The Role of Cutaneous Innervation in the Sensory Abnormalities Associated with Diabetic Neuropathy

Abstract

Diabetes-induced nerve damage results in cutaneous denervation, nerve conduction slowing, suppressed regenerative responses, and debilitating painful or insensate sensory symptoms. The increasing prevalence of diabetic neuropathy and its persistent treatment difficulties justify continued study into the mechanisms underlying the disease and the exploration of animal models useful for assessing its complications. The purpose of this study was to elucidate the relationship between cutaneous nerve fiber density and the presence of neuropathy and its accompanying symptoms. Experiments were conducted in various animal models of diabetes and in diabetic human patients. To explore mechanisms underlying insensate neuropathy, we evaluated behavioral responses to noxious stimuli in three animal models. Diabetic mice consistently displayed decreased sensitivity and hypoalgesia, indicative of insensate neuropathy. We also assessed peripheral innervation both indirectly, via spinal Fos expression, and directly, by quantifying footpad innervation. The diabetes-induced behavioral responses were paralleled by progressive reductions in dorsal horn activation by peripheral afferents. In contrast, direct quantification of peripheral fibers did not reveal deficits in total nerve fiber density. However, a subpopulation-specific reduction was found; peptidergic c-fibers were preferentially lost early in the diabetes progression, and their loss correlated with loss of nociceptive sensitivity. These results suggest that diabetes-induced behavioral deficits are more closely associated with the peptidergic, rather than the nonpeptidergic, subpopulation, underscoring the importance of peptidergic fibers in pain sensation. In addition, human diabetic and nondiabetic subjects were recruited and tested for measures of sensory nerve fiber function and evaluated for cutaneous nerve fiber density. While both epidermal and dermal nerve densities were highly specific and sensitive measures for diagnosing diabetic neuropathy, correlations with sensory measures were varied, and nerve fiber density could only account for at most 60% of the variability in any given neuropathic symptom. Collectively, results from the animal and human investigations demonstrate there are subtle disconnects in the degree to which nerve fiber density indicates pain sensitivity. Future studies should be directed toward understanding other mechanisms important in nociception during neuropathic disease. In addition, before nerve fiber density is relied upon as an outcome measure of treatment efficacy, the ability of cutaneous innervation to predict pain or lack of pain in human subjects should be addressed.