UC Berkeley

In the last 30 years there has been a sharp increase in obesity among children, and minority populations are particularly vulnerable. Although etiology of obesity is thought to be multifactorial with causes stemming from diet, environment, genetics and their interaction, no clear molecular pathways have been identified. Underlying obesity development are changes in critical energy balance hormones, adiponectin and leptin (adipokines), however their development and determinants over the childhood period remain poorly understood.

Previous studies indicate that certain features of the early life environment may have lasting effects on future child metabolic health and highlight the potential obesogenic role of Bisphenol A (BPA) - a high volume production chemical detectable in 93% of the United States population. Mechanisms of BPA action remain uncertain however a leading hypothesis argues that BPA exposure may result in epigenetic changes, such as altered deoxyribonucleic acid (DNA) methylation, affecting expression of adipogenic genes.

To address these data gaps, we proposed the following specific aims:

(1) To measure plasma adiponectin and leptin in Mexican-American children from the Center for Health Assessment of Mothers and Children of Salinas (CHAMACOS) cohort at birth and again at 2, 5, and 9 years, examining heterogeneity in adipokine growth patterns and their association with candidate perinatal factors.

(2) To determine whether maternal or concurrent urinary BPA concentrations are associated with adiponectin and/or leptin levels in children.

(3) To characterize DNA methylation structure of peroxisome proliferator-activated receptor gamma (PPARy) - the master regulator gene in adipogenesis, determine whether PPARy methylation is associated with child adipokines and/or body size and whether prenatal or concurrent BPA may influence PPARy methylation.

Our results highlight several developmental differences in adiponectin vs. leptin over the childhood period. While leptin levels closely and positively correlated with child body size at all ages, adiponectin had inverse and weaker associations with body mass index (BMI) at 2, 5, and 9 years. Further, adjusting for BMI, adiponectin reflected an improved lipid profile while leptin was directly related to systolic and diastolic blood pressure in 9-year-old children.

Of the candidate perinatal factors examined, we identified maternal consumption of sugar-sweetened beverages (SSB) during pregnancy and increased rate of growth during the first 6 months of life as significant risk factors for altered adiponectin levels during childhood. Further, children with greater birth weight had rapidly-rising leptin levels over the birth to 9-year period and highest BMI and waist circumference at 9 years.

Our BPA analyses indicated sexually dimorphic responses similar to those previously reported in animal studies. While BPA concentrations during early pregnancy were directly associated with adiponectin levels in 9-year-old girls (b=3.71, P=0.03, N=131), BPA concentrations during late pregnancy were associated with increased plasma leptin in 9-year-old boys (b=0.06, P=0.01, N=179), controlling for sociodemographics, dietary variables and child BMI.

Finally, using the Infinium Illumina 450K Array, we examined DNA methylation in 23 sites spanning the PPARy promoter and gene body region in discovery (N=117 at birth, N=108 at 9 years) and validation (N=116 at birth, N=131 at 9 years) sets of children. We report that methylation in site 1 was significantly and negatively associated with child size at birth (b=-2.5, P=0.04) and at 9 years (b=-4.8, P<0.001) in the discovery set, and these relationships were replicated in the validation set. Overall our research adds evidence in support of the hypothesis the children's metabolic health may be programmed during early life and suggests that epigenetic mechanisms may play an important role in determining child size.