BBI 2007

Brain, Behavior, and Immunity 21 (2007) 561–568

www.elsevier.com/locate/ybrbi

Cytokine regulation of MMP-9 in peripheral glia: Implications for pathological processes and pain in injured nerve

Sharmila Chattopadhyay a,b, Robert R. Myers a,b, Julie Janes a, Veronica Shubayev a,b,¤

a San Diego VA Healthcare System, USAb University of California, San Diego, School of Medicine, Department of Anesthesiology, La Jolla, CA, USA

Received 26 August 2006; received in revised form 20 October 2006; accepted 20 October 2006Available online 26 December 2006

Abstract

Matrix metalloproteinase-9 (MMP-9) is an extracellular protease that is induced in Schwann cells hours after peripheral nerve injuryand controls axonal degeneration and macrophage recruitment to the lesion. Here, we report a robust (90-fold) increase in MMP-9mRNA within 24 h after rat sciatic nerve crush (1 to 60 days time-course). Using direct injection into a normal sciatic nerve, we identifythe proinXammatory cytokines TNF-� and IL-1� as potent regulators of MMP-9 expression (Taqman qPCR, zymography). MyelinatingSchwann cells produced MMP-9 in response to cytokine injection and crush nerve injury. MMP-9 gene deletion reduced unstimulatedneuropathic nociceptive behavior after one week post-crush and preserved myelin thickness by protecting myelin basic protein (MBP)from degradation, tested by Western blot and immunoXuorescence. These data suggest that MMP-9 expression in peripheral nerve is con-trolled by key proinXammatory cytokine pathways, and that its removal protects nerve Wbers from demyelination and reduces neuro-pathic pain after injury.© 2006 Elsevier Inc. All rights reserved.

Keywords: Schwann cell; Matrix metalloproteinase; TNF-�; IL-1�; NGF; Glia; Myelination; MBP; Pain; Neuropathy

1. Introduction

Neuropathic pain is often a consequence of neuropatho-logical and molecular changes resulting from peripheralnerve damage. Complex interactions of injured peripheralnerve Wbers with activated glia (Schwann cells) andrecruited immune cells is regulated by a number of immu-nomodulatory and trophic factors. ProinXammatory cyto-kines, such as tumor necrosis factor alpha (TNF-�) andinterleukins (IL-1�, IL-6), have been implicated in the path-ogenesis of Wallerian degeneration and neuropathic pain,as they control axonal demyelination, degeneration, blood-nerve permeability, and immune cell recruitment (Stollet al., 2002), and thus, represent model therapeutic targetsin neuropathic pain (Myers et al., 2006). Recently, we have

* Corresponding author. Fax: +1 858 534 1445.E-mail address: [email protected] (V. Shubayev).

0889-1591/$ - see front matter © 2006 Elsevier Inc. All rights reserved.doi:10.1016/j.bbi.2006.10.015

shown that some critical actions of TNF-� in injured nerve,such as macrophage recruitment, are mediated by matrixmetalloproteinase-9 (MMP-9 or gelatinase B) (Shubayevet al., 2006).

MMP-9 belongs to a family of Zn2+-dependent extracel-lular proteases called matrix metalloproteinases (MMPs),that comprise collagenases, gelatinases, stromelysins, andmembrane-type MMPs (Woessner, 1994). In the nervoussystem, MMPs produce neuroinXammation by controllingneurovascular permeability, immune cell recruitment,demyelination, cell necrosis, and apoptosis (Yong et al.,1998; Kieseier et al., 1999b, Rosenberg, 2002; Lee et al.,2004a, Yong, 2005). MMP-9 is upregulated in experimentalperipheral neuropathy models (La Fleur et al., 1996; Kherifet al., 1998; Ferguson and Muir, 2000; Siebert et al., 2001;Hughes et al., 2002; Platt et al., 2003; Demestre et al., 2004)and in patients with symptomatic neuropathy (Leppertet al., 1999; Mawrin et al., 2003; Renaud et al., 2003; Gureret al., 2004). We have recently shown that MMP-9 gene

562 S. Chattopadhyay et al. / Brain, Behavior, and Immunity 21 (2007) 561–568

deletion or pharmacologic inhibition reduces injury-induced macrophage recruitment and protects nerves fromaxonal degeneration (Shubayev et al., 2006).

In the central nervous system, MMP-9 is fundamental tomyelination (Yong, 2005), in part, by degradation of myelinbasic protein (MBP) (Gijbels et al., 1993; Proost et al.,1993). While MBP constitutes only 10–20% of PNS myelin(Jacobs, 2005), it is critical to maintaining integrity andcompactness of peripheral nerve in development (Martiniand Schachner, 1997) and after injury (LeBlanc and Podu-slo, 1990). The importance of MMP-9 in peripheral nervedemyelination has been documented (Redford et al., 1995,1997; Kieseier et al., 1999a,b; Siebert et al., 2001), but themechanism of its action has not been clariWed.

The purpose of this study is to address whether activa-tion of peripheral glia by proinXammatory cytokinesinduces MMP-9 expression in vivo, and to analyze the roleof MMP-9 in controlling MBP levels and demyelinationafter peripheral nerve injury.

2. Methods

2.1. Animal surgery

Adult female Sprague–Dawley rats (nD 133; 250 g, Harlan Labs),MMP-9 knockout (n D 25, FVB.Cg-Mmp9tm1Tvu/J) and wild-type mice(n D 25, FVB/NJ), TNFR1 (n D 8, B6.129-Tnfrsf1atm1Mak/J), TNFR1/2knockout (n D 8, B6.129S-Tnfrsf1atm1Imx Tnfrsf1btm1Imx/J) and wild-type(n D 8, B6129SF2/J) mice were used. All mouse strains were obtained fromJackson Laboratory (Bar Harbor, ME). Anesthesia was induced with 4%IsoXurane (IsoSol; Vedco, St. Joseph, MO), the sciatic nerve was exposedunilaterally at the mid-thigh level, and crushed using Wne, smooth-surfaceforceps twice for 5 s each to produce nerve crush. Nerve injections weremade into uninjured rat sciatic nerves using a Hamilton syringe, a 30-gauge-needle and an injectate volume of 5 �l. Animals were sacriWcedusing an intraperitoneal injection of a cocktail containing sodium pento-barbital (Nembutal, 50 mg/ml; Abbott Labs, North Chicago, IL) diaze-pam (5 mg/ml, Steris Labs, Phoenix, AZ) and saline (0.9%, Steris Labs) in avolume proportion of 1:1:2, respectively. All procedures were performedaccording to protocols approved by the VA Healthcare System Commit-tee on Animal Research, and conform to the NIH Guidelines for AnimalUse.

2.2. Antibodies and proteins

Recombinant rat TNF-� (R&D Systems), IL-1� (Pierce) or NGF(Invitrogen) were delivered into sciatic nerve at 250 pg per rat, or bovineserum albumin (BSA, Sigma, 0.1%, vehicle) in 5 �l volume as previouslydescribed (Wagner and Myers, 1996). The following antibodies were usedfor immunodetection: rabbit anti-MMP-9 (Torrey Pines Labs, 1:500),mouse anti-MBP (Abcam, 1:50), rabbit anti-S100 (Dako, 1:2000), andmouse anti-�-actin (Sigma, 1:10,000). Respective normal serum or IgGwas used for negative control. All antibodies were diluted in 1% blockingserum in PBS.

2.3. Immunohistochemistry

ParaYn-embedded, 4% paraformaldehyde-Wxed nerve sections (10-�m-thick) were deparaYnized with xylenes, rehydrated in graded ethanolPBS and subjected to detection as previously described (Shubayev andMyers, 2002) and summarized below:

(1) Dilaminobenzidine (DAB): endogenous peroxidase was blocked with 3% H2O2, antigen retrieval (Dako, Carpinteria, CA) (5 min at 95 °C

and 20 min at room temperature), non-speciWc binding was blockedwith 10% normal goat serum, followed by a rabbit anti-MMP-9 anti-body incubation (see above) overnight at 4 °C, goat anti-rabbit IgG(Vector) and avidin-biotin complex (ABC Elite, Vector) application.Sections were developed with DAB (brown stain, Vector).

(2) ImmunoXuorescence: 0.5% sodium borohydride in 1% dibasic sodium phosphate buVer was applied for 5 min to block endogenous aldehydegroups, followed by Dako antigen retrieval, non-speciWc binding blockin 5% goat serum for 30 min, mouse anti-MBP antibody (see above)overnight at 4 °C, alexa goat anti-mouse 488 antibody for 1 h, andnuclear 4�,6-diamidino-2-phenylindole (DAPI) stain (MolecularProbes, 1:20000, blue) for 5 min. Sections were mounted using Slowf-ade gold antifade reagent (Molecular Probes). PBS was used for allwashes.

2.4. Real-time qPCR

Sciatic nerve fragments and L5/L4 DRG samples were pooled from 2rats and stored in RNA-later (Ambion) at ¡20 °C. Total RNA wasextracted with Trizol (Invitrogen) and treated with RNase-free DNAse I(Qiagen). The RNA purity was veriWed by OD260/280 absorption ratioof about 2.0. cDNA was synthesized using SuperScript II Wrst-strandRT-PCR kit (Invitrogen). Gene expression was measured by quantita-tive real-time polymerase chain reaction (qPCR, MX4000, Stratagene,La Jolla, CA) using 50 ng of rat cDNA and 2£ Taqman Universal PCRMaster Mix (Applied Biosystems) with a one-step program (95 °C for10 min, 95 °C for 30 s and 60 °C for 1 min for 50 cycles). Primers andTaqman probes for MMP-9 from Biosearch Technologies (Novato, CA)were optimized using injured sciatic nerve cDNA (ampliWcationeYciency of 100.1–100.3%), as reported earlier (Shubayev et al., 2006).Glyceraldehyde 3-phosphate dehydrogenase (GAPDH) gene was used asa normalizer, and its expression was conWrmed to be not regulated ininjured and uninjured nerves. Duplicate samples without cDNA (no-template control) for each gene showed no contaminating DNA. Rela-tive mRNA levels were normalized to GAPDH, Wve samples per groupwere quantiWed using the comparative Ct method (Livak and Schmitt-gen, 2001), and a fold change was determined by the MX4000 (PfaZ,2001).

2.5. Gelatin zymography

Nerves were lysed in non-reducing, protease-inhibitor-free Laemmlisample buVer, heated at 55 °C for 5 min and 50 �g tissue per well wasrun on 10% SDS polyacrylamide gel containing 1 mg/ml of gelatin at160 V for 90 min (Shubayev and Myers, 2000). The gels were washed in2.5% Triton X-100, developed at 37 °C overnight in 50 mM Tris–HCl,150 mM NaCl, 5 mM CaCl2, 1 �M ZnCl2, and 0.2 mM sodium azide(pH 7.6) and stained with colloidal blue (Invitrogen), indicating gelatin-olytic MMP activity as a clear band on a dark background of unde-graded gelatin. Inverted images are presented. Recombinant humanMMP-9 (Chemicon) was used for control. Zymograms were digitizedusing an EC3 Darkroom (UVP Imaging) and quantiWed by LabWorks4.5. Data are expressed as relative optical density (OD) of gelatinolyticactivity.

2.6. Western blotting

Nerves were lysed in Laemmli buVer containing 10 mM PMSF, 5 mMEDTA, protease inhibitor cocktail (Sigma) (pH 6.8) as previouslydescribed (Shubayev and Myers, 2000), reduced with 10% �-mercap-toethanol (Fisher), and 30–50 �g of protein (BSA Protein Assay, Pierce)was run on 15% SDS–PAGE in a Laemmli system. Proteins were trans-ferred to nitrocellulose at 50 V for 60 min in transfer buVer (12 mM Tris–base, 95 mM glycine, and 20% methanol, pH 8.3). Non-speciWc bindingwas blocked in 5% non-fat dry milk (Bio-Rad) followed by a primaryantibody incubation overnight at 4 °C, HRP-tagged goat anti-mouse oranti-rabbit IgG, and detection with enhanced chemiluminescence

S. Chattopadhyay et al. / Brain, Behavior, and Immunity 21 (2007) 561–568 563

(Amersham). Molecular weight was determined using HRP-taggedSDS–PAGE standards (Bio-Rad). Blots were digitized using an EC3Darkroom (UVP Imaging) and quantiWed by LabWorks 4.5. Data areexpressed as relative optical density (OD) ratios of experimental to con-trol proteins.

2.7. Spontaneous pain behavior

Spontaneous pain behavior was measured according to the methoddescribed by Attal et al. (Attal et al., 1990; Paulson et al., 2002) in MMP-9knockout (n D 10) and wild-type mice (n D 10) after sciatic nerve crush for2 weeks. Each animal was placed in a plexiglass cylinder (19 cm £ 31 cm)and allowed to habituate. One animal at a time was continuously observedfor 2 min. This was repeated 2 more times within the next 2 h. DiVerentpositions of the injured hind paw were continuously rated, according tothe following numerical scoring system: 0 D the paw is placed normally onthe Xoor, 1 D the paw is placed lightly on the Xoor and the toes are in aventroXexed position, 2 D only the internal edge of the paw is placed onthe Xoor, 3 D only the heel is placed on the Xoor and the hind paw is in aninverted position, 4 D the whole paw is elevated, and 5D the animal licksthe paw. During each 2 min (120 s) test period, measurements were takencontinuously by a tester blinded to the experimental groupings. In practi-cal terms, this was done by pressing one of six (0–5) numerical keys on acomputer keyboard. Only one key was pressed at a time, corresponding tothe instantaneous behavior of the animal. This resulted in a continuous120 s evaluation of the behavior that could be parsed oV-line into seconds/behavior during the experimental period. An index for noxious behaviorwas calculated by multiplying the amount of time the mice spent in eachbehavior multiplied by a weighting factor for that behavior, and dividedby the length of the observational period, using the formula:[0t0 + 1t1 + 2t2 + 3t3 + 4t4 + 5t5]/120 s, where t0–t5 are the durations in sec-onds spent in behaviors 0–5, respectively. The three values correspondingto three blocks of 120 s were averaged to determine the spontaneous painscore for each mouse.

3. Results

3.1. MMP-9 expression in crushed rat sciatic nerve and corresponding DRG

The patterns of MMP-9 mRNA expression were ana-lyzed during the course of Wallerian degeneration after ratsciatic nerve crush (Fig. 1). MMP-9 expression in nerve wasrobustly elevated (86.9§ 7.78-fold) at 1 day after crush, andgradually returned to baseline by 60 days post-crush. In thecorresponding DRG, MMP-9 expression was moderatelystable throughout the course of injury, showing a signiW-cant 2.65§ 0.28 increase only at 2 weeks post-crush.

3.2. Cytokines regulate MMP-9 expression in peripheral nerve

Pro-inXammatory cytokines activate glia after nerveinjury. MMP-9 in peripheral nerve is produced only afterinjury, predominantly by Schwann cells (Shubayev andMyers, 2000, 2002). Cytokines and trophic factors areknown inducers of MMP-9 (Nagase, 1997). Twenty-fourhours after we injected recombinant rat TNF-�, IL-1� orNGF proteins into normal sciatic nerve, MMP-9 mRNA(Fig. 2A) and proteolytic activity (Fig. 2B) were analyzed.Day 1 crushed and uninjured nerves served as positive andnegative controls, respectively. A signiWcant increase in

MMP-9 mRNA was observed after NGF, TNF-� and IL-1� injection relative to BSA (vehicle) injection and unin-jured nerve. However, BSA injection did cause someMMP-9 induction relative to uninjured nerve. Immuno-histochemical analysis of MMP-9 after TNF-� injectionparalleled the mRNA and protein expression data, andidentiWed myelinated Schwann cells as a chief source ofMMP-9 in response to cytokine injections. Again, weobserved a mild increase in MMP-9 after BSA injection,but a robust increase after TNF-� injection, comparableto that of Day 1 crush. Some axonal reactivity was notedin TNF-�-injected and crushed nerves, probably due toincreased neuronal-glial interaction. The overall histolo-pathological changes in cytokine-injected nerves weremild and comparable to that of crushed nerves.

To identify a speciWc pathway of TNF-�-mediatedMMP-9 induction, we assessed MMP-9 activity in TNF-�receptor 1 (TNFR1) knockout and TNFR1 and TNFR2double-knockout mouse nerves at Day 1 after crush(Fig. 3). Similar to TNF-� knockout (Shubayev et al.,2006), we observed only a mild decline of MMP-9 inTNFR1 and TNFR1/2 knockouts. There was no signiW-cant diVerence in MMP-9 activity between TNFR1knockouts and TNFR1/2 double-knockout mice, suggest-ing that TNFR1 is the main TNF-� receptor to mediateMMP-9 expression. These data suggest that high MMP-9levels in knockout cytokine nerve injury models is main-tained due to compensatory activation of related mecha-nisms, such as IL-1�. Together, these data support thehypothesis that Schwann cell activation by several impor-tant cytokine and trophic pathways results in MMP-9induction.

Fig. 1. MMP-9 mRNA expression after sciatic nerve crush. Real-timeTaqman qPCR for MMP-9 in nerve and ipsilateral DRG. Data areexpressed as the mean fold increase §SE in crushed relative to uninjuredgroups, n D 10/group, ¤p < 0.05, ¤¤p < 0.01, by one-way ANOVA andTukey’s post hoc. Note an 87-fold increase in MMP-9 mRNA in nerve at1 day that is gradually reduced by 60 days after crush.

564 S. Chattopadhyay et al. / Brain, Behavior, and Immunity 21 (2007) 561–568

3.3. MMP-9 inXuences neuropathic pain behavior

We sought to determine if MMP-9, as a cytokine-medi-ated factor, regulates neuropathic pain. Spontaneous painbehavior was scored in a blinded fashion in MMP-9 knock-out and wild-type animals for 2 weeks after nerve crush(Fig. 4). MMP-9 knockout mice expressed less pain, as indi-cated by a statistically signiWcant decline in the pain indexrelative to wild-type animals, at 2 days and at 8 and 10 daysafter crush. MMP-9 deletion, however, did not facilitaterecovery from neuropathic pain, demonstrating the samescore of 0.8 in both groups at 2 weeks after crush.

3.4. MMP-9 controls myelin protein content after nerve injury

While MMP-9 importance in regulating MBP turnover inCNS is well-accepted, its role in processing MBP in peripheralnerve has not been analyzed. MMP-9 knockout and wild-typemouse nerves were analyzed for MBP protein levels at 10days after crush (Fig. 5). At this time-point, animals displayreduced pain behavior (see Fig. 4), and MBP levels in wild-type injured sciatic nerve are normalized after initial demye-lination (Gupta et al., 1988; LeBlanc and Poduslo, 1990); weconWrmed the latter observation (not shown). MMP-9 genedeletion caused almost a 2-fold increase in unprocessed MBP

(52kDa) relative to wild-type (Fig. 5A and B). No change inS100 (common Schwann cell marker, 13kDa) or �-actin (pro-tein loading control, 42kDa) was seen. Calibration of MBP toS100 levels indicates that MBP protection in MMP-9 knock-out nerves is not related to the changes in Schwann cell viabil-ity. ImmunoXuorescence for MBP (green) and the nuclearstain, DAPI (blue) (Fig. 5C), paralleled observation of theWestern blot, showing preserved MBP levels and myelinthickness after MMP-9 deletion.

These data indicate that in the PNS, MMP-9 regulatesMBP turnover and myelin thickness, while MMP-9 genedeletion protects, concurrently, from neuropathic pain andmyelin degradation.

4. Discussion

This study demonstrates that in peripheral nerve MMP-9 is induced within a day after injury in response to proin-Xammatory cytokines, and that MMP-9 gene deletionreduces neuropathic pain behavior in concordance withpreserved myelin integrity.

MMP-9 increase within 1 day after nerve injury, preced-ing neuropathological evidence of degeneration, has been aconsistent observation (La Fleur et al., 1996; Kherif et al.,1998; Ferguson and Muir, 2000; Shubayev and Myers,2000, 2002, 2006; Platt et al., 2003). TNF-� induces MMP-9

Fig. 2. Cytokine-induced MMP-9 expression in sciatic nerve. (A) Real-time Taqman qPCR for MMP-9. Data are expressed as the mean fold increase §SEin injected relative to uninjured nerves, nD 10/group. Statistics: (¤p < 0.05, relative to uninjured nerve, #p < 0.05, relative to vehicle group by one-wayANOVA and Tukey’s post hoc). Crushed (Day 1) nerves were used for positive control. (B) Gelatin zymography (inverted image) demonstrates increasedMMP-9 activity in nerve after NGF, TNF-� and IL-1� injection. Uninjured and crushed (Day 1) nerves are used as negative and positive controls, respec-tively (n D 6/group). (C) MMP-9 immunoreactivity after TNF-� injection showing myelinated Schwann cell (arrow) reactivity is similar to the endogenousMMP-9 changes after crush. Micrographs are representative of four animals/group (100£ objective magniWcation).

S. Chattopadhyay et al. / Brain, Behav

in the CNS (Rosenberg et al., 1995), in injured sciatic nerve(Shubayev et al., 2006), and as shown here, in uninjured sci-atic nerve. While this study emphasizes the importance ofTNF-�, it also implicates IL-1� and NGF in MMP-9induction in peripheral nerve. IL-1� upregulates MMP-9 inoptic nerve (Zhang and Chintala, 2004) and brain (Vecilet al., 2000), and NGF is known to induce MMP-9 in cul-tured neurons (Muir, 1994; Shubayev and Myers, 2004).The ability of the vehicle injection to cause the increase inMMP-9 is consistent with observations of mild inXamma-tory response to sham surgeries (Kleinschnitz et al., 2005).

We observed that Schwann cells produce MMP-9 inresponse to TNF-� in vivo, in accordance with our earlierstudies in cultured primary Schwann cells (Shubayev et al.,2006). However, other endoneurial cells can upregulateMMP-9 (Shubayev and Myers, 2002), and may do so inresponse to cytokines, as has been shown for Wbroblasts(Singer et al., 1999) and endothelial cells (Genersch et al.,2000). It remains to be determined whether Schwann cellsof diVerent phenotypes equally respond to TNF-� chal-lenge by increasing MMP-9 production. Central micro- andmacroglia also produce MMP-9 in response to injury(Hughes et al., 2002; Rosenberg, 2002; Lee et al., 2004b).

MMP-9 is a critical mediator of demyelination in thecentral (Rosenberg, 2002) and peripheral (Kieseier et al.,1999b) nervous systems. It is known to control the break-down of MBP (Chandler et al., 1995), a late component ofmyelin formation that is produced by Schwann cells ininjured peripheral nerve (Gupta et al., 1988; LeBlanc and

Fig. 3. Partial reduction in MMP-9 in nerve after TNF-� receptor dele-tion. Gelatin zymography (inverted image) demonstrates partial reductionin MMP-9 activity in crushed (Day 1) TNFR1 knockout and TNFR1and 2 double-knockout nerves. Densitometry (graph, n D 6/group,mean § SE), ¤p < 0.05 knockout vs. wild-type by Student’s t-test.

Poduslo, 1990). While MMP-9-dependent degradation ofMBP has been shown in models of multiple sclerosis (Gij-bels et al., 1993; Proost et al., 1993) and cerebral ischemia(Asahi et al., 2001; Cho et al., 2006), this is the Wrst demon-stration of this relationship in the PNS. Other MMPs, suchas MMP-12 (Larsen et al., 2006) and MMP-3 (D’Souza andMoscarello, 2006), regulate MBP processing in the CNSand may play a role in peripheral nerve. MMP-9 is alsoinvolved in myelination via interaction with proteoglycansand growth factors (Yong, 2005). While MMP-9 promotesTNF-�-mediated macrophage recruitment into the injurednerve (Shubayev et al., 2006), neither TNF-� (Liefner et al.,2000) nor MMP-9 (Siebert et al., 2001) alter the myelinphagocytosing function of macrophages, suggesting thattheir roles in demyelination is not secondary to the abilityto modulate macrophage recruitment.

Activation of Schwann cells has been implicated in thepathogenesis of neuropathic pain (McMahon et al., 2005;Myers et al., 2006). Here, we observed a delayed, mild butstatistically signiWcant reduction in pain behavior afterMMP-9 gene deletion. The delay in mechanical allodyniais characteristic of other neuroprotective models, such asthe spontaneous WldS mutant mouse model of delayedWallerian degeneration (Sommer and Schafers, 1998),which fails to induce MMP-9 and TNF-� (Shubayevet al., 2006). The mild eVect may point to the secondaryrole of MMP-9 in pain or compensatory mechanisms ofMMP-9 knockout. To date, two other studies directlyassessed the eVect of MMP inhibition on neuropathicpain. MT5-MMP gene deletion virtually ablated mechan-ical allodynia associated with partial sciatic nerve liga-tion (Komori et al., 2004), and synthetic inhibitor TAPIsigniWcantly reduced thermal hyperalgesia and mechani-cal allodynia after chronic constriction injury in mice(Sommer et al., 1997). TAPI inhibits TNF-� activation bychelating TNF-� converting enzyme (TACE) and, at

Fig. 4. MMP-9 gene deletion reduces painful behavior. Unstimulated painscore was recorded in MMP-9¡/¡ and control FVB mice for 2 weeksafter sciatic nerve crush. DiVerent positions of the injured hind paw wererated in each animal for 15 minutes (3£ 300 s) using 0–5 numerical scale(see methods); data expressed as mean § SE. Note reduction in pain scorein MMP-9¡/¡ animals, n D 10/group, ¤p < 0.05 knockout vs. wild-type byone-way ANOVA and Tukey’s post hoc.

ior, and Immunity 21 (2007) 561–568 565

566 S. Chattopadhyay et al. / Brain, Behavior, and Immunity 21 (2007) 561–568

higher doses, MMP-9 and other MMPs. MMP inhibitionalso improves electrophysiologic nerve conduction andmotor performance (Leppert et al., 1999; Hsu et al.,2006).

Observation of unstimulated foot positioning is com-monly done in the formalin test, and is used here to moni-tor long-lasting or tonic pain, the most common features ofclinical painful neuropathy. This test correlates well withhyperalgesia to mechanical and thermal stimuli in majormodels of experimental neuropathy (Attal et al., 1990). Ourstudy suggests that MMP-9 role in demyelination relates tothe basic mechanisms of neuropathic pain. Demyelinationof injured aVerents is known to cause ectopic discharge andneuropathic nociception due to remodeling of the exposedaxonal membrane, such as sodium channel insertion that isnormally inhibited by myelin (Devor, 2006). Mechanicalallodynia (signaled by myelinated aVerents) associates withsciatic nerve crush (Lancelotta et al., 2003), but perhapsbetter assessed in models producing a more robust mechan-ical sensitization, e.g., spinal nerve ligation (Kim andChung, 1992).

The present study utilizes mouse and rat species.While changes in MMP-9 expression after cytokineinjections were correlated with the well-deWned para-digm of nerve crush all in rat sciatic nerve, the use ofmutant mouse nerve crush models allowed us to analyzespeciWc mechanisms of MMP-9 expression (TNF-�receptor knockout) and function (MMP-9 knockout).Earlier work shows highly consistent changes in cyto-

kine and MMP expression between mouse and rat nervecrush models, but certain signaling diVerences betweenthe species might exist.

In conclusion, this study suggests that MMP-9 is a sensi-tive biomarker of peripheral nerve injury that is regulatedby multiple cytokine pathways. MMP-9 deletion protectsnerve Wbers by preservation of MBP protein levels andmyelin thickness and reduces spontaneous pain behaviors.

Acknowledgments

The authors thank Jenny Dolkas, Amy Friedrich, andMila Angert for expert technical assistance. This work issupported by the Department of Veterans AVairs and theNIH Grant NS18715.

References

Asahi, M., Wang, X., Mori, T., Sumii, T., Jung, J.C., Moskowitz, M.A.,Fini, M.E., Lo, E.H., 2001. EVects of matrix metalloproteinase-9 geneknock-out on the proteolysis of blood–brain barrier and white mattercomponents after cerebral ischemia. J. Neurosci. 21, 7724–7732.

Attal, N., Jazat, F., Kayser, V., Guilbaud, G., 1990. Further evidence for‘pain-related’ behaviours in a model of unilateral peripheral mononeu-ropathy. Pain 41, 235–251.

Chandler, S., Coates, R., Gearing, A., Lury, J., Wells, G., Bone, E., 1995.Matrix metalloproteinases degrade myelin basic protein. Neurosci.Lett. 201, 223–226.

Cho, K.O., La, H.O., Cho, Y.J., Sung, K.W., Kim, S.Y., 2006. Minocyclineattenuates white matter damage in a rat model of chronic cerebral hyp-operfusion. J. Neurosci. Res. 83, 285–291.

Fig. 5. MMP-9 regulates MBP turnover. Western blot (A) and immunoXuorescence (B) for myelin basic protein (MBP) in crushed (day 10) MMP-9knockout nerves showed preserved MBP (myelinating Schwann cell marker) and no change in S100 (Schwann cell marker) or �-actin (loading control).Densitometry was done in nD 8/group. Dual-immunoXuorescence for MBP (green) and nuclear stain DAPI (blue) shows preserved MBP and myelinthickness after MMP-9 deletion (B). Micrographs are representative of n D 4/group (100£ objective magniWcation).

S. Chattopadhyay et al. / Brain, Behavior, and Immunity 21 (2007) 561–568 567

Demestre, M., Wells, G.M., Miller, K.M., Smith, K.J., Hughes, R.A., Gear-ing, A.J., Gregson, N.A., 2004. Characterisation of matrix metallopro-teinases and the eVects of a broad-spectrum inhibitor (BB-1101) inperipheral nerve regeneration. Neuroscience 124, 767–779.

Devor, M., 2006. Sodium channels and mechanisms of neuropathic pain. J.Pain 7, S3–S12.

D’Souza, C.A., Moscarello, M.A., 2006. DiVerences in susceptibility ofMBP charge isomers to digestion by stromelysin-1 (MMP-3) andrelease of an immunodominant epitope. Neurochem. Res..

Ferguson, T.A., Muir, D., 2000. MMP-2 and MMP-9 increase the neurite-promoting potential of schwann cell basal laminae and are upregulatedin degenerated nerve. Mol. Cell Neurosci. 16, 157–167.

Genersch, E., Hayess, K., Neuenfeld, Y., Haller, H., 2000. Sustained ERKphosphorylation is necessary but not suYcient for MMP-9 regulationin endothelial cells: involvement of Ras-dependent and -independentpathways. J. Cell Sci. 113 (Pt 23), 4319–4330.

Gijbels, K., Proost, P., Masure, S., Carton, H., Billiau, A., Opdenakker, G.,1993. Gelatinase B is present in the cerebrospinal Xuid during experi-mental autoimmune encephalomyelitis and cleaves myelin basic pro-tein. J. Neurosci. Res. 36, 432–440.

Gupta, S.K., Poduslo, J.F., Mezei, C., 1988. Temporal changes in PO andMBP gene expression after crush-injury of the adult peripheral nerve.Brain Res. 464, 133–141.

Gurer, G., Erdem, S., Kocaefe, C., Ozguc, M., Tan, E., 2004. Expression ofmatrix metalloproteinases in vasculitic neuropathy. Rheumatol. Int. 24,255–259.

Hsu, J.Y., McKeon, R., Goussev, S., Werb, Z., Lee, J.U., Trivedi, A., Noble-Haeusslein, L.J., 2006. Matrix metalloproteinase-2 facilitates woundhealing events that promote functional recovery after spinal cordinjury. J. Neurosci. 26, 9841–9850.

Hughes, P.M., Wells, G.M., Perry, V.H., Brown, M.C., Miller, K.M., 2002.Comparison of matrix metalloproteinase expression during Walleriandegeneration in the central and peripheral nervous systems. Neurosci-ence 113, 273–287.

Jacobs, E.C., 2005. Genetic alterations in the mouse myelin basic proteinsresult in a range of dysmyelinating disorders. J. Neurol. Sci. 228,195–197.

Kherif, S., Dehaupas, M., Lafuma, C., Fardeau, M., Alameddine, H.S.,1998. Matrix metalloproteinases MMP-2 and MMP-9 in dener-vated muscle and injured nerve. Neuropathol. Appl. Neurobiol. 24,309–319.

Kieseier, B.C., Clements, J.M., Gearing, A.J., Hartung, H.P., 1999a. Matrixmetalloproteinases exhibit diVerent expression patterns in inXamma-tory demyelinating diseases of the central and peripheral nervous sys-tem. Ann. N. Y. Acad. Sci. 878, 567–570.

Kieseier, B.C., Seifert, T., Giovannoni, G., Hartung, H.P., 1999b. Matrixmetalloproteinases in inXammatory demyelination: targets for treat-ment. Neurology 53, 20–25.

Kim, S.H., Chung, J.M., 1992. An experimental model for peripheral neu-ropathy produced by segmental spinal nerve ligation in the rat. Pain 50,355–363.

Kleinschnitz, C., BrinkhoV, J., Sommer, C., Stoll, G., 2005. Contralateralcytokine gene induction after peripheral nerve lesions: dependence onthe mode of injury and NMDA receptor signaling. Brain Res. Mol.Brain Res. 136, 23–28.

Komori, K., Nonaka, T., Okada, A., Kinoh, H., Hayashita-Kinoh, H.,Yoshida, N., Yana, I., Seiki, M., 2004. Absence of mechanical allo-dynia and Abeta-Wber sprouting after sciatic nerve injury in micelacking membrane-type 5 matrix metalloproteinase. FEBS Lett. 557,125–128.

La Fleur, M., Underwood, J.L., Rappolee, D.A., Werb, Z., 1996. Basementmembrane and repair of injury to peripheral nerve: deWning a potentialrole for macrophages, matrix metalloproteinases, and tissue inhibitorof metalloproteinases-1. J. Exp. Med. 184, 2311–2326.

Lancelotta, M.P., Sheth, R.N., Meyer, R.A., Belzberg, A.J., GriYn, J.W.,Campbell, J.N., 2003. Severity and duration of hyperalgesia in rat var-ies with type of nerve lesion. Neurosurgery 53, 1200–1208. discussion1208–1209.

Larsen, P.H., DaSilva, A.G., Conant, K., Yong, V.W., 2006. Myelin forma-tion during development of the CNS is delayed in matrix metallopro-teinase-9 and -12 null mice. J. Neurosci. 26, 2207–2214.

LeBlanc, A.C., Poduslo, J.F., 1990. Axonal modulation of myelin geneexpression in the peripheral nerve. J. Neurosci. Res. 26, 317–326.

Lee, S.R., Tsuji, K., Lee, S.R., Lo, E.H., 2004a. Role of matrix metallopro-teinases in delayed neuronal damage after transient global cerebralischemia. J. Neurosci. 24, 671–678.

Lee, S.R., Wang, X., Tsuji, K., Lo, E.H., 2004b. Extracellular proteolyticpathophysiology in the neurovascular unit after stroke. Neurol. Res.26, 854–861.

Leppert, D., Hughes, P., Huber, S., Erne, B., Grygar, C., Said, G.,Miller, K.M., Steck, A.J., Probst, A., Fuhr, P., 1999. Matrix metallo-proteinase upregulation in chronic inXammatory demyelinatingpolyneuropathy and nonsystemic vasculitic neuropathy. Neurology53, 62–70.

Liefner, M., Siebert, H., Sachse, T., Michel, U., Kollias, G., Bruck, W.,2000. The role of TNF-alpha during Wallerian degeneration. J. Neuro-immunol. 108, 147–152.

Livak, K.J., Schmittgen, T.D., 2001. Analysis of relative gene expressiondata using real-time quantitative PCR and the 2(-Delta Delta C(T))Method. Methods 25, 402–408.

Martini, R., Schachner, M., 1997. Molecular bases of myelin formation asrevealed by investigations on mice deWcient in glial cell surface mole-cules. Glia 19, 298–310.

Mawrin, C., Brunn, A., Rocken, C., Schroder, J.M., 2003. Peripheral neu-ropathy in systemic lupus erythematosus: pathomorphological featuresand distribution pattern of matrix metalloproteinases. Acta Neuropa-thol. (Berl.) 105, 365–372.

McMahon, S.B., CaVerty, W.B., Marchand, F., 2005. Immune and glialcell factors as pain mediators and modulators. Exp. Neurol. 192,444–462.

Muir, D., 1994. Metalloproteinase-dependent neurite outgrowth within asynthetic extracellular matrix is induced by nerve growth factor. Exp.Cell Res. 210, 243–252.

Myers, R.R., Campana, W.M., Shubayev, V.I., 2006. The role of neuroin-Xammation in neuropathic pain: mechanisms and therapeutic targets.Drug Discov Today 11, 8–20.

Nagase, H., 1997. Activation mechanisms of matrix metalloproteinases.Biol. Chem. 378, 151–160.

Paulson, P.E., Casey, K.L., Morrow, T.J., 2002. Long-term changes inbehavior and regional cerebral blood Xow associated with painfulperipheral mononeuropathy in the rat. Pain 95, 31–40.

PfaZ, M.W., 2001. A new mathematical model for relative quantiWcationin real-time RT-PCR. Nucleic Acids Res. 29, e45.

Platt, C.I., Krekoski, C.A., Ward, R.V., Edwards, D.R., Gavrilovic, J., 2003.Extracellular matrix and matrix metalloproteinases in sciatic nerve. J.Neurosci. Res. 74, 417–429.

Proost, P., Van Damme, J., Opdenakker, G., 1993. Leukocyte gelatinase Bcleavage releases encephalitogens from human myelin basic protein.Biochem. Biophys. Res. Commun. 192, 1175–1181.

Redford, E.J., Hall, S.M., Smith, K.J., 1995. Vascular changes and demye-lination induced by the intraneural injection of tumour necrosis factor.Brain 118 (Pt 4), 869–878.

Redford, E.J., Smith, K.J., Gregson, N.A., Davies, M., Hughes, P., Gearing,A.J., Miller, K., Hughes, R.A., 1997. A combined inhibitor of matrixmetalloproteinase activity and tumour necrosis factor-alpha process-ing attenuates experimental autoimmune neuritis. Brain 120 (Pt. 10),1895–1905.

Renaud, S., Erne, B., Fuhr, P., Said, G., Lacroix, C., Steck, A.J., Leppert,D., 2003. Matrix metalloproteinases-9 and -2 in secondary vasculiticneuropathies. Acta Neuropathol. (Berl.) 105, 37–42.

Rosenberg, G.A., 2002. Matrix metalloproteinases in neuroinXammation.Glia 39, 279–291.

Rosenberg, G.A., Estrada, E.Y., DencoV, J.E., Stetler-Stevenson, W.G.,1995. Tumor necrosis factor-alpha-induced gelatinase B causes delayedopening of the blood–brain barrier: an expanded therapeutic window.Brain Res. 703, 151–155.

568 S. Chattopadhyay et al. / Brain, Behavior, and Immunity 21 (2007) 561–568

Shubayev, V.I., Myers, R.R., 2000. Upregulation and interaction of TNF-alpha and gelatinases A and B in painful peripheral nerve injury. BrainRes. 855, 83–89.

Shubayev, V.I., Myers, R.R., 2002. Endoneurial remodeling by TNFalpha-and TNFalpha-releasing proteases. A spatial and temporal co-localiza-tion study in painful neuropathy. J. Peripher. Nerv. Syst. 7, 28–36.

Shubayev, V.I., Myers, R.R., 2004. Matrix metalloproteinase-9 promotesnerve growth factor-induced neurite elongation but not new sproutformation in vitro. J. Neurosci. Res. 77, 229–239.

Shubayev, V.I., Angert, M., Dolkas, J., Campana, W.M., Palenscar, K., Myers,R.R., 2006. TNFalpha-induced MMP-9 promotes macrophage recruit-ment into injured peripheral nerve. Mol. Cell Neurosci. 31, 407–415.

Siebert, H., Dippel, N., Mader, M., Weber, F., Bruck, W., 2001. Matrixmetalloproteinase expression and inhibition after sciatic nerve axot-omy. J. Neuropathol. Exp. Neurol. 60, 85–93.

Singer, C.F., Marbaix, E., Lemoine, P., Courtoy, P.J., Eeckhout, Y., 1999.Local cytokines induce diVerential expression of matrix metallopro-teinases but not their tissue inhibitors in human endometrial Wbro-blasts. Eur. J. Biochem. 259, 40–45.

Sommer, C., Schafers, M., 1998. Painful mononeuropathy in C57BL/Wldmice with delayed wallerian degeneration: diVerential eVects of cyto-kine production and nerve regeneration on thermal and mechanicalhypersensitivity. Brain Res. 784, 154–162.

Sommer, C., Schmidt, C., George, A., Toyka, K.V., 1997. A metallopro-tease-inhibitor reduces pain associated behavior in mice with experi-mental neuropathy. Neurosci. Lett. 237, 45–48.

Stoll, G., Jander, S., Myers, R.R., 2002. Degeneration and regeneration ofthe peripheral nervous system: from Augustus Waller’s observations toneuroinXammation. J. Peripher. Nerv. Syst. 7, 13–27.

Vecil, G.G., Larsen, P.H., Corley, S.M., Herx, L.M., Besson, A., Goodyer,C.G., Yong, V.W., 2000. Interleukin-1 is a key regulator of matrixmetalloproteinase-9 expression in human neurons in culture and fol-lowing mouse brain trauma in vivo. J. Neurosci. Res. 61, 212–224.

Wagner, R., Myers, R.R., 1996. Endoneurial injection of TNF-alpha pro-duces neuropathic pain behaviors. Neuroreport 7, 2897–2901.

Woessner Jr., J.F., 1994. The family of matrix metalloproteinases. Ann. N.Y. Acad. Sci. 732, 11–21.

Yong, V.W., 2005. Metalloproteinases: mediators of pathology and regen-eration in the CNS. Nat. Rev. Neurosci. 6, 931–944.

Yong, V.W., Krekoski, C.A., Forsyth, P.A., Bell, R., Edwards, D.R., 1998.Matrix metalloproteinases and diseases of the CNS. Trends Neurosci.21, 75–80.

Zhang, X., Chintala, S.K., 2004. InXuence of interleukin-1 beta inductionand mitogen-activated protein kinase phosphorylation on optic nerveligation-induced matrix metalloproteinase-9 activation in the retina.Exp. Eye Res. 78, 849–860.