Submitted to: Obesity
Publication Type: Abstract only
Publication Acceptance Date: 9/1/2008
Publication Date: 10/1/2008
Citation: Bray, M.S., Sailors, M., Kueht, M., McFarlin, B., Jackson, A. 2008. Genetic predictors of exercise adherence in the Tiger study [abstract]. Obesity. 16:7-OR (Suppl. 1). Interpretive Summary:
Technical Abstract: Obesity established in adolescence strongly predicts obesity for the remainder of adult life, suggesting this is a critical time in which to establish healthy diet and physical activity behaviors. Many studies have reported low levels of habitual activity among this age group, however. We hypothesize that genetic variation may influence physical activity adherence. Therefore, the purpose of this study was to investigate the association between genetic variation and adherence to 15 weeks of aerobic exercise training. We conducted an initial examination of genes related to neural signaling and energy metabolism for association to dropout status in the initial two cohorts of the TIGER study (n=130 males and 267 females). We analyzed a total of 260 SNPs (59 genes) for differences in allele frequencies between exercise drop-out and non-drop-out groups, and identified 19 SNPs with significant differences between these two groups. Permutation testing (500 iterations in each of 20 rounds of testing) was used to confirm these results, eliminating three of the initial SNPs identified. The significant SNPs were then analyzed in logistic regression models, controlling for baseline age, gender, race/ethnicity, BMI, DXA percent fat, baseline physical activity rating, and waist circumference. Eight SNPs in seven genes remained significantly associated with exercise drop-out status at 15 weeks after controlling for the covariates listed above: POMC, PCSK1, BDNF, CLOCK, UCP2, PPARGC1, and LIPC. Of these, POMC encodes a precursor protein that is cleaved by PCSK1, producing corticotropin, alpha-melanocyte stimulating hormone (associated with the anorectic effects of exercise), and beta-endorphin (a neuropeptide associated with the "runner’s high"). Genotyping of the POMC variant in the entire study cohort (n=1338) resulted in a significant association with 15-week drop out (p<0.048), after controlling for the covariates listed above. In race-specific analyses, these effects were strongest in the non-Hispanics white subjects (OR=0.66, CI=0.47,0.93, p<0.01). Our preliminary results suggest that common variation in POMC may be associated with exercise drop-out rate. These exciting findings provide the first evidence of a genetic basis for exercise adherence.