LAUSR

Growth differentiation factor 15 (GDF15) reduces food intake, dietary fat preference, and body weight in pre-clinical studies. Through recent human and in vitro studies, we have identified a role for GDF15 in the energy expenditure (exercise) side of the energy balance equation. Collectively, GDF15 has emerged as a novel candidate for the treatment of human obesity. However, whether GDF15 regulates food intake or food preference in humans remains to be elucidated. Herein, we hypothesized that fasting and exercise-induced GDF15 plasma concentrations would be related to habitual food intake and preference in adults with overweight/obesity. Samples previously collected from eleven adults (age: 54 ± 3, BMI: 32.4 ± 1.6, kg/m2, six women, five men) were subjected to secondary analysis. Fasting plasma GDF15 was quantified by ELISA, while habitual diet was measured from 3-day dietary records. Dietary data was analyzed by the Dietary Analysis Core at Pennington Biomedical Research Center using FNDDS (USDA 2015–2016). A sub-cohort (n = 9) further completed a 7-day aerobic exercise training for 60 min/day at ∼85% HRmax, GDF15 assessment was repeated post-training. At baseline, GDF15 showed trends in reduced total energy intake (kcals, R = −0.40, P = 0.22), and was inversely associated with fatty foods (oils, g/day) (R = −0.62, P < 0.05), consistent with animal models. Exercise increased GDF15 in six adults (responders), but decreased in three adults (non-responders). Responders habitually consumed 52% less total fat (g/1000kcals, P = 0.01) and 59% less oils (g/1000kcals, P = 0.01) compared to non-responders. In agreement, exercise-induced changes in GDF15 negatively associated with baseline fat (kcal/g, R = −0.68, P < 0.05) and cheese (serving/kcal, R = −0.67, P < 0.05) intake. This study provides preliminary evidence that GDF15 is linked to reduced fatty food preference in adults with overweight or obesity, suggesting a promising anti-obesity effect of GDF15 in humans. Larger human trials with repeated measurements are warranted to further elucidate the response. The study was supported by the National Institute of Health: UL1RR024989, U54GM104940, DK108089 (JPK) and T32AT004094 (JTM – trainee).