Maternal and Cord Blood Folate Concentrations Are Inversely Associated with Fetal DNA Hydroxymethylation, but Not DNA Methylation, in a Cohort of Pregnant Canadian Women.

BACKGROUND: Aberrancies in fetal DNA methylation programming may modify disease susceptibility of the offspring. Maternal folate status has potential to alter fetal DNA methylation. OBJECTIVES: We examined the association of maternal and cord blood concentrations of folate and unmetabolized folic acid (UMFA), vitamin B-12, vitamin B-6, and choline with fetal DNA methylation and hydroxymethylation and assessed potential modifying effects of 38 fetal genetic variants in 22 genes. METHODS: Nutrient blood concentrations were measured in 368 pregnant women in early pregnancy (12-16 wk of gestation) and at delivery (37-42 wk of gestation) and in cord blood. DNA methylation and hydroxymethylation in cord blood mononuclear cells were quantified by LC-MS/MS. Pearson partial correlations were used to determine the association between individual nutrients and DNA methylation and hydroxymethylation. RESULTS: Serum and RBC folate and plasma UMFA concentrations (primary outcomes) in early pregnancy, at delivery, and in cord blood were not significantly associated with fetal DNA methylation. In contrast, maternal RBC folate in early pregnancy (r = -0.16, P = 0.04) and cord plasma UMFA (r = -0.23, P = 0.004) were inversely correlated with fetal DNA hydroxymethylation. Neither maternal and cord blood concentrations of other nutrients nor fetal genotypes (secondary outcomes) were significantly associated with fetal DNA methylation or hydroxymethylation. Infants born to mothers with RBC folate concentrations in the highest quartile and serum vitamin B-12 concentrations in the lowest quartile in early pregnancy had significantly lower fetal DNA methylation and higher birth weight compared with those born to mothers with lower RBC folate and higher serum vitamin B-12 concentrations (P = 0.01). CONCLUSIONS: Maternal and cord blood folate concentrations are associated with fetal DNA hydroxymethylation, but not DNA methylation, in a cohort of pregnant Canadian women. The observation that high folate and low vitamin B-12 maternal status in early pregnancy may be associated with decreased fetal DNA methylation and higher birth weight warrants further investigation. This trial was registered at clinicaltrials.gov as NCT02244684.

Formate concentrations in maternal plasma during pregnancy and in cord blood in a cohort of pregnant Canadian women: relations to genetic polymorphisms and plasma metabolites.

BACKGROUND: One-carbon metabolism, responsible for purine and thymidylate synthesis and transmethylation reactions, plays a critical role in embryonic and fetal development. Formate is a key player in one-carbon metabolism. In contrast to other one-carbon metabolites, it is not linked to tetrahydrofolate, is present in plasma at appreciable concentrations, and may therefore be distributed to different tissues. OBJECTIVE: The study was designed to determine the concentration of formate in cord blood in comparison with maternal blood taken earlier in pregnancy and at delivery and to relate formate concentrations to potential precursors and key fetal genotypes. METHODS: Formate and amino acids were measured in plasma during early pregnancy (12-16 wk), at delivery (37-42 wk), and in cord blood samples from 215 mothers, of a prospective cohort study. Three fetal genetic variants in one-carbon metabolism were assessed for their association with cord plasma concentrations of formate. RESULTS: The formate concentration was ∼60% higher in the cord blood samples than in mothers' plasma. The maternal formate concentrations did not differ between the early pregnancy samples and those taken at delivery. Plasma concentrations of 4 formate precursors (serine, glycine, tryptophan, and methionine) were increased in cord blood compared with the maternal samples. Cord blood formate was influenced by fetal genotype, being ∼12% higher in infants harboring the MTHFR A1298C (rs1801131) AC or CC genotypes and 10% lower in infants harboring the MTHFD1 G1958A (rs2236225) GA or AA genotypes. CONCLUSIONS: The increased formate concentrations in cord blood may support the increased activity of one-carbon metabolism in infants. As such, it would support increased rates of purine and thymidylate synthesis and the provision of methionine for methylation reactions.

Fetal one-carbon nutrient concentrations may be affected by gestational diabetes.

Both insufficiency and excess of one-carbon nutrients (folate, choline, vitamins B6 and B12) during pregnancy have been associated with gestational diabetes mellitus (GDM). However, the precise nature of this association has not been clearly established. We hypothesized that GDM may affect one-carbon nutrients concentrations in the fetus, thus possibly participating in epigenetic programing of the offspring. Maternal blood was collected at recruitment (12-16 weeks). At delivery (28-42 weeks), both maternal and cord blood were collected. Blood concentrations of one-carbon nutrients and their metabolites were compared between the two groups. A total of 368 women were included in the study, of whom 19 (5.6%) were later diagnosed with GDM. No significant differences were found in maternal blood concentrations of one-carbon nutrients and their metabolites between the GDM and control groups at recruitment or at delivery. In cord blood, however, serum folate (87.7 [IQR 70.4-103.9] vs 66.6 [IQR 45.5-80.3] nmol/L, P = .025) and plasma TMAO (2.82 [IQR 1.3-3.2] vs 1.35 [IQR 1.0-2.0] µmol/L, P = .017) concentrations were higher, while plasma betaine concentrations were lower (17.5 [IQR 16.3-19.4] vs 21.1 [IQR 18.0-24.1] µmol/L, P = .019) in infants born to mothers with GDM compared with control. Our data suggest that while maternal blood concentrations of one-carbon nutrients and their metabolites may not affect the risk of GDM, GDM may alter concentrations of serum folate, plasma betaine and TMAO in cord blood. These alterations in one-carbon nutrient concentrations in fetal circulation may impact epigenetic programing, thereby contributing to physiologic changes and disease susceptibility in adulthood associated with GDM offspring.

Suboptimal maternal and cord plasma pyridoxal 5' phosphate concentrations are uncommon in a cohort of Canadian pregnant women and newborn infants.

Vitamin B6 is important in fetal development, but little is known of the vitamin B6 status of pregnant women and newborns in North America and potential modifying factors. This prospective study determined maternal and cord plasma concentrations of pyridoxal 5' phosphate (PLP; an indicator of vitamin B6 status) in a convenience sample of 368 Canadian pregnant women and their newborns. The association of maternal intake of vitamin B6 and fetal genetic variants with cord plasma PLP and homocysteine concentrations was also examined. Dietary and supplemental intakes of vitamin B6 were assessed in early and mid to late pregnancy. PLP concentrations were measured in maternal plasma in early pregnancy and at delivery, and in cord plasma. Six fetal variants of the MTHFR and CßS genes were assessed for their association with cord plasma PLP and homocysteine concentrations. Geometric mean (95% CI) PLP concentrations were 107 (98, 116) nmol/L in early pregnancy and 58 (53, 62) nmol/L at delivery, respectively, and 296 (275, 319) nmol/L in cord blood (p < .0001). During early pregnancy and at delivery, 3.6% and 5.5% of women had plasma PLP concentrations <20 nmol/L, respectively. Ninety eight percent of the women with supplemental B6 intake of at least the recommended dietary allowance had PLP concentrations >20 nmol/L. Fetal genetic variants were not associated with cord PLP and homocysteine concentrations. Vitamin B6 deficiency is uncommon in a cohort of Canadian pregnant women due largely to prevalent vitamin B6 supplement use.

Low Serum Vitamin B-12 Concentrations Are Prevalent in a Cohort of Pregnant Canadian Women.

BACKGROUND: Among Canadian women of reproductive age, 5% and 20% have serum vitamin B-12 concentrations indicative of deficiency (<148 pmol/L) and marginal status (148-220 pmol/L), respectively. Given the association between suboptimal vitamin B-12 and adverse pregnancy outcomes, an understanding of vitamin B-12 status during pregnancy, and factors that influence it, is required. OBJECTIVE: This prospective analysis from the PREFORM (PREnatal FOlic acid exposuRe on DNA Methylation in the newborn infant) study investigated 1) vitamin B-12 status in a cohort of Canadian pregnant women and their newborns, 2) the association of maternal dietary vitamin B-12 intake with maternal and cord blood concentrations of vitamin B-12 and its biomarkers, and 3) the association of fetal genetic polymorphisms with cord blood concentrations of vitamin B-12 and its biomarkers. METHODS: In pregnant Canadian women (n = 368; mean ± SD age: 32 ± 5 y), vitamin B-12 intakes were assessed in early (0-16 wk) and mid- to late (23-37 wk) pregnancy. Serum vitamin B-12 and plasma total homocysteine (tHcy) and methylmalonic acid (MMA) in maternal blood at 12-16 wk of pregnancy and at delivery (28-42 wk) and in cord blood were measured and compared by using regression analyses. The associations of 28 fetal genetic variants in vitamin B-12 metabolism and cord blood vitamin B-12, tHcy, and MMA concentrations were assessed by using regression analysis, with adjustment for multiple testing. RESULTS: A total of 17% and 38% of women had deficient and 35% and 43% had marginal serum vitamin B-12 concentrations at 12-16 wk of pregnancy and at delivery, respectively. Only 1.9-5.3% had elevated MMA (>271 nmol/L), and no women had elevated tHcy (>13 µmol/L). Maternal dietary vitamin B-12 intake during pregnancy was either weakly associated or not associated with maternal and cord blood vitamin B-12 (r(2) = 0.17-0.24, P < 0.0008), tHcy (P = NS) and MMA (r(2) = 0.05-0.11, P < 0.001). Fetal genetic polymorphisms were not associated with cord blood concentrations of vitamin B-12 and its biomarkers. CONCLUSIONS: Deficient and marginal serum vitamin B-12 concentrations are prevalent in Canadian pregnant women with the use of traditional cutoffs, despite supplement use. Given the growing interest among women to adhere to a vegetarian diet that may be lower in vitamin B-12, and vitamin B-12's importance in pregnancy, the functional ramifications of these observations need to be elucidated. This trial was registered at clinicaltrials.gov as NCT02244684.

Maternal Choline Status, but Not Fetal Genotype, Influences Cord Plasma Choline Metabolite Concentrations.

BACKGROUND: Choline deficiency during pregnancy can lead to adverse birth outcomes, including impaired neurodevelopment and birth defects. Genetic variants of choline and one-carbon metabolism may also influence birth outcomes by altering plasma choline concentrations. The effects of maternal ad libitum choline intake during pregnancy and fetal genetic variants on maternal and cord concentrations of choline and its metabolites are unknown. OBJECTIVES: This prospective study sought to assess the effect of 1) maternal dietary choline intake on maternal and cord plasma concentrations of choline and its metabolites, and 2) fetal genetic polymorphisms on cord plasma concentrations. METHODS: The dietary choline intake of 368 pregnant Canadian women was assessed in early (0-16 wk) and late (23-37 wk) pregnancy with the use of a food frequency questionnaire. Plasma concentrations of free choline and its metabolites were measured in maternal samples at recruitment and delivery, and in the cord blood. Ten fetal genetic variants in choline and one-carbon metabolism were assessed for their association with cord plasma concentrations of free choline and its metabolites. RESULTS: Mean maternal plasma free choline, dimethylglycine, and trimethylamine N-oxide (TMAO) concentrations increased during pregnancy by 49%, 17%, and 13%, respectively (P < 0.005), whereas betaine concentrations decreased by 21% (P < 0.005). Cord plasma concentrations of free choline, betaine, dimethylglycine, and TMAO were 3.2, 2.0, 1.3, and 0.88 times corresponding maternal concentrations at delivery, respectively (all P < 0.005). Maternal plasma concentrations of betaine, dimethylglycine, and TMAO (r(2) = 0.19-0.51; P < 0.0001) at delivery were moderately strong, whereas maternal concentrations of free choline were not significant (r(2) = 0.12; P = 0.06), predictors of cord plasma concentrations of these metabolites. Neither maternal dietary intake nor fetal genetic variants predicted maternal or cord plasma concentrations of choline and its metabolites. CONCLUSION: These data collectively indicate that maternal choline status, but not fetal genotype, influences cord plasma concentrations of choline metabolites. This trial was registered at clinicaltrials.gov as NCT02244684.