In diabetic animals and patients increased advanced glycation end products (AGE) covalently attach to intracellular, plasma and extracellular matrix proteins in the nerve, and other tissues (37,38). It has been postulated that such modifications might affect neural metabolism, axonal transport, and tissue repair, thus enhancing neural degeneration and impairing neural regeneration. The strongest evidence in support of this model is that administration of aminoguanidine, an AGE inhibitor, attenuated MNCV, and sensory NCV (SNCV) deficits in diabetic rats (39,40). However, besides inhibition of AGE formation, aminoguanidine is known to inhibit the activity of other enzymes including the inducible-, neuronal-, and endothelial-nitric oxide synthases, and the semicarbazide-sensitive amine oxidase (41). It is likely that there might be still other enzymes that are affected by this drug. Therefore, it is not clear if the beneficial effect of this drug on diabetic neuropathy is the consequence of inhibition of AGE formation.
One of the mechanisms by which AGE exerts its toxic effects is its interaction with its receptor for advanced glycation end products (RAGE) (42,43). This leads to the activation of a cascade of cytotoxic pathways, including the activation of the transcription factor, which is nuclear factor (NF)-kB, thought to contribute to diabetic neuropathy (44). Infusion into diabetic rats containing the soluble fragment of RAGE (sRAGE), which consists of the AGE-binding domain that attenuated vascular dysfunction, indicated that the AGE-RAGE interaction plays an important role in diabetic microvascular complications (45). Administration of sRAGE also appeared to be beneficial to diabetic neuropathy (44). When 3-months WT mice with diabetes were placed on 55°C hotplate, the latency or the time it takes for them to respond was significantly longer than that for the mice without diabetes, indicating impairment of pain perception as a result of diabetes. Treatment with sRAGE for 3 weeks completely normalized the delayed latency in mice with diabetes, indicating that RAGE is involved in the hyperglycemia-induced loss of pain perception. When the RAGE null mutant mice were induced to become diabetic, the nociceptive threshold was lower than that of the WT mice with diabetes but significantly higher than that of the WT or RAGE-null mice without diabetes, indicating that RAGE deficiency provides partial protection against diabetic sensory neuropathy (44). Examination of the footpad skin of these mice revealed that diabetes led to a significant decrease in the PGP9.5-positive small nerve fibers in this tissue in the WT mice. However, a similar decrease in nerve fibers also occurred in the footpad of the diabetic RAGE null mice, indicating that RAGE deficiency did not protect against diabetes-induced small fiber loss, although it partially restored the function of the remaining nerve fibers.
The RAGE null mice and the WT control mice in these experiments were offspring of SVEV129 x C57BL. Perhaps, because of their hybrid genetic background, diabetes did not cause a significant reduction in their MNCV or SNCV. Interestingly, diabetes led to a dramatic induction of NF-kB activity in the sciatic nerve of the WT mice as judged by electrophorectic mobility shift assay. Activation of NF-kB is thought to be the key contributor to hyperglycemia-induced tissue lesions. Yet the large increase in NF-kB in the sciatic nerve did not lead to any significant change in the MNCV or SNCV, indicating that this transcription factor might not play an important role in the pathogenesis of diabetic neuropathy. It would be interesting to cross the RAGE null mutation into pure C57BL genetic background where the WT mice exhibit significant reduction in MNCV and SNCV when induced to become diabetic. Then one can determine if RAGE deficiency has any protective effect on diabetes-induced MNCV and SNCV reduction.
It is interesting that while sRAGE completely restored the nociceptive threshold of the diabetic mice, RAGE deficiency only partially restored it. One explanation is that there are other AGE receptors such as the macrophage scavenger receptor and the galectin-3 that might have similar deleterious effect as RAGE when they interact with AGE (46). With the removal of RAGE, AGEs would bind to other receptors to exert their toxic effects. On the other hand, infusion of sRAGE might engage all the AGEs, making them unavailable to interact with their receptors. However, treatment with antibodies against RAGE that should only block RAGE, also has a strong protective effect similar to sRAGE, suggesting that there are other contributing factors to the diabetes-induced neural dysfunction. Further, intracellular AGE and AGE attached to extracellular matrix are also thought to contribute to diabetic lesions (47,48). However, these AGEs are not able to interact with cell surface RAGE or other AGE receptors. Removing RAGE is unlikely to block their deleterious effects.
Was this article helpful?