Discrepancies in the regulation of plasma adiponectin and TNF-alpha levels and adipose tissue gene expression in obese African Americans with glucose intolerance: a pilot study using rosiglitazone.

OBJECTIVES: We examined the effects of rosiglitazone: 1) on glucose homeostasis, insulin action, beta-cell function, and plasma adiponectin and TNF-alpha (TNF-alpha) levels; and 2) the expression of adipose tissue TNF-alpha and adiponectin mRNA in African Americans with parental history of type 2 diabetes and with varying degrees of glucose intolerance.

SUBJECTS AND METHODS: The study groups comprised 11 African Americans with normal glucose tolerance and six with diabetes and impaired glucose tolerance. The glucose-intolerant subjects received rosiglitazone (4-8 mg/day) every morning for 12 weeks. They underwent oral glucose tolerance test (OGTT) and subcutaneous adipose tissue biopsy (under local anesthesia) before and after 12 weeks of rosiglitazone therapy. Beta cell function and insulin resistance were calculated by using homeostasis model assessment (HOMA). Adipose tissue gene expression (mRNA) was measured by real-time polymerase chain reaction in both groups.

RESULTS: Rosiglitazone monotherapy improved both fasting and two-hour serum glucose levels during OGTT in the glucose-intolerant group. However, mean serum insulin and C-peptide levels did not change when compared with baseline. Rosiglitazone monotherapy improved insulin resistance but not overall beta-cell secretion. Mean adiponectin levels at fasting and two hours after oral glucose ingestion were significantly (50%) lower in the glucose-intolerant group than in the control group. Rosiglitazone monotherapy significantly increased plasma adiponectin levels at fasting and two hours after oral challenge by two-fold in the glucose-intolerant group. Mean plasma TNF-alpha levels were not significantly different at fasting and after two hours during OGTT. Rosiglitazone had no significant effect on plasma TNF-alpha levels during OGTT. No significant differences were seen in the expression of adipose tissue TNF-alpha and adiponectin mRNA in the groups at baseline. Rosiglitazone did not significantly change the adipose tissue adiponectin and TNF-alpha mRNA. Rosiglitazone was well tolerated, without experiencing weight gain, edema, and liver function test abnormality in the glucose intolerant subjects.

SUMMARY: Rosiglitazone improved glucose homeostasis and insulin resistance in high-risk African Americans. We found that adiponectin was lower in the glucose-intolerant group, while TNF-alpha was similar. While rosiglitazone increased plasma adiponectin, it had no effect on adipose tissue adiponectin mRNA. In addition, rosiglitazone had no effect on plasma TNF-alpha and adipose tissue TNF-alpha mRNA. We conclude that the metabolic effects of rosiglitazone could be mediated by adiponectin but not TNF-alpha in African Americans with glucose intolerance. Our study demonstrates that: 1) the role of adipocytokines in the etiology of type 2 diabetes in African Americans is complex; and 2) that adiponectin, but not TNF-alpha, could mediate the metabolic benefits of thiazolidinediones in African Americans with glucose intolerance.