Maternal obesity linked to childhood weight issues through DNA changes

In a recent study published in the journal BMC Medicine, researchers investigate the impact of maternal pre-pregnancy overweight or obesity (OWO) on newborn cord blood deoxyribonucleic acid methylation (DNAm) and its potential role in intergenerational obesity risk in a high-risk United States demographic.

Study: Maternal pre-pregnancy BMI, offspring epigenome-wide DNA methylation, and childhood obesity: findings from the Boston Birth Cohort. Image Credit: Mala Iryna / Shutterstock.com

Rising rates of childhood obesity

Childhood obesity in the U.S. rose from 18.5% between 2015-2016 to 19.7% between 2017-2020. This currently amounts to a total of 14.7 million children, with higher rates of childhood obesity observed in racial minorities. Maternal pre-pregnancy obesity, which occurs in about 25% of U.S. conceptions, is a leading risk factor for childhood obesity.

Despite suggestions that maternal obesity may cause epigenetic alterations that influence child obesity, there remains a lack of conclusive research supporting these claims. To date, only a few studies have linked maternal body mass index (BMI), fetal DNAm, and childhood BMI, whereas other studies have reported differing results and often overlooked notable risk factors like smoking.

Thus, additional research is needed to determine the epigenetics of childhood obesity due to maternal pre-pregnancy obesity, particularly in underrepresented racial groups in the U.S., to ultimately guide targeted interventions.

About the study

The present study utilized 903 mother-child pairs from the Boston Birth Cohort (BBC), which primarily encompasses urban and low-income minorities. Initiated in 1998, mothers delivering single births at Boston Medical Center were included, whereas mothers with twins, triplets, or infants born with significant congenital disabilities were excluded from the analysis.

BBC specifically targeted low birthweight and premature births for its study. Mothers provided data regarding their socio-demographics, lifestyle habits, and medical histories following delivery.

Cord blood samples were collected and subsequently fractionated into plasma, white blood cells, and red blood cells, all of which were stored at -80° C. DNA methylation was measured in 963 cord blood samples using a specific Illumina BeadChip, following which samples were profiled for methylation through rigorous quality control processes. After multiple filtering steps, 903 samples with specific cytosine-phosphate-guanine (CpG) sites were analyzed.

Data on maternal weight before pregnancy and smoking habits during pregnancy were obtained from the mothers. The BMI of all children between birth to 18 years was also analyzed, with OWO defined as BMI above the 85^th percentile for age and gender.

The analytical methodology involved comparing the demographics of newborns to maternal pre-pregnancy BMI, identifying differentially methylated sites linked to maternal BMI, and executing mediation analyses. The mediation analysis aimed to determine the extent to which the influence of maternal BMI on child obesity risk was due to methylation at a specific locus.

Study findings

A total of 903 mother-child duos were included in the final analysis, of whom 71% of mothers identified as Black. Boys comprised 52.9% of the child participants. The median pre-pregnancy BMI was 26.62 kg/m².

About 53% of mothers had a pre-pregnancy OWO status, whereas 63.2% of children were considered OWO between one and 18 years of age. Compared to non-OWO mothers, OWO mothers were more likely to be older, Black, and smokers. Children born to these OWO mothers were also more likely to be considered OWO between one and 18 years of age.

Eight CpG sites corresponded to six genes and were significantly associated with maternal BMI. The highest hypermethylated difference for each single unit increase in maternal BMI was related to the tumor protein P53 inducible nuclear protein 1 (TP53INP1) gene. Other notable hypermethylated CpG sites were observed in the phosphatase and tensin homolog (PTEN) and KILLIN genes.

Comparatively, hypomethylated genes identified in newborns born to mothers with higher BMIs were observed in the alpha- and gamma-adaptin binding protein (AAGAB), alpha kinase 1 (ALPK1), and excision repair cross-complementation group 8 (ERCC8) genes.

A total of 481 CpG sites with altered DNAm in newborns were associated with maternal BMI. The annotated genes of these CpG sites were associated with hematocrit, triglycerides, and tobacco use disorder. Of these 481 CpG sites, 26 were associated with newborn birth weight; however, none of these CpG sites had a significant mediation effect on the association between maternal BMI and birth weight.

Fourteen CpG sites had a significant mediation effect on the association between maternal BMI and child OWO in at least one age group. More specifically, between seven and nine years of age, eight CpG sites mediated this association, whereas five CpG sites mediated this association by 13 years of age. The mediating proportions in these age groups ranged from 4.38-25.21% and 3.99-10.57%, respectively.

Maternal smoking did not have a significant mediation effect on methylation at any of the identified CpG sites.