Studies of aldose reductase using neuronal cell culture and ligated rat sciatic nerve

Abstract

Sorbinil (CP 45,634), a potent aldose reductase (AR) inhibitor, has the ability to normalize both sorbitol levels and functional parameters such as orthograde axonal transport and motor nerve conduction velocity in peripheral nerves of diabetic rats, which implicates flux through the polyol pathway in the pathophysiology of diabetic neuropathy. In order to understand more fully the role of this enzyme, it is important to determine the major cellular location of AR in peripheral nerve. Experiments were designed that have taken advantage of the observation that peripheral nerve axons degenerate distal to an injury site, while Schwann cells remain viable. One sciatic nerve in each experimental rat was chronically ligated (up to 6 weeks) before inducing diabetes by an intravenous (iv) injection of streptozotocin (STZ; 65 mg/kg). Two days after the STZ injection, both sciatic nerves were removed from each animal, and the ligated nerve was divided into proximal (Schwann cells and axons) and distal (Schwann cells only) portions before being processed for sorbitol determinations. The intact nerves and the proximal protion of the ligated nerves (having both Schwann cells and axons) retained the ability to accumulate sorbitol after STZ injection, while the distal portion (having Schwann cells only) lost this capacity 4 days after ligation. This lack of ability to accumulate sorbitol was not due to failure of the substrate (glucose) to reach the distal nerve segment. Additionally, homogenates of whole nerves and of proximal portions of ligated nerves were able to form sorbitol from glucose in the presence of NADPH while homogenates of distal portions of ligated nerves had lost approximately 85% of this activity. In addition, when glyceraldehyde was used as the substrate, the distal segment homogenates lost 70% of the glycerol-synthesizing activity seen in whole nerve homogenates. These data contrast with previous reports that showed no loss of AR activity in distal ends of transected rabbit sciatic nerve. The fact the neuronal tissue has AR-like activity is demonstrated in studies utilizing both primary cultures of rat sympathetic neurons and the PC12 pheochromocytoma cell line; both of these accumulate sorbitol when grown in high-glucose medium. This accumulation is inhibited in a dose-dependent manner by the inclusion of sorbinil. It has been found that the distal portions of ligated nerves neither accumulate sorbitol nor retain the ability to make sorbitol under conditions where both whole nerves and proximal segments of ligated nerves do accumulate and synthesize this sugar. These findings are consistent with at least two interpretations that are not mutually exclusive. The first is that sorbitol-synthesizing activity (AR) may be located in axons rather than in Schwann cells, as previously reported. Alternatively, the sorbitol-synthesizing activity may be a differentiated function of Schwann cells, which is lost as the Schwann cells de-differentiate after ligation. The former interpretation of the data is supported by the observation that primary cultures of rat sympathetic neurons can make sorbitol. This interpretation lends support to the hypothesis that peripheral diabetic neuropathy is a primary dysfunction of axons. Further experimentation will attempt to distinguish between these two possibilities.