Partial preservation of pancreatic beta-cells by vanadium: evidence for long-term amelioration of diabetes.

Streptozotocin (STZ)-diabetic rats treated with vanadium can remain euglycemic for up to 20 weeks following withdrawal from vanadium treatment. In this study, we examined the effects of short-term vanadium treatment in preventing or reversing the STZ-induced diabetic state. Male Wistar rats were untreated (D) or treated (DT) with vanadyl sulfate for 1 week before administering STZ. Treatment was subsequently maintained for 3 days (DT3) or 14 days (DT14) post-STZ, after which vanadium was withdrawn. At 4 to 5 weeks post-STZ and following long-term withdrawal from vanadium, DT14 rats demonstrated levels of food and fluid intake and glucose tolerance that were not significantly different from those of age-matched untreated nondiabetic rats, and had significantly reduced glycemic levels in the fed state compared with D and DT3 groups. The proportion of animals that were euglycemic (fed plasma glucose < 9.0 mmol/L) was significant in DT14 (five of 10) relative to D (one of 10) and DT3 (one of 10) (P = .01). All euglycemic animals had an improved pancreatic insulin content that, albeit low (12% of control), was strongly linked to euglycemia in the fed state (r = -.91, P < .0001). Moreover, the highly significant correlation persisted with the analysis of untreated STZ-rats alone (r = -.95, P < .0001). Similarly, improvements in glucose tolerance and insulin secretory function in euglycemic rats were strongly correlated with small changes in residual insulin content. Hence, as vanadium pretreatment did not prevent STZ-induced beta-cytotoxicity, the vanadium-induced amelioration of the diabetic state appears to be secondary to the preservation of a functional portion of pancreatic beta cells that initially survived STZ toxicity. The partial preservation of pancreatic beta cells, albeit small in proportion to the total insulin store, was both critical and sufficient for a long-term reversal of the diabetic state. These results suggest that apparently modest effects in preserving residual pancreatic insulin content can have profound consequences on glucose homeostasis and may bear important implications for interventions that have "limited" protective effects on beta cells.