Role of glucosamine in development of diabetic neuropathy independent of the aldose reductase pathway

Long-term metabolic aberrations contribute to the development of diabetic neuropathy but the precise mechanism or mechanisms remains elusive. We have previously shown that aldose reductase-deficient mice exhibit delayed onset and progression of neuropathy following induction of diabetes, suggesting a role both for downstream metabolites of this enzyme and also for other unrelated pathways. In this study, we have utilized comprehensive metabolomics analyses to identify potential neurotoxic metabolites in nerve of diabetic mice and explored the mechanism of peripheral nerve injury. Aldose reductase knockout and control C57Bl/6J mice were made diabetic by injection of streptozotocin and followed for 8–16 weeks. Diabetic aldose reductase knockout mice exhibited delayed onset of nerve conduction slowing compared to diabetic wild-type mice. The sciatic nerves from aldose reductase knockout mice exposed to 12 weeks of diabetes were used for metabolomics analysis and compared with analyses of nerves from age-matched diabetic wild-type mice as well as non-diabetic aldose reductase knockout and wild-type mice. Neurotoxicity of candidate metabolites was evaluated using cultured Schwann cells and dorsal root ganglion neurons, and further confirmed in vivo. Metabolomics analysis identified elevated glucosamine levels in both diabetic aldose reductase knockout and diabetic wild mice. Exposure to glucosamine reduced survival of cultured Schwann cells and neurons accompanied by increased expression of cleaved caspase 3, CCAT-enhancer-binding homologous protein and mitochondrial hexokinase-I, along with ATP depletion. These changes were suppressed by siRNA to hexokinase-I or the ATP donor, inosine, but not by the antioxidant N-acetylcysteine or the endoplasmic reticulum-stress inhibitor 4-phenylbutyrate. The O-GlcNAcylation enhancer, O-(2-acetamido-2-deoxy-d-glucopyranosylidene) amino N-phenylcarbamate, did not augment glucosamine neurotoxicity. Single dose glucosamine injection into mice caused a reduction of sciatic nerve Na, K-ATPase activity, ATP content and augmented expression of hexokinase-I, which were suppressed by pretreatment with inosine but not with 4-phenylbutyrate. Mice implanted with a subcutaneous pump to infuse glucosamine for 12 weeks developed nerve conduction slowing and intraepidermal nerve fibre loss, recapitulating prominent indices of diabetic neuropathy. While acute glucosamine neurotoxicity is unlikely to contribute substantially to the slowly developing neuropathy phenotype in humans, sustained energy deprivation induced by glucosamine may well contribute to the pathogenesis of diabetic neuropathy. Our data thus identifies a novel pathway for diabetic neuropathy that may offer a potential new therapeutic target.