The sorbitol pathway of glucose metabolism consists of two reactions. First, glucose is reduced to its sugar alcohol sorbitol by NADPH-dependent AR. Then, sorbitol is oxidized to fructose by NAD-dependent SDH. Negative consequences of the sorbitol pathway hyperactivity under diabetic or hyperglycemic conditions include intracellular sorbitol accumulation and resulting osmotic stress, and generation of fructose, which is 10-times more potent glycation agent than glucose. One group reported that increased flux through SDH leads to so called "pseudohypoxia," i.e., an increased free cytosolic NADH/NAD+ (12) ratio, whereas others (36) did not find a relation between cytosolic or mitochondrial NAD+/NADH redox state and SDH activity in the peripheral nerve. Two groups obtained the results indicating that increased AR, but not SDH, activity contributes to PDN (37,38). Furthermore, SDH inhibition appeared detrimental rather than beneficial, for autonomic neuropathy (39).
The role for AR in PDN has been reviewed in detail (40). New evidence for the key role of AR in the pathogenesis of PDN has been generated in both experimental studies in animal and cell culture models of diabetes and clinical trials of ARIs. The results implicating increased AR activity in high glucose- and diabetes-induced oxidative-nitrosative stress (41-45) and downstream events such as mitogen-activated PK (MAPK) activation (46,47), PARP activation (45), COX-2 activation (Calcutt et al., unpublished), and activation of NF-kB (46,47) are of particular interest. In particular, it has been demonstrated that AR inhibition counteracts high glucose- and diabetes-induced superoxide formation in aorta (42), epineurial vessels (45), and endothelial cells (43-45); nitrotyrosine formation in peripheral nerve (45), vasa nervorum (46), kidney glomeruli and tubuli (48), endothelial cells (43) and mesangial cells (48), lipid peroxidation in peripheral nerve (41), and retina (44); loss of two major nonenzymatic antioxidants, reduced glutathione (GSH), and ascorbate in peripheral nerve (45), and downregulation of several major antioxidative defense enzymes in the retina (44).
The author's group has also demonstrated the key role for AR in diabetes-associated PARP activation in peripheral nerve, retina, kidney, human Schwann cells, and mesan-gial cells (45,48). Two groups have demonstrated the key role for AR in diabetes-induced MAPK activation in rat lens and DRG neurons (46,47). Both PARP activation and MAPK activation are involved in transcriptional regulation of gene expression, through the transcription factors NF-kB, activator protein-1, p53, and others (49,50). Activation of these transcription factors leads to upregulation of inducible nitric oxide synthase, COX-2, endothelin-1, cell-adhesion molecules, and inflammatory genes (49,51). Thus, the demonstration of a major contribution of AR to oxidative-nitrosative stress and PARP and MAPK activation in tissue-sites for diabetic complications allows to predict that in the near future the link between increased AR activity and altered tran-scriptional regulation and gene expression will be established. Any product of genes controlled through PARP- and MAPK-dependent transcription factors, regardless of how unrelated to the sorbitol pathway this product looks from a biochemical point of view, will be affected by a diabetes-associated increase in AR activity and amenable to control by AR inhibition. In accordance with this prediction, the most recent findings have shown that increased AR activity is responsible for diabetes-induced COX-2 upregulation in the spinal cord (Calcutt et al., unpublished), and NF-kB activation in high-glucose exposed vascular smooth muscle cells (52).
The role for AR in the pathogenesis of PDN is supported by findings obtained in AR-overexpressing and AR-knockout mice. Yagihashi et al. (53) demonstrated that induction of STZ-diabetes in the mice transgenic for human AR resulted in more severe peripheral nerve sorbitol and fructose accumulation, MNCV deficit, and nerve fibre atrophy than in their nontransgenic littermates. Treatment of diabetic transgenic mice with the ARI WAY121-509 significantly prevented the accumulation of sorbitol, the decrease in MNCV, and the increased myelinated fibre atrophy in diabetic transgenic mice. Similar findings had been obtained in another transgenic mouse model that over-expressed AR specifically in the Schwann cells of peripheral nerve under the control of the rat myelin protein zero promoter (54). The transgenic mice exhibited a significantly higher reduction in MNCV under both diabetic and galactosemic conditions than the nontransgenic mice with normal AR content. In contrast, AR-deficient mice appeared protected from motor nerve conduction slowing after 4 and 8 weeks of STZ-diabetes (55). These data lend further support to the important role of AR in functional, metabolic, and morphological abnormalities characteristic of PDN.
The findings in transgenic and knockout mouse models are in line with new studies with structurally diverse ARIs. Coppey et al. (56) implicated AR in diabetes-induced impairment of vascular reactivity of epineurial vessels, an early manifestation of PND, which precedes motor nerve conduction slowing. The author's group has demonstrated that established functional and metabolic abnormalities of, at least, early PDN, can be reversed with an adequate dose of ARI, i.e., the dose that completely suppressed diabetes-associated sorbitol pathway hyperactivity (41). Calcutt et al. (57) have shown that the ARI statil prevented thermal hypoalgesia in STZ-diabetic rats; furthermore, another ARI IDD 676, given from the onset of diabetes, prevented the development of thermal hyperalgesia and also, stopped progression to thermal hypoalgesia when delivered in the last 4 weeks of an 8-week period of diabetes. In a recent study (58), the ARI fidarestat partially prevented thermal hypoalgesia in type 2 diabetic ob/ob mice. Tactile allodynia was not prevented by an ARI treatment in either Calcutt et al. (57) study in STZ-diabetic rats or the study in ob/ob mice (58), although paw withdrawal thresholds in response to light touch with flexible von Frey filaments tended to increase in ob/ob mice treated with fidarestat in comparison with the corresponding untreated group.
New evidence supports the role of AR in the pathogenesis of advanced PDN. A 15-month AR inhibition with fidarestat dose-dependent corrected slowed F-wave, MNCV, and SNCV in STZ-diabetic rats (55). In the same study, diabetes-induced para-nodal demyelination, and axonal degeneration were reduced to the normal with such low dose of fidarestat as 2 mg/kg. Other manifestations of advanced PDN, such as axonal atrophy, distorted axon circularity, and reduction of myelin sheath thickness were also inhibited. The results of two double-blind placebo-controlled clinical trials of fidarestat in patients with type 1 and type 2 diabetes are also encouraging (60, and Arezzo et al., unpublished). Fidarestat improved electrophysiological measures of median and tibial MNCV, F-wave minimum latency, F-wave conduction velocity, and median SNCV (forearm and distal) as well as subjective symptoms of PDN, such as numbness, spontaneous pain, sensation of rigidity, paresthesia in the sole upon walking, heaviness in the foot, and hypesthesia. These results support the applicability of the AR concept to the pathogenesis of human PDN, and are consistent with the findings of another clinical trial with the ARI zenarestat, indicating that robust inhibition of AR in diabetic human nerve improves nerve physiology and fiber density (51). Recently, improvement of sensory nerve conduction velocity in patients with diabetic sensorimo-tor polyneuropathy was also found with the new ARI AS-3201 (62). The role for AR in human diabetic neuropathy is also supported by the genetic polymorphism data (40).
Was this article helpful?