We have previously shown that myo-inositol depletion in the embryonic tissue at a critical stage of organogenesis has a crucial role in hyperglycemia-induced embryopathy. This study tested whether myo-inositol depletion in early organogenesis contributes to the pathogenesis of streptozocin-induced diabetic embryopathy. Rats were made diabetic by streptozocin administration before conception, and the diabetic rats were treated with diet supplemented by 2% myo-inositol or insulin from 6 to 11 gestational days during the period of maximum teratological susceptibility. In each group on the 11th gestational day, growth retardation and incidence of malformations were recorded, and myo-inositol and sorbitol content in the embryonic and extraembryonic tissues were examined. In diabetic rats, the myo-inositol content of the embryos was decreased by 36% (P < 0.01) compared with control rats, and there was growth retardation (crown-rump length 3.37 +/- 0.04 vs. 3.87 +/- 0.03 mm, P < 0.01; somite no. 27.5 +/- 0.2 vs. 29.1 +/- 0.2, P < 0.01) and a significantly increased incidence of the neural lesions (17.6 vs. 1.9%, P < 0.01). Insulin treatment resulted in near normalization of maternal serum glucose and complete restoration of myo-inositol content in the embryos with significant improvement of the growth retardation (crown-rump length 3.55 +/- 0.06 vs. 3.37 +/- 0.04 mm, P < 0.05; somite no. 28.2 +/- 0.13 vs. 27.5 +/- 0.2, P < 0.05) and a significantly lowered incidence of neural lesions (2.5 vs. 17.6%, P < 0.01) compared with those of the untreated diabetic rats. Despite complete restoration of myo-inositol content in the embryos by oral supplementation with myo-inositol of the diabetic rats, the embryos showed no improvement of growth retardation and had partially lowered incidence of neural lesions (9.6 vs. 17.6%, P < 0.05) but still had higher incidence of neural lesions than controls (9.6 vs. 1.9%, P < 0.05). We demonstrated that metabolic improvement of maternal diabetes by insulin treatment could prevent early growth retardation and dysmorphogenesis, whereas correction of only myo-inositol depletion did not. These data suggest that diabetes-induced growth retardation and dysmorphogenesis during early organogenesis may be caused by multifactorial factors and cannot be explained solely by the myo-inositol depletion hypothesis.