We explored the specific impact of polyol pathway hyperactivity on dorsal root ganglia (DRG) using transgenic mice that overexpress human aldose reductase because DRG changes are crucial for the development of diabetic sensory neuropathy. Littermate mice served as controls. Half of the animals were made diabetic by streptozotocin injection and followed for 12 weeks. After diabetes onset, diabetic transgenic mice showed a significant elevation of pain sensation threshold after transient decrease and marked slowing of motor and sensory nerve conduction at the end of the study, while these changes were modest in diabetic littermate mice. Protein kinase C (PKC) activities were markedly reduced in diabetic transgenic mice, and the changes were associated with reduced expression of membrane PKC-␣ isoform that was translocated to cytosol. Membrane PKC-␤II isoform expression was contrariwise increased. Calcitonin gene-related peptide-and substance P-positive neurons were reduced in diabetic transgenic mice and less severely so in diabetic littermate mice. Morphometric analysis disclosed neuronal atrophy only in diabetic transgenic mice. Treatment with an aldose reductase inhibitor (fidarestat 4 mg ⅐ kg -1 ⅐ day -1, orally) corrected all of the changes detected in diabetic transgenic mice. These findings underscore the pathogenic role of aldose reductase in diabetic sensory neuropathy through the altered cellular signaling and peptide expressions in DRG neurons. Diabetes 53:3239 -3247, 2004 P olyol pathway hyperactivity has been extensively studied for the mechanisms of diabetic neuropathy, where aldose reductase is a key regulating enzyme (1,2). In animal models, diabetes-induced peripheral nerve conduction deficits, neurometabolic imbalances, altered nerve blood flow, and morphologic abnormalities are prevented by structurally diverse aldose reductase inhibitors (ARIs) (3-5). Past clinical trials of ARIs were not, however, convincingly successful (6,7), and specific effects of polyol pathway hyperactivity on the clinical and structural aspects of diabetic sensory neuropathy is yet to be clear (8,9). The development of a transgenic animal model provides an ideal tool to identify the specific role of single molecules in disease mechanisms. We have established a strain of transgenic mice that overexpress human aldose reductase (10 -12). In this model, we confirmed that polyol pathway hyperactivity was indeed related to the severity of motor nerve conduction delay and nerve fiber atrophy. More recently, a new transgenic model that overexpresses aldose reductase, specifically in Schwann cells by use of myelin protein Po promoter, has enabled researchers to address more precise mechanisms of polyol pathway in the development of neuropathy in diabetes (13). Nevertheless, from these studies, it was not shown which measures were comparable with those found in human diabetic neuropathy, and it still remains obscure what could be the most appropriate target for the intervention with a potent and specific inhibitor for human aldos...

To explore the relationship between polyol pathway and protein kinase C (PKC), we examined PKC activities and expressions of PKC isoforms separately in endoneurial and vessel-rich epineurial tissues in diabetic mice transgenic for human aldose reductase (Tg). Tg and littermate control mice (Lm) were made diabetic by streptozotocin at 8 weeks of age and treated orally with aldose reductase inhibitor (ARI) (fidarestat 3-5 mg/kg/day) or placebo for 12 weeks. At the end, compared with non-diabetic state, sorbitol contents were increased 6.4-fold in endoneurium and 5.1-fold in epineurium in diabetic Tg, whereas the increase was detected only in endoneurium in diabetic Lm. Endoneurial PKC activity was significantly reduced in diabetic Tg. By contrast, epineurial PKC activity was increased in both diabetic Lm and diabetic Tg and there was no significant difference between the two groups. These changes were all corrected by ARI treatment. Consistent with the changes of PKC activities, diabetic Tg showed decreased expression of PKCa in endoneurium, whereas there was an increased expression of PKCbII in epineurium in both diabetic Tg and diabetic Lm. These findings suggest the presence of dichotomous metabolic pathway between neural and vascular tissues in the polyol-PKC-related pathogenesis of diabetic neuropathy.