1. To investigate the pharmacological effects of T-1095, this novel derivative of phlorizin was administered to GK rats for 8 weeks. T-1095 treatment significantly lowered plasma glucose and glycosylated haemoglobin (HbA1c) levels, but did not significantly affect bodyweight. 2. T-1095 treatment did not affect 3.3 mmol/L glucose-induced insulin secretion in the isolated perfused pancreas of GK rats. 3. The peak insulin release in T-1095-treated GK rats was significantly higher during the first phase than in untreated GK rats (3-4 min after beginning 16.7 mmol/L glucose perfusion). The total amount of insulin secreted during the first phase in T-1095-treated GK rats was significantly higher than in untreated GK rats (35.3 +/- 1.4 vs. 27.3 +/- 2.5 ng in T-1095-treated compared with untreated rats, respectively). 4. During the second phase, insulin release in T-1095-treated GK rats was somewhat higher than in untreated GK rats (7-30 min after beginning 16.7 mmol/L glucose perfusion). The total amount of insulin secreted during the second phase in T-1095-treated GK rats was significantly higher than in untreated GK rats (88.2 +/- 6.1 vs. 68.1 +/- 5.7 ng, respectively). 5. The total amount of insulin secreted during perfusion in T-1095-treated GK rats was significantly higher than in untreated GK rats (123.5 +/- 7.3 vs. 95.4 +/- 7.7 ng, respectively). 6. These data show that the metabolic indices, plasma glucose and HbA1c levels and insulin secretion are significantly improved by T-1095 treatment in GK rats, which are spontaneously diabetic rats, suggesting its usefulness as a novel oral therapeutic antidiabetic agent.