4. Discussion
We found that adjusted OR for childhood wheezing at 2 years age according to quartile of maternal folate was significantly increased compared with that of the lowest quartile but not for 1, 4, and 7 years of age. In a case-control study of childhood wheeze for DNA methylation, which are related asthma at three CpG sites of IKZF3 andORMDL3 , no association of maternal folate with DNA methylation was noticed. This study provided novel evidence that the negative effect of maternal folate on an infant’s wheezing did not remain at 7 years, and maternal folate cannot modify DNA methylation at three CpG sites ofIKZF3 and ORMDL3 of 7-year-old children.
In this study, results differ by the child’s age because the infant’s high sensibility to infections due to immature immune system and respiratory function. Infections are strong risk factors for infant’s allergy. Furthermore, the frequency of an infant’s infections until 2 years of age among wheezing children was approximately 2–3 fold higher than that without wheezing at 2 years of age (Supplementary Table 5). As children grow, respiratory structure and immune function mature and allergic symptoms become more stable, the negative effect of maternal folate on infant’s wheezing is not observed over 4 years of age in the current study.
Medians of detected maternal serum folate levels among previous human four studies were reported as 21.5 nmol/l (9.5 ng/ml) in South Korea (Kim et al., 2015), 21.2 nmol/l (9.36 ng/ml) in The Netherlands (van der Valk et al., 2013), and 51.2 nmol/l (22.6 ng/ml) (Roy et al., 2018) and 43.5 nmol/l (19.2 ng/ml) (Vereen et al., 2018) in the USA. Although previous studies have various sample sizes and deferent timing of maternal sample collection (first, second, or third trimester) with an outcome range of 1 to 7 years of age, these studies reported diminished or insignificant association between maternal folate levels and risk of wheezing, asthma, and respiratory symptoms in childhood. Additionally, more than 80% of participants had optimal folate of maternal intracellular folate concentration, which is a recommended biomarker for folate status than serum levels (WHO, 2012), showing decreased asthma risk at 6–7 years of age in a dose-dependent manner (Magdelijns et al., 2011). Inconsistently, in the Norwegian Study, negative effect of maternal folate (median 8.7 nmol/l) were observed on respiratory tract infections up to 18 months, asthma at 3 and 7 years of age (Parr et al., 2017). The median maternal folate in the current study was 16.7 nmol/l (7.40 ng/ml), which was lower than previous studies reporting a protective or null effect (Kim et al., 2015; Roy et al., 2018; van der Valk et al., 2013; Vereen et al., 2018) and higher than the Norwegian study reporting a negative effect (Parr et al., 2017). These observations signify the negative effect among the population is reasonably due to lower maternal folate levels. Therefore, negative effect at 2 years of age in this study can be caused by insufficient maternal folate status among our participants. The current study suggests that maternal folate status needs to improve.
In a case-control study, three CpG sites were not associated with maternal folate (Table 4). A previous systematic review specifies the necessity to detect modified DNA methylation change by maternal folate in the human study (Brown et al., 2014). This study is the first to report that individual DNA methylation was detected at three CpG sites of IKZF3 and ORMDL3 -related asthma at 7 years of age by pyrosequencing and it does not differ by maternal folate during pregnancy. Only one study has described methylation beta-values of CpG sites related to asthma, including IKZF3 and ORMDL3 using the Infinium 450 K assay, and asthmatic children had lower methylation than control children (Acevedo et al., 2015). In this study, same direction was observed that is inversely associated with childhood wheeze and DNA methylation of IKZF3 (Supplementary Table 4). DNA hyper-methylation of long interspersed nucleotide element –1 (LINE–1 ) from the buccal cell at 6–­­17 years age children, was associated with maternal intake of methyl donating nutrients including folate, but unassociated with quality of life measures among children with asthma (Montrose, 2017). LINE–1 is considered an indicator of global methylation, so dietary nutrients can affect not onlyLINE–1 but also a wide range differentiation DNA methylation among several genes. No association between maternal folate and DNA methylation at three CpG sites was observed, however, maternal folate and dietary nutrients during pregnancy can modify DNA methylation. Further study is needed to examine the modified effects of global epigenetic differentiation including the epigenome wide association study on the association between folate levels and childhood wheeze or asthma.
The advantages of this study included the large-scale perspective birth cohort for minimum bias in the data. Furthermore, the biological assessment of folate levels to detect maternal folate in serum during pregnancy was performed. In addition, assessment of wheezing symptoms using the ISAAC questionnaire, which is an international standard questionnaire for the assessment of children’s allergies was undertaken. This study had some limitations. DNA methylation for only three CpG sites were analyzed due to which global epigenetic differentiation was not covered. Also, the modified effect of DNA methylation at age 2 was not evaluated, despite its significant association with maternal folate. Finally, in this study, maternal serum folate levels were measured at a single sampling point, not intracellular folate concentration, while estimated half-life of folate was assessed at approximately 6 h (Sauberlich, 1999). This limitation can cause the misclassification of maternal folate levels. However, serum folate assay was commonly conducted in large epidemiological studies (Parr et al., 2017; Roy et al., 2018). Moreover, studies in Japan have reported that maternal diet habit was maintained across pregnancy (Jaw et al., 2016), stating relative stability of maternal folate levels.