LB-FP-5 Novel Markers of Obesity: a Population-Based Proteomic Approach

Program: Late-Breaking Abstracts
Session: SAT-LB-Late-Breaking Poster Session 1
Bench to Bedside
Saturday, June 15, 2013: 1:45 PM-3:45 PM
Expo Halls ABC (Moscone Center)

Poster Board SAT-LB-04
Christine G Lee*1, Aaron Baraff2, Richard Smith3, Erin Baker3, Vladislav Petyuk3, Peggy M Cawthon4, Douglas C Bauer5, David Gibbs1, Arie Baratt1, Shannon McWeeney1, Jodi Lapidus1 and Eric S Orwoll1
1Oregon Health & Science University, 2University of Washington, 3Pacific Northwest National Laboratory, 4California Pacific Medical Center, San Francisco, CA, 5UCSF, San Francisco, CA
Introduction: Despite concerns about the public health burden of obesity, there are few targets for preventing or treating obesity. The discovery of markers and mediators of obesity is needed for elucidating mechanisms of obesity and its complications. Given the high prevalence of obesity in older adults, we developed a new population-based proteomic approach to identify peptides and proteins associated with obesity in older men.

Methods: We performed a cross-sectional analysis of 2473 ambulatory, community-dwelling men ages ≥65 years enrolled in the Osteoporotic Fractures in Men Study from 6 sites in the U.S. The categorization of obese (BMI≥30), overweight (BMI 25.0-29.9) and normal weight (BMI 18.5-24.9) was based on WHO criteria. High-throughput quantitative proteomic analysis was performed on serum samples using a multi-dimensional approach coupling liquid chromatography, ion-mobility separation, and mass spectrometry (LC-IMS-MS). Peptides that were differentially abundant in obese versus normal weight men were identified using analysis of covariance adjusted for multiple comparisons using the Storey method with a false discovery rate (FDR) q-value < 0.05. Models also included adjustments for age, site, comorbidities, lifestyle factors, and medication use.  Meta-analytic methods accounting for correlated metrics were used to generate protein-obesity association rankings from averaged differential abundance and combined p-values for peptides of each protein.

Results: Among the older men, 536 (21.7%) were obese and 650 (26.3%) were of normal weight. Of 18485 identified serum peptides, 1237 were associated with obesity in fully adjusted models with a q-value<0.05, and these peptides mapped to 169 proteins with a q-value < 0.05. Among these proteins were well-known markers of obesity: obese men had a higher abundance of C-reactive protein (q-value=0.003) and a lower abundance of adiponectin (q-value=0.005). The 5 top-ranking proteins associated with obesity were zinc-alpha-2 glycoprotein (ZAG2), glutathione peroxidase 3 (GPX3), vitamin D-binding protein (DBP), putative zinc-alpha-2-glycoprotein-like 1 (ZAGL1), and afamin. Obese men had 20-26% lower abundance of ZAG2, GPX3, DBP, and ZAGL1, and 34% higher abundance of afamin compared to normal weight men, q-values<10-10.

Conclusion: We have demonstrated a rapid and broad assessment of peptides and proteins associated with obesity using a novel, population-based proteomic approach. Traditional obesity markers like C-reactive protein were identified, and our findings support prior reports of decreased ZAG2 and GPX3 expression in adipose tissue of obese subjects and gene polymorphisms in DBP associated with high adiposity. Among the top ranking 5 proteins were 2 not previously reported with obesity, afamin and ZAGL1. Further research to understand the biologic roles of these proteins in obesity is needed.

Nothing to Disclose: CGL, AB, RS, EB, VP, PMC, DCB, DG, AB, SM, JL, ESO

*Please take note of The Endocrine Society's News Embargo Policy at

Sources of Research Support: The Osteoporotic Fractures in Men (MrOS) Study is supported by National Institutes of Health funding. The following institutes provide support: the National Institute ofArthritis and Musculoskeletal and Skin Diseases (NIAMS), the National Institute on Aging (NIA),the National Center for Research Resources (NCRR), and NIH Roadmap for Medical Researchunder the following grant numbers: U01 AR45580, U01 AR45614, U01 AR45632, U01 AR45647,U01 AR45654, U01 AR45583, U01 AG18197, U01-AG027810, and UL1 RR024140. Population proteomic technology, informatics and statistical methods were supported by NIH/NCRR grant number UL1 RR024140-04S2.