Delayed peripheral Nerve Degeneration, Regeneration, and pain in Mice Lacking Inducible Nitric Oxide Synthase (original) (raw)

Abstract

Inducible nitric oxide synthase (iNOS) may be a critical factor in the repair of injured tissues. In mice lacking iNOS we observed abnormalities in how the peripheral nerve responds to each of 3 fundamental types of injury: chronic constriction partial nerve injury (a model of neuropathic pain), nerve crush, and nerve transection. In each type of injury, mice lacking iNOS had evidence of a regenerative delay, preceded by slowing of myelinated fiber Wallerian degeneration (WD). In wild-type mice, iNOS immunoreactivity and the presence and upregulation of its mRNA were demonstrated distal to injury, but neither was observed in the knockout mice. Slowed WD was suggested by the abnormal persistence of apparent myelinated fiber profiles distal to the injury zones in mice lacking iNOS compared to wild-type controls. In mice lacking iNOS there were fewer regenerating myelinated fibers, smaller caliber regenerating fibers, and slowed reinnervation of muscle endplates distal to the injury zone. Slowed degeneration was also associated with normal initiation but delayed expression of neuropathic pain. Our findings highlight important relationships among nitric oxide, WD, neuropathic pain, and axon regeneration.

Introduction

Injury to peripheral nerve results in degeneration of the distal nerve stump in a process known as Wallerian degeneration (WD). Although peripheral nerve axons are capable of regeneration, many neuropathies are associated with poor axonal regrowth resulting in muscle weakness, loss of sensation, and debilitating neuropathic pain, often refractory to analgesia. Normal WD, including myelin breakdown and clearance, and its associated microenvironment, allows successful regrowth of nerve fibers from the proximal nerve segment to target tissues (3–6). The generation of neuropathic pain following injury is also intimately associated with the progress of WD (3–5, 7). Since both regeneration and neuropathic pain rely on the WD process and clearance of myelin debris, understanding its mechanism is crucial for the development of relevant forms of treatment.

Following peripheral nerve injury, rapid recruitment of macrophages accompanies WD. Infiltrating macrophages and other non-neuronal peripheral nerve constituents, such as Schwann cells and fibroblasts, begin to express pro-inflammatory molecules such as interleukin 1-β(IL-1β), interleukin-6 (IL-6), tumor necrosis-α (TNF-α), and nerve growth factor (NGF)(8–10). These mediators and their associated inflammatory response may play an important role in determining the success of regeneration and neuronal repair (11, 12). Injury-related cytokines regulate the transcription of the inflammatory or inducible nitric oxide synthase (iNOS) in macrophages, glia, and other cells (13). The toxicfree radical nitric oxide (NO), produced by iNOS in macrophages, participates in an early non-specific immunological reaction associated with cytotoxicity (14). iNOS is thus a major contributing enzyme in wound healing (15) and tissue regeneration (16). In central nervous system (CNS) cerebral ischemia, traumatic brain injury or MPTP-induced dopaminergic neurodegeneration(17–20), iNOS is upregulated in glial cells and infiltrating macrophages. iNOS expression is also increased in degenerative Alzheimer disease and human immunodeficiency virus (HIV)-infected brains (21, 22). iNOS expression and local elaboration of NO may either promote or protect against neuronal damage (17, 19, 23).

Peripheral nerve injury is also associated with local upregulation of iNOS in macrophages and Schwann cells(24–26) with subsequent NO release. NO participates in the response of the peripheral nerve to injury by generating rises in nerve blood flow within the injured nerve trunk (25). Local actions of NO, or NO-related species such as peroxynitrite, may be of critical importance during WD by contributing to the breakdown and removal of injured axons and myelin (27, 28). Alternatively, local NO may also impede regeneration by promoting growth cone collapse (29).

In this work we used iNOS-null (KO) mice to address the role of iNOS-NO in WD, axon regeneration, and neuropathic pain. By using 3 fundamental models of peripheral nerve injury, chronic constriction partial nerve injury, a model of neuropathic pain, crush of the nerve trunk, and complete nerve transection, we show that mice lacking iNOS have a regenerative delay, preceded by slowed WD. In addition, slowed breakdown of myelinated fibers in the chronic constriction model was associated with normal initiation but delayed expression of behavioral features of neuropathic pain. The findings highlight the importance of iNOS in WD and myelin clearance, the reliance of regenerative events on prior degeneration, and the relationship of neuropathic pain to these events.

Materials and Methods

Animals

_iNOS_-KO mice, generated by gene targeting in embryonic stem cells, as described previously (30), were used throughout the study. The mutants were from a mixed background of 129/SvEv × C57BL/6. Mice from the same background were used as wild-type (WT) controls. For all experiments both male and female mice (8–14-wk-old) were used (with a similar sex distribution among the groups; gender did not influence the results presented). Animals were group housed (4–6 animals per cage) in a temperature (21 ± 2°C), humidity (50%–65%), and light (12-h light dark cycle, lights on 7:00 am) controlled vivarium in a pathogen-free environment. Animals had free access to food and water. All experiments were approved by the local Animal Care Committee adhering to guidelines of the Canadian Council of Animal Care. Nociceptive testing also conformed to the guidelines of the International Association for the Study of pain (31).

RT-PCR Analysis of iNOS-mRNA Expression

Isolation of total RNA was done as described (32), using Trizol® reagent (Life Technologies, Burlington, ON, Canada). Final RNA concentration was determined in all samples using GeneQuant® (Amersham pharmacia Biotech, piscataway, NJ). RT-PCR was performed according to the “primer dropping” approach using methods and primers previously described (33, 34) for iNOS and GAPDH, in turn yielding 499 and 302 bp products, respectively. pCR reactions were performed in 50 μl reaction tubes containing 5 μl 1XPCR buffer (10 mM Tris-HCL, pH 9.0. 50 mM KCl, 1.5 mM MgCl), 2 μl of RT reaction product, 20 pmol of each 5′ and 3′ primer pair, 2 mM of each dNTP and 2 units of Taq DNA polymerase (Gibco-BRL, Burlington, ON, Canada). cDNAs for iNOS and GAPDH were co-amplified for 32 and 26 cycles, respectively.

All experiments also included negative RT controls where superscript II was omitted to exclude the possibility of genomic DNA amplification. After amplification, pCR products were separated on a 2% agarose gel (cast in the presence of 0.5 μg/ml ethidium bromide). For evaluation of changes in iNOS-mRNA expression, gel bands were analyzed using a fluorescent Gel Imaging system (Gel Doc1000, Bio-Rad, Mississauga, ON, Canada) and image analysis software (Adobe photoshop 5.0).

iNOS Immunohistochemistry

Ten-micrometer cryostat sections taken distal to the injury site were immunostained as described (25, 35) using a rabbit antibody against iNOS (1:500, Transduction Laboratories, Lexington, KY) as the primary antibody. Incubation with FITC conjugated goat anti-rabbit antibody (1:500, Sigma, Mississauga, ON, Canada) was used to visualize iNOS. The specificity of the antibody employed was previously addressed (25) using Western blot and immunochemistry in samples of intact or chronic constriction rat sciatic nerve (the polyclonal antibody was raised against rat but is similarly sensitive to mouse iNOS as shown by Laubach et al [36]). A single ∼130 kDa band was identified and cells labeled with iNOS co-labeled with markers for subpopulations of Schwann cells and macrophages.

Nerve Injury Models

Three separate models of peripheral nerve injury were studied: partial nerve injury induced by a chronic constriction injury (CCI) of the nerve associated with neuropathic pain (37), crush of the nerve-trunk, and complete nerve transection. Different and selective approaches tailored to the type of injury were used for each model to reduce the use of animals of limited availability. For CCI, we evaluated iNOS mRNA and protein expression, quantitative myelinated fiber morphometry out to 21 days following injury, and behavioral indices of neuropathic pain. For nerve crush we evaluated myelinated fiber morphometry 14 days and 6 wk following crush. For nerve transection, we evaluated morphometry of myelinated fibers 10 wk following the injury, and carried out serial electrophysiological recordings of motor fiber reinnervation over the same period of time in the same mice. For all procedures mice were anesthetized with a mixture of sodium pentobarbital (65 mg/ml; 2 ml), diazepam (5 mg/ml; 2 ml), and saline (7 ml) injected ip at a dose of 0.5 ml/100 g. In all models the right sciatic nerve was exposed at the mid-thigh level by a blunt dissection. Nerve crush was performed by pinching the nerve with a jeweller forceps for 2 × 10 s about 10 mm proximal to the trifurcation. A clean scalpel cut at a similar location was used to perform nerve transection. The stumps were not resutured, but left separated to eliminate the influence of suturing on the model. CCI was prepared according to the original model with modification for mice (37, 38). CCI is associated with preferential loss of large myelinated fibers and relative sparing on the unmyelinated fiber population (39, 40).

Light Microscopy and Quantitative Morphometry

Epon-embedded, semi-thin transverse sections stained with toluidine-blue from both WT and KO mice were used to count and size numbers of persistent fibers, not yet degenerated and, at later times, the numbers of degenerating and regenerating myelinated fiber profiles. Degenerating myelinated fibers were recognized by standard pathological criteria: the formation of myelin ovoids, loss of the axon profile, and their association with macrophages (41). Regenerating profiles were recognized by their small caliber and their grouping, and by their reappearance in the setting of complete or near complete loss of the original myelinated fiber profiles. Under ×100 magnification using a microcomputer-based digital image analysis set-up (42), the numbers, size profiles, and densities of myelinated fibers were counted at a fixed distance (5 mm) from the site of nerve injury. Uninjured (intact) profiles of myelinated fibers in sham operated nerves were also counted.

Nerve Conduction Studies

Regeneration of motor fibers, reflecting functional motor recovery, was assessed by recording compound muscle action potentials (M waves) from the sciatic-tibial innervated interosseous muscles of the hindpaw while stimulating proximal (notch) and distal (knee) to the injury site as described (43, 44). Briefly, the sciatic nerve was stimulated by near nerve platinum subdermal EEG electrodes (E2; Grass Instruments, Astro-Med, West Warwick, RI), with a stimulation duration of 0.05–0.15 ms and voltage of 1–15 V to achieve a supramaximal compound muscle action potential.

Nociceptive Behavioral Testing

To examine the effect of delayed WD and regeneration on the development of pain hypersensitivity (neuropathic pain) in KO mice, we evaluated responses to both thermal and mechanical nociceptive stimuli. All nociceptive testing was performed on mice acclimatized to the testing environment and stimuli. Development of thermal hypersensitivity was evaluated by applying thermal stimuli using a heat source to the plantar surface of the hindpaws as described (45). Mechanical hypersensitivity (tactile allodynia) was measured using a calibrated set of von-Frey monofilaments of increasing stiffness that exert different forces (0.2–29 g) when pushed against the tested paw. A difference score (in grams) was assigned with negative scores indicating allodynia. In both testing paradigms mice were not restrained.

Statistical Analysis

All data were presented as mean ± SEM. Data were analyzed by using either a 1-way analysis of variance (ANOVA), Student _t_-test, Newman-Keuls, or the Mann-Whitney _U_-test when appropriate. A value of p < 0.05 was considered significant.

Results

Intact Nerve

Intact sham operated nerves without injury had low-level iNOS mRNA expression identified by RT-PCR. Low-level iNOS expression was consistent with the mild trauma of sham nerve exposure. No iNOS mRNA expression was detected in _iNOS_-KO mice (Fig. 1). Intact sciatic nerves taken from KO mice had a similar distribution and density of myelinated axons as their WT littermates (KO 20,322 ± 1,402 fibers/mm2, n = 5; WT 21,464 ± 881 fibers/mm2, n = 4; p = not significant [NS]) (Fig. 2).

Fig. 1.

Wallerian myelinated fiber degeneration-induced iNOS upregulation. a, b: Distal segments of injured sciatic nerve (following nerve constriction) taken 14 days (d) post-surgery from wild-type mice contain iNOS-immunoreactive cells (b, arrows). Similar sections from KO mice show no iNOS expression (a). c, d, e: RT-PCR analysis of iNOS-mRNA expression following nerve injury shows an upregulation of iNOS-mRNA in injured nerve segments between 5 d and 14 d following injury in WT mice (c, e), but not in KO mice (d, e). * = p < 0.05 compared to sham operated nerves. Analysis of Variance with 2-tailed Newman-Keuls post-hoc test (n = 4 for each group for Western blot or RT-PCR). Scale bar: a, b = 40 μm

Fig. 2.

Fiber size distribution histograms of myelinated fibers in _iNOS_-knockout (KO) mice and wild-type controls. The upper left panel indicates control sham-exposed uninjured sciatic nerves. Note that at 5 days (d) (right upper), KO mice have persistent and apparently intact myelinated fiber profiles, whereas these have largely disappeared in WT mice. By 14 d (left lower), these myelinated fiber profiles have only started to decrease in KO mice, unlike WT Mice. By 21 d (right lower), there are large numbers of new small myelinated fibers representing regenerating fibers in the WT mice, but not in the KO mice. (n = 4 for each time point and group of rats)

Chronic Constriction Nerve Injury (CCI)

In WT rats undergoing CCI there was a dramatic local increase in iNOS expression in the segment of the nerve distal to the injury, both at the transcription and translation levels, indicating upregulation of both mRNA and protein (24, 25). When compared to sham operated nerves, injured nerves taken from WT mice had a rise in iNOS-mRNA expression 5 days following the injury that reached a peak at 14 days and started to decline after that. Upregulation of iNOS-mRNA was also associated with a local increase in iNOS protein, detected by immunohistochemical labeling, as previously described in rats with CCI (24). _iNOS_-KO mice failed to express iNOS-mRNA or protein after CCI. Results of RT-PCR and immunohistochemistry are provided in Figure 1.

Morphometric studies addressed the extent of apparent myelinated fiber breakdown and fiber regeneration (Figs. 2, 3). Five days following CCI constriction, KO mice, unlike WT, had not yet broken down their small myelinated fibers: there was a similar number and density of small caliber axons (2–9-μm fiber diameter) compared to sham operated uninjured nerves (sham 13,231 ± 1,091 fibers/mm2, n = 4; KO-/- at 5 days 11,194 ± 840 fibers/mm2; p = NS, n = 4). The number and density of large caliber axons (10–20-μm fiber diameter) was mildly reduced (sham 7,091 ± 214 fibers/mm2; KO-/- 5,338 ± 531 fibers/mm2; p < 0.05, 1-way ANOVA, n = 4). Conversely, in WT mice there were far fewer surviving intact large and small caliber myelinated axons (Fig. 2; p < 0.01, 1-way ANOVA, n = 4) than in nerves taken from the _iNOS_-KO mice. By 14 days, numbers of myelinated fibers in KO mice had begun to decline but to a much lesser extent than in WT mice. Regenerative myelinated sprouts appeared distal to the sutures in partially injured nerves of the CCI model by 21 days after surgery in the WT mice. However, at the same time, similar sprouts had not appeared in KO mice.

Fig. 3.

Delayed Wallerian myelinated fiber degeneration in _iNOS_-knockout mice. Transverse semi-thin sections taken distal to CCI or sham operated sciatic nerves of wild-type and _iNOS_-knockout mice. All sections were stained with toluidine blue. Sham operated nerves taken from wild-type (A) and knockout (B) had a similar number and density of myelinated fibers. At 5 and 14 days following CCI (D and F, respectively), there were significantly more residual intact myelinated fibers (not yet broken down) in the knockout mice than in nerves taken from wild-type mice at the same time points (C and E, respectively). The bottom panels are slightly higher power views of 21-day sections illustrating abundant new regenerating myelinated fibers in wild-type mice (G) but very few in knockout mice (H). Scale bars: A–F = 50 μm; G, H = 30 μm

We addressed the behavioral features of neuropathic pain in the partial nerve injury CCI model by comparing nociceptive responses to thermal and mechanical stimuli applied to the hindpaw. In WT mice, thermal and tactile hypersensitivity were first observed by 48 h following injury, persisted through the second week after surgery (14 days) and then diminished, as expected in this model (37, 38). Conversely, despite an early trend toward hypersensitivity at 48 h in KO mice, thermal and tactile hypersensitivity disappeared between day 5 and day 14, and at these times were not significantly different from values prior to the CCI injury (Fig. 4). Furthermore, while these behavioral indices of neuropathic pain in the WT mice had diminished by 3 wk following injury, KO mice at this time began to exhibit delayed thermal hypersensitivity and tactile allodynia. The nociceptive behavioral differences between KO and WT mice coincided with the morphological evidence of earlier myelinated fiber breakdown (associated with earlier neuropathic pain) and more rapid regeneration (associated with resolution of neuropathic pain) in WT compared to KO mice.

Behavioral nociceptive testing of the thermal (A) and mechanical (B) responses of KO and WT mice undergoing CCI, indicating an attenuation in pain hypersensitivity in KO mice at 5 days (d) and 14 d following CCI at times when WT mice had persistent hypersensitivity. * = p < 0.01. Student t-test (n = 8 in each group)

Fig. 4.

Nerve Crush, Transection

Similar to CCI, the number of degenerating axon profiles 14 days following a complete crush injury also demonstrated apparent slowing of myelinated fiber breakdown. While in WT mice, distal to the crush zone, there was a rapid clearance of degenerative myelinated profiles associated with axonal regeneration (see below), KO mice had a higher number of persistent degenerating myelinated fiber profiles than WT mice at this time point (Table; Fig. 5), indicating delayed WD and myelin clearance. The number and density of new myelinated axonal sprouts distal to a crush site at 14 days were significantly higher in WT than KO mice (Table; Fig. 5). Moreover, regenerated myelinated fibers in WT mice were larger in caliber (data not shown), indicating greater maturity. By 6 wk following crush injury, fiber numbers, size distributions, and densities were similar in the KO and WT mice.

TABLE

Morphometric Studies of Sciatic Myelinated Fibers Distal to Crush at 14 Days

TABLE

Morphometric Studies of Sciatic Myelinated Fibers Distal to Crush at 14 Days

Fig. 5.

Delayed myelinated fiber degeneration and regeneration following complete nerve crush in _iNOS_-KO mice and WT controls 2 wk after injury. The top panels are transverse semi-thin sections of sciatic nerves distal to crush injury in a WT mouse (A) compared to an _iNOS_-KO mouse (B). Note the larger number of regenerating fibers (arrows) and fewer number of degenerating myelinated fiber profiles (asterisks) in WT compared to KO mice. The fiber size histogram in (C) illustrates the smaller number of regenerating myelinated fibers in KO compared to WT mice (* = p < 0.01; ANOVA). In (D), the higher number of degenerating myelinated fiber profiles in KO mice is indicated (* = p < 0.05; Student _t_-test). (n = 4 for each group). Scale bar: A = 50 μm

Morphometric studies following nerve transection were carried out 10 wk after injury since regenerative events are much more delayed in this type of injury than following crush. At this time, as noted at earlier times following crush or CCI, there was evidence of a maturational delay of regenerating axons in the KO mice with myelinated fibers in the KOs that were smaller than their WT littermates (Fig. 6A, B). We further characterized motor fiber regeneration by recording compound muscle action potentials (M waves) from the sciatic-tibial innervated interosseous muscles. The amplitude of the M wave is directly proportional to number of motor axons reinnervating interosseous endplates. As observed in Figure 6C, there were both delays in the reappearance of the M wave (the first detectable small amplitude M wave is generated by the first motor unit that is established during regeneration) from stimulation proximal to the injury site and in the subsequent recovery of the amplitude of the M wave (reflecting further motor unit reconnections). The former finding indicated a delay in the time required for the first regenerating motor axon to establish a functional motor unit at the endplate and the latter finding indicated impaired regeneration of subsequent motor axons arriving at the endplate that contribute to the M wave amplitude.

Fig. 6.

Delayed myelinated fiber regeneration following complete nerve transection in _iNOS_-KO mice and WT controls 10 wk after injury. A, B: The fiber size histograms illustrate a shift of fibers to smaller size categories in KO (* = p < 0.05 Student _t_-test at 6, 7 μm size category; n = 6 for each group). C: The appearance and recovery of tibial foot interosseous M wave compound muscle action potentials after transection is illustrated. The amplitude of the M wave is proportional to the numbers of reinnervating motor axons. In KO mice compared to WT mice, there is a delay in the appearance of the M wave after injury (observed at 4 weeks in WT but not KO mice) and in the recovery of its amplitude (* = p < 0.05; n = 5 for each group)

Discussion

The major findings from this work are as follows: (i) mice undergoing chronic constriction partial nerve injury upregulate iNOS mRNA and protein, a reaction that is absent in _iNOS_-KO mice; (ii) mice lacking iNOS have a delay in the apparent breakdown of myelinated fibers following both CCI and complete nerve crush; (iii) mice lacking iNOS have a delay in myelinated fiber regeneration after CCI, crush, or complete nerve transection; (iv) mice lacking iNOS have a delay in the expression of neuropathic pain following CCI; and (v) mice lacking iNOS have a delay in the motor axon reinnervation of muscle endplates.

An inflammatory reaction following injury is a crucial part of the healing process in tissue. Our finding that mice lacking iNOS have a pronounced delay in fiber breakdown distal to a peripheral nerve injury site and in fiber regeneration, highlights the importance of inflammatory iNOS and NO release following injury (14, 19, 20, 33, 46, 47). While we conclude that _iNOS_-KO animals had a delay in apparent WD of myelinated fibers, we have not strictly distinguished breakdown of axons and their myelin sheath in this work, events that are usually closely associated. It may be, for example, that the delays we have observed related more strictly to “clearance” of myelin sheaths and debris than axon dissolution. In separate work using an alternative _iNOS_-KO genetic model we have observed relatively minor differences between KO mice and WT in the rapid loss of axonal excitability following transection that occurs as axons disintegrate (Kennedy, Levy, and Zochodne, unpublished observation). This would then differ from the Ola mouse where both axons and their myelin sheath persist abnormally following transection (3).

In the CCI model, where strangulation of fibers beneath the sutures renders distal axonal degeneration, fiber breakdown “normally” occurs gradually. This model had abnormal persistence of myelinated profiles in mice lacking iNOS, eventually being resorbed in a delayed fashion. By 21 days in the CCI model, when behavioral indices of neuropathic pain had diminished, new axonal sprouts had emerged in WT mice. pain behavior thus appeared to be most prominent during actual fiber breakdown and subsequently subsided during regeneration. In contrast, KO mice by 21 days had only begun to develop increased pain behavior later when substantial fiber breakdown had finally ensued; yet no new axonal sprouts had appeared.

Apparent WD and myelin clearance proceeded much more rapidly after crush than after CCI. Some myelinated fiber profiles undergoing degeneration that had not yet been resorbed were still evident by 2 wk in both KO and WT mice. Many more of these degenerating profiles, however, persisted in the KOs. WT mice also had greater numbers of more mature, larger regenerating axons by 2 wk. In transected nerves, regenerating fibers were of smaller caliber in mice lacking iNOS, and there was delayed distal muscle endplate reinnervation.

The actions of NO in the peripheral nervous system parallel its role in healing following injury in other tissues (15, 16, 46, 48). During inflammation, in the presence of superoxide, NO forms peroxynitrite, a powerful oxidant capable of initiating lipid peroxidation (49, 50). This action is likely to assist in the dissolution of the myelin sheath or the phagocytosis of myelin by macrophages (50). Secretions of pro-inflammatory cytokines IL-1β and TNF-α, and apolipoprotein-E by macrophages (1) may further regulate the degeneration and regeneration processes. It is also possible that NO-mediated rises in nerve blood flow (24) are essential to support the metabolic requirements of newly sprouting axons. Following nerve injury, lack of NO might be detrimental to normal degeneration and regeneration by failing to support these requirements through the local blood supply. The influence of NO on WD and myelin clearance may thus indirectly facilitate axonal regeneration, a response that is highly dependent upon the rapid initiation of WD and a successful clearance of axonal and myelin debris (3–5).

Pain hypersensitivity, the clinical manifestation of neuropathic pain following nerve injury, also depends on rapid WD (6, 7). Local production of TNF-α and NGF by macrophages and Schwann cells during WD promotes the sensitization of injured nociceptive afferents (51, 52). NGF may also promote sympathetic sprouting in injured dorsal root ganglia (53, 54), and this is thought to contribute to nociceptive changes following nerve injury (53, 55–57). Delayed expression of these inflammatory molecules may have contributed to changes in the timing of neuropathic pain we observed in mice lacking iNOS. An absence of hyperalgesia was also recently observed in _iNOS_-KO mice with local Zymosan paw injection, a model of thermal hyperalgesia (58).

In summary, we suggest that the local release of NO following peripheral nerve injury by iNOS is a critical feature of WD. Our findings also support the concept that the subsequent success of axonal regeneration in the peripheral nervous system depends on how WD, including myelin clearance, proceeds. They support similar ideas that have been demonstrated in the Ola mouse, a strain of mice in which WD and the subsequent regeneration are impaired, though for reasons that are not certain (3–5). Finally, we show that the maintenance and resolution of neuropathic pain following nerve injury coincide with the development of WD and subsequent axon regeneration, respectively. Manipulation of NO action may offer interesting therapeutic options for the treatment of human peripheral nerve injuries.

Acknowledgment

Brenda Boake provided expert secretarial assistance. DWZ is a Senior Scholar of the Alberta Heritage Foundation for Medical Research. The work was supported by an operating grant from the Medical Research Council of Canada.

References

Macrophage systems in peripheral nerves

A review. J Neuropathol Exp Neurol

1993

;

553

–

Nerve injury, axonal degeneration and neural regeneration: Basic insights

Brain pathol

1999

;

313

–

Delayed wallerian degeneration in sciatic nerves of C57BL/Ola mice is associated with impaired regeneration of sensory axons

Brain Res

1990

;

530

117

–

Macrophage dependence of peripheral sensory nerve regeneration: possible involvement of nerve growth factor

Neuron

1991

;

359

–

Impaired motor axon regeneration in the C57BL/Ola mouse

J Comp Neurol

1993

;

333

449

–

Reduced hyperalgesia in nerve-injured WLD mice: Relationship to nerve fiber phagocytosis, axonal degeneration, and regeneration in normal mice

Exp Neurol

1996

;

141

–

101

Wallerian degeneration is required for both neuropathic pain and sympathetic sprouting into the DRG

Pain

1997

;

–

Interleukin-1 regulates synthesis of nerve growth factor in non-neuronal cells of rat sciatic nerve

Nature

1987

;

330

658

–

Interleukin 6 in intact and injured mouse peripheral nerves

Eur J Neurosci

1996

;

530

–

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

Neuroscience

1996

;

625

–

Inflammation and axonal regeneration

J Neurol

1994

;

242

S57

–

Sensory impairments and delayed regeneration of sensory axons in interleukin-6-deficient mice

J Neurosci

1999

;

4305

–

Nitric oxide synthases: Gene structure and regulation

Adv pharmacol

1995

;

–

Regulation and function of inducible nitric oxide synthase during sepsis and acute inflammation

Adv pharmacol

1995

;

155

–

et al. .

Reversal of impaired wound repair in iNOS-deficient mice by topical adenoviral-mediated iNOS gene transfer

J Clin Invest

1998

;

101

967

–

et al. .

Impaired liver regeneration in inducible nitric oxide synthase deficient mice

Proc Natl Acad Sci USA

1998

;

13829

–

Inducible nitric oxide synthase gene expression in brain following cerebral ischemia

J Cereb Blood Flow Metab

1995

;

378

–

Interaction between inducible nitric oxide synthase and cyclooxygenase-2 after cerebral ischemia

Proc Natl Acad Sci USA

1998

;

10966

–

et al. .

Inducible nitric oxide synthase is an endogenous neuroprotectant after traumatic brain injury in rats and mice

J Clin Invest

1999

;

104

647

–

et al. .

Inducible nitric oxide synthase stimulates dopaminergic neurodegeneration in the MPTP model of parkinson disease

Nat Med

1999

;

1403

–

Inducible nitric oxide synthase immunoreactivity in the Alzheimer disease hippocampus: Association with Hirano bodies, neurofibrillary tangles, and senile plaques

J Neuropathol Exp Neurol

1999

;

1163

–

et al. .

Human immunodeficiency virus infection, inducible nitric oxide synthase expression, and microglial activation: pathogenetic relationship to the acquired immunodeficiency syndrome dementia complex

Ann Neurol

1999

;

207

–

Nitric oxide synthase inhibition delays axonal degeneration and promotes the survival of axotomized retinal ganglion cells

Exp Neurol

1999

;

158

366

–

Local nitric oxide synthase activity in a model of neuropathic pain

Eur J Neurosci

1998

;

1846

–

Local expression of inducible nitric oxide synthase in an animal model of neuropathic pain

Neurosci Lett

1999

;

260

207

–

Expression of three forms of nitric oxide synthase in peripheral nerve regeneration

J Neurosc Res

1999

;

198

–

207

Reactive oxygen species are required for the phagocytosis of myelin by macrophages

J Neuroimmunology

1998

;

–

Contrasting roles for nitric oxide and peroxynitrite in the peroxidation of myelin lipids

J Neuroimmunol

1999

;

–

Neuronal growth cone collapse and inhibition of protein fatty acylation by nitric oxide

Nature

1993

;

366

562

–

et al. .

Altered responses to bacterial infection and endotoxic shock in mice lacking inducible nitric oxide synthase

Cell

1995

;

641

–

Ethical guidelines for investigations of experimental pain in conscious animals

Pain

1983

;

109

–

Single-step method of RNA isolation by acid guanidinium thiocyanate-phenol-chloroform extraction

Anal Biochem

1987

;

162

156

–

et al. .

Inducible nitric oxide synthase-deficient mice have enhanced leukocyte-endothelium interactions in endotoxemia

FASEB J

1997

;

955

–

Monitoring mRNA expression by polymerase chain reaction: The “primer-dropping” method

Anal Biochem

1994

;

223

251

–

et al. .

Evidence for nitric oxide and nitric oxide synthase activity in proximal stumps of transected peripheral nerves

Neuroscience

1999

;

1515

–

Mice lacking inducible nitric oxide synthase are not resistant to lipopolysaccharide-induced death

Proc Natl Acad Sci USA

1995

;

10688

–

A peripheral mononeuropathy in rat that produces disorders of pain sensation like those seen in man

Pain

1988

;

–

107

Transient action of the endothelial constitutive nitric oxide synthase (ecNOS) mediates the development of thermal hypersensitivity following peripheral nerve injury

Eur J Neurosci

2000

;

2323

–

The spectrum of fiber loss in a model of neuropathic pain in the rat: An electron microscopic study

Pain

1991

;

359

–

Is large myelinated fiber loss associated with hyperalgesia in a model of experimental peripheral neuropathy in the rat?

Pain

1993

;

233

–

Biopsy diagnosis of peripheral neuropathy

Butterworth-Heinemann

Toronto,

1995

Auer RN Automated nerve fibre size and myelin sheath measurement using microcomputer-based digital image analysis: Theory, method and results

J Neurosci Methods

1994

;

229

–

Diabetes increases sciatic nerve susceptibility to endothelin-induced ischemia

Diabetes

1996

;

627

–

Inhibition of nitric oxide synthase enhances peripheral nerve regeneration in mice

Neurosci Lett

1997

;

228

–

A new and sensitive method for measuring thermal nociception in cutaneous hyperalgesia

Pain

1988

;

–

The selective beneficial effects of nitric oxide inhibition in experimental colitis

Am J physiol

1995

;

268

G673

–

G684

et al. .

Expression of inducible nitric oxide synthase and its involvement in pulmonary granulomatous inflammation in rats

Am J pathol

1996

;

149

2005

–

Inducible nitric oxide synthase plays a critical role in resolving intestinal inflammation

Gastroenterology

1997

;

112

1022

–

Peroxynitrite-induced membrane lipid peroxidation: The cytotoxic potential of superoxide and nitric oxide

Arch Biochem Biophys

1991

;

288

481

–

et al. .

Nitric oxide regulation of superoxide and peroxynitrite-dependent lipid peroxidation

Formation of novel nitrogen-containing oxidized lipid derivatives. J Biol Chem

1994

;

269

26066

–

Tumour necrosis factor-alpha induces ectopic activity in nociceptive primary afferent fibres

Neuroscience

1997

;

255

–

Neutralization of endogenous NGF prevents the sensitization of nociceptors supplying inflamed skin

Eur J Neurosci

1999

;

1698

–

1704

Adrenergic innervation of rat sensory ganglia following proximal or distal painful sciatic neuropathy: Distinct mechanisms revealed by anti-NGF treatment

Eur J Neurosci

1999

;

837

–

Nerve growth factor-induced sprouting of mature, uninjured sympathetic axons

J Comp Neurol

1992

;

326

327

–

Adrenergic excitation of cutaneous pain receptors induced by peripheral nerve injury

Science

1991

;

251

1608

–

Expression of alpha2-adrenergic receptors in rat primary afferent neurones after peripheral nerve injury or inflammation

J physiol

1999

;

515

533

–

Rapid sprouting of sympathetic axons in dorsal root ganglia of rats with a chronic constriction injury

Pain

1997

;

237

–

et al. .

Suppressed injury-induced rise in spinal prostaglandin E2 production and reduced early thermal hyperalgesia in iNOS-deficient mice

J Neurosci

2000

;

6714

–

Author notes

Current address for Dan Levy: Department of Anesthesia, Harvard Medical School and Beth Israel Deaconess Medical Center, Boston, Massachusetts