CYP19A1 promoter methylation in saliva associated with milestones of pubertal timing in urban girls

BACKGROUND Childhood obesity and early puberty are intermediate risk factors for later metabolic and reproductive disorders including diabetes, polycystic ovarian syndrome (PCOS), and breast cancer. Atypical methylation patterns in genes related to hormone and adipose metabolism, such as CYP19A1 (aromatase) and PPARG (peroxisome proliferator-activated receptor gamma), are associated with alterations in gene expression which may contribute to pathogenesis of these diseases. If present in early life, it is conceivable similar methylation aberrations may result in hormone perturbations that alter pubertal timing. METHODS We used Cox proportional hazard models to investigate whether promoter methylation of CYP19A1 and PPARG, independently or in concert with body weight, was associated with age at breast (B2) or pubic hair development (PH2) when assayed in saliva DNA collected from a cohort of New York City, Black and Hispanic girls (N = 130) enrolled in a study of pubertal timing between 6-8 years of age. RESULTS An inverse association between CYP19A1 methylation and risk of early PH2 was suggested (HR = 0.95, 95% CI = 0.90-1.00, p = 0.05). CYP19A1 methylation also appeared to modify risk of early B2 associated with body weight. Specifically, compared to normal weight girls with 'high' CYP19A1 methylation, significantly increased risk of early B2 was observed in overweight girls with 'low' but not 'high' CYP19A1 methylation (HR = 2.15; 95% CI = 1.23- 3.76). However, in formal tests for effect modification, the interaction between body weight and methylation did not reach statistical significance (p for interaction = 0.085). PPARG methylation was not significantly associated with PH2 or B2. CONCLUSIONS Though limited by sample size, our findings suggest methylation of CYP19A1, a critical gene in estrogen biosynthesis, may influence timing of breast development in overweight girls. Consistent with emerging reports, these data support the notion that epigenetic marks in surrogate tissues may improve risk prediction when added to standard plasma and anthropometric indicators, and warrant further study.