MATERIALS AND METHODS

Adult male C57BL/6J (H-2b) mice (The Jackson Laboratory, Bar Harbor, ME) were used as both donors and recipients. Diabetes was induced in recipient mice by a single intraperitoneal (IP) injection of streptozotocin (STZ), 220 mg/kg body weight (Upjohn Laboratories, Kalamazoo, Mich). Only mice showing hyperglycemia for at least 4 to 5 days and with nonfasting blood glucose levels above 350 mg/dL for at least two consecutive days were used as islet recipients. Pancreatic islets were isolated by intraductal injection of collagenase followed by stationary digestion at 37 C for 36 minutes. After dextran gradient purification, 200 islets per recipient were handpicked free of contaminating tissue. The islets were implanted into the renal subcapsular space. Twelve islet recipients received CNI-1493 (10 mg/kg/d, twice daily, IP), beginning on the day preceding transplantation and twice daily thereafter for 10 days. Transplanted STZ-induced diabetic mice that did not receive CNI-1493 but were injected with saline were used as controls (n 7). Nonfasting blood glucose levels were determined with test strips (LifeScan, Milpitas, Calif) daily for the first week posttransplantation and three times a week thereafter until achievement of blood glucose levels less than 250 mg/dL on two consecutive days, then weekly until sacrifice after 100 days. Data analysis was carried out using the one-tailed Mann-Whitney test and differences were considered significant when P. 05. IP TNF concentrations in control and CNI-treated islet isograft recipients were measured at days 3, 5, and 8 posttransplantation. Three animals were used at each time interval. The TNF concentration was determined by enzyme-linked immunosorbent assay (ELISA) and plated in triplicate in 96-well microtiter plates according to the manufacturer’s instructions. The TNF concentration represents the mean of results obtained from three animals in each group.