-specific changes in DNA methylation and expression in rodents (5-7). Furthermore, methyl nutrient metabolism and prenatal exposure to maternal mood are independently associated with gene-specific changes in DNA methylation in humans (8-11). The overall objective of this thesis is to investigate whether disturbances in methyl nutrient metabolism and depressed mood during pregnancy affect methylation and gene expression (mRNA levels) of the serotonin transporter in mothers and their newborns. As steps toward achieving the objective, I performed DNA, RNA, and biochemical analyses on blood samples obtained from a cohort of pregnant women and their newborn infants. A schematic representation of my thesis is shown in Figure 1.2, and more details will be provided in Chapters 3-5. In this introduction, relevant background information and literature review 1 pertaining to depression, serotonin, developmental programming, and methyl nutrient metabolism will be provided. I will then present a rationale in support of my thesis project. 2 Figure 1.1 Methyl nutrient metabolism. Abbreviations: THF, tetrahydrofolate; DHF, dihydrofolate; 5-MTHF, 5-methyltetrahydrofolate; 5,10-MTHF, 5,10-methylenetetrahydrofolate; AdoMet, S-adenosylmethionine; AdoHcy, Sadenosylhomocysteine; Met-DNA, methylated DNA; Hcy, homocysteine; B12, vitamin B12; dTMP, deoxythymidine monophosphate; TYMS, thymidylate synthase; DHFR, dihydrofolate reductase; MTHFD, methylenetetrahydrofolate dehydrogenase; SHMT, serine hydroxymethyltransferase; MAT, methionine adenosyltransferase; MTHFR, methylenetetrahydrofolate reductase; MS, methionine synthase; DNMT, DNA methyltransferase; SAHH, AdoHcy hydrolase. 3 Figure 1.2 Schematic representation of thesis project. 4 1.2 Depression and the Brain 1.2.1 Depression Depression is the 4th leading cause of disease worldwide (12), with a life-time prevalence of 12.2% in Canada (13). This mental illness is characterized by sadness, guilt, lack of pleasure, low energy, and disturbed sleep or appetite (14). Depression occurs more frequently in women and in older individuals, and is likely under-reported (13). There are many hypotheses regarding the etiology of depression, such as dysfunctions in neurotransmitter systems, dysfunctions in the hypothalamic-pituitary-adrenal (HPA) axis, or imbalances in the immune system (15). However, it remains unclear how these factors contribute to depression. Depression is mainly treated by psychiatric medications. The most frequently prescribed class of antidepressants is SSRI, and the less common classes include tricyclic antidepressants and monoamine oxidase inhibitors (16). However, inadequate or non-response to treatment of depression remains a concern (17). 1.2.2 Serotonergic System One type of neurotransmitter system suggested to be involved in depression is the serotonergic system. Serotonin is a neurotransmitter that is involved in the modulation of stress responses, and acts as a trophic factor (cell division, synaptogenesis, etc.) during neurodevelopment (18). The cell bodies of 5-HT neurons are mainly located in the raphe nuclei in the brain stem (15). The axons of 5-HT neurons extend throughout the entire brain, where 5HT is released and could be taken up by 16 different subtypes of 5-HT receptors on the postsynaptic membranes (19). 5 Serotonin is also present in the blood, mainly in platelets (20). It has been demonstrated in rats that elevated levels of 5-HT in the brain are reflected in the peripheral circulation (via administration of 5-hydroxytryptophan, a precursor of 5-HT) (21). As such, whole blood serves as an easier and more useful source to measure body 5-HT levels than the brain/neural tissue. Decreased levels of tryptophan in plasma (22) and 5-hydroxyindoleacetic acid (metabolite of 5HT) in cerebral spinal fluid (23) are often reported in depressed patients. Furthermore, low levels of whole blood 5-HT have been associated with depression in children (24). This suggests that having a “serotonergic vulnerability” may lead to depressive symptoms (15). A modulator of the amount and duration of 5-HT in the synaptic cleft is the serotonin transporter (5-HTT). Serotonin transporter is a membrane-bound protein that governs the reuptake of 5-HT, thereby decreasing extracellular 5-HT levels. The transporter is mainly found on presynaptic neurons and platelets (25). SSRIs target 5-HTT and raise the level of extracellular 5-HT (26). The gene that encodes 5-HTT, SLC6A4, is located on chromosome 17. Methylation of the SLC6A4 promoter has been shown to be associated with SLC6A4 expression in lymphoblast cell lines (27). Although, the methylation status seems to depend on the genotype of a 44 base-pair (bp) insertion/deletion variant (referred to as the 5-HTT-linked polymorphic region, or 5HTTLPR, variant) in the promoter region (28). The long (l) allele has a frequency of 0.6 in a Caucasian population, and contains 16 repeats of a 22 base-pair (bp) repeat sequence (29). The short (s) allele has an allele frequency of 0.4 in a Caucasian population and contains 14 repeats (29). Homozygosity for the short allele (ss) is associated with increased SLC6A4 promoter methylation (27). Using luciferase plasmid constructs, it was shown that the 5-HTTLPR s allele has lower basal promoter activity than the l allele in human placental choriocarcinoma cell lines 6 (28). The ss genotype is associated with anxiety (29), depression in adults after exposure to stressful life events (such as threat, loss, humiliation, or defeat) (30), depression in adults after exposure to childhood maltreatment (30), and prolonged salivary cortisol response to stressors (31,32). As well, the s allele has been shown to unfavorably affect SSRI treatment in the elderly (33). Intriguingly, cortisol may increase 5-HTT expression (34). However, this increase is dependent on the 5HTTLPR genotype (34). More recently, a variant in the l allele of the 5HTTLPR has been identified (A→G; rs 25531; presence of the minor G allele in the 5HTTLPR l variant (Lg) creates a binding site for AP2, a repressor), and shown to be associated with obsessive-compulsive disorder (35). 1.2.3 Depression During Pregnancy Antenatal depression occurs in approximately 15% of mothers during pregnancy (1) and one third of depressed mothers take SSRIs during pregnancy (1). Maternal depression and SSRI exposure during pregnancy appear to be associated with adverse outcomes in the newborn, and this raises concerns regarding the long-term consequences of exposure to antenatal depression and SSRIs in the development of the offspring (2-4). However, it is difficult to distinguish between the consequences of prenatal exposure to maternal depression and prenatal exposure to SSRIs. Mothers who were depressed during pregnancy have higher levels of cortisol and lower levels of 5-HT in urine than mothers who were not depressed (2). Prenatal exposure to maternal depression was associated with higher levels of cortisol and lower levels of 5-HT in the urine of newborns (2). Further, newborns from mothers who were depressed during pregnancy exhibited less optimal motor, orientation, and habituation scores than infants of mothers who were not 7 depressed (2). Rats that were exposed to adverse maternal care during early postnatal development [a time period suggested to be similar to the fetus in the 3rd trimester of pregnancy in humans (26)] showed higher plasma corticosterone (dominant form of glucocorticoid in rodents) responses to acute stress during adulthood than rats that were not exposed to adverse maternal care (36). Infants with prenatal exposure to SSRIs showed reduction of early evening basal salivary cortisol levels compared to non-SSRI exposed infants at 3 months of age (3). Furthermore, prenatal SSRI exposure was associated with neurobehavioural disturbances in newborns, such as increased jitteriness, feeding disturbances, irritability, and respiratory disturbances (4). These findings suggest that prenatal exposure to “shifts in environment”, such as maternal mood and SSRIs, may influence offspring 5-HT activity and behaviour. 1.3 Developmental Programming 1.3.1 Overview of Developmental Programming Developmental programming refers to the concept that environmental exposures during prenatal and early postnatal development may program an individual’s metabolic and physiological response to environmental shifts later in life and contribute to the development of chronic diseases. This theory was postulated by Dr. David Barker in the 1980’s (37). Barker and colleagues found that geographical regions of the United Kingdom with high rates of low birth weight in the early 1900’s also had high rates of deaths due to coronary heart disease between 1968 and 1978 (38,39). Many human epidemiological and animal studies thus followed to support the concept of developmental programming. 8 1.3.2 Human Epidemiological Evidence Support for the concept of developmental programming came from studies of a cohort conceived during the World War II Dutch famine. The famine caused severe stress and nutritional inadequacy, where energy intakes decreased to less than 1000 kcal/day (less than 4200 kJ/day) per person (40). Prenatal exposure to the famine in the 2nd trimester was associated with affective psychosis in both men and women as adults (40,41), Antisocial Personality Disorder in men at 18 year of age (42), and addiction in men at ~60 years of age (43). Early gestational exposure to the Dutch famine was also associated with schizophrenia at 24-48 years of age in both men and women (44). Other epidemiological studies showed that early gestational exposure to the 1959-1961 Chinese famine was associated with increased risk for schizophrenia later in life (45). Prenatal and early postnatal (up to 3 months old infants) exposure to maternal stress imposed by the 1967 Six-Day War in Israel was associated with increased risks for developmental delays and social withdrawal behaviour in boys at elementary school age (46), and schizophrenia in females after 30 years of age (47). Perhaps the most intriguing studies were those that showed prenatal exposure to maternal anxiety [often comorbid with depression (48)] have adverse consequences on the offspring. Children prenatally exposed to maternal anxiety during the 32nd week of gestation had higher salivary awakening cortisol levels, and displayed behavioural and emotional problems at 10 years old (49,50). Moreover, a longitudinal study showed that antenatal depression was associated with antisocial behaviour in the offspring in adolescence (51). Taken together, in utero exposure to maternal stress suggests that adverse outcomes surface later in life. 9 1.3.3 Developmental Programming in Animal Models In addition to human studies, many animal studies support the concept of developmental programming. In rats, numerous researchers have studied neonatal pups exposed to different maternal care in the early postnatal period. This included mothers that exhibited high levels of licking/grooming-arched back nursing (LG/ABN) towards their young (assumed good maternal behaviour), or mothers that exhibited lower LG/ABN (36). In the first 10 days of life, offspring exposed to low maternal LG/ABN care had higher levels of plasma adrenocorticotropic hormone and corticosterone in response to stressors (restraint test) than offspring exposed to high maternal LG/ABN care (36). Early postnatal exposure to low maternal LG/ABN also led to more fearful responses in novel environments in adult (~postnatal day 100) offspring (52). Female rats exhibited similar maternal behaviour towards their own offspring as the female rats’ mothers were to them when they were pups (53). To take it a step further, researchers conducted crossfostering studies and showed that offspring conceived by low licking/nursing dams, but nursed by high licking/nursing foster dams showed high licking/nursing behaviour towards their own offspring (54,55). Similarly, offspring from female rhesus monkeys exposed to stressors during late pregnancy exhibited abnormal social behaviour at adolescence (four years of age) (56). These epidemiological and animal studies show a clear pattern between prenatal or early postnatal exposure to shifts in environment, and the possible programming of behaviour that may have life-long health consequences. The molecular mechanism underlying developmental programming is not well understood but is thought to be mediated through epigenetic mechanisms. 10 1.4 Epigenetics 1.4.1 Epigenetics Defined Epigenetics refers to changes in gene expression that occur without changes in DNA sequence (57). Epigenetic processes include DNA methylation, chromatin modifications (acetylations, methylation, phosphorylation, etc.), and miRNA (57). Epigenetic patterns are heritable, but are responsive to shifts in environment, particularly during development (5,6,55). DNA methylation takes place at CpG dinucleotides, specifically the 5’ position of cytosines (58). DNA methyltransferases (DNMTs) are ubiquitous enzymes responsible for DNA methylation (59). Maintenance methylation is performed by DNMT1, which is responsible for methylation during somatic cell replication (58). De novo methylation is accomplished by DNMT3a and DNMT3b during embryonic development (59). DNA methylation in the promoter is often associated with transcriptional silencing (60). It is responsible for maintaining many biological roles such as silencing of repetitive elements (60), X-chromosome inactivation (61) and genomic imprinting (61). 1.4.2 Epigenetics as a Potential Programming Mechanism Early studies to suggest that epigenetic processes play a role in developmental programming showed that early postnatal variations in maternal care altered the methylation status of Nr3cl (gene that encodes the glucocorticoid receptor, GR) and Bdnf (gene that encodes the brain derived neurotrophic factor) in adult rats (6,55). Neonatal pups were exposed to either mothers that displayed normal care (high levels of nursing/licking), or adverse care (low levels of nursing/licking). Offspring that were exposed to adverse maternal care during the early postnatal period [postnatal day (PND) 1-7] had increased Nr3c1 and Bdnf promoter methylation 11 in the hippocampus or prefrontal cortex (6,55). These DNA methylation patterns were present in the 1st day of life, were not present in cross-fostered pups, were shown to persist into adulthood (PND 90), and were associated with elevated corticosterone levels (6). Gene-specific DNA methylation has been associated with prenatal exposure to stress in humans. In utero exposure to maternal depression was associated with increased methylation of NR3C1 promoter in cord blood (9). Interestingly, this region of NR3C1 is analogous to the Nr3c1 region in rats where prenatal exposure to adverse maternal care giving altered methylation in the brain (6). Further, elevation of infant salivary cortisol in response to stress at 3 months of age was associated with methylation of NR3C1 promoter (9). Increased methylation has been observed in NR3C1 in post-mortem hippocampal tissue from suicide victims who had experienced abuse in childhood, compared to suicide victims with no abuse history, or hippocampal tissue from adults who had died from other causes (10). NR3C1 mRNA levels in the hippocampus were also decreased in suicide victims who experienced childhood abuse, compared to adults who had no history of child abuse and died of accidental causes (10). Another study showed that maternal 2nd trimester mood was inversely associated with SLC6A4 methylation in the blood of both mothers and their newborns (11). Taken together, these studies provide evidence that prenatal and early postnatal exposure to adverse environments can alter the methylation status of DNA, which may program gene expression and contribute to the development of chronic disease. 12 1.5 Methyl Nutrient Metabolism DNA methylation is closely linked to the methyl nutrient metabolism (Figure 1.1; page 2). Some components of this metabolism include methyl nutrients (such as folate and vitamin B12) and the enzyme methylenetetrahydrofolate reductase (MTHFR). 1.5.1 Folate Folate is a water soluble essential B vitamin (62). The predominant forms of dietary folate are polyglutamylated, and are found in leafy vegetables, potatoes, and fruits (63). Naturally occurring folate is also found in the monoglutamylated form, in milk, eggs, and meats (63). In addition, polyglutamylated forms of folate are synthesized by the bacterial microflora in the large intestine (64). Following studies that proved the success of folate in reducing rates of neural tube defects (NTDs) (65,66), the Canadian government mandated folic acid-fortification of grain products in 1998, and recommended folic-acid-containing prenatal vitamin supplement usage in pregnant women (67,68). The Recommended Dietary Allowance (RDA) of folate is 400 ug per day for the general population and 600 ug per day for pregnant women (69). Folic acid is the monoglutamate folate form generally found in fortified foods [white flour, cornmeal and enriched pasta products (68)] and supplements because it is more stable than naturally occurring folates (70,71). Since fortification, a typical North American diet in a day contains roughly 400 ug of folate (72). High folic acid intake after fortification leads to elevated unmetabolized folic acid in the circulation (73), and this is likely due to low DHFR activity in humans (74,75). The average Tolerable Upper Intake Level for folic acid is 1 mg per day in adults (69). 13 In the duodenum and upper jejunum, the polyglutamylated folate is hydrolyzed to its monoglutamylated form by the enzyme glutamate carboxypeptidase II (GCP II), located at the intestinal apical brush border membrane (62).These monoglutamylated folates are absorbed by the proton-coupled folate transporter into the intestinal mucosa. From there, folate, in the form of tetrahydrofolate (THF) becomes reduced to 5-methyltetrahydrofolate (5-MTHF) within the enterocyte (76). Folic acid, on the other hand, needs to be first reduced to THF by dihydrofolate reductase (DHFR) in the enterocyte prior to converting to 5-MTHF (77). The reduction of folate to 5-MTHF may occur in the liver as well. 5-MTHF is the predominant circulating form of folate in the bloodstream and is taken into cells via carriers or receptors, such as reduced folate carrier and folate receptor α (62). In order for folate to remain intracellular and to participate in the methyl nutrient metabolism, folate becomes polyglutamylated (77). Rodent studies have shown that during pregnancy, plasma and hepatic folate levels were much higher in the fetus than the mother (78), suggesting a preferential transfer of folate from mother to fetus during gestation. Similarly in humans, blood from the umbilical cord taken at birth had much higher levels of total folate than maternal blood taken 1-12 hours prior to delivery (75). Most of this difference in folate levels between the mother and the newborn was due to higher cord blood levels of 5-MTHF, and not unmetabolized folic acid (75). Although folic acid supplementation is effective at reducing NTD risk (79) and may even enhance antidepressant response (80), there are also concerns about excess folic acid intake. One major concern is that excess folate intake may mask vitamin B12 deficiency (81,82). 14 1.5.2 Vitamin B12 Vitamin B12, or cobalamin, is a large cobalt-containing water soluble B vitamin (83,84). Vitamin B12 is synthesized by bacteria and is mainly found in animal products, such as milk, eggs, meats, and fish (84). Vitamin B12 has also been fortified in cereals (84). A typical North American diet in a day contains 3-30 ug of vitamin B12 (85). Vitamin B12 from the diet is initially bound to a salivary protein, haptocorrin, in the stomach (86). After migrating to the duodenum, vitamin B12 is released from haptocorrin and subsequently binds to a gastric protein, intrinsic factor (IF), in the ileum (86). The IF-bound Vitamin B12 is then absorbed by the intestinal mucosal cells via receptor-mediated endocytosis (84). Generally, 50% of dietary vitamin B12 is absorbed in healthy adults (84). However, malabsorption is common in the elderly and in patients with gastric dysfunctions (87). Once in the enterocyte, vitamin B12 is released from the IF and binds to transcobalamin II, forming a complex called holotranscobalamin (HoloTC) (88). HoloTC accounts for about 30% of total circulating vitamin B12 and is the biologically active form taken up by cells (88). The RDA for vitamin B12 is 2.4 ug per day for adults and 2.6 ug per day for pregnant women (69). Given that cord blood from newborns and the placenta contain higher levels of vitamin B12 than blood from mothers at delivery (89), vitamin B12 may be preferentially transferred to the fetus during pregnancy. 1.5.3 Methyl Nutrient Metabolism Both folate and vitamin B12 are involved in the methyl nutrient metabolism. As part of the folate cycle, 5-MTHF functions to donate methyl groups for the remethylation of homocysteine (Hcy) to methionine, producing THF. This reaction is catalyzed by the 15 ubiquitously expressed enzyme, methionine synthase (MS) (90). MS requires vitamin B12 as an enzymatic cofactor (90). Serine hydroxymethyltransferase, a vitamin B6 dependent enzyme, converts THF to 5,10-methyleneTHF using serine as a one-carbon donor (62). Alternatively, methylenetetrahydrofolate dehydrogenase converts THF through a series of reactions to 5,10methyleneTHF as well (62). The intermediate product of the series of reactions is 10formylTHF, and 10-formylTHF can donate a carbon for purine synthesis (62). 5,10methyleneTHF is a cofactor for thymidylate synthase, which converts deoxyuridine monophosphate to deoxythymidine monophosphate (dTMP) for thymine and later on pyrimidine synthesis (62). 5,10-methyleneTHF is subsequently reduced to 5-MTHF by the vitamin B2 dependent enzyme, MTHFR (62). Subsequently, 5-MTHF donates a methyl group for the formation of methionine (62). The folate cycle is metabolically linked to the methionine cycle. After the remethylation of Hcy to methionine, methionine is converted to AdoMet by methionine adenosyltransferase (62). AdoMet is a universal methyl donor for the methylation reactions of DNA, RNA, lipids, and proteins, catalyzed by methyltransferases such as DNMTs. S-adenosylhomocysteine (AdoHcy) is produced following methyl donation by AdoMet, and AdoHcy is then hydrolyzed to adenosine and Hcy by AdoHcy hydrolase (62). High levels of AdoHcy have been shown to inhibit methyl transferase activity (91,92). Hcy can be remethylated to methionine by MS, completing the methionine cycle. Alternatively Hcy can be remethylated by the liver and kidney specific betaine homocysteine methyltransferase, which uses betaine as the methyl donor (62). As well, Hcy can partake in the transsulfuration pathway to form cysteine, via vitamin B6 dependent enzymes cystathionine β-synthase and cystathionine γ-lyase (62). 16 1.5.4 Methylenetetrahydrofolate Reductase Polymorphism The gene that encodes for MTHFR, MTHFR, is located on chromosome 1 and has a common variant (C677T; rs1801133) in exon 4. The MTHFR variant results in the conversion of an alanine amino acid to a valine residue in the catalytic domain of the protein at a frequency of 38% (93). This substitution results in a thermolabile enzyme with 70% reduced activity in subjects with the homozygous TT genotype (93). The T allele has a frequency of approximately 0.38 in Caucasian populations (93). The MTHFR 677TT genotype has been associated with increased risk for NTDs (94-97), colorectal carcinomas (when folate is limiting) (98), and stroke (when folate is limiting) (99). Studies have shown that the genetic variant is associated with an increased risk of psychiatric disorders such as unipolar depression, schizophrenia, bipolar disorder, and anxiety (100-102). Most noteworthy is the recent study which found that MTHFR 677TT genotype is associated with antenatal depression in the 2nd trimester (11). 1.5.5 Deficiencies in Folate and Vitamin B12 Deficiencies in methyl nutrients may disrupt the equilibrium of the methyl nutrient metabolism and result in adverse consequences in health. Folate deficiency is defined as < 6 nmol/L in serum or plasma, although this value varies (103). The definition of vitamin B12 deficiency varies as well, but is generally around <150 pmol/L in serum or plasma (87). HoloTC deficiency has been suggested to be <35 pmol/L in plasma (104). Low levels of folate have been shown to lead to megaloblastic anemia (77) and elevated concentrations of plasma Hcy (a risk factor for cardiovascular disease) (105). As well, folate supplementation has been shown to 17 decrease the incidences of NTDs (65,66). Importantly, low folate levels have been associated with depression in older adults (106). The folate status of Canadians, including women of child-bearing age, has improved significantly since the commencement of mandatory fortification (107,108). However, insufficient vitamin B12 status in Canadians had been reported in the elderly and in pregnant women from Ontario (109,110). Excess folic acid could mask vitamin B12 deficiency, because MS remains inactive in the absence of vitamin B12 (77). As a consequence, folate becomes “trapped” in the 5-MTHF form, creating a functional folate deficiency (111). This functional folate deficiency results in impaired DNA synthesis, the cause of megaloblastic anemia (111). While vitamin B12 deficiency leads to megaloblastic anemia in the presence of low folate, excess folic acid could mask the haematological manifestations of low vitamin B12, and delay diagnosis of vitamin B12 deficiency (77). The autoimmune disease, pernicious anemia, leads to vitamin B12 deficiency due to vitamin B12 malabsorption (112). Vitamin B12 deficiency has been associated with colorectal cancer (113), NTDs (114), and cognitive deficits (115,116). Notably, vitamin B12 was inversely associated with depression in the elderly (106,117). 1.5.6 Methyl Nutrients and DNA Methylation Methyl nutrients have been shown to be capable of altering DNA methylation. For example, in a study, men with hyperhomocysteinaemia (HHcy) exhibited DNA hypomethylation and bi-allelic expression of the imprinted gene, H19 (expressed only on the maternal allele) in blood (118). However, after 5-MTHF supplementation, H19 shifted back to monoallelic expression (118). In mice with HHcy, there was decreased Igf2/H19 differentially methylated region (DMR) methylation in liver, yet increased Igf2/H19 DMR methylation in the brain and 18 aorta (Igf2 is an imprinted gene expressed by the paternal allele, and encodes for insulin-like growth factor 2) (119), and methylation-silencing of Fads2 in liver (120). Furthermore, there is a well established mutation in the Agouti gene caused by the insertion of a retrotransposon, creating an ectopic promoter for the Agouti gene. Mice supplemented with folic acid, vitamin B12, choline and betaine during pregnancy had pups with increased methylation at the retrotransposon, leading to the darkening of offspring coat colour (5). Pregnant rats fed folic acid in the absence of vitamin B12 showed reduced global DNA methylation in the placenta (121). Additionally, the effects methyl nutrients have on DNA methylation may be mediated by genetics. For instance, folate status correlates with total genomic DNA methylation only in subjects with the MTHFR 677TT genotype, and not the MTHFR 677CC genotype in blood cells from Italian adults (8). 1.6 Rationale and Hypothesis Altered methyl nutrient levels, such as low plasma folate and vitamin B 12, and high plasma Hcy, are associated with depressive disorders in the elderly (100,106,122). The MTHFR 677TT genotype is associated with depression in both the non-pregnant population and pregnant women in their 2nd trimester (11,101). Although folate status in pregnant women has improved since 1998 due to folic acid fortification (107,108), poor vitamin B12 status was shown in pregnant women in Ontario (110). No studies have assessed the role of methyl nutrient status in the etiology of depression during late pregnancy. Serotonin not only plays an important role in mood regulation, but also serves as a trophic signal directing early brain development and function (26). Prenatal exposure to disturbed maternal mood and SSRI has been associated with adverse outcomes in the offspring 19 (4,49). The serotonin transporter governs the availability of extracellular 5-HT, and SLC6A4 expression have been shown to be influenced by DNA methylation (27). DNA methylation has been suggested as an underlying mechanism contributing to the phenomenon of developmental programming. Studies in rodent models have shown differential Nr3c1 methylation and expression following early postnatal exposure to adverse maternal care giving (6,55). Similarly in humans, it has been shown that prenatal exposure to maternal depressed mood during the 2 nd trimester was associated with the methylation status of NR3C1 in blood leukocytes of newborns and SLC6A4 in blood leukocytes of both the mothers and their newborns (9,11). DNA methylation is metabolically linked to the methyl nutrient metabolism. Studies showed that folate, vitamin B12, and the MTHFR 677TT genotype are associated with gene-specific and global changes in DNA methylation (8,123). However, little is known about the collective roles of maternal methyl nutrient metabolism and maternal mood on methylation and gene expression of SLC6A4 in mothers and their newborns. Based on previous findings, I hypothesize that disturbances in methyl nutrient metabolism contribute to maternal depressed mood during pregnancy, and prenatal exposure to maternal depressed mood affects SLC6A4 methylation and expression in mothers and their newborns. The hypothesis will be addressed based on the following specific aims: 1. To determine the effects of methyl nutrient metabolism (folate, vitamin B12, holoTC, and MTHFR C677T genotype) on mood during the 3rd trimester, and to determine whether maternal mood and methyl nutrient status are associated with SLC6A4 methylation and expression in blood collected during the 3rd trimester of pregnancy. 20 2. To determine whether prenatal exposure to maternal methyl nutrient metabolism (folate, vitamin B12, holoTC, and MTHFR C677T genotype) and maternal mood is associated with SLC6A4 methylation and expression in newborns (cord blood). 3. To determine whether newborn methyl nutrient metabolism (folate, vitamin B12, holoTC and MTHFR C677T genotype) is associated with SLC6A4 methylation and expression in newborns (cord blood). 21 2 CHAPTER 2: Materials and Methods 2.1 Study Population Ninety women were recruited during their early 3rd trimester of pregnancy (26-28 weeks gestation) between October 2006 and October 2009 at the Women’s and Children’s Health Center, and family physician offices (Vancouver, BC, Canada). The pregnant women were recruited as part of a longitudinal cohort study to examine the effects of prenatal SSRI exposure on neurobehavioural outcomes during infancy and childhood. As such, some women were receiving SSRIs treatment. 2.2 Maternal Mood Assessments Mood was assessed in the early third trimester of pregnancy (26-28 weeks gestation). Clinician administered questionnaires used were the Hamilton Rating Scale for Depression (HAM-D) (124) and Hamilton Rating Scale for Anxiety (HAM-A) (125). HAM-D, designed to assess depressive mood, is a 21-item questionnaire with a score range of 0-61. Higher scores on the HAM-D denote more depressive mood symptoms, and a score of 10 or more indicates depression. HAM-A is a 14-item questionnaire with a score range of 0-56, with higher scores indicating more anxious symptoms. HAM-A is designed to measure psychic anxiety (psychological distress and agitation) and somatic anxiety (physical complaints caused by anxiety). A HAM-A score of 14 or more indicates anxiety (14-17 indicates mild anxiety, 18-24 indicates moderate anxiety, and 25-30 indicates severe anxiety). In addition, maternal mood was assessed by the self-reported questionnaire, Edinburgh Postnatal Depression Scale (EPDS) (126). 22 EDPS is a 10-item questionnaire with a score range of 0-30, designed to assess depressive mood severity. An EPDS score of 10 or more signifies depression. 2.3 Subject Samples Whole blood samples were collected in EDTA-coated tubes (BD Vacutainer from Becton, Dickinson and Co., Franklin Lakes, NJ) from approximately 90 women at 33-36 weeks gestation. Newborn cord blood was also collected in the same manner at birth. EDTA, (ethylenediaminetetraacetic acid) serves as an anti-coagulant (chelates to the calcium in the blood) (127). Serum was also collected from the women at delivery and newborn cord blood at birth. To extract the serum, blood was collected, allowed to coagulate at room temperature for 15 minutes, centrifuged (3,000g for 8 minutes), and the serum (top layer) was collected. Whole blood and serum samples were stored at -80ºC in the Devlin laboratory at the Child and Family Research Institute (Vancouver, BC, Canada) for subsequent analyses. 2.4 Nucleic Acid Extraction Half milliliter of blood was aliquoted for monocyte extraction, and subsequent DNA extraction (for DNA methylation assays). Another 0.5ml of blood was aliquoted into 1.38ml of PAXgene blood RNA stabilizing solution (Qiagen Inc., Mississauga, ON) for total RNA extraction and genomic DNA extraction (for gene expression and genotyping assays, respectively). 23 2.4.1 DNA Isolation from Monocytes The relationship between SLC6A4 methylation status and 3rd trimester maternal mood was determined previously in whole blood, a heterogeneous population of cells (11). To take this one step further, I focused on cell-specific differences in SLC6A4 methylation status. As such, I chose to look at SLC6A4 methylation in monocytes, which represent 2-8% of the leukocyte population (128). Monocytes were extracted from maternal and newborn cord blood samples using EasySep Human CD14 (monocyte-specific marker (129)) positive selection kit and EasySep magnet (StemCell Technologies, Vancouver, BC). Genomic DNA was extracted from the monocytes using DNeasy Blood & Tissue kit (Qiagen Inc., Mississauga, ON), and treated with RNase A (Qiagen Inc., Mississauga, ON) to remove contaminating RNA. 2.4.2 RNA Isolation from Whole Blood Initial attempts at extracting RNA from monocytes using TRIzol (Invitrogen, Carlsbad, CA) with UltraPure Glycogen (Invitrogen, Carlsbad, CA) as carrier for RNA, or using the RNeasy Plus Mini kit (Qiagen Inc., Mississauga, ON) were unsuccessful. The trial experiments revealed that the integrity and amount of RNA extracted from monocytes was insufficient to accurately quantify mRNA levels. Hence, SLC6A4 mRNA was not quantified in monocytes. Alternatively, I quantified SLC6A4 mRNA levels in whole blood. Whole blood was incubated in PAXgene blood RNA stabilizing solution overnight at room temperature (20 hours) (130). The solution was centrifuged at 3000g for 10 minutes at 4ºC and the pellet was resuspended with 2ml RNase-free water. Two hundred microlitres of solution was taken for DNA extraction (see section 2.4.3). The remaining solution was used for RNA 24 extraction following manufacturer’s protocol for PAXgene Blood RNA Kit (Qiagen Inc., Mississauga, ON), and treated with DNase I (Qiagen Inc., Mississauga, ON) to remove contaminating genomic DNA. 2.4.3 DNA Isolation from Whole Blood DNA was extracted from the blood-PAXgene solution using the DNeasy Blood & Tissue kit (Qiagen Inc., Mississauga, ON) following manufacturer’s instructions, and treated with RNase A to remove contaminating RNA. 2.4.4 Assessment of Nucleic Acid Quality and Quantity Quality of DNA and RNA were assessed on an agarose gel. DNA quality was confirmed by visualization of high-molecular weight nucleic acid, and RNA quality was confirmed by visualization of intact 18S and 28S rRNA bands. Concentration and purity of DNA and RNA were determined using the Nanovue NanoSpec spectrometer (General Electric Inc., Fairfield, CT). An A260/A280 ratio (ratio of absorbance at 260nm and 280nm) of 1.8-2.1 was deemed an acceptable purity reading for DNA and RNA, as per manufacturer’s instructions. A lower ratio value indicates protein contamination (proteins absorb light at 280nm). 2.5 DNA Methylation Assay 2.5.1 Target Gene I chose SLC6A4 as my target gene for two reasons. Firstly, serotonin transporter plays a critical role in stress regulation. Secondly, methylation of the promoter region of SLC6A4 had recently been associated with SLC6A4 mRNA level (27). Philibert et al. had analyzed a CpG rich 25 region surrounding exon 1A. In that region, three CpG sites upstream of the transcription start site (TSS) were shown to be associated with mRNA levels. I chose to focus on a portion of that CpG rich region. The region of focus encompasses 10 CpG sites between -471bp and -374bp upstream of the TSS. The CpG 8 that I analyzed in my study corresponds to one of the three CpG sites that Philibert et al. had shown to be associated with gene expression (the CpG at location 872 in the Philibert et al. study) (27). My DNA methylation assay was focused on a 203bp fragment containing the 10 CpG sites (Figure 2.1). 2.5.2 Bisulphite Pyrosequencing Monocyte-specific genomic DNA (50ng) was bisulphite-treated using EZ DNA Methylation-Gold kit (Zymo Research, Irvine, CA) following manufacturer’s protocol. The process of bisulphite conversion converts unmethylated cytosines to uracils through deamination (131). The CpG-rich region of SLC6A4 promoter (11) was amplified by PCR using customdesigned primers (PyroMark Assay Design software, version 2.0), and HotStar Taq Polymerase (Qiagen Inc., Mississuaga, ON). The biotin labeled primer was ordered from Integrated DNA Technologies, Inc. (Coralville, IA) and non-biotin-labeled primer was ordered from Invitrogen (Carlsbad, CA) (Table 2.1). The mastermix consisted of 1x PCR buffer, 1x Q-solution, 0.2mM of dNTP, 0.2uM of the forward and reverse primers, and 2 units of Hotstart Taq. The cycling condition was 95°C for 15 minutes, 45 cycles of 94°C (1 minute), 60°C (1 minute), and 72°C (1 minute), 72°C for 10 minutes, and held infinite at 4°C. Quantification of methylation was performed using the PyroMark MD Pyrosequencing System (Qiagen Inc., Mississuaga, ON) (132). Sequencing primers were designed using the PyroMark Assay Design software (version 2.0) and ordered from Invitrogen (Carlsbad, CA). 26 PCR samples (10-15ul) were prepared for pyrosequencing. Each sample was incubated with 38ul of binding buffer, 2ul of Streptavidin Sepharose High Performance Beads, and water to make up a total volume of 80ul. The biotinylated PCR products were isolated using the PyroMark Vacuum Prep Workstation (Qiagen Inc., Mississuaga, ON). Pyrosequencing reactions were prepared containing 0.3mM of sequencing primer, annealing buffer, and the purified biotinylated PCR product. The percent methylation at each CpG site for the gene of interest was quantified using the Pyro Q-CpG software (version 1.0.9, 2006, Biotage, AB). 27 Figure 2.1 Schematic representation of the SLC6A4 promoter region analyzed for methylation status. The CpGs are underlined and numbered. Numbering of the gene sequence is relative to the transcriptional start site. Adapted from Devlin et al., 2010, with permission (11). 28 Table 2.1 Primers for bisulphite pyrosequencing. Gene PCR Product Size (bp) 203 Serotonin Transporter (SLC6A4) PCR Primers: SLC6A4-F (forward) 5’-biotin-GTATTGTTAGGTTTTAGGAAGAAAGAGAGA-3’ SLC6A4-R (reverse) 5’-AAAAATCCTAACTTTCCTACTCTTTAACTT-3’ Sequencing Primer: SLC6A4-S 5’-AAACTACACAAAAAAACAAAT-3’ 29 2.6 Gene Expression Assay Total RNA (500ng) was reverse transcribed using the High Capacity cDNA Reverse Transcription Kit (Applied Biosystems, Foster City, CA). The reaction contained 15.8 ul water, 8 ul RT buffer, 3.2 ul dNTP mix, 8 ul primer mix, and 4 ul reverse transcriptase for each sample, following manufacturer’s instructions. The cDNA was diluted by 1 in 5 for SLC6A4 mRNA quantification by real-time PCR. The ΔΔCt method of relative quantification (133) was used to quantify SLC6A4 mRNA levels with commercially available TaqMan primers and probes for SLC6A4 (FAM-dye labeled; SLC6A4 assay ID: Hs00169010_m1*) and the ABI 7500 real-time PCR system (Applied Biosystems, Foster City, CA). The gene, SLC6A4, contains 14 exons (134) and has at least four spliced variants (135). This Taqman probe spans exons 8-9, and detects all but one minor splice variant of SLC6A4 (GenBank ID: AY902473.1 not detected). The endogenous control for the ΔΔCt method was 18s rRNA (Applied Biosystems, Foster City, CA). The genes were quantified using 10ul Taqman Gene Expression Master Mix, 2x primers, 4ul of water, and 5ul of diluted cDNA. Calibrator for the maternal samples was comprised of 8 pooled samples from women who had low early 3rd trimester depressive mood scores (HAM-D and EPDS) and were not using SSRIs. The calibrator for the newborn samples was comprised of 8 pooled samples from newborns who were not prenatally exposed to SSRIs and whose mothers had low early 3rd trimester depressive mood scores (HAM-D and EPDS). Each sample was run in duplicate. 2.7 Genotyping Assays The 5-HTTLPR insertion/deletion variant was genotyped by PCR and agarose gel electrophoresis (136) (Table 2.2). Genotyping of the MTHFR C677T and the 5-HTTLPR La/Lg 30 variants were accomplished using TaqMan SNP genotyping assay primers and MGB probes (FAM-dye and VIC-dye labeled), Taqman Genotyping Mastermix (Applied Biosystems, Foster City, CA), and the ABI 7500 real-time PCR system (Applied Biosystems, Foster City, CA). The primers and probes for the MTHFR C677T variant are standard oligonucleotides by Applied Biosystems (Assay ID: C___1202883_20). The primers and probes for the 5HTTLPR La/Lg variants were custom-made by Applied Biosystems (35). The genotyping contained 9 ul of Taqman Genotyping Mastermix, 0.45ul of primers and probes, 6.6ul of water, and 1.5ul of DNA. 7-deaza-dGTP (New England Biolabs, Pickering, ON) was used in a 1:1 ratio with dGTP due to the nature of the PCR reaction. 7-deaza-dGTP is commonly used when a GC rich region is amplified by PCR. 7-deaza-dGTP acts like dATP through less hydrogen bonding to the complementary nucleotide, and this helps prevent compressions. The cycling conditions were 2 minutes at 50°C, 10 minutes at 95°C, followed by 40 cycles of 15s at 96°C and 90s at 62.5°C. 31 Table 2.2 Primers for genotyping. Gene Product Size (bp) 5HTTLPR (l/s) 529 for l allele; 485 for s allele PCR Primers: 5-HTTLPR-F 5’-GGCGTTGCCGCTCTGAATGC-3’ (forward) 5-HTTLPR-R 5’-GAGGGACTGAGCTGGACAACCAC-3’ (reverse) 142 5HTTLPR (A→G; rs25531) PCR Primers: 5-HTTLPR-SNP 5’-CTCCTAGGATCGCTCCTGCAT-3’ (forward) 5-HTTLPR-SNP 5’-GATGCTGGAAGGGCTGCA-3’ (reverse) Reporter Sequences: Reporter 1 (VIC dye) 5’-CCCCGGCATCCCCCCT-3’ Reporter 2 (FAM dye) 5’-CCCCAGCATCCCCCCT-3’ 32 2.8 Methyl Nutrient Quantification Vitamin B12 was initially quantified in maternal serum samples by a Lactobacillus delbrueckii microbiological assay (ALPCO Diagnostics, Salem, NH). However, the vitamin B12 levels obtained were much lower than expected. I suspected that this result was caused by intrapartum antibiotic usage by women at delivery, as confirmed by review of maternal clinical charts. Therefore I decided to proceed with an alternative method. Folate, vitamin B12, and HoloTC was quantified using an AxSYM autoanalyzer (Abbott, North Chicago, IL) via a chemiluminescence method, following manufacturer’s instructions. 2.9 Exclusions To ensure that all the biochemical analyses performed were from blood or serum taken at the same time-points for each subject, I chose to only include in my study 1) maternal blood that was taken at 33-36 weeks of gestation, 2) newborn cord blood taken at delivery, 3) maternal serum taken between 33 weeks of gestation and delivery, and 4) newborn serum taken at delivery. As such, I chose to exclude six maternal blood samples that were taken at time of delivery or post-delivery. A source of newborn blood for a small subset of my cohort was taken more than 24 hours after birth (from heel pricks). Therefore, these fifteen newborn blood samples were also excluded. I chose to exclude the methyl nutrient measurements of five women and three newborns, because they were taken between 24 to 48 hours post-delivery or after birth. 2.10 Statistical Analyses In order to explore my three specific aims, general linear models (GLMs) were used for both categorical and continuous variables. Details of statistical analyses are presented in 33 Chapters 3-5. Least Significant Difference (LSD) post-tests were performed to account for multiple testing. P-values less than 0.05 were chosen to be considered significant. Data was checked for normality via histograms. Given that the distribution of SLC6A4 mRNA levels and methyl nutrients had a positive skew, all analyses using mRNA levels, folate, and holoTC as dependent factors were carried out after natural logarithmic (ln) transformation. All statistical analyses were conducted using PASW Statistics, version 18.0 (IBM, Armonk, NY) for Windows. 34 3 CHAPTER 3: Mothers My first specific aim was to determine the effects of methyl nutrient status (folate, vitamin B12, HoloTC) and MTHFR C677T genotype on mood during the 3rd trimester, and to determine whether maternal mood and methyl nutrient status are associated with SLC6A4 methylation and expression during the 3rd trimester of pregnancy. This specific aim was addressed by the following questions (see Figure 3.1 for schematic representation): a. What is the relationship between maternal methyl nutrient metabolism and maternal mood during the 3rd trimester of pregnancy? b. What is the influence of mood and methyl nutrient metabolism on SLC6A4 methylation in the 3rd trimester of pregnancy? c. What are the effects of methyl nutrient status and SLC6A4 methylation on SLC6A4 mRNA levels in the 3rd trimester of pregnancy? d. Is the 5HTTLPR variant associated with depressive mood during pregnancy, SLC6A4 methylation and SLC6A4 mRNA levels? Figure 3.1 Specific aim 1. 35 3.1 Maternal Methyl Nutrient Metabolism and Mood Maternal demographic characteristics were examined according to MTHFR C677T genotype and the characteristics were found to not vary significantly between the genotypes (Table 3.1). The genotype frequencies for the MTHFR C677T variant were approximately 52.9% for CC, 33.3% for CT, and 13.8% for TT. This is similar to previous reports in Caucasian populations (11,93). Using GLM, I determined that the maternal MTHFR genotype, adjusted for SSRI usage, was not associated with maternal depressive or anxious mood (dependent variable) during the 3rd trimester of pregnancy (P>0.05). Given that early 3rd trimester EPDS scores and HAM-D scores were highly associated (P<0.001) (Figure 3.2), only EPDS scores were used in the subsequent analyses. EPDS was chosen over HAM-D because a previous study showed that the EPDS score from the 2nd trimester of pregnancy was positively associated with the MTHFR 677TT genotype (11). In order to determine whether disturbances in methyl nutrient status are associated with depressive mood during the 3rd trimester, GLM was performed with depressive mood as a continuous dependent variable. The covariates were set as maternal folate status and maternal holoTC status; the fixed factors were set as maternal MTHFR C677T variant and SSRI usage; and the dependent factor was set as mood (as represented by EPDS or HAM-A scores). Vitamin B12 levels were highly associated with holoTC levels in the mothers (P<0.001) (Figure 3.3), and holoTC has been suggested to be a more biologically useful indicator of vitamin B 12 status (137). Therefore, maternal holoTC was chosen to represent maternal vitamin B12 status in subsequent analyses. No significant associations between methyl nutrient status and mood in the 3 rd trimester were found (P>0.05). 36 A GLM was then conducted to determine whether SSRI usage (adjusted for mood) influences folate and holoTC levels in the pregnant women in their 3 rd trimester of pregnancy. No significant differences were found (P>0.05). Furthermore, the women were categorized into a “depressed” group, and a “non-depressed” group according to their early 3rd trimester EPDS score. An EPDS score of 10 or higher indicated depressive mood, and 9 or lower indicated nondepressive mood. No differences in methyl nutrient status were detected between the two mood groups, with or without adjustments for SSRI usage (P>0.05). 37 Table 3.1 Maternal MTHFR C677T genotype and demographic data of the pregnant women. Maternal Characteristics Maternal MTHFR C677T Genotype CC (n=46) CT (n=29) TT (n=12) Maternal age at birth, years (SD) 33.7 (5.7) 34.1 (4.9) 34 (6.2) Maternal education, years (SD) 17.7 (3.5) 18 (3.4) 17.9 (4.6) Delivery type, % caesarian-section 20 24 25 SRI treated during pregnancy, % 32 48 25 Maternal alcohol consumption in single drinks during whole pregnancy 5.5 2.5 9.5 Tobacco Use, % 4 4 0 SD = Standard Deviation 38 Figure 3.2 Relationship between early 3rd trimester EPDS scores and HAM-D scores. R2 = 0.657, P<0.001 Figure 3.3 Relationship between maternal vitamin B12 and holoTC status. R2 = 0.787, P< 0.001 39 3.2 Maternal Methyl Nutrient Metabolism, Mood, and Methylation GLM was used to determine if mood and methyl nutrient metabolism affect SLC6A4 methylation in the 3rd trimester of pregnancy. The dependent factor was SLC6A4 methylation at specific CpG sites. The covariates were set as maternal mood (either EPDS scores or HAM-A scores), maternal folate levels, maternal holoTC levels, and maternal age at delivery. The fixed factors were set as maternal MTHFR C677T genotype and SSRI usage. Serum folate levels were associated with SLC6A4 methylation at CpGs 1, 4 and 8 (P<0.05) (Tables 3.2 and 3.3). In order to display the data in a more biologically significant manner, I multiplied the estimated increase in percent SLC6A4 methylation (β value) and its 95% confidence interval (CI) by five, to show the percent methylation increase per 5 nmol/L increase in folate levels. When adjusted for EPDS scores, every 5 nmol/L increase in folate level was associated with 0.967, 0.390, and 0.477 % increase in SLC6A4 methylation at CpG sites 1, 4, and 8, respectively (P≤0.05) (Table 3.2). When adjusted for HAM-A scores, every 5 nmol/L increase in folate level was associated with 0.928, 0.414, and 0.453 % increase in SLC6A4 methylation at CpG sites 1, 4, and 8, respectively (P<0.05) (Table 3.3). Maternal 3rd trimester EPDS predicted a decrease in SLC6A4 methylation at CpG 10, such that every unit increase in EPDS was associated with 0.163 % decrease in methylation (P<0.05) (Table 3.4). Maternal 3rd trimester HAM-A also predicted a decrease in SLC6A4 methylation at CpG 10, such that every unit increase in HAM-A was associated with 0.170 % decrease in methylation (P<0.05) (Table 3.5). 40 Table 3.2 Relationship between maternal folate levels and maternal SLC6A4 methylation (adjusted for EPDS scores). Estimated difference in percent SLC6A4 CpG Methylation (β) Variable Serum folate level (per 5 nmol/L increase) 95% CI 0.280 P Partial eta squared (ƞ2) CpG 1 0.967 ( , 1.655 ) 0.007 0.133 CpG 2 0.153 ( -0.125 , 0.430 ) 0.276 0.023 CpG 3 0.348 ( -0.495 , 1.190 ) 0.412 0.013 CpG 4 0.390 ( , 0.780 ) 0.050 0.072 CpG 5 0.150 ( -0.330 , 0.630 ) 0.535 0.007 CpG 6 -0.002 ( -0.445 , 0.445 ) 0.992 0.000 CpG 7 -0.259 ( -0.990 , 0.475 ) 0.482 0.010 CpG 8 0.477 ( , 0.875 ) 0.019 0.101 CpG 9 -0.285 ( -1.000 , 0.430 ) 0.428 0.012 CpG 10 0.135 ( -0.520 , 0.790 ) 0.681 0.003 Mean CpG 0.207 ( -0.050 , 0.460 ) 0.109 0.049 0.000 0.080 CI; Confidence Interval Estimates adjusted for EPDS score, maternal holoTC levels, maternal age at delivery, maternal MTHFR C677T genotype, and SSRI exposure. 41 Table 3.3 Relationship between maternal folate levels and maternal SLC6A4 methylation (adjusted for HAM-A scores). Estimated difference in percent SLC6A4 CpG Methylation (β) Variable Serum folate level (per 5 nmol/L increase) 95% CI 0.240 P Partial eta squared (ƞ2) CpG 1 0.928 ( , 1.620 ) 0.009 0.123 CpG 2 0.160 ( -0.120 , 0.440 ) 0.260 0.024 CpG 3 0.294 ( -0.540 , 1.125 ) 0.482 0.010 CpG 4 0.414 ( , 0.810 ) 0.040 0.078 CpG 5 0.144 ( -0.350 , 0.635 ) 0.560 0.007 CpG 6 -0.006 ( -0.455 , 0.445 ) 0.980 0.000 CpG 7 -0.259 ( -1.005 , 0.485 ) 0.488 0.009 CpG 8 0.453 ( , 0.860 ) 0.029 0.089 CpG 9 -0.253 ( -0.995 , 0.490 ) 0.498 0.009 CpG 10 0.043 ( -0.600 , 0.685 ) 0.894 0.000 Mean CpG 0.192 ( -0.060 , 0.445 ) 0.134 0.043 0.020 0.050 CI; Confidence Interval Estimates adjusted for HAM-A score, maternal holoTC levels, maternal age at delivery, maternal MTHFR C677T genotype, and SSRI exposure. 42 Table 3.4 Relationship between maternal 3rd trimester EPDS scores and maternal SLC6A4 methylation. Estimated difference in percent SLC6A4 CpG Methylation (β) Variable EPDS score (per unit increase) 95% CI P Partial eta squared (ƞ2) CpG 1 -0.079 ( -0.245 , 0.087 ) 0.345 0.017 CpG 2 0.029 ( -0.038 , 0.097 ) 0.387 0.014 CpG 3 -0.017 ( -0.221 , 0.187 ) 0.871 0.001 CpG 4 0.066 ( -0.029 , 0.160 ) 0.170 0.036 CpG 5 -0.078 ( -0.194 , 0.039 ) 0.186 0.033 CpG 6 -0.031 ( -0.138 , 0.077 ) 0.570 0.006 CpG 7 0.105 ( -0.072 , 0.283 ) 0.239 0.027 CpG 8 -0.079 ( -0.175 , 0.017 ) 0.103 0.050 CpG 9 0.170 ( -0.003 , 0.343 ) 0.054 0.069 CpG 10 -0.163 ( -0.321 , -0.004 ) 0.045 0.075 Mean CpG -0.008 ( -0.069 , 0.807 0.001 0.054 ) CI; Confidence Interval Estimates adjusted for maternal folate levels, maternal holoTC levels, maternal age at delivery, maternal MTHFR C677T genotype, and SSRI exposure. 43 Table 3.5 Relationship between maternal 3rd trimester HAM-A scores and maternal SLC6A4 methylation. Estimated difference in percent SLC6A4 CpG Methylation (β) Variable HAM-A score (per unit increase) 95% CI P Partial eta squared (ƞ2) CpG 1 -0.069 ( -0.207 , 0.068 ) 0.315 0.019 CpG 2 0.010 ( -0.046 , 0.066 ) 0.729 0.002 CpG 3 -0.111 ( -0.277 , 0.055 ) 0.186 0.033 CpG 4 0.042 ( -0.037 , 0.121 ) 0.290 0.022 CpG 5 -0.001 ( -0.099 , 0.097 ) 0.984 0.000 CpG 6 -0.003 ( -0.092 , 0.087 ) 0.953 0.000 CpG 7 -0.017 ( -0.166 , 0.131 ) 0.816 0.001 CpG 8 -0.039 ( -0.119 , 0.042 ) 0.341 0.017 CpG 9 0.043 ( -0.105 , 0.191 ) 0.559 0.007 CpG 10 -0.170 ( -0.298 , -0.042 ) 0.010 0.120 Mean CpG -0.031 ( -0.082 , 0.216 0.029 0.019 ) CI; Confidence Interval Estimates adjusted for maternal folate levels, maternal holoTC levels, maternal age at delivery, maternal MTHFR C677T genotype, and SSRI exposure. 44 3.3 Maternal Methyl Nutrient Metabolism, Methylation, and Gene Expression The effects of methyl nutrient status and SLC6A4 methylation on SLC6A4 mRNA levels in the 3rd trimester of pregnancy were determined. GLMs were conducted to determine whether maternal folate levels, maternal holoTC levels, maternal MTHFR C677T genotype, mood during the 3rd trimester, and SSRI usage were independently associated with mRNA levels. I did not find significant differences in any of the above (P>0.05). I also used GLMs to determine whether SLC6A4 methylation at each of the CpG sites is associated with SLC6A4 mRNA levels. For every percent increase in mean CpG methylation and methylation at CpG sites 5, 7, 8, 10, a unit increase of 0.221, 0.126, 0.059, 0.128, and 0.098 in SLC6A4 expression was detected (respectively) (P<0.05) (Table 3.6). To take this a step further, I conducted a GLM with SLC6A4 mRNA level as the dependent factor. The covariates were set for site-specific maternal SLC6A4 methylation, maternal folate levels, maternal holoTC levels, and maternal age at delivery. The fixed factor was maternal MTHFR C677T genotype. After adjusting for methyl nutrient status, I found that the mean CpG methylation and methylation at CpG sites 5, 7, and 10 predicted increases in SLC6A4 mRNA levels (P<0.05) (Table 3.7). Every percent increase in the mean level of SLC6A4 methylation and methylation at CpGs 5, 7, and 10 was associated with 0.221, 0.115, 0.065, and 0.084 unit increase in mRNA expression. No associations were found between folate, holoTC, MTHFR C677T genotype, or maternal age and SLC6A4 mRNA levels. 45 Table 3.6 Estimated difference in maternal SLC6A4 gene expression according to maternal SLC6A4 methylation sites (unadjusted). Unadjusted Variable Estimated difference in SLC6A4 gene expression (β) CpG 1 -0.005 ( -0.067 , 0.058 CpG 2 0.035 ( -0.122 , CpG 3 0.025 CpG 4 P Partial eta squared (ƞ2) ) 0.877 0.000 0.192 ) 0.657 0.004 ( -0.030 , 0.080 ) 0.364 0.016 -0.072 ( -0.182 , 0.037 ) 0.189 0.033 CpG 5 0.126 ( , 0.214 ) 0.006 0.137 CpG 6 0.075 ( -0.024 , 0.175 ) 0.136 0.042 CpG 7 0.059 ( 0.001 , 0.117 ) 0.047 0.073 CpG 8 0.128 ( 0.027 , 0.229 ) 0.014 0.111 CpG 9 0.015 ( -0.046 , 0.075 ) 0.627 0.005 CpG 10 0.098 ( 0.035 , 0.160 ) 0.003 0.160 Mean CpG 0.221 ( 0.062 , 0.380 ) 0.007 0.130 95% CI 0.038 CI; Confidence Interval. β and 95% CI represent ln-transformed SLC6A4 expression values. 46 Table 3.7 Estimated difference in maternal SLC6A4 gene expression according to maternal SLC6A4 methylation sites (adjusted). Adjusted Variable Estimated difference in SLC6A4 gene expression (β) CpG 1 -0.005 ( -0.074 CpG 2 0.077 CpG 3 P Partial eta squared (ƞ2) , 0.065 ) 0.897 0.000 ( -0.091 , 0.244 ) 0.361 0.018 0.030 ( -0.027 , 0.087 ) 0.290 0.024 CpG 4 -0.090 ( -0.205 , 0.025 ) 0.124 0.050 CpG 5 0.115 ( 0.023 , 0.206 ) 0.015 0.119 CpG 6 0.090 ( -0.013 , 0.193 ) 0.085 0.062 CpG 7 0.065 ( 0.007 , 0.124 ) 0.030 0.096 CpG 8 0.108 ( -0.006 , 0.221 ) 0.062 0.072 CpG 9 0.010 ( -0.053 , 0.073 ) 0.753 0.002 CpG 10 0.084 ( 0.019 , 0.150 ) 0.012 0.126 Mean CpG 0.221 ( 0.054 , 0.387 ) 0.010 0.131 95% CI CI; Confidence Interval. Estimates adjusted for maternal folate levels, maternal holoTC levels, maternal age at delivery, and maternal MTHFR C677T genotype. β and 95% CI represent ln-transformed SLC6A4 expression values. 47 3.4 5HTTLPR The genotype frequencies for the maternal 5HTTLPR insertion/deletion variant in this study population were determined to be 37.0% for l/l, 49.3% for l/s, and 13.7% for s/s, which is similar to previous reports in non-pregnant women and men (30). To determine whether the maternal 5HTTLPR insertion/deletion variant is associated with maternal SLC6A4 mRNA levels, a GLM was performed with mRNA levels as the dependent variable. SLC6A4 mRNA levels did not vary with 5HTTLPR insertion/deletion genotype (P>0.05). Unfortunately the effect of the 5HTTLPR A→G variant on SLC6A4 mRNA levels could not be examined in my study population due to small sample size (n=68). The distribution of the 5HTTLPR A→G genotypes amongst the pregnant women was 20.6% La/La, 8.8% La/Lg, 45.6% La/S, and 25% S/S. Also, no women in my cohort had the Lg/Lg or Lg/S genotypes. In order to determine whether the 5HTTLPR insertion/deletion variant is associated with SLC6A4 methylation in the 3rd trimester of pregnancy, I conducted a GLM with the maternal 5HTTLPR insertion/deletion variant as the fixed factor, and maternal SLC6A4 CpG sites as dependent variables. I did not detect any associations between the 5HTTLPR genotype and methylation at each CpG site (P>0.05). Furthermore, I conducted a GLM with the maternal 5HTTLPR insertion/deletion variant as the fixed factor, and the mood score (either EPDS or HAM-A) as the dependent variable. I also adjusted for SSRI usage. No relationships were found between 3rd trimester EPDS or HAMA scores and the 5HTTLPR insertion/deletion variant (P>0.05). 48 4 CHAPTER 4: Maternal Influences on Newborns My second specific aim was to determine whether prenatal exposure to maternal methyl nutrient status (folate, vitamin B12, HoloTC, and MTHFR C677T genotype) and maternal mood are associated with SLC6A4 methylation and expression in newborns. This specific aim was addressed by the following questions (see Figure 4.1 for schematic representation): a. What are the effects of maternal late gestation depressive mood and maternal methyl nutrient status on SLC6A4 methylation in newborns? b. What is the relationship between maternal methyl nutrient status and SLC6A4 mRNA levels in newborns? c. What is the relationship between maternal mood and SLC6A4 mRNA levels in newborns? Figure 4.1 Specific aim 2. 49 4.1 Maternal Methyl Nutrient Metabolism, Maternal Mood, and Newborn Methylation Newborn clinical characteristics were examined according to their mothers’ MTHFR C677T genotype, and these clinical characteristics did not vary significantly with maternal MTHFR C677T genotype (Table 4.1). In addition, maternal SLC6A4 methylation was not associated with newborn SLC6A4 methylation (P>0.05). To determine the association between maternal late gestation depressive mood and maternal methyl nutrient status on SLC6A4 methylation in newborns, I conducted a GLM. The covariates were set as maternal folate status, maternal holoTC status, maternal age at delivery, and mood. The fixed factors were set as maternal MTHFR C677T genotype, sex of newborn and SSRI exposure. The dependent factor was set for the methylation of each individual newborn SLC6A4 CpG site. As shown in Table 4.2, after adjustments for maternal methyl nutrient metabolism (folate, holoTC, and MTHFR genotype), maternal age at delivery, SSRI exposure and EPDS scores, SLC6A4 methylation at CpG 8 was lower in newborns with mothers with the MTHFR 677TT genotype, but only compared to newborns with mothers with the CT genotype (P<0.05). Similar results were obtained after adjusting for maternal methyl nutrient metabolism, maternal age at delivery, SSRI exposure and HAM-A scores, such that percent SLC6A4 methylation at CpG 8 was lower in newborns with mothers carrying the MTHFR 677TT genotype, but only compared to newborns with mothers carrying the CT genotype (P<0.05) (Table 4.3). I also found that prenatal exposure to maternal 3rd trimester EPDS scores predicted a decrease in newborn SLC6A4 methylation at CpG 10, such that every unit increase in EPDS score was associated with a 0.254 % decrease in methylation (P<0.05) (Table 4.4). 50 Table 4.1 Maternal MTHFR C677T genotype and clinical data of their newborns. Newborn Characteristics Maternal MTHFR C677T Genotype CC (n=46) CT (n=29) TT (n=12) Prenatal SSRI Exposure, days (SD)* 248.1 (60) 257.9 (47.9) 223.3 (101.6) Birth weight, g (SD) 3404 (462) 3434 (602) 3400 (371) Head Circumference, cm (SD) 34.9 (1.4) 34.3 (1.4) 34.7 (1.3) Birth length, cm (SD) 51.4 (2.4) 51.1 (2.7) 50.6 (2.2) Gestational age at birth, weeks (SD) 39.5 (1.6) 39.2 (1.7) 39.6 (1.2) Sex, % Males 57 55 50 Apgar score at 1 minute (SD) 7.9 (1.6) 8.2 (1.4) 8.6 (1.0) Apgar score at 5 minute (SD) 9.0 (0.4) 9.0 (0.5) 9.0 (0.4) * Only amongst newborns who were exposed to SSRIs 51 Table 4.2 Relationship between maternal MTHFR C677T genotype and newborn SLC6A4 methylation (adjusted for EPDS scores). Newborn SLC6A4 Methylation (%) Maternal MTHFR Genotype CC (n = 31) CT (n = 22) TT (n = 7) CpG 1 4.87 ± 0.32 4.55 ± 0.40 4.47 ± 0.63 CpG 2 1.70 ± 0.24 1.91 ± 0.31 2.10 ± 0.49 CpG 3 4.82 ± 0.36 4.02 ± 0.45 4.69 ± 0.71 CpG 4 1.49 ± 0.24 1.88 ± 0.31 1.11 ± 0.48 CpG 5 6.00 ± 0.21 6.14 ± 0.26 5.40 ± 0.41 CpG 6 2.12 ± 0.22 1.72 ± 0.28 1.17 ± 0.44 CpG 7 3.38 ± 0.37 3.25 ± 0.47 4.18 ± 0.74 CpG 8 6.40 ± 0.24 6.82 ± 0.31 5.66 ± 0.48* CpG 9 4.71 ± 0.37 4.74 ± 0.47 4.46 ± 0.74 CpG 10 10.00 ± 0.56 10.01 ± 0.71 9.89 ± 1.11 Mean CpG 4.55 ± 0.13 4.51 ± 0.16 4.31 ± 0.25 * P<0.05 as compared to the CT genotype. Data represent mean ± Standard Error of the Mean (SEM). Estimated means are adjusted for maternal folate levels, maternal holoTC levels, maternal age at delivery, EPDS mood score, sex of newborn, and SSRI exposure. Corrected for multiple comparisons using LSD. 52 Table 4.3 Relationship between maternal MTHFR C677T genotype and newborn SLC6A4 methylation (adjusted for HAM-A scores). Newborn SLC6A4 Methylation (%) Maternal MTHFR Genotype CC (n = 31) CT (n = 22) TT (n = 7) CpG 1 4.88 ± 0.32 4.56 ± 0.41 4.43 ± 0.63 CpG 2 1.71 ± 0.25 1.88 ± 0.31 2.15 ± 0.49 CpG 3 4.90 ± 0.36 4.01 ± 0.45 4.62 ± 0.71 CpG 4 1.51 ± 0.24 1.88 ± 0.30 1.07 ± 0.47 CpG 5 5.99 ± 0.21 6.16 ± 0.27 5.38 ± 0.41 CpG 6 2.07 ± 0.22 1.75 ± 0.27 1.18 ± 0.42 CpG 7 3.38 ± 0.37 3.25 ± 0.47 4.17 ± 0.73 CpG 8 6.40 ± 0.24 6.83 ± 0.31 5.65 ± 0.48* CpG 9 4.64 ± 0.38 4.73 ± 0.48 4.56 ± 0.75 CpG 10 10.17 ± 0.59 10.03 ± 0.74 9.64 ± 1.15 Mean CpG 4.56 ± 0.13 4.51 ± 0.16 4.28 ± 0.25 * P<0.05 as compared to the CT genotype. Data represent mean ± SEM. Estimated means are adjusted for maternal folate levels, maternal holoTC levels, maternal age at delivery, HAM-A mood score, sex of newborn, and SSRI exposure. Corrected for multiple comparisons using LSD. 53 Table 4.4 Relationship between prenatal exposure to maternal 3rd trimester mood and newborn SLC6A4 methylation. Estimated difference in percent SLC6A4 CpG Methylation (β) Variable EPDS score (per unit increase) 95% CI P Partial eta squared (ƞ2) CpG 1 -0.043 ( -0.157 , 0.071 ) 0.451 0.012 CpG 2 0.039 ( -0.049 , 0.126 ) 0.377 0.017 CpG 3 -0.082 ( -0.210 , 0.046 ) 0.204 0.035 CpG 4 -0.038 ( -0.123 , 0.047 ) 0.375 0.017 CpG 5 -0.017 ( -0.091 , 0.058 ) 0.657 0.004 CpG 6 0.025 ( -0.054 , 0.103 ) 0.529 0.009 CpG 7 -0.006 ( -0.139 , 0.127 ) 0.927 0.000 CpG 8 -0.004 ( -0.090 , 0.083 ) 0.935 0.000 CpG 9 0.095 ( -0.038 , 0.228 ) 0.157 0.043 CpG 10 -0.254 ( -0.454 , -0.054 ) 0.014 0.124 Mean CpG -0.028 ( -0.074 , 0.017 0.215 0.033 ) CI; Confidence Interval Estimates adjusted for maternal folate levels, maternal holoTC levels, maternal age at delivery, maternal MTHFR C677T genotype, sex of newborn, and SSRI exposure. 54 4.2 Maternal Methyl Nutrient Metabolism and Newborn Gene Expression The relationship between maternal methyl nutrient status and SLC6A4 mRNA levels in newborn whole blood was determined. I conducted a GLM and the covariates were set as maternal folate levels and maternal holoTC levels. The fixed factors were set as maternal MTHFR C677T genotype and sex of newborn. The dependent factor was set as newborn SLC6A4 mRNA levels. No associations between maternal methyl nutrient status and SLC6A4 mRNA levels in cord blood from newborns were detected (P>0.05). 4.3 Maternal Mood and Newborn Gene Expression Prenatal exposure to maternal depressive mood and SSRIs did not directly influence SLC6A4 mRNA levels in newborns (P>0.05). This was shown by a GLM with the dependent variable set as newborn SLC6A4 mRNA levels, the covariate set as mood (EPDS or HAM-A), and the fixed factor set as SSRI exposure. I also conducted a GLM wherein newborn gene expression was categorically grouped by mothers who were depressed during pregnancy or mothers who were not depressed during pregnancy (EPDS scores of 10 or more, and EPDS scores of 9 or less, respectively). I further adjusted for SSRI exposure in this analysis, and found no associations (P>0.05). 55 5 CHAPTER 5: Newborns My third specific aim was to determine whether newborn methyl nutrient metabolism (folate, holoTC and MTHFR C677T genotype) is associated with SLC6A4 methylation and mRNA levels in cord blood from newborns. This specific aim was addressed by the following questions (see Figure 5.1 for schematic representation): a. Is there an association between maternal methyl nutrient status and newborn methyl nutrient status? b. What is the relationship between newborn methyl nutrient status and methylation of SLC6A4 in newborns? c. What are the effects of newborn methyl nutrient metabolism and SLC6A4 methylation on SLC6A4 mRNA levels in newborns? d. Is the 5HTTLPR variant in newborns associated with newborn SLC6A4 methylation and mRNA levels? Figure 5.1 Specific aim 3. 56 5.1 Maternal and Newborn Methyl Nutrients The methyl nutrient status of mothers at delivery and their newborns at delivery are shown in Table 5.1. None of the women in my study population were deficient in folate at delivery (defined as <6 nmol/L). Although, 29.2% of my cohort of women were considered deficient in vitamin B12 (defined as <150 pmol/L), only 2.8% of my cohort were deficient in holoTC (defined as <35 pmol/L). I found no associations between maternal and newborn folate levels (P>0.05) (Figure 5.2A). Maternal vitamin B12 and holoTC levels were closely associated with newborn vitamin B12 and holoTC levels, respectively (P<0.001 and P=0.001) (Figure 5.2B and Figure 5.2C). Newborn vitamin B12 levels were also highly associated with newborn holoTC levels (P<0.001) (Figure 5.3). Therefore, newborn holoTC status was chosen to represent newborn vitamin B12 status in subsequent analyses. 57 Table 5.1 Methyl nutrient status in mothers and newborns. n Median (1st Quartile, 3rd Quartile) Minimum Level Maximum Level Maternal Folate (nmol/L) 72 37.4 (35.8, 39.4) 17.5 65.0 Maternal HoloTC (pmol/L) 72 78.8 (56.6, 99.8) 26.6 664.9 Maternal Vitamin B12 (pmol/L) 72 197.9 (139.7, 268.5) 59.4 2003.7 Newborn Folate (nmol/L) 69 37.8 (36.5, 39.6) 32.4 56.6 Newborn HoloTC (pmol/L) 69 179.2 (117.8, 279.4) 32.5 647.8 Newborn Vitamin B12 (pmol/L) 69 390.4 (253.9, 592.5) 73.4 2352.2 58 A) B) C) Figure 5.2 Relationships between maternal and newborn methyl nutrient levels. A) Folate; R2 = 0.039, P> 0.05. B) Vitamin B12; R2 = 0.591, P<0.001. C) HoloTC; R2 = 0.429, P= 0.001. Figure 5.3 Relationship between newborn vitamin B12 and holoTC levels. R2 = 0.647, P< 0.001. 59 5.2 Newborn Methyl Nutrient Metabolism and DNA Methylation A GLM was used to determine the relationship between newborn methyl nutrient metabolism (folate, holoTC, and MTHFR genotype) and newborn SLC6A4 methylation. The covariates were set as newborn folate levels and newborn holoTC levels. The fixed factors were set as newborn MTHFR C677T genotype and sex of newborn. The dependent factor was set as methylation at each newborn SLC6A4 CpG site. As shown in Table 5.2, newborn MTHFR C677T genotype was associated with percent SLC6A4 methylation at CpGs 6 and 10 (P<0.05). Specifically, newborns with the TT genotype had lower SLC6A4 methylation at CpG 6 compared to newborns with the CC and CT genotypes, and lower methylation at CpG 10 compared to newborns with the CT genotype. 60 Table 5.2 Relationship between newborn MTHFR C677T genotype and newborn SLC6A4 methylation. Newborn SLC6A4 Methylation (%) Newborn MTHFR Genotype CC (n = 26) CT (n = 24) TT (n = 11) CpG 1 4.40 ± 0.31 4.64 ± 0.32 5.37 ± 0.47 CpG 2 1.86 ± 0.26 1.27 ± 0.27 1.63 ± 0.39 CpG 3 4.75 ± 0.40 4.61 ± 0.41 4.05 ± 0.61 CpG 4 1.69 ± 0.25 1.46 ± 0.26 1.94 ± 0.38 CpG 5 5.98 ± 0.23 6.13 ± 0.24 5.58 ± 0.36 CpG 6 1.82 ± 0.21 2.21 ± 0.22 0.97 ± 0.32*,# CpG 7 3.38 ± 0.39 3.30 ± 0.41 3.79 ± 0.60 CpG 8 6.29 ± 0.26 6.64 ± 0.27 5.97 ± 0.40 CpG 9 4.59 ± 0.38 5.05 ± 0.39 5.13 ± 0.57 CpG 10 10.17 ± 0.58 10.74 ± 0.60 8.15 ± 0.88* Mean CpG 4.50 ± 0.13 4.61 ± 0.13 4.26 ± 0.20 * P<0.05 as compared to CT genotype. # P<0.05 as compared to CC genotype. Data represent mean ± SEM. Estimated means are adjusted for newborn folate, newborn holoTC, and sex of newborn. Corrected for multiple comparisons using LSD. 61 5.3 Newborn Methyl Nutrient Metabolism, DNA Methylation, and Gene Expression To examine the effects of newborn methyl nutrient metabolism and SLC6A4 methylation on SLC6A4 mRNA levels, I first conducted GLMs to determine whether newborn folate levels, newborn holoTC levels, newborn MTHFR C677T genotype, and sex of newborn were independently associated with newborn SLC6A4 mRNA level. I did not find significant differences in any of the above (P>0.05). Next, I conducted GLMs to determine whether newborn SLC6A4 methylation at each of the CpG sites was associated with newborn SLC6A4 expression. I found that for every percent increase in methylation at CpG site 7, there was a decrease in 0.189 unit of newborn SLC6A4 mRNA level (P<0.05) (Table 5.3). Similarly, for every percent increase in methylation at CpG site 4, there was a 0.207 unit decrease in newborn SLC6A4 mRNA level (P<0.05) (Table 5.3). Further, I conducted a GLM in order to take into account the effects of newborn methyl nutrient metabolism (folate, holoTC, and MTHFR genotype), sex of newborn and newborn SLC6A4 methylation on gene expression. The model had site-specific SLC6A4 methylation, newborn folate levels, and newborn holoTC levels as covariates, and had newborn MTHFR C677T genotype and sex of newborn as fixed factors. SLC6A4 mRNA level in newborns was set as the dependent factor. As shown in Table 5.4, CpG site 7 predicted a decrease in SLC6A4 methylation, such that with every percent increase in SLC6A4 methylation at CpG 7, there was a 0.155 unit decrease in SLC6A4 gene expression (P<0.05). 62 Table 5.3 Relationship between newborn SLC6A4 methylation and newborn SLC6A4 expression (unadjusted). Unadjusted Variable Estimated difference in SLC6A4 gene expression (β) P Partial eta squared (ƞ2) CpG 1 0.099 ( -0.076 , 0.274 ) 0.262 0.026 CpG 2 0.079 ( -0.127 , 0.286 ) 0.443 0.012 CpG 3 0.092 ( -0.042 , 0.226 ) 0.175 0.037 CpG 4 -0.207 ( -0.409 , -0.005 ) 0.045 0.080 CpG 5 0.076 ( -0.159 , 0.311 ) 0.519 0.009 CpG 6 0.027 ( -0.230 , 0.284 ) 0.831 0.001 CpG 7 -0.189 ( -0.310 , -0.068 ) 0.003 0.167 CpG 8 0.104 ( -0.110 , 0.317 ) 0.334 0.019 CpG 9 -0.086 ( -0.230 , 0.058 ) 0.234 0.029 CpG 10 0.067 ( -0.037 , 0.170 ) 0.200 0.033 Mean CpG 0.009 ( -0.419 , 0.437 ) 0.967 0.000 95% CI CI; Confidence Interval. β and 95% CI represent ln-transformed SLC6A4 expression values. 63 Table 5.4 Relationship between newborn SLC6A4 methylation and newborn SLC6A4 expression (adjusted). Estimated difference in Variable SLC6A4 gene expression (β) Adjusted 95% CI P Partial eta squared (ƞ2) CpG 1 0.092 ( -0.106 , 0.289 ) 0.355 0.020 CpG 2 0.037 ( -0.173 , 0.247 ) 0.722 0.003 CpG 3 0.047 ( -0.098 , 0.192 ) 0.514 0.010 CpG 4 -0.126 ( -0.346 , 0.094 ) 0.256 0.031 CpG 5 0.006 ( -0.231 , 0.243 ) 0.961 0.000 CpG 6 -0.009 ( -0.275 , 0.258 ) 0.947 0.000 CpG 7 -0.155 ( -0.283 , -0.027 ) 0.019 0.125 CpG 8 0.108 ( -0.104 , 0.320 ) 0.308 0.025 CpG 9 -0.007 ( -0.162 , 0.147 ) 0.923 0.000 CpG 10 0.057 ( -0.047 , 0.160 ) 0.276 0.028 Mean CpG 0.030 ( -0.394 , 0.454 ) 0.886 0.000 CI; Confidence Interval. Estimates adjusted for newborn folate, newborn holoTC, newborn MTHFR C677T genotype, and sex of newborn. β and 95% CI represent ln-transformed SLC6A4 expression values. 64 5.4 5HTTLPR Lastly, to determine whether the 5HTTLPR insertion/deletion variant in newborns is associated with newborn SLC6A4 mRNA levels, a GLM was performed with mRNA levels as the dependent variable. The newborn 5HTTLPR variant was set as the categorical variable and newborn mRNA level was set as the dependent factor. Newborn SLC6A4 mRNA level did not vary with newborn 5HTTLPR genotype (P>0.05). The Genotype frequencies for the newborn 5HTTLPR insertion/deletion variant in my study population were 30.4% for l/l, 45.7% for l/s, and 16.3% for s/s, similar to that previous reported in adults (30). SLC6A4 mRNA levels were not grouped by the newborn 5HTTLPR A→G variant due to small sample size (n=61). The distribution of the 5HTTLPR A→G genotypes amongst the newborns was 27.9% La/La, 9.8% La/Lg, 41.0% La/S, 1.6% Lg/S, and 19.7% S/S. In my cohort, no newborns had the Lg/Lg genotypes. In order to determine whether the 5HTTLPR insertion/deletion variant is associated with SLC6A4 methylation in newborns, I conducted a GLM with the newborn 5HTTLPR insertion/deletion variant as the fixed factor, and newborn SLC6A4 CpG sites as dependent variables. I did not detect any associations between the 5HTTLPR genotype and methylation at each CpG sites in newborns (P>0.05). 65 6 CHAPTER 6: General Discussion 6.1 Discussion of Results 6.1.1 Methyl Nutrient Status Results from this thesis indicated that the MTHFR 677TT genotype was not associated with increased depressive or anxious mood during the 3rd trimester. This is surprising given that previous findings showed that the MTHFR C677T variant is associated with depression not only in the general population, but also in pregnant women in their 2nd trimester (Devlin et al.) (11,101). An explanation for this may simply be small sample size (n=83). Another explanation may involve gene-environment interactions. The function of the MTHFR enzyme is to reduce various forms of folate to 5-MTHF. Given that the MTHFR 677TT variant leads to reduced enzymatic activity (93), less folate is reduced to the 5-MTHF form in individuals with the variant. However, in a population with high levels of folate such as my study population, the reduced MTHFR activity may be overcome and adequate levels of 5-MTHF may be produced. In addition, MTHFR activity is dependent on riboflavin as a cofactor (62). A study had shown that riboflavin intake is inversely associated with colorectal adenomas, and the association was more pronounced in adults with the MTHFR 677TT genotype than the 677CC or 677CT genotypes (138). Elderly females with low levels of riboflavin and the MTHFR 677TT genotype had higher risks for fractures compared to elderly females with low levels of riboflavin and the MTHFR 677CC genotype (139). The inverse association between plasma riboflavin and Hcy was much more significant in men with the MTHFR 677TT genotype than the 677CC genotype (140). On the contrary, high levels of plasma riboflavin (as a result of supplementation) had been shown to increase the amount of 5-MTHF in mucosal cells of individuals (with colorectal polyps) with the MTHFR 677CT or 677TT genotypes (141). These studies suggest that 66 riboflavin interacts with the MTHFR C677T genotype, such that the presence of low riboflavin levels may exacerbate the effects of the MTHFR C677T variant, and high riboflavin levels may enhance MTHFR activity in individuals with the MTHFR C677T variant. I suspect that riboflavin status in my cohort of pregnant women may influence any associations between the MTHFR C677T variant and mood. However, riboflavin was not quantified in my study population. I do not expect that riboflavin will be low in my study population as riboflavin is likely included in prenatal supplements (Materna contains 1.4mg) and may be found to be fortified in cereals, rice, cornmeal, and flour (142). Even though the cohort of pregnant women from Devlin et al. were also taking prenatal supplements containing folic acid (11), serum methyl nutrients such as folate or riboflavin were not quantified in that study population. My study is the first to look at whether folate and vitamin B12 status are associated with depression during late pregnancy. Folate was not found to be different between mothers who were depressed (EPDS score of 10 or above) during pregnancy or not depressed (EPDS score of 9 or below) during pregnancy. This is not surprising, because as a consequence of folic acid supplementation and the usage of prenatal supplements (such as “Materna”) by all the women, the women in my cohort all had similar but elevated folate status (median = 37.4 nmol/L) at delivery. Thus, this did not allow me to determine any associations of depression or anxiety in pregnant women with low folate status. It may be interesting to conduct a similar study in a population with no government mandated folic acid fortification, such as Malaysia or New Zealand. Furthermore, holoTC was not found to be different between mothers who were depressed during pregnancy or not depressed during pregnancy. This is not what I expected, given that previous studies in the elderly have shown that vitamin B12 is inversely associated with 67 depressive mood (106,117) However, the elderly are at greater risk for vitamin B12 deficiency due to malabsorption of vitamin B12 in the intestine (87). Similar to my results, a recent Japanese study (n = 86) also found no relationships between depression during the 1st trimester of pregnancy and either serum folate levels or vitamin B12 intake levels (143). I expected that the vitamin B12 status in my cohort of pregnant women to be low, as a study had shown previously that Ontario women in early gestation had low vitamin B12 status (geometric mean = 249 pmol/L; n = 10,622) (110). The vitamin B12 level in my current cohort of women (median = 198 pmol/L) was lower than the pregnant women from Ontario (110), a group of non-pregnant 18-25 year old women from Manitoba (median = 400 pmol/L; n = 95) (144), and populations of non-pregnant women of reproductive age in Vietnam (mean = 494 pmol/L; n = 245) (145) and the Netherlands (median = 245 pmol/L; n = 53) (146). A closer look at my data indicated that almost 30% of women were deficient in vitamin B12 during delivery. The vitamin B12 status of the women in my study at delivery was highly associated with their holoTC status. However, I found that less than 3% were deficient in holoTC. The women in my cohort had similar holoTC levels (median = 79 pmol/L) compared to pregnant women in Ontario (geometric mean = 81 pmol/L) (114), non-pregnant women of reproductive age in Vietnam (mean = 78 pmol/L) (145), and Latino elderly living in California (median = 80 pmol/L; n = 1209) (104). This discrepancy in the percent of women deficient in vitamin B12 and percent deficient in holoTC highlights the fact that vitamin B12 may be preferentially metabolized to holoTC to ensure adequate levels of the active form of vitamin B12 required by cells. A study had shown that 9-80% of the total serum vitamin B12 is in the form of holoTC in the elderly (147). This proportion was dependent on vitamin B12 status, with higher percentage of vitamin B12 in the holoTC form when vitamin B12 levels are low (148). 68 My results showed that newborns had similar and even higher levels of folate, vitamin B12, and holoTC than their mothers at delivery. This provides support that there may be preferential transfer of folate and vitamin B12 to the fetus during pregnancy (78,89). A preferential transfer may also explain the slightly lower levels of vitamin B12 in my group of pregnant women compared to various non-pregnant populations (144-146). A recent study from Spain showed that the levels of both folate and vitamin B12 in pregnant women gradually declined with the progression of pregnancy (149), and this may be attributable to hemodilution during pregnancy (69). Although the decline in folate levels during pregnancy may not be present in women in North America due to prolonged exposure to folic acid fortification, this trend may be present in vitamin B12 levels. As such, this may be another explanation as to why the vitamin B12 status in my cohort of women, taken at delivery, is lower than various nonpregnant populations (144-146). There were a few women with very elevated vitamin B12 (eg. 1479 pmol/L) and holoTC (eg. 627 pmol/L) levels at delivery in my study population, compared to the median of 198 pmol/L and 79 pmol/L, respectively. The elevated levels were reflected in the women’s newborns at birth (holoTC = 296 pmol/L; vitamin B12 = 2352 pmol/L). Although we did not record dietary intake data, it would be interesting to know how the process of pregnancy may lead to changes in appetite, and thus preferences for certain diets, in my cohort of women. As well, mood may affect eating patterns. These could affect nutritional status in this current cohort and explain some variations. Furthermore, genetic variants in key enzymes may affect absorption of methyl nutrients into the body. For instance, a variant in the GCP II gene, GCPII, affects absorption of dietary folate into the intestine (150). There may be a genetic variant in the 69 receptor responsible for absorption of vitamin B12, such that much more vitamin B12 is absorbed or retained in cells. Taking it a step further, I examined the characteristics of mothers and their newborns with lower methyl nutrient levels than the rest of my cohort. I did not see any distinct characteristics (eg. mood, level of education, number of pregnancies, MTHFR C677T or 5HTTLPR insertion/deletion genotypes, number of alcoholic drinks during pregnancy, etc.) that distinguished the mothers and their newborns from the rest of the cohort. However, I did see that the three women with the lowest folate levels (18.8 nmol/L, 18.9 nmol/L, and 20.0 nmol/L) also were vitamin B12 deficient (59.4 pmol/L, 132.1 pmol/L, and 87.5 pmol/L, respectively) and had lower levels of holoTC than the rest of the cohort (34.8 pmol/L, 54.2 pmol/L, and 26.63 pmol/L, respectively). Given that compliance for the consumption of prenatal supplements was not recorded, this finding could be due to low compliance with supplement usage during pregnancy. As quantified by the ΔΔCt method of relative quantification (RQ), the SLC6A4 mRNA levels in the women with low folate levels (no RNA available for one woman, and RQ values were 0.326 and 0.892 for the other two women), were not outliers compared to the rest of the study population (median = 0.777; mean = 0.935; SD = 0.622). The SLC6A4 % methylation at CpG sites 1-10 and the mean CpG % methylation in the women with low folate levels were not outliers compared to the rest of the cohort’s % methylation at corresponding sites. 6.1.2 Discussion of First Aim – Mothers As a step in determining the connections between maternal mood, maternal methyl nutrient status, and epigenetic regulation of SLC6A4 during pregnancy, I found that folate was associated with increases in methylation at CpG sites 1, 4 and 8 for SLC6A4. Given the role of 70 folate in the methyl nutrient metabolism and its connection to the production of AdoMet, my result seems to support the notion that folate alters gene-specific methylation by the donation of methyl groups. Taking into account the fact that the folate status in my population of pregnant women were all very similar (the majority were between 30-40 nmol/L), the association between folate and methylation suggests that gene-specific methylation may be very sensitive to small changes in folate levels. The influence of folate in methylation had previously been shown in imprinted genes (altered methylation of H19 in men treated with 5-MTHF) (118). North American grain products are fortified with folic acid, and it is now very rare to be folate-deficient in Canada (107). Concerns have been raised regarding whether high levels of folate, commonly seen in populations with folic acid fortification (107), may play a role in current health risks. These concerns include the inactivation of tumour-suppressor genes and the promotion of the proliferation of pre-neoplasms that could lead to cancers such as colorectal carcinomas (151,152). Unmetabolized folic acid is also associated with decreased natural killer cells’ cytotoxicity (153). Furthermore, high folate levels may mask vitamin B12 deficiency (81,82). Folate-vitamin B12 imbalance has been associated with cognitive impairments in the elderly (154) and insulin resistance in 6-year old children (155). Altered methylation status may be a possible mechanism behind disorders associated with high folate levels. High folate levels may lead to increased levels of 5-MTHF. 5-MTHF serves as the methyl group donor for the synthesis of methionine (62). Under circumstances of high folate levels, more Hcy is expected to be remethylated to methionine. This may lead to increased production of AdoMet and increased capacity for methylation reactions. This could also potentially decrease the expression of enzymes involved, such as MS or MAT, such that any decrease in expression of these enzymes could counterbalance the likelihood for elevated levels 71 of folate to increase methylation. The potential increase in methylation capacity imposed by high folate levels is also dependent on the availability of other components of the methyl metabolism, such as vitamin B12, riboflavin, or vitamin B6. For instance, MS does not function in the absence of adequate vitamin B12 (156). This would elevate Hcy in the body. Riboflavin is a cofactor for MTHFR, and the enzyme (methionine synthase reductase) that maintains the activity of MS (62). Deficiency in riboflavin in the presence of high folate would not only elevate non-reduced forms of folate (forms that are not 5-MTHF), but also elevate Hcy in the blood. Hcy may then be reconverted to AdoHcy. This would decrease the AdoMet/AdoHcy ratio and decrease methylation capacity (119,120). A portion of Hcy could go through the transulfuration pathway to form cysteine, but this reaction is vitamin B6 dependent (62). Thus, low levels of vitamin B6 in the presence high folate could further exacerbate the alteration in methylation capacity through the synthesis of AdoHcy. Although my study population consisted of women with high folate levels and suboptimal vitamin B12 levels, the holoTC status of the women and their newborns were not low. Given that holoTC represents the proportion of vitamin B12 required by cells (137), the problem with folate-vitamin B12 imbalance may not be present in my cohort. However, it is important to quantify other methyl nutrients, such as riboflavin and vitamin B6, in order to better identify the relationship between high folate levels and methylation. Interestingly, both EPDS and HAM-A scores were inversely associated with SLC6A4 CpG 10 methylation in monocytes. A previous study from our lab had shown that EPDS was inversely associated with SLC6A4 methylation in leukocytes at sites which corresponded to CpGs 1, 4, 5, 6, 7, and 9 in my study (11). However, that study did not control for folate and holoTC levels. Taken together, these results suggest that methylation levels may be very 72 different between cell populations. Further, these results suggest that the relationship between EPDS or HAM-A scores and SLC6A4 promoter methylation may be influenced by folate and holoTC levels. After adjusting for methyl nutrient status and MTHFR C677T genotype, I found that SLC6A4 CpGs 5, 7, 8, 10, and mean methylation levels were directly associated with SLC6A4 mRNA levels. This is contrary to what I expected, because a previous study by Philibert et al. in lymphoblast cell lines demonstrated that methylation of the SLC6A4 promoter was inversely associated with SLC6A4 mRNA levels (27). The authors looked at 81 CpG sites, of which four were associated with mRNA levels. CpG 8 in my study corresponds to one of the four sites. Given that my methylation analyses were performed in monocytes, and my gene expression analyses were performed in leukocytes, the contradicting results between the studies might be due to cell-specific differences between lymphoblasts, monocytes, and leukocytes. Another explanation for this discrepancy is that a different method (bisulphite pyrosequencing) for quantifying DNA methylation was used in this thesis compared to that used by Philibert et al. Bisulphite pyrosequencing allows for more sensitive detection of methylation than bisulphite sequencing using clones. The s allele of the 5HTTLPR insertion/deletion variant had previously been shown to be associated with higher SLC6A4 methylation than the l allele by Philibert et al. (27). The 5HTTLPR is located in the promoter of SLC6A4, approximately 1,000bp upstream of the CpGrich region where methylation was analyzed in this thesis. A possible mechanism behind the relationship between the 5HTTLPR insertion/deletion variant and SLC6A4 methylation may be via the state of the chromatin (57). Specifically, the s allele may induce methylation at the CpG sites within the 5HTTLPR due to the shorter length of the region compared to the l allele. This 73 may lead to the binding of methyl CpG binding protein 2 (MeCP2) (157) and the methylation of lysine 9 on histone 3 (H3K9) (158). Subsequently, there may be recruitment of histone modifying enzymes such as histone deacetylase (HDAC) (158,159). These changes may cause positive feedback for CpG methylation and may lead to more condensed chromatin structure in the 5HTTLPR region (57). It is possible that CpG methylation begins at the 5HTTLPR, but will extend to surrounding regions (including the SLC6A4 promoter region that I analyzed) (160). Ultimately, this is all speculative. No studies have been carried out that examined the functional protein interactions in the SLC6A4 promoter region. In my current study, I did not find an association between 5HTTLPR genotype and SLC6A4 methylation. This may be due to two reasons. Firstly, the total number of CpG sites varied between my study and the study conducted by Philibert et al. (27). Philibert et al. examined the average methylation level of 81 CpG sites, and found an association between average methylation and the 5HTTLPR insertion/deletion variant. In my study, I examined 10 CpG sites in the 5’ portion of the region analyzed by Philibert and colleagues. I did not see associations between the 5HTTLPR insertion/deletion variant and methylation (either at individual CpG sites or mean CpG methylation). Secondly, the methylation technique conducted by Philibert et al. and my study differ. Philibert and colleagues calculated methylation via the average of 10 clones (bisulphite sequencing). I conducted my study via bisulphite pyrosequencing, which as mentioned previously, is a much more quantitative technology that allows detection of small changes in DNA methylation. Further studies to examine the relationship between the 5HTTLPR insertion/deletion variant and SLC6A4 methylation is warranted via the bisulphite pyrosequencing method. 74 The fact that pregnant women homozygous for the short allele of the 5HTTLPR insertion/deletion variant did not have more depressive mood during pregnancy than women with the l/l or l/s genotypes in my cohort was not surprising. This is because although previous studies had shown that the s/s genotype was associated with depression, the association was only present after exposure to childhood maltreatment or major stressful life events (29,30). In my current study, I did not collect information regarding early life adversity or occurrence of major life events (such as loss, humiliation, threat, or defeat). Perhaps a relationship between depression and the s/s genotype would become evident in the context of previous exposure to adverse life events. Therefore, obtaining information on the emotional backgrounds of the women in my study population may be important in influencing the outcomes of my study. A previous study also did not find an association between 5HTTLPR and women in the late 2nd trimester, however, they did not control for stressful life events or childhood adversity either (11). In addition, I did not observe a relationship between 5HTTLPR insertion/deletion genotype and mRNA levels of SLC6A4. This is unexpected because previous studies had shown that the 5HTTLPR insertion/deletion variant is associated with SLC6A4 gene expression in lymphoblast cell lines (161). This could be due to differences in the region of mRNA transcript detected in my study and the previous study. The commercially available probe used in my study spans exons 8-9 and detects most splice variants of SLC6A4. Whereas the previously study only showed the association between 5HTTLPR genotype and SLC6A4 gene expression with a probe that recognized exon 1A (and not the alternative exon 1B) in SLC6A4. The previous study also did not find an association between genotype and SLC6A4 mRNA levels when they used a probe that bridged exons 8-9 (161). 75 6.1.3 Discussion of Second Aim – Maternal Influences on Newborns Newborns with mothers with the MTHFR 677TT genotype had lower methylation at CpG site 8. This trend, although not significant, was previously shown in leukocytes (11). It is conceivable that the maternal MTHFR 677TT genotype may decrease the level of 5-MTHF in mothers, leading to less transfer of 5-MTHF to the newborns. A study showed that the dominant form of folate in cord blood is 5-MTHF (75). Hence, maternal MTHFR 677TT genotype may influence the levels of 5-MTHF available in the newborn. Any decrease in 5-MTHF may alter the methylation capacity of AdoMet in the newborns. Previous literature had shown that altered methyl nutrient levels during pregnancy led to changes in offspring DNA methylation (5,162,163). A study in rats showed that dams fed a low protein (LP) diet gave birth to offspring with decreased gene-specific methylation. However, prenatal exposure to LP diet in conjunction with folate supplementation did not cause altered methylation profiles in the same genes (162). In humans, folate supplementation had been shown to be associated with increased IGF2 methylation in the newborn (163). This alludes to the fact that prenatal exposure to imbalances in the methyl nutrient metabolism may play important roles in gene-specific DNA methylation. I also analyzed the effects of maternal methyl nutrients and mood on SLC6A4 methylation and mRNA levels in newborns. Prenatal exposure to 3rd trimester EPDS scores was inversely associated with newborn SLC6A4 methylation at CpG 10. The direction of this association between EPDS and CpG 10 was the same as the direction of association between EPDS and maternal SLC6A4 CpG 10 methylation. Given that maternal methylation status of SLC6A4 was not associated with newborn methylation status of SLC6A4, it is possible that the effect of EPDS on CpG 10 methylation was due to a “programming” effect of mood, and not the heritability of methylation patterns from mother to newborn. My results are in agreement with 76 previous literature that had shown that antenatal mood influences newborn DNA methylation of stress regulatory genes, including SLC6A4 (9,11). Interestingly, HAM-A scores did not have an effect on the methylation of CpG 10, or any other CpG sites in the newborn. This suggests that although depression and anxiety are often comorbid (48), they may still differ in their etiology and have different effects on SLC6A4 methylation. Prenatal exposure to maternal depression and anxiety may lead to disturbances in brain development and behaviour, given that 5-HT acts as a trophic factor during early development (18,49,164). SSRIs function to block 5-HTT, leading to an increase in the level of 5-HT in the extracellular space (26). In the mothers, SSRIs serve to increase the serotonergic tone. SSRIs taken by the mothers can also cross the placenta and the blood brain barrier (165), thereby potentiallyaltering the serotonergic tone in the newborns as well. This may be detrimental in influencing the development of the sertotonergic system in the newborns. Although not measured in my cohort, the mothers and newborns with exposure to SSRIs may have altered 5HT levels compared to mothers and newborns without exposure to SSRIs. Any changes in 5-HT levels caused by SSRI exposure could influence homeostasis mechanisms in the body, such that components of the serotonergic system (eg. 5-HTT) may be altered. Specifically, SSRI exposure could potentially alter SLC6A4 methylation. Given that SSRIs block 5-HTTs, perhaps SLC6A4 methylation would subsequently change in a direction such that there is more SLC6A4 expression, and consequently more 5-HTT protein expression to counterbalance the changes caused by SSRIs. Studies had shown that fluoxetine (a type of SSRI) increases protein and mRNA levels of methyl-CpG binding domain 1(MBD1) protein and MeCP2 in the brain (dorsal caudateputamen, frontal cortex and hippocampus) in adult rats (166,167). Fluoxetine also increased 77 HDAC mRNA levels and decreased histone 3 acetylation (associated with open chromatin structure) in the dorsal caudate-putamen, frontal cortex and hippocampus of adult rats (166). On the contrary, studies suggested that fluoxetine upregulates histone 3 acetylation in the hippocampus of traumatic-brain-injured rats (167), and at the Bdnf promoter in the hippocampus of methylmercury exposed mice (168). This suggests that the direction of alterations in histone acetylation caused by fluoxetine exposure is context-dependent. Taken together, these rodent studies provide evidence towards underlying mechanisms for how SSRIs may influence methylation. However, I did not find an association between SSRI exposure and SLC6A4 methylation in either mothers or newborns in my study. Although, this does not necessarily mean that SSRIs do not have an effect on SLC6A4 methylation. Intriguingly, no effects of maternal mood or SSRI exposure on newborn SLC6A4 mRNA levels were found in my cohort. This is contrary to what was expected because a recent study had shown that both maternal depression and anxiety were associated with higher SLC6A4 mRNA levels in the placenta, even when the women were treated with SSRIs (169). Further studies are necessary to determine whether mood or SSRI exposure could alter SLC6A4 methylation or mRNA levels in other cell types or tissues. 6.1.4 Discussion of Third Aim – Newborns As a step in determining the role of newborn methyl nutrient metabolism in newborn methylation and expression of stress-regulatory genes, I analyzed SLC6A4 methylation and mRNA levels in newborns. Newborn MTHFR 677TT genotype was inversely associated with SLC6A4 CpG 6 and 10 methylation levels in newborns. This suggests that methyl nutrient metabolism in newborns may play a role in DNA methylation during development. In particular, 78 because the MTHFR 677TT genotype results in a decrease in functional activity of the enzyme MTHFR (93), newborns with the 677TT genotype may not have the capacity to produce as much 5-MTHF as newborns with the 677CC or 677CT genotypes. While the remethylation of homocysteine to methionine may be accomplished by other methyl donors, such as betaine in liver, this pathway is not present in monocytes. As such, the MTHFR 677TT genotype may decrease methylation capacity. Further, less production of 5-MTHF may lead to elevated Hcy concentrations in serum. A consequence of this may be increased AdoHcy level, because Hcy can be reconverted back to AdoHcy (62). Elevated levels of AdoHcy have been suggested to be an inhibitor of methyltransferases, thus further decreasing methylation capacity (91,92). Studies have shown that individuals with the MTHFR 677TT genotype have lower global DNA methylation than individuals with the 677CC or 677CT genotypes in several populations (123,170,171). Furthermore, mice with the targeted disruption of the Mthfr gene (Mthfr +/-) have lower AdoMet in the liver and higher AdoHcy in the brain than Mthfr +/+ mice (172). Both Mthfr +/- and Mthfr -/- mice exhibited global DNA hypomethylation in the brain and ovaries compared to Mthfr +/+ mice (173). Given that all newborns had adequate folate levels in my cohort, there may be decreased 5-MTHF availability and increased availability of other forms of folate in the cells. For instance, perhaps more folates are in the forms of 5,10-MethyleneTHF or 10-formylTHF. This may lead to altered dTMP or purine synthesis (174). Newborn SLC6A4 CpG 4 and 7 methylation was inversely associated with SLC6A4 mRNA level. However, this inverse relationship only remained for CpG 7 methylation when I controlled for newborn folate levels, newborn holoTC levels, newborn MTHFR C677T variant, and sex of newborn. This result is interesting because a previous study had shown that 2nd 79 trimester EPDS score was inversely associated with newborn SLC6A4 methylation at CpGs 4 and 7 in leukocytes (11). If methylation patterns in monocytes reflect that of leukocytes, decreased SLC6A4 methylation at CpGs 4 or 7 in newborn monocytes could potentially lead to increased serum 5-HTT levels via increased SLC6A4 mRNA levels in leukocytes. As such, an increase in 5-HTT levels could effectively decrease the availability of extracellular 5-HT. This could mean that prenatal exposure to maternal depressive mood may lead to a “serotonergic vulnerability” in the developing infant (15). Additionally, 5-HT had been shown to induce GR gene expression in hippocampal cells (175). Lower 5-HT protein levels could lead to less GR gene expression and elevated cortisol levels in the infant, and thus affect an infant’s stress response. However, it is not known whether this induction process occurs in monocytes, or other blood cells. 6.2 Limitations, Strengths, and Future Directions 6.2.1 Limitations This study has several limitations. Firstly, the mothers’ whole blood and their newborns’ cord blood were not separated into leukocytes, erythrocytes, and platelets prior to storage. This resulted in difficulty in isolating monocytes. Unfortunately, I was not able to obtain good quality RNA from monocytes for gene expression analyses. Although DNA for methylation analyses was isolated from monocytes, RNA was isolated from whole blood, a heterogeneous population of cells. As a result, this did not allow comparison of DNA methylation and mRNA levels in the same cell type. Furthermore, it is conceivable that the amount and quality of RNA extracted heavily depends on the length of time that the whole blood and cord blood were at room temperature prior to storage in the freezer. This is because RNA is highly unstable and 80 intracellular RNAse is still active at room temperature. In retrospect, perhaps a better way for RNA storage is aliquoting the blood into an RNA stabilizing solution (such as the PAXgene solution) prior to freezing. Secondly, methylation analysis in just one cell type does not allow for an extension of the observations in my study to other cell or tissue types. In order to determine whether the results from my study is a cell-specific effect or not, further methylation analyses should be performed in other cell types, such as lymphocytes, or other tissue types, such as the placenta (important for fetal development and growth). Thirdly, a larger sample size would allow for more robust statistical results. It was previously determined that 90 mother-newborn pairs would be sufficient sample size to detect clinically important relationships of at least 0.24-0.28 with 80% power and alpha of 0.05. Due to the exclusions I made (because of when the blood and serum were taken for the mothers and their newborns, inability to isolate monocyte-specific DNA from small blood volumes, and inability to extract intact RNA for some samples), my sample size decreased to approximately 75, depending on the type of analyses performed. As such, the categorical variables used in my analyses, such as the MTHFR or 5-HTTLPR variants have low sample size in certain groups. For instance, only seven study subjects with MTHFR 677TT genotype were analyzed for the influence of maternal MTHFR variant on newborn SLC6A4 methylation. In addition, not enough study subjects represented all genotypes of the 5-HTTLPR A→G substitution variant. Further, food frequency questionnaires designed to assess folate and vitamin B12 intakes should be administered to the women during pregnancy, allowing for assessments of methyl nutrient intake (may be different between the women in response to the side effects of 81 pregnancy). Results from food frequency questionnaires, in combination with serum methyl nutrient quantification, may be a better indicator of methyl nutrient levels. In addition, my study required multiple comparisons both within statistical tests and between statistical models. I conducted the LSD post-test to account for multiple comparisons within statistical tests. For instance, I conducted the LSD post-test when I determined the relationship between the MTHFR C677T genotypes and methylation. However, for between statistical models (eg. determining the effect of mood on SLC6A4 methylation at CpG 1, versus determining the effect of mood on SLC6A4 methylation at CpG 2), I did not account for multiple comparisons. As a consequence, there is a possibility that some of the significances I observed in my results are type I errors. Given that this is a study that focused on exploring the trends between methyl nutrients, mood, and serotonin transporter methylation and mRNA levels, it was of more interest to determine 1) any potential associations between the components, and 2) the direction of any trends. To take it a step further, correction for multiple comparisons between statistical models is needed. This could be achieved using the Benjamini and Hochberg false discovery rate calculation (176). Lastly and most importantly, the level of differences observed in methylation levels (if truly significant) was very small. One may wonder whether small changes in methylation are biologically meaningful. Given the complexity and dynamics of the genome (57,177), it is not clear whether small changes are relevant but it certainly may be. This is because any small changes in methylation may lead to modifications in chromatin structure (through the recruitment of transcription factors or through modifications of histones). Hence, any small changes in DNA methylation may amplify into larger changes. These minute changes in DNA methylation may also represent the beginning of deviations from normal methylation patterns, 82 and these changes could be exacerbated into larger changes with further exposure to similar environmental factors. One way to test whether minute changes in SLC6A4 methylation may be biologically relevant is to quantify whether downstream changes, such as 5-HTT protein levels, occur due to differences in methylation or mRNA levels. Another way to test whether small alterations in SLC6A4 methylation are biologically meaningful is to look at protein interactions with the CpG-rich or surrounding regions. This could be accomplished via chromatin immunoprecipitation. Putative binding sites for transcription factors could be determined via in silico search on the MatInspector software (178). 6.2.2 Strengths Despite the limitations of this study, there are many strengths offered by this project that made the research remarkable. Firstly, this is the first study of its kind to examine gene-nutrient interactions through measuring methyl nutrient levels and SLC6A4 methylation and mRNA levels. This study allowed for the determination of how 5-HTT is influenced by methyl nutrients and mood on two different genomic levels – SLC6A4 methylation and gene expression. Secondly, this study permitted the usage of bisulphite pyrosequencing, a state-of-the-art technology that allows precise quantification of DNA methylation at CpG sites. Lastly, and most importantly, this study was performed with human subjects and this allows for direct translational value into the healthcare setting. 6.2.3 Future Directions To take this project a step further, methylation and expression of NR3C1 will be analyzed to determine its relationships with mood and methyl nutrient metabolism in mothers and their 83 newborns. Further, this could potentially provide evidence at the gene level for any 5-HTT and GR interactions in blood cells. There is plasma currently in storage that would permit future assessments of 5-HTT protein levels (and GR protein levels) in mothers and their newborns. There is also in storage the portion of leukocytes that does not contain monocytes. SLC6A4 methylation should be assessed in lymphocytes to determine whether there are cell-specific differences in SLC6A4 methylation patterns. Additional studies should also be undertaken to assess alternative CpG-rich regions of the SLC6A4 promoter where methylation was shown to be associated with gene expression (27). Furthermore, my thesis project is a branch of a large prospective study looking at the effects of prenatal exposure to maternal depressive mood and SSRIs on the developmental outcomes of children. Behaviour in early infancy and childhood could be assessed to determine whether prenatal exposure to maternal depressive mood, SSRIs, and alterations in the methyl nutrient metabolism may “program” or impact growth and development of the children. 6.3 Concluding Remarks The World Health Organization had recognized depression as a major cause of morbidity worldwide (14). The occurrence of depression is especially concerning during pregnancy, due to the possibility that altered in utero environment imposed by maternal depressive mood or SSRI usage have adverse consequences in development of the offspring (2,4). The methyl nutrient metabolism may be the link between depression and gene expression. In this thesis, I determined the associations between maternal mood, methyl nutrient status, SLC6A4 methylation and gene expression of SLC6A4 in mothers and their newborns. The main findings of this thesis are as follows: 84 1. Maternal folate status was positively associated with maternal 3rd trimester SLC6A4 methylation at CpGs 1, 4, and 8. 2. Maternal 3rd trimester mood was inversely associated with SLC6A4 CpG 10 methylation in the mothers and their newborns. 3. In newborns, decreased methylation at SLC6A4 CpG 8 was associated with maternal MTHFR 677TT genotype, and decreased methylation at CpGs 6 and 10 were associated with newborn MTHFR 677TT genotype. 4. Maternal SLC6A4 mRNA level was positively associated with mean maternal SLC6A4 methylation and methylation at CpG sites 5, 7, 8, and 10. Yet, newborn SLC6A4 mRNA level was inversely associated with newborn SLC6A4 methylation at CpG sites 4 and 7. 5. Homozygosity for the minor alleles of the MTHFR C677T and the 5-HTTLPR insertion/deletion variants in mothers were not found to influence maternal 3 rd trimester mood. In conclusion, the results in my thesis provide evidence that methyl nutrient metabolism may influence methylation of SLC6A4, a gene involved in stress-regulation in both mothers and their newborns. Further, the results suggest that antenatal maternal depressive mood during pregnancy may affect SLC6A4 methylation in both mothers and their newborns. This may lead to, or “program” changes in the serotonergic system via DNA methylation. In addition, SLC6A4 methylation may be one type of epigenetic mechanism that affects SLC6A4 gene expression. Taken together, early life environment, genetics and epigenetics may serve as mechanisms that influence development. As such, alterations in maternal mood and methyl nutrient metabolism leading to changes in SLC6A4 expression may set up life-long health consequences in the 85 newborn. My thesis work serves as an important step in providing proof of gene-environment interactions during pregnancy, and contributes to the exciting field of developmental programming. 86 Bibliography (1) Oberlander TF, Warburton W, Misri S, Aghajanian J, Hertzman C. Neonatal outcomes after prenatal exposure to selective serotonin reuptake inhibitor antidepressants and maternal depression using population-based linked health data. Arch Gen Psychiatry 2006 Aug;63(8):898906. (2) Field T, Diego M, Dieter J, Hernandez-Reif M, Schanberg S, Kuhn C, et al. Prenatal depression effects on the fetus and the newborn. Infant Behavior and Development 2004 5;27(2):216-229. (3) Oberlander TF, Grunau R, Mayes L, Riggs W, Rurak D, Papsdorf M, et al. Hypothalamicpituitary-adrenal (HPA) axis function in 3-month old infants with prenatal selective serotonin reuptake inhibitor (SSRI) antidepressant exposure. Early Hum Dev 2008 Oct;84(10):689-697. (4) Moses-Kolko EL, Bogen D, Perel J, Bregar A, Uhl K, Levin B, et al. Neonatal signs after late in utero exposure to serotonin reuptake inhibitors: literature review and implications for clinical applications. JAMA 2005 May 18;293(19):2372-2383. (5) Waterland RA, Jirtle RL. Transposable elements: targets for early nutritional effects on epigenetic gene regulation. Mol Cell Biol 2003 Aug;23(15):5293-5300. (6) Weaver IC, Cervoni N, Champagne FA, D'Alessio AC, Sharma S, Seckl JR, et al. Epigenetic programming by maternal behavior. Nat Neurosci 2004 Aug;7(8):847-854. (7) Lillycrop KA, Slater-Jefferies JL, Hanson MA, Godfrey KM, Jackson AA, Burdge GC. Induction of altered epigenetic regulation of the hepatic glucocorticoid receptor in the offspring of rats fed a protein-restricted diet during pregnancy suggests that reduced DNA methyltransferase-1 expression is involved in impaired DNA methylation and changes in histone modifications. Br J Nutr 2007 Jun;97(6):1064-1073. (8) Friso S, Choi SW, Girelli D, Mason JB, Dolnikowski GG, Bagley PJ, et al. A common mutation in the 5,10-methylenetetrahydrofolate reductase gene affects genomic DNA methylation through an interaction with folate status. Proc Natl Acad Sci U S A 2002 Apr 16;99(8):5606-5611. (9) Oberlander TF, Weinberg J, Papsdorf M, Grunau R, Misri S, Devlin AM. Prenatal exposure to maternal depression, neonatal methylation of human glucocorticoid receptor gene (NR3C1) and infant cortisol stress responses. Epigenetics 2008 Mar-Apr;3(2):97-106. (10) McGowan PO, Sasaki A, D'Alessio AC, Dymov S, Labonte B, Szyf M, et al. Epigenetic regulation of the glucocorticoid receptor in human brain associates with childhood abuse. Nat Neurosci 2009 Mar;12(3):342-348. 87 (11) Devlin AM, Brain U, Austin J, Oberlander TF. Prenatal exposure to maternal depressed mood and the MTHFR C677T variant affect SLC6A4 methylation in infants at birth. PLoS One 2010 Aug 16;5(8):e12201. (12) Ustun TB, Ayuso-Mateos JL, Chatterji S, Mathers C, Murray CJ. Global burden of depressive disorders in the year 2000. Br J Psychiatry 2004 May;184:386-392. (13) Patten SB, Wang JL, Williams JV, Currie S, Beck CA, Maxwell CJ, et al. Descriptive epidemiology of major depression in Canada. Can J Psychiatry 2006 Feb;51(2):84-90. (14) World health Organization. Depression. Available at: http://www.who.int/topics/depression/en/. Accessed June/03, 2011. (15) Jans LA, Riedel WJ, Markus CR, Blokland A. Serotonergic vulnerability and depression: assumptions, experimental evidence and implications. Mol Psychiatry 2007 Jun;12(6):522-543. (16) Yonkers KA, Wisner KL, Stewart DE, Oberlander TF, Dell DL, Stotland N, et al. The management of depression during pregnancy: a report from the American Psychiatric Association and the American College of Obstetricians and Gynecologists. Gen Hosp Psychiatry 2009 Sep-Oct;31(5):403-413. (17) Kessler RC, Berglund P, Demler O, Jin R, Koretz D, Merikangas KR, et al. The epidemiology of major depressive disorder: results from the National Comorbidity Survey Replication (NCS-R). JAMA 2003 Jun 18;289(23):3095-3105. (18) Gaspar P, Cases O, Maroteaux L. The developmental role of serotonin: news from mouse molecular genetics. Nat Rev Neurosci 2003 Dec;4(12):1002-1012. (19) Naughton M, Mulrooney JB, Leonard BE. A review of the role of serotonin receptors in psychiatric disorders. Hum Psychopharmacol 2000 Aug;15(6):397-415. (20) Kremer HP, Goekoop JG, Van Kempen GM. Clinical use of the determination of serotonin in whole blood. J Clin Psychopharmacol 1990 Apr;10(2):83-87. (21) Nakatani Y, Sato-Suzuki I, Tsujino N, Nakasato A, Seki Y, Fumoto M, et al. Augmented brain 5-HT crosses the blood-brain barrier through the 5-HT transporter in rat. Eur J Neurosci 2008 May;27(9):2466-2472. (22) Cowen PJ, Parry-Billings M, Newsholme EA. Decreased plasma tryptophan levels in major depression. J Affect Disord 1989 Jan-Feb;16(1):27-31. (23) Asberg M, Thoren P, Traskman L, Bertilsson L, Ringberger V. "Serotonin depression"--a biochemical subgroup within the affective disorders? Science 1976 Feb 6;191(4226):478-480. 88 (24) Rogeness GA, Mitchell EL, Custer GJ, Harris WR. Comparison of whole blood serotonin and platelet MAO in children with schizophrenia and major depressive disorder. Biol Psychiatry 1985 3;20(3):270-275. (25) Lesch KP, Wolozin BL, Murphy DL, Reiderer P. Primary structure of the human platelet serotonin uptake site: identity with the brain serotonin transporter. J Neurochem 1993 Jun;60(6):2319-2322. (26) Oberlander TF, Gingrich JA, Ansorge MS. Sustained neurobehavioral effects of exposure to SSRI antidepressants during development: molecular to clinical evidence. Clin Pharmacol Ther 2009 Dec;86(6):672-677. (27) Philibert R, Madan A, Andersen A, Cadoret R, Packer H, Sandhu H. Serotonin transporter mRNA levels are associated with the methylation of an upstream CpG island. Am J Med Genet B Neuropsychiatr Genet 2007 Jan 5;144B(1):101-105. (28) Heils A, Teufel A, Petri S, Stober G, Riederer P, Bengel D, et al. Allelic variation of human serotonin transporter gene expression. J Neurochem 1996 Jun;66(6):2621-2624. (29) Lesch KP, Bengel D, Heils A, Sabol SZ, Greenberg BD, Petri S, et al. Association of anxiety-related traits with a polymorphism in the serotonin transporter gene regulatory region. Science 1996 Nov 29;274(5292):1527-1531. (30) Caspi A, Sugden K, Moffitt TE, Taylor A, Craig IW, Harrington H, et al. Influence of life stress on depression: moderation by a polymorphism in the 5-HTT gene. Science 2003 Jul 18;301(5631):386-389. (31) Gotlib IH, Joormann J, Minor KL, Hallmayer J. HPA axis reactivity: a mechanism underlying the associations among 5-HTTLPR, stress, and depression. Biol Psychiatry 2008 May 1;63(9):847-851. (32) Way BM, Taylor SE. The serotonin transporter promoter polymorphism is associated with cortisol response to psychosocial stress. Biol Psychiatry 2010 Mar 1;67(5):487-492. (33) Murphy GM,Jr, Hollander SB, Rodrigues HE, Kremer C, Schatzberg AF. Effects of the serotonin transporter gene promoter polymorphism on mirtazapine and paroxetine efficacy and adverse events in geriatric major depression. Arch Gen Psychiatry 2004 Nov;61(11):1163-1169. (34) Glatz K, Mossner R, Heils A, Lesch KP. Glucocorticoid-regulated human serotonin transporter (5-HTT) expression is modulated by the 5-HTT gene-promotor-linked polymorphic region. J Neurochem 2003 Sep;86(5):1072-1078. (35) Hu XZ, Lipsky RH, Zhu G, Akhtar LA, Taubman J, Greenberg BD, et al. Serotonin transporter promoter gain-of-function genotypes are linked to obsessive-compulsive disorder. Am J Hum Genet 2006 May;78(5):815-826. 89 (36) Liu D, Diorio J, Tannenbaum B, Caldji C, Francis D, Freedman A, et al. Maternal care, hippocampal glucocorticoid receptors, and hypothalamic-pituitary-adrenal responses to stress. Science 1997 Sep 12;277(5332):1659-1662. (37) Barker DJ. Fetal programming of coronary heart disease. Trends Endocrinol Metab 2002 Nov;13(9):364-368. (38) Barker DJ, Osmond C. Infant mortality, childhood nutrition, and ischaemic heart disease in England and Wales. Lancet 1986 May 10;1(8489):1077-1081. (39) Barker DJ. The origins of the developmental origins theory. J Intern Med 2007 May;261(5):412-417. (40) Brown AS, Susser ES, Lin SP, Neugebauer R, Gorman JM. Increased risk of affective disorders in males after second trimester prenatal exposure to the Dutch hunger winter of 194445. Br J Psychiatry 1995 May;166(5):601-606. (41) Brown AS, van Os J, Driessens C, Hoek HW, Susser ES. Further evidence of relation between prenatal famine and major affective disorder. Am J Psychiatry 2000 Feb;157(2):190195. (42) Neugebauer R, Hoek HW, Susser E. Prenatal exposure to wartime famine and development of antisocial personality disorder in early adulthood. JAMA 1999 Aug 4;282(5):455-462. (43) Franzek EJ, Sprangers N, Janssens AC, Van Duijn CM, Van De Wetering BJ. Prenatal exposure to the 1944-45 Dutch 'hunger winter' and addiction later in life. Addiction 2008 Mar;103(3):433-438. (44) Susser E, Neugebauer R, Hoek HW, Brown AS, Lin S, Labovitz D, et al. Schizophrenia after prenatal famine. Further evidence. Arch Gen Psychiatry 1996 Jan;53(1):25-31. (45) St Clair D, Xu M, Wang P, Yu Y, Fang Y, Zhang F, et al. Rates of adult schizophrenia following prenatal exposure to the Chinese famine of 1959-1961. JAMA 2005 Aug 3;294(5):557-562. (46) Meijer A. Child psychiatric sequelae of maternal war stress. Acta Psychiatr Scand 1985 Dec;72(6):505-511. (47) Malaspina D, Corcoran C, Kleinhaus KR, Perrin MC, Fennig S, Nahon D, et al. Acute maternal stress in pregnancy and schizophrenia in offspring: a cohort prospective study. BMC Psychiatry 2008 Aug 21;8:71. (48) Fawcett J, Kravitz HM. Anxiety syndromes and their relationship to depressive illness. J Clin Psychiatry 1983 Aug;44(8 Pt 2):8-11. 90 (49) O'Connor TG, Heron J, Golding J, Beveridge M, Glover V. Maternal antenatal anxiety and children's behavioural/emotional problems at 4 years. Report from the Avon Longitudinal Study of Parents and Children. Br J Psychiatry 2002 Jun;180:502-508. (50) O'Connor TG, Ben-Shlomo Y, Heron J, Golding J, Adams D, Glover V. Prenatal anxiety predicts individual differences in cortisol in pre-adolescent children. Biol Psychiatry 2005 Aug 1;58(3):211-217. (51) Hay DF, Pawlby S, Waters CS, Perra O, Sharp D. Mothers' antenatal depression and their children's antisocial outcomes. Child Dev 2010 Jan-Feb;81(1):149-165. (52) Caldji C, Tannenbaum B, Sharma S, Francis D, Plotsky PM, Meaney MJ. Maternal care during infancy regulates the development of neural systems mediating the expression of fearfulness in the rat. Proc Natl Acad Sci U S A 1998 Apr 28;95(9):5335-5340. (53) Myers MM, Brunelli SA, Shair HN, Squire JM, Hofer MA. Relationships between maternal behavior of SHR and WKY dams and adult blood pressures of cross-fostered F1 pups. Dev Psychobiol 1989 Jan;22(1):55-67. (54) Francis D, Diorio J, Liu D, Meaney MJ. Nongenomic transmission across generations of maternal behavior and stress responses in the rat. Science 1999 Nov 5;286(5442):1155-1158. (55) Roth TL, Lubin FD, Funk AJ, Sweatt JD. Lasting epigenetic influence of early-life adversity on the BDNF gene. Biol Psychiatry 2009 May 1;65(9):760-769. (56) Clarke AS, Soto A, Bergholz T, Schneider ML. Maternal gestational stress alters adaptive and social behavior in adolescent rhesus monkey offspring. Infant Behavior and Development 1996 12;19(4):451-461. (57) Jaenisch R, Bird A. Epigenetic regulation of gene expression: how the genome integrates intrinsic and environmental signals. Nat Genet 2003 Mar;33 Suppl:245-254. (58) Bestor TH. The DNA methyltransferases of mammals. Hum Mol Genet 2000 Oct;9(16):2395-2402. (59) Okano M, Bell DW, Haber DA, Li E. DNA methyltransferases Dnmt3a and Dnmt3b are essential for de novo methylation and mammalian development. Cell 1999 Oct 29;99(3):247257. (60) Rodenhiser D, Mann M. Epigenetics and human disease: translating basic biology into clinical applications. CMAJ 2006 Jan 31;174(3):341-348. (61) Verona RI, Mann MR, Bartolomei MS. Genomic imprinting: intricacies of epigenetic regulation in clusters. Annu Rev Cell Dev Biol 2003;19:237-259. 91 (62) Blom HJ. Folic acid, methylation and neural tube closure in humans. Birth Defects Res Part A Clin Mol Teratol 2009 Apr;85(4):295-302. (63) Melse-Boonstra A, de Bree A, Verhoef P, Bjorke-Monsen AL, Verschuren WM. Dietary monoglutamate and polyglutamate folate are associated with plasma folate concentrations in Dutch men and women aged 20-65 years. J Nutr 2002 Jun;132(6):1307-1312. (64) Rong N, Selhub J, Goldin BR, Rosenberg IH. Bacterially synthesized folate in rat large intestine is incorporated into host tissue folyl polyglutamates. J Nutr 1991 Dec;121(12):19551959. (65) MRC Vitamin Study Research Group. Prevention of neural tube defects: results of the Medical Research Council Vitamin Study. MRC Vitamin Study Research Group. Lancet 1991 Jul 20;338(8760):131-137. (66) Berry RJ, Li Z, Erickson JD, Li S, Moore CA, Wang H, et al. Prevention of neural-tube defects with folic acid in China. China-U.S. Collaborative Project for Neural Tube Defect Prevention. N Engl J Med 1999 Nov 11;341(20):1485-1490. (67) Health Canada. Nutrition for a healthy pregnancy: national guidelines for childbearing years. Ottawa: Minister of Public Works and Government Services. 1999. (68) Health Canada. Food and drug regulations, amendment Schedule no. 1066. 1997. (69) Institute of Medicine. Dietary Reference Intakes for Thiamin, Riboflavin, Niacin, Vitamin B6, Folate, Vitamin B12, Pantothenic Acid, Biotin, and Choline. . http://www.nap.edu/openbook.php?record_id=6015&page=306 ed. Washington, DC: Institute of Medicine: National Academy Press; 1998. p. 306-356. (70) Copp AJ, Greene ND, Murdoch JN. The genetic basis of mammalian neurulation. Nat Rev Genet 2003 Oct;4(10):784-793. (71) Blom HJ, Shaw GM, den Heijer M, Finnell RH. Neural tube defects and folate: case far from closed. Nat Rev Neurosci 2006 Sep;7(9):724-731. (72) Shakur YA, Garriguet D, Corey P, O'Connor DL. Folic acid fortification above mandated levels results in a low prevalence of folate inadequacy among Canadians. Am J Clin Nutr 2010 Oct;92(4):818-825. (73) Kalmbach RD, Choumenkovitch SF, Troen AM, D'Agostino R, Jacques PF, Selhub J. Circulating folic acid in plasma: relation to folic acid fortification. Am J Clin Nutr 2008 Sep;88(3):763-768. (74) Bailey SW, Ayling JE. The extremely slow and variable activity of dihydrofolate reductase in human liver and its implications for high folic acid intake. Proc Natl Acad Sci U S A 2009 Sep 8;106(36):15424-15429. 92 (75) Obeid R, Kasoha M, Kirsch SH, Munz W, Herrmann W. Concentrations of unmetabolized folic acid and primary folate forms in pregnant women at delivery and in umbilical cord blood. Am J Clin Nutr 2010 Dec;92(6):1416-1422. (76) Qiu A, Jansen M, Sakaris A, Min SH, Chattopadhyay S, Tsai E, et al. Identification of an intestinal folate transporter and the molecular basis for hereditary folate malabsorption. Cell 2006 Dec 1;127(5):917-928. (77) Pietrzik K, Bailey L, Shane B. Folic acid and L-5-methyltetrahydrofolate: comparison of clinical pharmacokinetics and pharmacodynamics. Clin Pharmacokinet 2010 Aug 1;49(8):535548. (78) Engeham SF, Haase A, Langley-Evans SC. Supplementation of a maternal low-protein diet in rat pregnancy with folic acid ameliorates programming effects upon feeding behaviour in the absence of disturbances to the methionine-homocysteine cycle. Br J Nutr 2010 Apr;103(7):9961007. (79) 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 Jul 12;357(2):135-142. (80) Coppen A, Bailey J. Enhancement of the antidepressant action of fluoxetine by folic acid: a randomised, placebo controlled trial. J Affect Disord 2000 Nov;60(2):121-130. (81) Dickinson CJ. Does folic acid harm people with vitamin B12 deficiency? QJM 1995 May;88(5):357-364. (82) Mills JL, Von Kohorn I, Conley MR, Zeller JA, Cox C, Williamson RE, et al. Low vitamin B-12 concentrations in patients without anemia: the effect of folic acid fortification of grain. Am J Clin Nutr 2003 Jun;77(6):1474-1477. (83) McCracken C. Challenges of long-term nutrition intervention studies on cognition: discordance between observational and intervention studies of vitamin B12 and cognition. Nutr Rev 2010 Nov;68 Suppl 1:S11-5. (84) Watanabe F. Vitamin B12 sources and bioavailability. Exp Biol Med (Maywood) 2007 Nov;232(10):1266-1274. (85) Eldridge AL. Comparison of 1989 RDAs and DRIs for Water-Soluble Vitamins. Nutr Today 2004 Mar;39(2):88-93. (86) Russell-Jones GJ, Alpers DH. Vitamin B12 transporters. Pharm Biotechnol 1999;12:493520. (87) Dali-Youcef N, Andres E. An update on cobalamin deficiency in adults. QJM 2009 Jan;102(1):17-28. 93 (88) Chatthanawaree W. Biomarkers of cobalamin (vitamin B12) deficiency and its application. J Nutr Health Aging 2011 Mar;15(3):227-231. (89) Giugliani ER, Jorge SM, Goncalves AL. Serum vitamin B12 levels in parturients, in the intervillous space of the placenta and in full-term newborns and their interrelationships with folate levels. Am J Clin Nutr 1985 Feb;41(2):330-335. (90) Greene ND, Stanier P, Copp AJ. Genetics of human neural tube defects. Hum Mol Genet 2009 Oct 15;18(R2):R113-29. (91) Deguchi T, Barchas J. Inhibition of transmethylations of biogenic amines by Sadenosylhomocysteine. Enhancement of transmethylation by adenosylhomocysteinase. J Biol Chem 1971 May 25;246(10):3175-3181. (92) Zappia V, Zydek-Cwick R, Schlenk F. The specificity of S-adenosylmethionine derivatives in methyl transfer reactions. J Biol Chem 1969 Aug 25;244(16):4499-4509. (93) Frosst P, Blom HJ, Milos R, Goyette P, Sheppard CA, Matthews RG, et al. A candidate genetic risk factor for vascular disease: a common mutation in methylenetetrahydrofolate reductase. Nat Genet 1995 May;10(1):111-113. (94) Christensen B, Arbour L, Tran P, Leclerc D, Sabbaghian N, Platt R, et al. Genetic polymorphisms in methylenetetrahydrofolate reductase and methionine synthase, folate levels in red blood cells, and risk of neural tube defects. Am J Med Genet 1999 May 21;84(2):151-157. (95) Naushad SM, Devi AR. Role of parental folate pathway single nucleotide polymorphisms in altering the susceptibility to neural tube defects in South India. J Perinat Med 2010;38(1):63-69. (96) Volcik KA, Blanton SH, Tyerman GH, Jong ST, Rott EJ, Page TZ, et al. Methylenetetrahydrofolate reductase and spina bifida: evaluation of level of defect and maternal genotypic risk in Hispanics. Am J Med Genet 2000 Nov 6;95(1):21-27. (97) Dalal A, Pradhan M, Tiwari D, Behari S, Singh U, Mallik GK, et al. MTHFR 677C-->T and 1298A-->C polymorphisms: evaluation of maternal genotypic risk and association with level of neural tube defect. Gynecol Obstet Invest 2007;63(3):146-150. (98) Guerreiro CS, Carmona B, Goncalves S, Carolino E, Fidalgo P, Brito M, et al. Risk of colorectal cancer associated with the C677T polymorphism in 5,10-methylenetetrahydrofolate reductase in Portuguese patients depends on the intake of methyl-donor nutrients. Am J Clin Nutr 2008 Nov;88(5):1413-1418. (99) Shen CD, Zhang WL, Sun K, Wang YB, Zhen YS, Hui RT. Interaction of genetic risk factors confers higher risk for thrombotic stroke in male Chinese: a multicenter case-control study. Ann Hum Genet 2007 Sep;71(Pt 5):620-629. 94 (100) Bjelland I, Tell GS, Vollset SE, Refsum H, Ueland PM. Folate, vitamin B12, homocysteine, and the MTHFR 677C->T polymorphism in anxiety and depression: the Hordaland Homocysteine Study. Arch Gen Psychiatry 2003 Jun;60(6):618-626. (101) Gilbody S, Lewis S, Lightfoot T. Methylenetetrahydrofolate reductase (MTHFR) genetic polymorphisms and psychiatric disorders: a HuGE review. Am J Epidemiol 2007 Jan 1;165(1):113. (102) Peerbooms OL, van Os J, Drukker M, Kenis G, Hoogveld L, MTHFR in Psychiatry Group, et al. Meta-analysis of MTHFR gene variants in schizophrenia, bipolar disorder and unipolar depressive disorder: Evidence for a common genetic vulnerability? Brain Behav Immun 2010 Dec 24. (103) Gilsing AM, Crowe FL, Lloyd-Wright Z, Sanders TA, Appleby PN, Allen NE, et al. Serum concentrations of vitamin B12 and folate in British male omnivores, vegetarians and vegans: results from a cross-sectional analysis of the EPIC-Oxford cohort study. Eur J Clin Nutr 2010 Sep;64(9):933-939. (104) Miller JW, Garrod MG, Rockwood AL, Kushnir MM, Allen LH, Haan MN, et al. Measurement of total vitamin B12 and holotranscobalamin, singly and in combination, in screening for metabolic vitamin B12 deficiency. Clin Chem 2006 Feb;52(2):278-285. (105) Homocysteine Studies Collaboration. Homocysteine and risk of ischemic heart disease and stroke: a meta-analysis. JAMA 2002 Oct 23-30;288(16):2015-2022. (106) Kim JM, Stewart R, Kim SW, Yang SJ, Shin IS, Yoon JS. Predictive value of folate, vitamin B12 and homocysteine levels in late-life depression. Br J Psychiatry 2008 Apr;192(4):268-274. (107) Colapinto CK, O'Connor DL, Tremblay MS. Folate status of the population in the Canadian Health Measures Survey. CMAJ 2011 Feb 8;183(2):E100-6. (108) Shuaibi AM, House JD, Sevenhuysen GP. Folate status of young Canadian women after folic acid fortification of grain products. J Am Diet Assoc 2008 Dec;108(12):2090-2094. (109) Ray JG, Vermeulen MJ, Langman LJ, Boss SC, Cole DE. Persistence of vitamin B12 insufficiency among elderly women after folic acid food fortification. Clin Biochem 2003 Jul;36(5):387-391. (110) Ray JG, Goodman J, O'Mahoney PR, Mamdani MM, Jiang D. High rate of maternal vitamin B12 deficiency nearly a decade after Canadian folic acid flour fortification. QJM 2008 Jun;101(6):475-477. (111) Shane B, Stokstad EL. Vitamin B12-folate interrelationships. Annu Rev Nutr 1985;5:115141. 95 (112) Toh BH, van Driel IR, Gleeson PA. Pernicious anemia. N Engl J Med 1997 Nov 13;337(20):1441-1448. (113) Dahlin AM, Van Guelpen B, Hultdin J, Johansson I, Hallmans G, Palmqvist R. Plasma vitamin B12 concentrations and the risk of colorectal cancer: a nested case-referent study. Int J Cancer 2008 May 1;122(9):2057-2061. (114) Ray JG, Wyatt PR, Thompson MD, Vermeulen MJ, Meier C, Wong PY, et al. Vitamin B12 and the risk of neural tube defects in a folic-acid-fortified population. Epidemiology 2007 May;18(3):362-366. (115) Wang HX, Wahlin A, Basun H, Fastbom J, Winblad B, Fratiglioni L. Vitamin B(12) and folate in relation to the development of Alzheimer's disease. Neurology 2001 May 8;56(9):11881194. (116) Smith AD, Refsum H. Vitamin B-12 and cognition in the elderly. Am J Clin Nutr 2009 Feb;89(2):707S-11S. (117) Robinson DJ, O'Luanaigh C, Tehee E, O'Connell H, Hamilton F, Chin AV, et al. Associations between holotranscobalamin, vitamin B12, homocysteine and depressive symptoms in community-dwelling elders. Int J Geriatr Psychiatry 2011 Mar;26(3):307-313. (118) Ingrosso D, Cimmino A, Perna AF, Masella L, De Santo NG, De Bonis ML, et al. Folate treatment and unbalanced methylation and changes of allelic expression induced by hyperhomocysteinaemia in patients with uraemia. Lancet 2003 May 17;361(9370):1693-1699. (119) Devlin AM, Bottiglieri T, Domann FE, Lentz SR. Tissue-specific changes in H19 methylation and expression in mice with hyperhomocysteinemia. J Biol Chem 2005 Jul 8;280(27):25506-25511. (120) Devlin AM, Singh R, Wade RE, Innis SM, Bottiglieri T, Lentz SR. Hypermethylation of Fads2 and altered hepatic fatty acid and phospholipid metabolism in mice with hyperhomocysteinemia. J Biol Chem 2007 Dec 21;282(51):37082-37090. (121) 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 Mar 10;6(3):e17706. (122) Devlin AM, Clarke R, Birks J, Evans JG, Halsted CH. Interactions among polymorphisms in folate-metabolizing genes and serum total homocysteine concentrations in a healthy elderly population. Am J Clin Nutr 2006 Mar;83(3):708-713. (123) Castro R, Rivera I, Ravasco P, Camilo ME, Jakobs C, Blom HJ, et al. 5,10methylenetetrahydrofolate reductase (MTHFR) 677C-->T and 1298A-->C mutations are associated with DNA hypomethylation. J Med Genet 2004 Jun;41(6):454-458. 96 (124) HAMILTON M. A rating scale for depression. J Neurol Neurosurg Psychiatry 1960 Feb;23:56-62. (125) HAMILTON M. The assessment of anxiety states by rating. Br J Med Psychol 1959;32(1):50-55. (126) Cox JL, Holden JM, Sagovsky R. Detection of postnatal depression. Development of the 10-item Edinburgh Postnatal Depression Scale. Br J Psychiatry 1987 Jun;150:782-786. (127) Lam NY, Rainer TH, Chiu RW, Lo YM. EDTA is a better anticoagulant than heparin or citrate for delayed blood processing for plasma DNA analysis. Clin Chem 2004 Jan;50(1):256257. (128) Medline Plus. Blood Differential. 2011; Available at: http://www.nlm.nih.gov/medlineplus/ency/article/003657.htm. Accessed June/10, 2011. (129) Ziegler-Heitbrock L, Ancuta P, Crowe S, Dalod M, Grau V, Hart DN, et al. Nomenclature of monocytes and dendritic cells in blood. Blood 2010 Oct 21;116(16):e74-80. (130) Beekman JM, Reischl J, Henderson D, Bauer D, Ternes R, Pena C, et al. Recovery of microarray-quality RNA from frozen EDTA blood samples. J Pharmacol Toxicol Methods 2009 Jan-Feb;59(1):44-49. (131) Grigg GW. Sequencing 5-methylcytosine residues by the bisulphite method. DNA Seq 1996;6(4):189-198. (132) Dupont JM, Tost J, Jammes H, Gut IG. De novo quantitative bisulfite sequencing using the pyrosequencing technology. Anal Biochem 2004 Oct 1;333(1):119-127. (133) Livak KJ, Schmittgen TD. Analysis of relative gene expression data using real-time quantitative PCR and the 2(-Delta Delta C(T)) Method. Methods 2001 Dec;25(4):402-408. (134) Lesch KP, Balling U, Gross J, Strauss K, Wolozin BL, Murphy DL, et al. Organization of the human serotonin transporter gene. J Neural Transm Gen Sect 1994;95(2):157-162. (135) NCBI AceView. Homo Sapien gene: SLC6A4. Available at: http://www.ncbi.nlm.nih.gov/IEB/Research/Acembly/av.cgi?db=human&c=Gene&l=SLC6A4. Accessed June 24, 2011. (136) Oberlander TF, Papsdorf M, Brain UM, Misri S, Ross C, Grunau RE. Prenatal effects of selective serotonin reuptake inhibitor antidepressants, serotonin transporter promoter genotype (SLC6A4), and maternal mood on child behavior at 3 years of age. Arch Pediatr Adolesc Med 2010 May;164(5):444-451. (137) Herzlich B, Herbert V. Depletion of serum holotranscobalamin II. An early sign of negative vitamin B12 balance. Lab Invest 1988 Mar;58(3):332-337. 97 (138) van den Donk M, Buijsse B, van den Berg SW, Ocke MC, Harryvan JL, Nagengast FM, et al. Dietary intake of folate and riboflavin, MTHFR C677T genotype, and colorectal adenoma risk: a Dutch case-control study. Cancer Epidemiol Biomarkers Prev 2005 Jun;14(6):1562-1566. (139) Yazdanpanah N, Uitterlinden AG, Zillikens MC, Jhamai M, Rivadeneira F, Hofman A, et al. Low dietary riboflavin but not folate predicts increased fracture risk in postmenopausal women homozygous for the MTHFR 677 T allele. J Bone Miner Res 2008 Jan;23(1):86-94. (140) Caudill MA, Dellschaft N, Solis C, Hinkis S, Ivanov AA, Nash-Barboza S, et al. Choline intake, plasma riboflavin, and the phosphatidylethanolamine N-methyltransferase G5465A genotype predict plasma homocysteine in folate-deplete Mexican-American men with the methylenetetrahydrofolate reductase 677TT genotype. J Nutr 2009 Apr;139(4):727-733. (141) Powers HJ, Hill MH, Welfare M, Spiers A, Bal W, Russell J, et al. Responses of biomarkers of folate and riboflavin status to folate and riboflavin supplementation in healthy and colorectal polyp patients (the FAB2 Study). Cancer Epidemiol Biomarkers Prev 2007 Oct;16(10):2128-2135. (142) Health Canada. Addition of Vitamins and Minerals to Foods, 2005. 2005; Available at: http://www.hc-sc.gc.ca/fn-an/nutrition/vitamin/fortification_final_doc_1-eng.php. Accessed June 28, 2011. (143) Watanabe H, Suganuma N, Hayashi A, Hirowatari Y, Hirowatari T, Ohsawa M. No relation between folate and homocysteine levels and depression in early pregnant women. Biosci Trends 2010 Dec;4(6):344-350. (144) Shuaibi AM, Sevenhuysen GP, House JD. Validation of a food choice map with a 3-day food record and serum values to assess folate and vitamin B-12 intake in college-aged women. J Am Diet Assoc 2008 Dec;108(12):2041-2050. (145) Vu TT, Nguyen TL, Nguyen CK, Nguyen TD, Skeaff CM, Venn BJ, et al. Folate and vitamin B12 status of women of reproductive age living in Hanoi City and Hai Duong Province of Vietnam. Public Health Nutr 2009 Jul;12(7):941-946. (146) Verkleij-Hagoort AC, de Vries JH, Stegers MP, Lindemans J, Ursem NT, SteegersTheunissen RP. Validation of the assessment of folate and vitamin B12 intake in women of reproductive age: the method of triads. Eur J Clin Nutr 2007 May;61(5):610-615. (147) McCracken C, Hudson P, Ellis R, McCaddon A, Medical Research Council Cognitive Function and Ageing Study. Methylmalonic acid and cognitive function in the Medical Research Council Cognitive Function and Ageing Study. Am J Clin Nutr 2006 Dec;84(6):1406-1411. (148) Refsum H, Johnston C, Guttormsen AB, Nexo E. Holotranscobalamin and total transcobalamin in human plasma: determination, determinants, and reference values in healthy adults. Clin Chem 2006 Jan;52(1):129-137. 98 (149) Ubeda N, Reyes L, Gonzalez-Medina A, Alonso-Aperte E, Varela-Moreiras G. Physiologic changes in homocysteine metabolism in pregnancy: A longitudinal study in Spain. Nutrition 2011 Mar 1. (150) Devlin AM, Ling EH, Peerson JM, Fernando S, Clarke R, Smith AD, et al. Glutamate carboxypeptidase II: a polymorphism associated with lower levels of serum folate and hyperhomocysteinemia. Hum Mol Genet 2000 Nov 22;9(19):2837-2844. (151) Cole BF, Baron JA, Sandler RS, Haile RW, Ahnen DJ, Bresalier RS, et al. Folic acid for the prevention of colorectal adenomas: a randomized clinical trial. JAMA 2007 Jun 6;297(21):2351-2359. (152) Lawrance AK, Deng L, Rozen R. Methylenetetrahydrofolate reductase deficiency and low dietary folate reduce tumorigenesis in Apc min/+ mice. Gut 2009 Jun;58(6):805-811. (153) Troen AM, Mitchell B, Sorensen B, Wener MH, Johnston A, Wood B, et al. Unmetabolized folic acid in plasma is associated with reduced natural killer cell cytotoxicity among postmenopausal women. J Nutr 2006 Jan;136(1):189-194. (154) Morris MS, Jacques PF, Rosenberg IH, Selhub J. Folate and vitamin B-12 status in relation to anemia, macrocytosis, and cognitive impairment in older Americans in the age of folic acid fortification. Am J Clin Nutr 2007 Jan;85(1):193-200. (155) Yajnik CS, Deshpande SS, Jackson AA, Refsum H, Rao S, Fisher DJ, et al. Vitamin B12 and folate concentrations during pregnancy and insulin resistance in the offspring: the Pune Maternal Nutrition Study. Diabetologia 2008 Jan;51(1):29-38. (156) Banerjee RV, Matthews RG. Cobalamin-dependent methionine synthase. FASEB J 1990 Mar;4(5):1450-1459. (157) Georgel PT, Horowitz-Scherer RA, Adkins N, Woodcock CL, Wade PA, Hansen JC. Chromatin compaction by human MeCP2. Assembly of novel secondary chromatin structures in the absence of DNA methylation. J Biol Chem 2003 Aug 22;278(34):32181-32188. (158) Fuks F, Hurd PJ, Wolf D, Nan X, Bird AP, Kouzarides T. The methyl-CpG-binding protein MeCP2 links DNA methylation to histone methylation. J Biol Chem 2003 Feb 7;278(6):4035-4040. (159) Sengupta N, Seto E. Regulation of histone deacetylase activities. J Cell Biochem 2004 Sep 1;93(1):57-67. (160) Wheeler BS, Blau JA, Willard HF, Scott KC. The impact of local genome sequence on defining heterochromatin domains. PLoS Genet 2009 Apr;5(4):e1000453. (161) Philibert RA, Sandhu H, Hollenbeck N, Gunter T, Adams W, Madan A. The relationship of 5HTT (SLC6A4) methylation and genotype on mRNA expression and liability to major 99 depression and alcohol dependence in subjects from the Iowa Adoption Studies. Am J Med Genet B Neuropsychiatr Genet 2008 Jul 5;147B(5):543-549. (162) Lillycrop KA, Phillips ES, Jackson AA, Hanson MA, Burdge GC. Dietary protein restriction of pregnant rats induces and folic acid supplementation prevents epigenetic modification of hepatic gene expression in the offspring. J Nutr 2005 Jun;135(6):1382-1386. (163) Steegers-Theunissen RP, Obermann-Borst SA, Kremer D, Lindemans J, Siebel C, Steegers EA, et al. Periconceptional maternal folic acid use of 400 microg per day is related to increased methylation of the IGF2 gene in the very young child. PLoS One 2009 Nov 16;4(11):e7845. (164) Ansorge MS, Morelli E, Gingrich JA. Inhibition of serotonin but not norepinephrine transport during development produces delayed, persistent perturbations of emotional behaviors in mice. J Neurosci 2008 Jan 2;28(1):199-207. (165) Kim J, Riggs KW, Misri S, Kent N, Oberlander TF, Grunau RE, et al. Stereoselective disposition of fluoxetine and norfluoxetine during pregnancy and breast-feeding. Br J Clin Pharmacol 2006 Feb;61(2):155-163. (166) Cassel S, Carouge D, Gensburger C, Anglard P, Burgun C, Dietrich JB, et al. Fluoxetine and cocaine induce the epigenetic factors MeCP2 and MBD1 in adult rat brain. Mol Pharmacol 2006 Aug;70(2):487-492. (167) Wang Y, Neumann M, Hansen K, Hong SM, Kim S, Noble-Haeusslein LJ, et al. Fluoxetine increases hippocampal neurogenesis and induces epigenetic factors but does not improve functional recovery after traumatic brain injury. J Neurotrauma 2011 Feb;28(2):259268. (168) Onishchenko N, Karpova N, Sabri F, Castren E, Ceccatelli S. Long-lasting depression-like behavior and epigenetic changes of BDNF gene expression induced by perinatal exposure to methylmercury. J Neurochem 2008 Aug;106(3):1378-1387. (169) Ponder KL, Salisbury A, McGonnigal B, Laliberte A, Lester B, Padbury JF. Maternal depression and anxiety are associated with altered gene expression in the human placenta without modification by antidepressant use: Implications for fetal programming. Dev Psychobiol 2011 May 5. (170) Graziano F, Kawakami K, Ruzzo A, Watanabe G, Santini D, Pizzagalli F, et al. Methylenetetrahydrofolate reductase 677C/T gene polymorphism, gastric cancer susceptibility and genomic DNA hypomethylation in an at-risk Italian population. Int J Cancer 2006 Feb 1;118(3):628-632. (171) Axume J, Smith SS, Pogribny IP, Moriarty DJ, Caudill MA. The MTHFR 677TT genotype and folate intake interact to lower global leukocyte DNA methylation in young Mexican American women. Nutr Res 2007 Jan;27(1):1365-1317. 100 (172) Devlin AM, Arning E, Bottiglieri T, Faraci FM, Rozen R, Lentz SR. Effect of Mthfr genotype on diet-induced hyperhomocysteinemia and vascular function in mice. Blood 2004 Apr 1;103(7):2624-2629. (173) Chen Z, Karaplis AC, Ackerman SL, Pogribny IP, Melnyk S, Lussier-Cacan S, et al. Mice deficient in methylenetetrahydrofolate reductase exhibit hyperhomocysteinemia and decreased methylation capacity, with neuropathology and aortic lipid deposition. Hum Mol Genet 2001 Mar 1;10(5):433-443. (174) Stover PJ. One-carbon metabolism-genome interactions in folate-associated pathologies. J Nutr 2009 Dec;139(12):2402-2405. (175) Meaney MJ, Diorio J, Francis D, LaRocque S, O'Donnell D, Smythe JW, et al. Environmental regulation of the development of glucocorticoid receptor systems in the rat forebrain. The role of serotonin. Ann N Y Acad Sci 1994 Nov 30;746:260-73; discussion 274, 289-93. (176) Benjamini Y, Hochberg Y. Controlling the False Discovery Rate: A Practical and Powerful Approach to Multiple Testing. Journal of the Royal Statistical Society.Series B (Methodological) 1995;57(1):pp. 289-300. (177) Schneider R, Grosschedl R. Dynamics and interplay of nuclear architecture, genome organization, and gene expression. Genes Dev 2007 Dec 1;21(23):3027-3043. (178) Cartharius K, Frech K, Grote K, Klocke B, Haltmeier M, Klingenhoff A, et al. MatInspector and beyond: promoter analysis based on transcription factor binding sites. Bioinformatics 2005 Jul 1;21(13):2933-2942. 101
Featured Collection
UBC Theses and Dissertations
Maternal mood during pregnancy, methyl nutrient metabolism, and serotonin transporter Wei, Julia Jia 2011
Item Metadata
Download
- Media
- 24-ubc_2011_fall_wei_julia.pdf [ 1.32MB ]
- Metadata
- JSON: 24-1.0105086.json
- JSON-LD: 24-1.0105086-ld.json
- RDF/XML (Pretty): 24-1.0105086-rdf.xml
- RDF/JSON: 24-1.0105086-rdf.json
- Turtle: 24-1.0105086-turtle.txt
- N-Triples: 24-1.0105086-rdf-ntriples.txt
- Original Record: 24-1.0105086-source.json
- Full Text
- 24-1.0105086-fulltext.txt
- Citation
- 24-1.0105086.ris
Full Text
Cite
Citation Scheme:
Usage Statistics
Share
Embed
Customize your widget with the following options, then copy and paste the code below into the HTML
of your page to embed this item in your website.
<div id="ubcOpenCollectionsWidgetDisplay">
<script id="ubcOpenCollectionsWidget"
src="{[{embed.src}]}"
data-item="{[{embed.item}]}"
data-collection="{[{embed.collection}]}"
data-metadata="{[{embed.showMetadata}]}"
data-width="{[{embed.width}]}"
async >
</script>
</div>
Our image viewer uses the IIIF 2.0 standard.
To load this item in other compatible viewers, use this url: http://iiif.library.ubc.ca/presentation/dsp.24.1-0105086/manifest