Aim: Evidence suggests that folic acid intake affects birth weight and that these effects may be mediated via the fetal epigenome. Our previous array data indicate that methylation in human cord blood at gene-specific CpGs is associated with birth weight percentile (BWP). Our aims were to investigate associations with BWP in specific CpGs identified by the array analysis in a significantly larger cohort and investigate the effects of other relevant factors on this association.
Materials & methods: Methylation status was examined in candidate CpGs in 129 cord blood samples using Pyrosequencing™. The effects of other potentially important factors; maternal smoking, folate-related metabolite levels and genetic variation in the MTHFR gene, were examined. Linear and logistic regression analyses were used to identify relationships between BWP and methylation levels in the context of other key factors.
Results: Increased cord methylation at CpGs in GSTM5 and MAP2K3 was associated with a reduced risk of having a birth weight below the 50th percentile (p = 0.010; odds ratio [OR]: 0.33 and p = 0.024; OR: 0.24, respectively) while higher methylation levels in APOB were associated with an increased risk (p = 0.023; OR: 2.56). Smoking during pregnancy modified the effect of methylation on BWP. Thus, compared with nonsmokers with a GSTM5 methylation level of >25% (median BWP: 54.7%), those who had smoked during pregnancy and whose GSTM5 methylation was <25% had the lowest median BWP (12.0%; p = 0.001). Furthermore, this latter group had the highest proportion of cases with BWPs below 50% (92.9 compared with 47.8% in nonsmokers with a GSTM5 methylation level of >25%; p = 0.013; OR: 14.2). Similar results were identified for MAP2K3, while the link with APOB reflected the inverse relationship between methylation at this locus and BWP.
Conclusion: Our data suggest that gene-specific methylation of cord DNA is associated with BWP and this methylation provides an additional effect on BWP to that of smoking during pregnancy.