Context: Zinc-α2-glycoprotein (ZAG) is a soluble protein similar to the class I major histocompatibility complex heavy chain, which has been implicated in lipid catabolism. We hypothesized that ZAG mRNA expression in adipose tissue maybe linked with lipolytic and adipokine gene expression and have a close relationship with clinical phenotype. Objectives: The objective of the study was to analyze ZAG gene expression in human adipose tissue from lean and obese subjects. ZAG circulating plasma levels and its relationship with cardiometabolic risk factors were also studied. Design: Seventy-three Caucasian (43 male and 30 female) subjects were included. Plasma and adipose tissue [sc (SAT) and visceral (VAT)] from the same patient were studied. mRNA of PPARγ, hormone-sensitive lipase (HSL), adipose triglyceride lipase, adiponectin, omentin, visfatin, and ZAG were quantified. Plasma concentrations of ZAG were determined with ELISA. Results: ZAG plasma levels showed a negative correlation with insulin (r = -0.39; P = 0.008) and the homeostasis model assessment for insulin resistance index (r = -0.36; P = 0.016). No differences in ZAG circulating levels according to body mass index classification were observed. ZAG expression in SAT was significantly reduced in overweight and obese individuals compared with lean subjects (P < 0.001 and P = 0.007, respectively). ZAG mRNA expression in both SAT and VAT depots were negatively correlated with many clinical and metabolic cardiovascular risk factors. After multiple linear regression analysis, SAT ZAG was mainly predicted by adiponectin mRNA expression (B = 0.993; P < 0.0001) and plasma triglyceride levels (B = -0.565; P = 0.006). VAT ZAG expression was predicted by adiponectin expression (B = 0.449; P < 0.0001), and HSL VAT expression (B = 0.180; P = 0.023). Conclusions: The present study provides evidence of a role of ZAG gene in adipose tissue metabolism, with a close association with adiponectin gene expression in sc and visceral fat. Copyright © 2009 by The Endocrine Society.