INTACT PROINSULIN, DES-31,32 PROINSULIN, AND SPECIFIC INSULIN CONCENTRATIONS AMONG NONDIABETIC AND DIABETIC SUBJECTS IN POPULATIONS AT VARYING RISK OF TYPE-2 DIABETES

OBJECTIVE - To examine hyperinsulinemia, insulin secretion, and beta-c ell function in Pima Indians, South Asians, and whites, populations at varying risk of diabetes. RESEARCH DESIGN AND METHODS-We investigated 136 Pima Indian, 98 Asian, and 80 white nondiabetic and 172 Pima Indi an, 40 Asian, and 49 white diabetic subjects. Highly specific assays f or insulin, intact proinsulin, and des 31,32 proinsulin were used. Ins ulin secretion was assessed using ratio of increment (0 to 30 min) in insulin to glucose concentrations during an oral glucose tolerance tes t (OGTT). RESULTS-Nondiabetic Pima Indians were significantly more obe se than Asians and whites. Pima Indian subjects had significantly high er (P < 0.01) fasting insulin concentrations (median 109 pmol/l, range 40-250) than Asian (37 pmol/l, range 17-91) and white (30 pmol/l, ran ge 10-82) subjects. These differences remained significant when contro lled for obesity Nondiabetic Pima Indians also had higher fasting C-pe ptide concentrations and higher early insulin secretion during an OGTT Fasting concentrations of intact proinsulin and des 31,32 proinsulin were also significantly higher in Pima Indians (P < 0.01). However, th e proportion of proinsulin-like molecules was significantly lower (P < 0.01) in Pima Indians (median 7.9% vs. 12.7% for South Asians and 12. 2% for whites). Subjects with diabetes from the three ethnic groups sh owed significantly higher fasting insulin concentrations but lower 30- min insulin and lower ratios of increment (0-30 min) in insulin to glu cose concentrations than did nondiabetic subjects, The proportion of p roinsulin-like molecules was not significantly different in diabetic s ubjects from the three ethnic groups. CONCLUSIONS - These specific ass ays for insulin indicate that after adjusting for obesity, nondiabetic Pima Indians are truly hyperinsulinemic, which is consistent with the ir insulin resistance as measured by other methods. Hyperinsulinemia i n this population with a high risk of diabetes is likely to be due to enhanced insulin secretion. Furthermore, in Pima Indians, the predomin ant beta-cell secretory product is insulin and not its precursors. We conclude that the differences in the risk of diabetes among these thre e groups are not due to differences in insulin secretion or insulin pr ocessing. Subjects with type 2 diabetes have defective early insulin s ecretion during OGTTs but show fasting hyperinsulinemia even when spec ific assays for insulin are used.