A Small Molecule Modulator of Hsp90 Improves Experimental Diabetic Neuropathy

Abstract

Background: Current Hsp90 inhibitors are therapeutically problematic. Although they induce a pro-survival heat shock response that promotes the refolding of damaged proteins, a confounding issue is that at these concentrations the inhibitors are cytotoxic, due to their ability to decrease the maturation of newly synthesized client proteins. KU-32 contains a novobiocin-based scaffold that binds to the C-terminal of Hsp90 and induces a pro-survival heat shock response at a concentration ~10,000 fold lower than that needed to induce neurotoxicity. This creates an optimal therapeutic window in which to operate, providing promise towards the development of novel neuroprotective agents. Objective: To evaluate whether the induction of the heat shock response through Hsp90 modulation could decrease or reverse the pathophysiological progression of diabetic peripheral neuropathy in Type-1 diabetic mice. Hypothesis: A small molecule modulator of Hsp90 will improve experimental diabetic neuropathy. Methods: After 8-12 weeks of diabetes induced by streptozotocin, the effects of weekly doses of KU-32 on several standard indices of diabetic neuropathy were measured. Results: Initial toxicity studies employing the weekly intraperitoneal administration of 2 or 20 mg/kg KU-32 to non-diabetic mice over 6 week duration did not alter motor or sensory nerve conduction velocity (MNCV/SNCV), mechanical or thermal sensitivity, or intra-epidermal nerve fiber density. Thus, the drug alone had no effect on altering common measures of neuropathy. In a 12-week intervention study, wild-type C57 Bl/6 animals receiving a weekly treatment regimen of 20 mg/kg KU-32 for 6 weeks exhibited a steady recovery to control levels in thermal and mechanical sensitivity, MNCV, and SNCV. KU-32 did not alter metabolic control. As Hsp70 is hypothesized to be a major target for KU-32, its necessity in neuroprotection was examined using Hsp70 double knockout mice (Hsp70.1/Hsp70.3). In a 12-week intervention study, Hsp70 knockout mice receiving a weekly treatment regimen of 20 mg/kg for 6 weeks displayed no improvements in thermal and mechanical sensitivity, MNCV, and SNCV. In 8-week intervention studies, animals demonstrated recoveries in sensory hypoalgesia and nerve conduction velocity deficits in a dose-dependent manner. KU-32 did not alter sensory nerve fiber innervation. Conclusions: These data suggest that hyperglycemia may adversely impact the ability of neurons to promote refolding or decrease unfolding of mildly damaged proteins. C-terminal Hsp90 modulators can improve several standard clinical indices of negative symptoms associated with small and large fiber dysfunction in the absence of improving overall metabolic control. The affects of KU-32 appear to be dose-dependent and require the presence of inducible Hsp70 for efficacy. Inducible Hsp70 is not required for the pathophysiological progression of diabetic neuropathy.