Aims/hypothesis: This study was designed to determine whether diabetes in rats is associated with phosphorylation of c-Jun N-terminal kinase (JNK) and one of its transcription factors, c-Jun, in sensory neurones innervating the lower limb. We also sought to determine which neuronal phenotypes are affected and to examine the effect of aldose reductase inhibition on JNK and c-Jun phosphorylation.
Methods: Diabetes was induced in rats using streptozotocin. Phosphorylation of JNK and c-Jun in lumbar dorsal root ganglia was determined by binding of phospho-specific antibodies using western blots and immunocytochemistry. Neuronal phenotypes were characterised by binding of N52 (an antibody that recognises the heavy neurofilament protein; for large-diameter mechanoceptors) and of calcitonin gene-related peptide and the plant glycoprotein lectin IB4 (for nociceptors). The efficacy of the aldose reductase inhibitor fidarestat was determined by measuring polyol pathway metabolites in sciatic nerve, and functionally by measuring the conduction velocity of motor and sensory nerves.
Results: In control rats, activated JNK and c-Jun were primarily associated with mechanoceptors; in diabetes this was increased, but a greater increase was seen in nociceptors. Phosphorylation was prevented in all cells by fidarestat, which normalised polyol pathway metabolites as well as motor nerve and sensory nerve conduction velocity.
Conclusions/interpretation: Fidarestat-sensitive phosphorylation of JNK and c-Jun occurs in fast-conduction mechanoceptors-the same class of neurones that registers the changes in sensory nerve conduction velocity-and in nociceptors. This supports the notion that mitogen-activated protein kinase phosphorylation, via the polyol pathway, may convert the direct effects of raised glucose into impaired nerve conduction and neuropathic pain. For proof of this we await the availability of specific JNK antagonists formulated for in vivo use.