Hyperglycemia is known to reduce dehydroepiandrosterone (DHEA) circulating levels; however, the mechanism by which hyperglycemia decreases DHEA is not elucidated. In this study, serum DHEA and DHEA sulfate (DHEA-S) levels were compared in 50 men with non-insulin-dependent diabetes mellitus (NIDDM) and 50 age-matched men with impaired glucose tolerance (IGT) receiving only diet therapy. Serum concentrations of DHEA and DHEA-S in the NIDDM group were significantly lower than in the IGT group (7.8 and 9.7 nmol/l vs 3.4 and 4.9 mumol/l, respectively; p < 0.01) but there was no significant difference in immunoreactive insulin between the two groups. When the results from both groups were combined, HbA1C was significantly inversely related to DHEA (r = -0.243, p < 0.01) and DHEA-S (r = -0.305, p < 0.01). Immunoreactive insulin showed no correlation with DHEA and DHEA-S. Multiple regression analysis showed that HbA1C was independently negatively related to both DHEA and DHEA-S. We conclude that hyperglycemia may decrease serum DHEA and DHEA-S in Japanese men with NIDDM, but the depression of DHEA(-S) is independent of serum insulin level.

Objective: To clarify the in¯uence of insulin therapy on body weight and fat distribution, we compared these parameters in ®ve non-insulin dependent diabetes mellitus (NIDDM) patients, with secondary sulfonylurea failure, before and after insulin therapy. Body weight increased signi®cantly after instituting insulin treatment. However, the visceral to subcutaneous fat (VaS) ratio decreased signi®cantly due to a marked increase in S-fat without a change in V-fat. Insulin therapy necessitated by sulfonylurea failure does not appear to accelerate the atherogenic process in NIDDM patients as there is no increase in visceral fat.

Abstract. Recent studies indicate that experimentally induced hyperinsulinemia may reduce serum dehydroepiandrosterone (DHEA) and dehydroepiandrosterone-sulfate (DHEA-S). Serum DHEA and DHEA-S decrease in diabetic patients, but the mechanism by which hyperglycemia decreases DHEA and DHEA-S is unknown. In this study, we investigated the effect of hyperglycemia on DHEA and DHEA-S in impaired glucose tolerance (IGT) by means of the 75g-oral glucose tolerance test (OGTT). We selected 30 male IGT patients receiving diet therapy only, whose insulinogenic Index was under 0.3. Oral glucose challenge significantly reduced DHEA (P=0.0001) and DHEA-S (P<0.05) at 60 and 120 min after OGTT. Setting the value of DHEA and DHEA-S at time zero as 100%, we calculated the DHEA and DHEA-S values at 60 and 120 min after OGTT as %DHEA(-S) 60 min and %DHEA(-S)120 min, respectively.DHEA and DHEA-S at time zero showed no correlation with BMI, HbAlc, the sum of insulin values (~IRI) or the area under the curve of plasma glucose (AUC). We found decreases in %DHEA 60 min (r=-0.411, P<0.05), %DHEA-S 60 min (r= -0.508, P<0.01) and %DHEA-S 120 min (r=-0.393, P<0.05) as AUC increased, but ~IRI showed no correlation with %DHEA(-S) 60 min or %DHEA(-5)120 min. We conclude that the depression of DHEA and DHEA-S after OGTT is attributable to hyperglycemia in male Japanese IGT with low insulin response.

Abstract. Macrophages from recent onset non-obese diabetic (NOD) mice showed cytotoxicity against the NOD mouse derived beta-cell line, MIN6N-9a. In this report, we examined whether nitric oxide is associated with beta-cell destruction. Peritoneal exudate cells (PEC), obtained from cyclophosphamide treated NOD mice showed higher cytotoxicity against MIN6N-9a compared to PECs from saline injected NOD mice (P<0.01). This effect was suppressed in cells incubated with 0.5 mmol/l NG-methyl-L-arginine, a nitric oxide synthase inhibitor (P<0.001). In addition, the nitrite concentration of the co-culture medium, as an index of nitric oxide production, increased in MIN6N-9a cells co-cultured with peritoneal exudate cells from cyclophosphamide injected NOD mice but not in co-culture with saline injected NOD mice (P<0.05). Thus, nitric oxide plays an important role in beta-cell line destruction of macrophages obtained from NOD mice.