Robert R. Myers

Schwann cells produce tumor necrosis factor alpha : Expression in injured and non-injured nerves

We investigated the production of tumor necrosis factor alpha, a pleotropic cytokine, in the rat sciatic nerve using immunohistochemistry for the protein and in situ hybridization for specific messenger RNA sequences. Results demonstrated that in the uninjured sciatic nerve, few endothelial cells, fibroblasts, and Schwann cells were immunoreactive for tumor necrosis factor. In situ hybridization did not yield visible results in this control tissue. Following a compression injury, the number of immunoreactive cells in the nerve fascicle was increased. In situ hybridization showed positive staining in Schwann cells and endothelial cells at seven days. These observations are the first to demonstrate the production of this proinflammatory cytokine by peripheral nerve glia, and further support other studies from this laboratory suggesting that tumor necrosis factor has a pathogenic role in nerve injury. Because this cytokine is produced by Schwann cells in intimate contact with nerve fibers, tumor necrosis factor may also have a role in the hyperalgesia associated with neuropathic pain.

Degeneration and regeneration of the peripheral nervous system: From Augustus Waller's observations to neuroinflammation

Abstract  This review article on the degeneration and regeneration of peripheral nerve fibers was presented as a Plenary Lecture at the 2001 meeting of the Peripheral Nerve Society. It is accompanied by a reprint of Augustus Wallers 1850 article, which gave rise to the pathologic process termed Wallerian degeneration. This review is focused on the role of neuroinflammation in Wallerian degeneration and how immune mediators contribute to both axonal degeneration and regeneration. Similarities and differences between the PNS and CNS in terms of inflammation and microglial activation after nerve injury are discussed, and point towards progress in understanding the failure of nerve fiber regeneration in the CNS.

Anti-inflammatory interleukin-10 therapy in CCI neuropathy decreases thermal hyperalgesia, macrophage recruitment, and endoneurial TNF-α expression

&NA; The chronic constriction injury model of mononeuropathy is a direct, partial nerve injury yielding thermal hyperalgesia. The inflammation that results from this injury is believed to contribute importantly to both the neuropathological and behavioral sequelae. This study involved administering a single dose (250 ng) of interleukin‐10 (IL‐10), an endogenous anti‐inflammatory peptide, at the site and time of a chronic constriction injury (CCI) lesion to determine if IL‐10 administration could attenuate the inflammatory response of the nerve to CCI and resulting thermal hyperalgesia. In IL‐10‐treated animals, thermal hyperalgesia was significantly reduced following CCI (days 3, 5 and 9). Histological sections from the peripheral nerve injury site of those animals had decreased cell profiles immunoreactive for ED‐1, a marker of recruited macrophages, at both times studied (2 and 5 days post‐CCI). IL‐10 treatment also decreased cell profiles immunoreactive for the pro‐inflammatory cytokine tumor necrosis factor alpha (TNF‐&agr;) at day 2, but not day 5. Qualitative light microscopic assessment of neuropathology at the lesion site did not suggest substantial differences between IL‐10 and vehicle‐treated sections. The authors propose that initial production of TNF‐&agr; and perhaps other proinflammatory cytokines at the peripheral nerve lesion site importantly influences the long‐term behavioral outcome of nerve injury, and that IL‐10 therapy may accomplish this by downregulating the inflammatory response of the nerve to injury.

Pathogenesis of sciatic pain : role of herniated nucleus pulposus and deformation of spinal nerve root and dorsal root ganglion

Abstract The basic pathophysiologic mechanisms related to disc herniation and sciatica are poorly understood. Recently it was demonstrated that nucleus pulposus from an intervertebral disc could induce structural and functional changes in adjacent nerve roots when applied epidurally, however, it is not known if such changes are painful. In a model for inducing disc herniation in the rat, we found that puncture of a lumbar disc with subsequent herniation of nucleus pulposus without nerve root compression, or chronic displacement of the 4th lumbar nerve root and ganglion, did not individually induce significant changes in thresholds for mechanical or thermal stimulation compared to sham‐operated animals. However, the combination of disc puncture and displacement induced a reduction of the threshold for thermal stimulation, indicating hyperalgesia, that was present 2 days after surgery and gradually recovered during a 14‐day period. These data and the associated description of this new model for experimental disc herniation may increase our understanding of the pathophysiologic events leading to sciatica and help in evaluating new modalities for diagnosis and treatment of disc herniation and sciatica.

The effect of thalidomide treatment on vascular pathology and hyperalgesia caused by chronic constriction injury of rat nerve

&NA; Tumor necrosis factor alpha (TNF) may be involved in the pathogenic mechanisms of neuropathic pain by affecting endothelial cells and by upregulation of receptor sensitivity in afferent nerve fibers. To test the hypothesis that TNF plays a role in the vascular changes and the pain‐related behavior in an experimental painful neuropathy in rats produced by tying loosely constrictive ligatures around one sciatic nerve, we investigated the effect of thalidomide, a selective blocker of TNF‐production in activated macrophages. In rats in which treatment with thalidomide was started preoperatively, there was diminished mechanical allodynia and thermal hyperalgesia during the early stage of the disease. TNF immunohistochemistry revealed reduced endoneurial immunoreactivity on day 5 post surgery as compared to sham‐treated animals. The pathologic vascular changes were also reduced in thalidomide‐treated rats. Starting treatment with thalidomide at a time point when hyperalgesia was already present did not alter the course of the pain‐related behavior. We conclude that preemptive treatment with a substance that blocks production of TNF reduces pain‐related symptoms and pathologic vascular changes in the chronic constriction injury model of neuropathic pain.

The role of neuroinflammation in neuropathic pain: mechanisms and therapeutic targets

Neuroinflammation is a proinflammatory cytokine-mediated process that can be provoked by systemic tissue injury but it is most often associated with direct injury to the nervous system. It involves neural-immune interactions that activate immune cells, glial cells and neurons and can lead to the debilitating pain state known as neuropathic pain. It occurs most commonly with injury to peripheral nerves and involves axonal injury with Wallerian degeneration mediated by hematogenous macrophages. Therapy is problematic but new trials with anti-cytokine agents, cytokine receptor antibodies, cytokine-signaling inhibitors, and glial and neuron stabilizers provide hope for future success in treating neuropathic pain.

Upregulation and interaction of TNFα and gelatinases A and B in painful peripheral nerve injury

Chronic constriction injury (CCI) to peripheral nerve causes a painful neuropathy in association with a process of axonal degeneration and endoneural remodeling that involves macrophage recruitment and local increase in extracellular proteases and tumor necrosis factor alpha (TNF-alpha). Cell surface activation of TNF-alpha from its transmembrane precursor, as well as sequestration of TNF-alpha receptors II and I, is performed by the zinc-dependent endopeptidase family of matrix metalloproteinases (MMPs). Among TNF-alpha-converting MMPs, basal lamina degrading gelatinases are thought to play a role in sciatic nerve injury. In the present study, we determined the forms of TNF-alpha involved in the development of CCI neuropathy in rats, using Western blot analysis, and the temporal correlation of TNF-alpha and TNFRI protein profiles with gelatinases activity at the site of peripheral nerve injury. We observed two peaks in TNF-alpha protein during the first week of CCI that correspond to previously reported peaks in painful behavior. We propose that the first peak at 6 h post-CCI is due to the local expression of the cytotoxic transmembrane 26 kDa TNF-alpha protein released by resident Schwann cells, mast cells and macrophages. This peak in TNF-alpha protein expression corresponds to an increase in gelatinase B (MMP-9) activity, which is greatly upregulated as early as 3 h following CCI to rat sciatic nerve. The second peak occurs at 5 days post-CCI, and may represent TNF-alpha protein released by hematogenously recruited macrophages. This peak is marked by the increase in active soluble 17 kDa TNF-alpha and by gelatinase A (MMP-2) upregulation. These observations suggest that there is a pathogenic role for the TNF-alpha-converting function of MMP-2 in painful CCI neuropathy. We conclude that severe nerve injury induces MMPs, TNF-alpha and TNFRI, which interactively control the privileged endoneurial environment and the pathogenesis of the painful neuropathies associated with the macrophage-dependent processes of Wallerian degeneration.

Axonal transport of TNF-α in painful neuropathy: distribution of ligand tracer and TNF receptors

Tumor necrosis factor-alpha (TNF-alpha) is a key player in peripheral nerve injury. In the inflammatory chronic constriction injury (CCI) model of sciatic neuropathy, upregulation of TNF-alpha mRNA and protein at the site of nerve injury has been associated with pain. We now report the distribution of endogenous TNF-alpha protein and its receptors along normal and CCI-injured sciatic nerves, and within the corresponding lumbar dorsal root ganglia (DRG). Using Western blotting, TNF-alpha was found to be distinctly increased at the injury site, as well as in the axons just distal to the corresponding DRG. The TNF-alpha signal between the injury site and DRG (midaxonal) was induced between 2 and 5 days post-CCI, suggesting activation of TNF-alpha axonal transport. Endogenous TNF-alpha was localized in small-diameter, presumably nociceptive, and large-diameter, presumably mechanoceptive, DRG sensory neurons in both normal and CCI animals. Intraneural microinjection of biotin-labeled TNF-alpha showed specific axonal uptake at the injection site, as detected by avidin-biotin-peroxidase histochemistry, and confirmed by co-localization with neurobiotin tracer. In control animals, fast retrograde transport of biotinylated TNF-alpha to both L4 and L5 DRG neurons was apparent 6 h following injection. TNF receptors TNFRI and TNFRII co-localized with biotinylated TNF-alpha tracer along the nerve trunk, suggesting that TNF-alpha transport may be receptor-mediated. In animals with CCI neuropathy, uptake of biotinylated TNF-alpha by neuronal soma was inhibited. Instead, there was signal accumulation in the axons immediately distal to the DRG, and TNFRI and RII were increased at this same anatomic location. These findings highlight a dynamic process of TNF-alpha protein and receptor regulation throughout the peripheral neural axis that bears on both the normal function of DRG neurons and the pathogenesis of painful neuropathies.

Exogenous erythropoietin protects against dorsal root ganglion apoptosis and pain following peripheral nerve injury.

Erythropoietin (Epo) has been shown to have potent anti‐apoptotic activity in central nervous system neurons in animal models of ischaemic injury. Recently, Epo and its receptor (EpoR) have been identified in the peripheral nervous system [Campana & Myers (2001), FASEB J., 15, 1804–1806]. Herein, we demonstrate that in painful neuropathy caused by L5 spinal nerve crush (SNC), therapy with recombinant human Epo (rhEpo) reduced dorsal root ganglion (DRG) apoptosis and pain behaviours. Quantification of both DRG neurons and satellite cells revealed that vehicle‐treated, crush‐injured DRGs had 35.5 ± 8.3% apoptotic neurons and 23.5 ± 2.36% satellite cells compared with 7.5 ± 6.3% apoptotic neurons and 6.4 ± 3.94% satellite cells in rhEpo‐treated, crush‐injured DRGs (P < 0.05). While rhEpo‐treated animals were not initially protected from mechanical allodynia associated with L5 SNC, rhEpo did significantly improve recovery rates compared to vehicle‐treated animals (P < 0.01). Systemic rhEpo therapy increased JAK2 phosphorylation, a key anti‐apoptotic signalling molecule for Epo‐induced neuroprotection, in DRGs after crush. Dual immunofluorescence demonstrated Epo‐induced JAK2‐p was associated with both neuronal and glial cells. JAK2‐p was associated with NF200‐positive large neurons and with smaller neurons. This population of small neurons did not colocalize with IB4, a marker of nonpeptidergic, glial derived growth factor‐responsive neurons. The findings link anti‐apoptosis activities of Epo/EpoR/JAK2 in DRG neurons capable of inducing protracted pain states with reductions in pain behaviours, and therefore support a role for Epo therapy in the treatment of neuropathic pain.

Pathology of experimental compression neuropathy producing hyperesthesia.

An experimental neuropathy in rats produced by tying loosely constrictive ligatures around one sciatic nerve has recently been shown to produce pain-related behavior that follows a reproducible time course. In the present study, we assessed the degree of thermal hyperesthesia and examined the sciatic nerves by light and electron microscopy at different time points from 1 day to 12 weeks after surgery. Edema was the initial pathologic change seen in the neuropathy and was associated with early Wallerian degeneration on day 1. Injury to the connective tissue sheaths with interruption of the perineurial layer, infiltration of macrophages into the endoneurium, and proliferation of endothelial cells were observed during the first week. Endothelial cells hypertrophied and changed to a rhomboid shape with gaps between adjacent cells. Most large myelinated and many small myelinated fibers underwent Wallerian degeneration. Unmyelinated fiber numbers were reduced to one-third of the normal value from day 5 to day 14. Axonal sprouts were numerous after 1 week and grew into the segment distal to the ligatures by 4 weeks. Aberrant sprouts in minifascicles outside the perineurium were present from 4 weeks on. The original ligatures were rapidly surrounded by large amounts of fibrous tissue and mostly absorbed by 12 weeks; the initial fascicular compression caused by the ligatures was maintained by the fibrous tissue. We conclude that, whereas neuroma formation may contribute to the pain-related behavior in the later stages of this neuropathy, acute changes in the endoneurial micro-environment are important for its initial development.