Insulin Binding and Glucose Transport in Adipocytes in Neonatal Streptozocin-Injected Rat Model of Diabetes Mellitus

The neonatal streptozocin (STZ)-injected rat (NSIR) model of diabetes mellitus resembles human non-insulin-dependent diabetes mellitus (NIDDM) with respect to abnormalities in insulin secretory responses. The suggestion that insulin deficiency leads to insulin resistance, a prominent feature of human NIDDM, led us to examine insulin binding and glucose transport in the NSIR during the development of hyperglycemia. Male Wistar rats were injected at 2 days of age with STZ (90 mg/kg i.p.) or vehicle alone. Mild insulin deficiency, reflected by minimally decreased fed plasma insulin concentrations, was apparent at 4 wk (mean ± SE, control vs. NSIR, 2.32 ± 0.19 vs. 1.75 ± 0.21 ng ml) and at 8 wk. Pancreatic insulin content was dramatically reduced in NSIR to 12 and 5% of control values at 4 and 8 wk, respectively (P < .001). Fed plasma glucose concentrations increased in the NSIR between 4 and 5 wk and were significantly elevated at 8 wk (251 ± 25 vs. 527 ± 52 mg/dl, P < .001). ¹²⁵I-labeled insulin binding showed a progressive increase as a function of adipocyte volume in control and NSIR. Epididymal fat pad weights and adipocyte volumes were significantly decreased in the NSIR. Thus, insulin binding did not differ when expressed per cell number but was increased in NSIR when corrected for cell size (percent specific binding × 10², 8.49 ± 0.96 vs. 11.56 ± 1.08/μl cell vol; P < .05, all ages combined). Insulin-stimulated 3-O-methylglucose transport was similar at 4 wk but significantly decreased in the NSIR at 8 wk (clearance, 272.6 ± 69.2 vs. 106.8 ± 21.0 fl · cell⁻¹ · s⁻¹, P < .05). However, when expressed as fold stimulation above basal or corrected for cell volume, no differences were observed. The abnormalities in insulin action that develop in the NSIR model of diabetes mellitus are characteristic of an insulin-deficient state. Comparison of the features of the insulin resistance demonstrated in this rodent model with those described in human NIDDM reveals both similarities and differences. We suggest that the NSIR model provides an incomplete representation of human NIDDM.