Neurotrophic factor changes in the rat thick skin following chronic constriction injury of the sciatic nerve

Abstract

Background: Cutaneous peripheral neuropathies have been associated with changes of the sensory fiber innervation in the dermis and epidermis. These changes are mediated in part by the increase in local expression of trophic factors. Increase in target tissue nerve growth factor has been implicated in the promotion of peptidergic afferent and sympathetic efferent sprouting following nerve injury. The primary source of nerve growth factor is cells found in the target tissue, namely the skin. Recent evidence regarding the release and extracellular maturation of nerve growth factor indicate that it is produced in its precursor form and matured in the extracellular space. It is our hypothesis that the precursor form of nerve growth factor should be detectable in those cell types producing it. To date, limitations in available immunohistochemical tools have restricted efforts in obtaining an accurate distribution of nerve growth factor in the skin of naïve animals and those with neuropathic pain lesions. It is the objective of this study to delineate the distribution of the precursor form of nerve growth factor to those cell types expressing it, as well as to describe its distribution with respect to those nerve fibers responsive to it.

Results: We observed a decrease in peptidergic fiber innervation at 1 week after the application of a chronic constriction injury (CCI) to the sciatic nerve, followed by a recovery, correlating with TrkA protein levels. ProNGF expression in CCI animals was significantly higher than in sham-operated controls from 1-4 weeks post-CCI. ProNGF immunoreactivity was increased in mast cells at 1 week post-CCI and, at later time points, in keratinocytes. P75 expression within the dermis and epidermis was significantly higher in CCI-operated animals than in controls and these changes were localized to neuronal and non-neuronal cell populations using specific markers for each.

Conclusions: We describe proNGF expression by non-neuronal cells over time after nerve injury as well as the association of NGF-responsive fibers to proNGF-expressing target tissues. ProNGF expression increases following nerve injury in those cell types previously suggested to express it.

Figure 1

Distribution of proNGF immunoreactivity in naïve glabrous skin and following application of CCI. A) In sham operated controls, proNGF (brown precipitate) was found in the upper dermis, particularly in wall of blood vessels, and in small patches of keratinocytes (arrowhead). B) 1 week post-injury, proNGF immunolabeling was detected in mast cells (arrows; identified based on toluidine blue metachromatic counterstaining), in blood vessels in the upper dermis as well as in keratinocytes. C) At 2 weeks post-injury, note the intense immunostaining of keratinocytes, the labeling of some unidentified structures in dermis and a lower number of immunostained mast cells. D) At 4 weeks post-injury, note mast cell immunolabeling. E) Preincubation of antibody with control peptide abolished all specific staining. F) proNGF quantification was done by Western Blot analysis (n = 6) from sham, and animals 1 - 4 week post-injury. OD was normalized against β-actin loading controls. ***p < 0.001, **p < 0.01, *p < 0.05; +SEM.

Figure 2

Changes in p75 expression on S100-IR Schwann cells. A) In sham-operated rats Schwann cells (green) were detected by means of S100 immunoreactivity in small nerves, with immunoreactivity of varying intensity from bright (arrow), lower in the dermis, to dim, along the dermo-epidermal junction. P75 immunoreactivity (red) was evident surrounding all S100-IR Schwann cells. B) One week following nerve injury, p75 immunostaining was dramatically upregulated in Schwann cells; the intense red color masked the mixture of red and green stainings which could be detected by analysing the separately the S100 and p75 stainings (not shown). C-D) 2 & 4 weeks post-injury a decrease in p75 intensity was observed in that the yellow indicative of S100 co-labelling was able to be visualized. E) proNGF (red) S100 (green) and p75 (blue) triple labelling to demonstrate the relative distribution of Schwann cells with respect to proNGF; note a limited distribution of Schwann cells with proNGF and faint immunoreactivity for p75 in sham-operated controls. F) 2 weeks post-injury, the clear upregulation of p75 immunoreactivity associated with S100 (arrow) was observed, which wrapped around proNGF-IR blood vessels.

Figure 3

Quantification of p75 protein content in glabrous skin. Glabrous skin was taken from the same region sampled for immunocytochemistry. A single band corresponding to that of the PC12 cell culture supernatant used as positive control was observed. Note the increased p75 protein levels particularly at 1-2 weeks post-lesion. OD was normalized against β-actin loading controls. N = 6; ***p < 0.001, **p < 0.01, *p < 0.05; +SEM.

Figure 4

Sympathetic and peptidergic nerve fibre association with proNGF. A) DβH-IR sympathetic efferents (red) were closely associated with proNGF (green) in sham operated control animals. B-C) At 1 and 2 weeks post-injury, no detectable difference was observed in the pattern of sympathetic efferent distribution with respect to proNGF. D) At 4 weeks post-lesion sprouted sympathetic efferents retained their association with proNGF, and continued to sprout towards the dermo-epidermal junction. E) In sham-operated rats, peptidergic afferents were distributed along the dermo-epidermal junction in which the occasional group of proNGF immunoreactive cells was found. F) At 1 week following nerve injury, very few peptidergic afferents were observed. G) At 2 weeks post-injury, peptidergic afferents were found along the dermo-epidermal junction as well as along proNGF immunoreactive cells (arrow). H) At 4 weeks post-lesion, peptidergic afferents increased in numbers were loosely associated with proNGF-IR cells (arrow).

Figure 5

Distribution of p75 immunoreactivity following nerve injury and relationship to nerve fibers and proNGF. A) In sham-operated animals, p75 was distributed (red) around and with PGP 9.5-IR nerve fibers (green). P75 staining was found more clearly around large cutaneous PGP 9.5-IR nerve fiber bundles and smaller fibers along the dermo-epidermal junction. Where nerve fibers crossed the dermo-epidermal junction into the epidermis, the yellow color representing p75 associating with nerve fibers was lost (arrow). B) At 1 week post-injury, virtually all PGP-9.5-IR nerve fibers disappeared from the upper dermis and epidermis; immunostaining in p75-IR Schwann was very intense. C) At 2 weeks post-injury, a low number of PGP-9.5-IR fibers were detected and were associated with p75 Schwann cells (yellow), however most of the PGP-9.5-IR was restricted to Langerhans cells in epidermis (arrow) D) At 4 weeks post-injury, p75 immunoreactivity decreased co-incidentally with the increase in PGP-9.5 immunoreactivityin the upper dermis (yellow). E-H) ProNGF and p75 association in sham-operated controls and in lesioned animals was loose in that most proNGF immunoreactivity was segregated from that for p75 and there was no obvious co-localization (arrows).

Figure 6

Quantification of TrkA protein content in glabrous skin. TrkA protein was sampled from glabrous skin of same region used for immunocytochemistry. Two bands were recognized by the monoclonal anti-TrkA antibody, a 110 kDa and 140 kDa form corresponding to the immature, unglycosylated form and mature form respectively. TrkA levels decreased significantly at 1 week post-injury and remained significantly lower at 2 weeks post-injury until 4 weeks, at which point no difference was observed compared to sham-operated controls. OD is normalized against β-actin loading controls. ***p < 0.001, **p < 0.01, *p < 0.05; +SEM.