Diabetic retinopathy (DR) is a serious complication of diabetes. It is driven by the loss of pericytes, cells which control vascular function, due to excess formation of sugar alcohols by the enzyme aldose reductase (AR). AR has been shown to be present in pericytes in retinal capillaries in human and animal studies. Several animal models for DR have been described including dogs, rats and mice. However, the pathways and concentrations of AR activity and production of sugars appears to differ between species. While AR inhibitors were effective in dogs they had only minor benefits in humans. Moreover, accumulation of retinal sugars in streptozotocin-induced diabetic rats was significantly higher than in diabetic mice. Therefore, new model systems in mice with lower AR activity are required. In this paper transgenic mice were produced expressing human AR and green fluorescent protein under the control of the α-smooth muscle actin promoter (SMAA-GFP-hAR). Diabetes was induced by crossing these mice to the naturally diabetic C57BL/6-Ins2Akita/J mice (AK-SMAA-GFP-AR). The results showed that sorbitola levels measured by HPLC were higher in diabetic mice compared to controls. AK-SMAA-GFP-AR mice showed induction of VEGF, IGF-1, bFGF and TGFβ, all vascular angiogenic factors, as well as complementary signalling molecules. A significant increase in acellular capillaries was also seen compared to control mice. The development of this model has several advantages over current animal studies. Manipulating mouse genomes is significantly more advanced than in rats and more antibodies available for analysis. Moreover mice are substantially cheaper to breed and maintain than rats and dogs. The potential for this mouse model to be used in testing new AR inhibitors or other drugs is an obvious benefit of this study.