Do nerve growth factor-related mechanisms contribute to loss of cutaneous nociception in leprosy?

While sensory loss in leprosy skin is the consequence of invasion by M. leprae of Schwann cells related to unmyelinated fibres, early loss of cutaneous pain sensation, even in the presence of nerve fibres and inflammation, is a hallmark of leprosy, and requires explanation. In normal skin, nerve growth factor (NGF) is produced by basal keratinocytes, and acts via its high affinity receptor (trk A) on nociceptor nerve fibres to increase their sensitivity, particularly in inflammation. We have therefore studied NGF- and trk A-like immunoreactivity in affected skin and mirror-site clinically-unaffected skin from patients with leprosy, and compared these with non-leprosy, control skin, following quantitative sensory testing at each site. Sensory tests were within normal limits in clinically-unaffected leprosy skin, but markedly abnormal in affected skin. Sub-epidermal PGP 9.5- and trk A- positive nerve fibres were reduced only in affected leprosy skin, with fewer fibres contacting keratinocytes. However, NGF-immunoreactivity in basal keratinocytes, and intra-epidermal PGP 9.5-positive nerve fibres, were reduced in both sites compared to non-leprosy controls, as were nerve fibres positive for the sensory neurone specific sodium channel SNS/PN3, which is regulated by NGF, and may mediate inflammation-induced hypersensitivity. Keratinocyte trk A expression (which mediates an autocrine role for NGF) was increased in clinically affected and unaffected skin, suggesting a compensatory mechanism secondary to reduced NGF secretion at both sites. We conclude that decreased NGF- and SNS/PN3-immunoreactivity, and loss of intra-epidermal innervation, may be found without sensory loss on quantitative testing in clinically-unaffected skin in leprosy; this appears to be a sub-clinical change, and may explain the lack of cutaneous pain with inflammation. Sensory loss occurred with reduced sub-epidermal nerve fibres in affected skin, but these still showed trk A-staining, suggesting NGF treatment may restore pain sensation.

A morphological and functional assessment of Mycobacterium leprae-induced nerve damage in a guinea-pig model of leprous neuritis.

Nerve damage, resembling that caused by Mycobacterium leprae in man, was created by the injection of cobalt-irradiated M. leprae organisms into the tibial nerve of guinea-pigs. Assessment of nerve damage was made by clinical, electrophysiological and morphometric means at intervals up to 13 weeks after injection. Quantitative immunohistochemical analysis of neuropeptide-containing fibres in the skin of the foot was also carried out. Significant nerve damage occurred 3 weeks after injection of M. leprae organisms. Motor and sensory functional loss peaked at 5 weeks after injection, and there was a significant decrease of peptide-immunoreactive nerves in all skin compartments. The nerve damage was self-limiting and functional recovery had occurred by 13 weeks. The model shows many of the features found in the nerve damage of treated leprosy patients.

Denatured muscle grafts for nerve repair in an experimental model of nerve damage in leprosy. 2. Recovery of peripheral peptide-containing nerves assessed by quantitative immunohistochemical study.

A marked depletion of neuropeptide-immunoreactive nerves, a consequence of the nerve damage which is commonly found in leprosy, has been reported in peripheral tissues of leprosy patients and of a leprosy animal model. The aim of this study was to investigate peripheral reinnervation following a denatured autologous muscle graft in an animal model of leprosy nerve damage. Possible reinnervation of the foot-pad skin was studied by immunohistochemistry using antisera to the neuronal marker protein gene product 9.5 (PGP), the neuropeptides calcitonin gene-related peptide (CGRP), substance P (SP), vasoactive intestinal peptide (VIP), and the C-flanking peptide of neuropeptide Y (CPON). The extent of the reinnervation process was assessed by image analysis quantification at different time points. At 8 weeks after muscle grafting, there were small numbers of immunoreactive nerves (p < 0.05). At 12, 16, and 20 weeks postoperatively there was a gradual increase in all immunostaining. At 20 weeks, no significant difference was found for PGP-, CGRP-, and SP-immunoreactive nerves in the epidermal and subepidermal layers compared to control (contralateral) tissue. In experimental tissue the recovery of immunoreactive nerves around sweat glands took longer (up to 12 weeks) than in other skin compartments, but after that time the recovery was rapid and at 20 weeks no difference was measured for VIP-immunoreactive nerves in comparison with controls. Around blood vessels, the recovery of CGRP- and CPON-immunoreactive fibers was slow, and at 20 weeks a difference with control samples (p < 0.01) was noted. In the same area, there was no significant difference for PGP immunoreactivity between controls and tissues at 20 weeks. In contrast, the immunoreactive nerve bundles in the dermis showed a faster recovery than nerves in other skin areas, with amounts similar to controls at 20 weeks. The significant recovery of immunoreactive nerves, in particular of those containing sensory neuropeptide, is consistent with the described functional recovery.