Experimental Neurology 230 (2011) 273–279

Contents lists available at ScienceDirect

Experimental Neurology

j ourna l homepage: www.e lsev ie r.com/ locate /yexnr

NaV1.7 accumulates and co-localizes with phosphorylated ERK1/2 within transectedaxons in early experimental neuromas

Anna-Karin Persson, Andreas Gasser, Joel A. Black, Stephen G. Waxman ⁎Department of Neurology and Center for Neuroscience and Regeneration Research, Yale University School of Medicine, New Haven, CT 06510, USARehabilitation Research Center, Veterans Affairs Connecticut Healthcare System, West Haven, CT 06516, USA

⁎ Corresponding author at: Neuroscience and RegeConnecticut Healthcare System, 950 Campbell Avenue, BUSA. Fax: +1 203 937 3801.

E-mail addresses: Anna-Karin.Persson@yale.edu (A.-Andreas.Gasser@yale.edu (A. Gasser), Joel.Black@yale.edStephen.Waxman@yale.edu (S.G. Waxman).

0014-4886/$ – see front matter © 2011 Elsevier Inc. Aldoi:10.1016/j.expneurol.2011.05.005

a b s t r a c t

a r t i c l e i n f o

Article history:Received 9 December 2010Revised 4 April 2011Accepted 6 May 2011Available online 13 May 2011

Keywords:Neuropathic painNerve injuryNeuromaVoltage-gated sodium channelsNaV1.7Mitogen-activated protein kinasesERK1/2p38

Peripheral nerve injury can result in formation of a neuroma, which is often associated with heightenedsensitivity to normally innocuous stimuli as well as spontaneous dysesthesia and pain. The onset andpersistence of neuropathic pain have been linked to spontaneous ectopic electrogenesis in axons withinneuromas, suggesting an involvement of voltage-gated sodium channels. Sodium channel isoforms NaV1.3,NaV1.7 and NaV1.8 have been shown to accumulate in chronic painful human neuromas, while, to date, onlyNaV1.3 has been reported to accumulate within experimental neuromas. Although recent evidence stronglysupport a major contribution for NaV1.7 in nociception, the expression of NaV1.7 in injured axons within acuteneuromas has not been studied. The current study examinedwhether NaV1.7 accumulates in experimental ratneuromas. We further investigated whether activated (phosphorylated) mitogen-activated protein (MAP)kinase ERK1/2, which is known to modulate NaV1.7 properties, is co-localized with NaV1.7 within axons inneuromas. We demonstrate increased levels of NaV1.7 in experimental rat sciatic nerve neuromas, 2 weeksafter nerve ligation and transaction. We further show elevated levels of phosphorylated ERK1/2 withinindividual neuroma axons that exhibit NaV1.7 accumulation. These results extend previous descriptions ofsodium channel and MAP kinase accumulation within experimental and human neuromas, and suggest thattargeted blockade of NaV1.7 or ERK1/2 may provide a strategy for amelioration of chronic pain that oftenfollows nerve injury and formation of neuromas.

neration Research Center, VAldg. 34, West Haven, CT 06516,

K. Persson),u (J.A. Black),

l rights reserved.

© 2011 Elsevier Inc. All rights reserved.

Introduction

Injury to peripheral nerves often results in the formation ofneuromas, which are tangled masses of blind-ending axons, prolifer-ating connective tissue and invading immune cells (Cravioto andBattista, 1981; Fried, et al., 1991; Frisen, et al., 1993). Thedevelopment of neuromas can lead to painful dysesthesias andheightened sensitivity to normally innocuous stimuli (Sorkin andYaksh, 2009) that are often refractory to therapeutic intervention,resulting in chronic pain and a reduction in the quality of life(Nikolajsen, et al., 2010). While it has been shown that axons withinneuromas can be sensitized to external stimuli (Rivera, et al., 2000)and can exhibit spontaneous ectopic activity (Wall and Devor, 1983;Welk, et al., 1990), the molecular mechanisms responsible for theaxonal hypersensitivity thought to underlie neuropathic pain associ-ated with neuromas are not fully understood.

Peripheral and central mechanisms have been suggested to partici-pate in thegenesis andpersistenceof chronicpain inneuromas.However,particular attentionhasbeen focusedon the contributionof voltage-gatedsodium channels in the pathophysiological changes leading to neuro-pathic pain. Nine sodium channel isoforms (NaV1.1–NaV1.9) have beencloned, each exhibiting distinct expression patterns, pharmacologicalsensitivities, and electrophysiological properties (Catterall, et al., 2005).The ensemble of channels expressed largely establishes the electro-responsiveness of neurons (Chahine, et al., 2005; Rush, et al., 2007). Theexpression of sodium channels within neurons is not invariable, but onthe contrary is plastic, with injury to nerves associated with up- anddown-regulation of specific sodium channel isoforms (Baker and Wood,2001; Dib-Hajj, et al., 2009; Waxman, et al., 2002).

Multiple sodium channels have been suggested to contribute tothe pathogenesis of pain with distinct mechanisms (Dib-Hajj andWaxman, 2010). For instance, NaV1.7, which is expressed in ~70% offunctionally-identified nociceptive neurons in dorsal root ganglia(DRG; (Djouhri, et al., 2003)), produces a depolarizing response tosmall, slow stimuli, such as generator potentials (Dib-Hajj, et al.,2007). Thus, NaV1.7 plays a critical role in setting the gain on theresponsiveness of nociceptors (Waxman, 2006). A major contributionfor NaV1.7 in nociception is supported by recent studies demonstrat-ing that gain-of-functionmutations of NaV1.7 can produce severe pain

274 A.-K. Persson et al. / Experimental Neurology 230 (2011) 273–279

(Dib-Hajj, et al., 2007; Waxman and Dib-Hajj, 2005), while loss-of-function mutations of NaV1.7 produce insensitivity to pain (Ahmad, etal., 2007; Cox, et al., 2006).

Functional properties of sodium channels can be directly modulatedby interactionswithpartner/accessorymolecules (Dib-Hajj andWaxman,2010;Patinoand Isom,2010) andbykinases (Cantrell andCatterall, 2001;Chahine, et al., 2005; Hudmon, et al., 2008; Stamboulian, et al., 2010).Phosphorylation of sodium channels by kinases can provide rapid post-translational modification of channel properties, often in an isoform-specific manner. While most work has focused on the actions of proteinkinase A (PKA) and protein kinase C (PKC) on sodium currents (Cantrelland Catterall, 2001), recent work has established a role for mitogen-activated protein (MAP) kinases in modulating functional properties ofsodium channels. MAP kinases, which are actively involved in theregulation of multiple cellular functions (Seger and Krebs, 1995) havebeen shown to directly phosphorylate specific sodium channel isoformsand alter their electrophysiological properties. ERK1/2 MAP kinase hasbeen shown to regulate the firing properties of DRG neurons by shiftingNaV1.7 activation and inactivation, thereby modulating neuronal excit-ability (Stamboulian, et al., 2010). MAP kinase p38 is also a knownmodulator of sodium current density by direct interaction with differentsodium channel isoforms (Hudmon, et al., 2008; Wittmack, et al., 2005).Elevated levels of activated ERK1/2 and p38 were previously demon-strated inpainfulhumanneuromas, suggesting apossible role inneuromapain (Black, et al., 2008).

Previous studies have demonstrated an accumulation of NaV1.7 inpainful human neuromas (Black, et al., 2008; Kretschmer, et al., 2002),suggesting a contribution to neuroma hyperexcitability and pain. Onestudy however, concluded that the expression levels of NaV1.7were notdirectly related to the level of neuroma pain (Bird, et al., 2007). In thisrespect, increased levels of the channel may not be the sole contributorto hyperexcitability within neuromas, since modulation of channelproperties may alter its contribution to electrogenesis. Since there isevidence demonstrating that phosphorylation ofNaV1.7 byERK1/2MAPkinases can alter functional responses, and therebymodulate nociceptorexcitability (Hudmon, et al., 2008; Stamboulian, et al., 2010), we askedwhether NaV1.7 and phosphorylated ERK1/2 are co-expressed withinearly experimental neuromas and, if so, whether NaV1.7 and theseMAPkinases are co-localized within individual axons.

Methods

Animals

Adult male Sprague–Dawley rats (250–300 g, Harlan, Indianapolis,IN) were housed under a 12 h light/dark cycle in a pathogen-free areawith ad libitum access to water and food. The experimental procedureswere approved by the VA Connecticut Healthcare System InstitutionalAnimal Care andUse Committee, in accordancewithNIH guidelines andguidelines of the Committee for Research and Ethical Issues of IASP.

Surgical procedures

Under general anesthesia (Ketamine/Xylazine, 75/5 mg/kg, i.p.), theleft sciatic nerve was exposed at mid-thigh, double-ligated with 4-0 nylon sutures and cut approximately 3 mm distal of the ligation.Approximately 2 mmof the distal nerve stumpwas excised tominimizeregeneration of the transected fibers into the distal nerve stump. Thewound was closed in layers and the rats returned to their home cage.

Primary antibodies

A polyclonal antibody against sodium channel isoform NaV1.7(rabbit; Y083, (Black, et al., 2008)) was used. Staining patterns of PNSand CNS tissue sections were used to validate the antibody, showingexpected labeling pattern in DRG and a lack of labeling in CNS tissue.

MAP kinases were detected using antibodies directed against the total(phosphorylated and non-phosphorylated form) ERK1/2 (TERK1/2,mouse; Invitrogen) as well as the phosphorylated forms of p38 (Pp38,rabbit; Cell Signaling, T180/Y182, #4511S XP) and ERK1/2 (PERK1/2,rabbit, Cell Signaling, Thr202/Tyr204, #9101S; mouse, Abcam,Cambridge, MA). Pp38 and PERK1/2 antibodies were validated underconditions previously shown to activate these MAP Kinases (Hudmon,et al., 2008; Stamboulian, et al., 2010): anisomycin-treated microgliaand NGF-treated DRG neurons exhibited increased immunoreactivityfor Pp38 and PERK1/2, respectively (data not shown). Anti-peripherin(chicken; Aves Labs Inc., Tigard, OR) and anti-neurofilament (NF) (NF-L, NF-M, NF-H, Encore, Gainsville, FL) were used as neuronal markersand CD68 (mouse, Serotec, Raleigh, NC) as a marker for macrophages.

Western blot

Rats underwent CO2 narcosis and decapitation. Neuromas andcontralateral nerves were dissected and rapidly homogenized in20 mM Tris, pH 7.5, 150 mM NaCl, 1 mM EDTA, 1% Triton X-100,Protease inhibitor minus EDTA (Roche Diagnostics GmbH, Mannheim,Germany) and phosphatase inhibitor (RocheDiagnostics). Proteinswereseparated on NUPAGE Bis–Tris 4–12% gels (Invitrogen) using MES asrunning buffer (Invitrogen). Proteins were transferred to nitrocellulosemembranes for 2 h at 30 V using the XCell II system (Invitrogen) andblocked for 1 h at room temperature in TBST (Tris-buffered salineplus 0.1% Tween 20)/10% nonfat milk buffer. Membranes were incuba-ted over night at 4 °C in primary antibodies anti-TERK1/2 (1:1000) andanti-PERK1/2 (#9101S, 1:1000) and for 1 h at room temperature insecondary antibodies goat anti-mouse HRP conjugate (1:10000, Dako,Glostrup, Denmark) or goat anti-rabbit HRP conjugate (1:10000, Dako).Membranes were washed in TBST for 3×10 min after primary andsecondary antibody incubations. Chemiluminescent detection wasperformed using Lightning Plus-ECL (PerkinElmer Life and AnalyticalSciences) and detected by X-ray film exposure (BioMax XAR, Kodak,Rochester, NY).

Immunohistochemistry

Rats were deeply anesthetized with Ketamine/Xylazine (75/5 mg/kgb.w., i.p.) and transcardially perfused with phosphate buffered saline(PBS) followed by a 4% paraformaldehyde solution in 0.14 M Sorensen'sphosphate buffer. Neuromas and contralateral sciatic nerves weredissected and cryoprotected with 30% sucrose in 0.01 M PBS overnightat 4 °C, cryostat sectioned (12 μm) and mounted on glass slides (FischerSuperFrost Plus) prior to staining. For individual rats, neuroma andcontralateral (un-injured) sciatic nerve were mounted and processed onthe same slide. Tissue from all animals was processed simultaneously aspreviously described (Fjell, et al., 2000). Briefly, sections were blockedwith PBS containing 5% fish skin gelatin (Sigma), 6% normal donkeyserum, 0.6% Triton X-100, and 0.02% sodium azide for 1 h at roomtemperature. Subsequently, slides were incubated individually or incombination with primary antibodies NaV1.7 (1:200), Pp38 (1:800),mouse PERK1/2 (1:100), chicken peripherin (1:100), NF (NF-L+NF-M+NF-H, all 1:1000), CD68 (1:100) 1–2 days at 4 °C, washed with PBS andthen secondary antibodies donkey anti-mouse IgG-488 Dylight (JacksonImmunoResearch, West Grove, PA), donkey anti-rabbit IgG-Cy3 (JacksonImmunoReasearch) and donkey anti-chicken IgG-Cy5 (Millipore Biosci-ence Research Reagents, Bedford, MA) 6 h at room temperature.

Image acquisition and immunolabeling quantification

Images were captured using a Nikon C1si confocal microscope(Nikon USA, Melville, NY) and EZ-C1 software (Nikon USA). Under thesame settings for control and neuroma tissues, random images wereaccrued (6–10 images/rat) from regions containing nerve fibers(peripherin- or NF-immunoreactive). For neuromas, images were

275A.-K. Persson et al. / Experimental Neurology 230 (2011) 273–279

taken within the region 200–1000 μm proximal to the ligation, i.e.within the most distal region within the neuroma where single axonscould be discriminated (close to the ligation, the density of the axonaltangle prevented single axons from being discerned). Image analysiswas performed using MetaMorph 7.0 (Molecular Devices, Downing-town, PA). To ensure signal quantification over axons, and not non-neuronal cells, each image was initially thresholded based onperipherin/NF immunosignal. A detection mask with region-of-interest objects was created, and within each region of interest(profile of axon), the percentage of positive pixels (above a thresholdthat excluded background) for the sodium channel or MAP kinase wasmeasured. Quantification was carried out in a blinded manner. Datafrom individual regions of interest were then averaged for each imageand multiple images averaged for each animal.

Statistics

Data are presented as mean±SEM. Student's unpaired t-test(Excel, Microsoft) was used for statistical analyses, with the criterionfor statistical significance set at *pb0.05 or **pb0.01.

Fig. 1. Sodium channel NaV1.7 accumulates in neuromas.A) Top panel. Control- and neuroma seis displayed throughout the neuroma adjacent to the ligature (~1000–1500 μm) compared toIncreased magnification of control and neuroma sections demonstrates NaV1.7 (red) imlocalization (yellow) of NaV1.7 and peripherin. Scale bar: 50 μm. B) Quantification of NaV1pixels within peripherin-labeled nerve fibers was calculated; data are presented as mean±

Results

Sodium channel NaV1.7 accumulates in acute rat sciatic nerve neuromas

Increased sodium channel expression in neuromas has beensuggested to underlie ectopic firing and contribute to chronic painthat has been observed (Nystrom and Hagbarth, 1981;Wall and Devor,1983) in neuromas. Accumulation of NaV1.7 sodium channels has beenpreviously reported in painful human neuromas, in which painpersisted for 19–192 months (Black, et al., 2008; Nikolajsen, et al.,2010). However, whether NaV1.7 accumulates within neuromas ofshorter duration of formation has yet to be established. As shown inFig. 1A, 14 days following mid-thigh transection and ligation of the ratsciatic nerve, increased levels of NaV1.7 immunolabelingwere exhibitedthroughout the extent of the neuroma (extending ~1500 μm proximalto the ligature) compared to more proximal regions of the transectednerve and to the contralateral nerve. At increased magnification,double-immunolabeling of the neuroma for peripherin and NaV1.7demonstrates that NaV1.7 is widely expressed in small unmyelinatedsensory nerve fibers.

ctions immunostained for NaV1.7 (scale bar 250 μm). Increased NaV1.7 immunolabelingmore proximal regions of the nerve and to control (contralateral) sections. Lower panel.munostaining in peripherin-labeled nerve fibers (green). Merged images show co-.7 accumulation in control and neuroma tissue. Percentage of NaV1.7 immunopositiveSEM, n=4, *pb0.05; unpaired Student's t-test.

276 A.-K. Persson et al. / Experimental Neurology 230 (2011) 273–279

Quantification of NaV1.7 within control and neuroma tissue isshown in Fig. 1B. Neuromas exhibited a ~4-fold increase in the area ofperipherin-positive axons that express NaV1.7 compared to controlfibers (pb0.05; n=4, Fig. 1B).

Neuromas exhibit elevated levels of active ERK1/2 that co-localizes withNaV1.7 within individual axons

It has been shown that activated ERK1/2 (PERK1/2) can phosphor-ylate NaV1.7, thereby modulating the gating properties of the channeland increasing the excitability of sensory neurons (Stamboulian, et al.,2010). In initial studies, we utilized Western blot methods todetermine the levels of total ERK1/2 (TERK1/2) and of phosphory-

Fig. 2. Levels of activated ERK1/2 are increased within neuromas. A) Left panel. Western blobetween control and neuroma tissue but substantially increased levels of activated ERK1Western blots of PERK1/2 and TERK1/2 levels from neuroma and control tissue. Data areB) Neuroma sections exhibit increased levels of activated ERK1/2 (green) in peripherin/Nmagnification image of peripherin/NF positive (red) neuroma axons co-labeled with ERK1(yellow) within individual axons. Scale bar, 50 μm (upper and middle panel), 10 μm (lowe

lated ERK1/2 (PERK1/2) within control and neuroma tissues. Asshown in Fig. 2A, control and neuroma tissues exhibited similar levelsof TERK1/2. However, levels of PERK1/2 were substantially increasedwithin neuroma tissues compared to control nerves, similar toprevious studies (Sheu, et al., 2000). Quantification of immunoblot-signal intensities for PERK1/2 showed a ~9-fold increase in the ratioPERK1/2/TERK1/2 in neuromas compared to control nerve (pb0.01;n=3, Fig. 2A).

To determine whether activated ERK1/2 was present in axonswithin neuromas, we performed double-immunolabeling for PERK1/2and axonal markers (neurofilament and peripherin) on neuromaand control tissue sections. As shown in Fig. 2B, minimal PERK1/2labeling was exhibited by axons within control nerves. In contrast,

t analysis demonstrates no change in the levels of total ERK1/2 (TERK1/2) (top panel)/2 (PERK1/2) in neuromas compared to control nerves. Right panel. Quantification ofnormalized and presented as mean±SEM, n=3, **pb0.01; unpaired Student's t-test.F-positive nerve fibers (red) compared to control nerve. Lower panel displays a high/2 (green). Merged images demonstrate co-localization of PERK1/2 and peripherin/NFr panel).

277A.-K. Persson et al. / Experimental Neurology 230 (2011) 273–279

substantially increased PERK1/2 immunolabeling was displayed byaxons, identified by staining for peripherin and neurofilament, withinneuromas (Fig. 2B). PERK1/2 immunolabeling was also detected innon-neuronal cells, likely reflecting PERK1/2 in Schwann cells thathave been previously demonstrated in injured rat sciatic nerves (Seo,et al., 2009).

Since levels of PERK1/2 are increased within neuromas andPERK1/2 can modulate NaV1.7 gating properties, we determinedwhether activated ERK1/2 displays increased expression in NaV1.7accumulating axons. Double-immunolabeling studies demonstrateco-localization of enhanced levels of PERK1/2 and the accumulation ofNaV1.7 within neuromas (Upper and middle panel, Fig. 3). BothPERK1/2 and NaV1.7 display similar patterns of immunolabeling in theregion adjacent to the ligature. The lower panel in Fig. 3 shows that, atincreased magnification, individual axons exhibit co-localization ofPERK1/2 and NaV1.7.

Active (phosphorylated) p38 levels are not increased within axons inneuromas

To determine whether the MAP kinase activated p38 (Pp38),which has been shown to modulate functional properties of sodiumchannels (Hudmon, et al., 2008; Wittmack, et al., 2005), is alsoaccumulated in acute experimental neuromas, we reacted control andneuroma tissue sections with an antibody to Pp38. As exemplified inFig. 4, the level of Pp38-immunoreactivity was not increased in axonswithin the neuromas (n=4) compared to control tissue. Strong Pp38immunosignal was exhibited, however, in infiltrating macrophagesidentified by CD68-labeling (Fig. 4, inset).

Discussion

Our results demonstrate, for the first time, the accumulation ofsodium channel NaV1.7 within early experimental neuromas. Wefurther show elevated levels of phosphorylated MAP kinase ERK1/2within individual neuroma axons that exhibit NaV1.7 immunolabel-ing. This observation may be functionally important, since activatedERK1/2 has been shown to phosphorylate NaV1.7 and enhance itsexcitability (Stamboulian, et al., 2010), and NaV1.7 has beenshown to play a major role as a driver of electrogenesis in DRGneurons (Dib-Hajj, et al., 2007; Rush, et al., 2007). The present resultsextend previous descriptions of sodium channel accumulation withinexperimental and human neuromas, and suggest that targeted block

Fig. 3. Activated ERK1/2 and NaV1.7 co-localize within axons in neuromas. Sections from coPERK1/2 (green) and NaV1.7 (red). Merged images demonstrate co-localization (yellow). Upsections. Scale bar, 500 μm. Lower panel. Increased magnification image demonstrates co-lobar, 10 μm.

of NaV1.7 or ERK1/2 may provide avenues for amelioration of chronicpain that often follows nerve injury and formation of neuromas.

Neuromas result from injury to nerves caused by trauma,transection and compression, and are composed of disorganizedmasses of blind-ending axons and infiltrating immune cells encapsu-lated in proliferating connective tissue (Fried, et al., 1991; Frisen, etal., 1993). In commonwith other forms of neuropathic pain, neuromascan be associated with hypersensitivity to mechanical and chemicalstimuli (Rivera, et al., 2000) and with debilitating spontaneouschronic pain that is often refractory to therapeutic intervention(Nikolajsen, et al., 2010). Both experimental and human neuromasexhibit spontaneous ectopic electrogenesis (Burchiel, 1984; Devorand Govrin-Lippmann, 1983; Wall and Gutnick, 1974), which hasbeen linked to neuropathic paresthesias and pain (Devor, et al., 1993;Nordin, et al., 1984; Nystrom and Hagbarth, 1981; Ochoa andTorebjork, 1980). In the search for underlying mechanisms contrib-uting to ectopic impulse generation in neuromas, particular attentionhas focused on voltage-gated sodium channels. Sodium channels areresponsible for action potential electrogenesis and conduction innormal nerves, and thus are likely to contribute to ectopic firing.Consistent with a strong contribution of sodium channels to ectopicdischarge within neuromas, sodium channel blockers includinglidocaine, phenytoin, carbamazepine and tetrodotoxin ameliorateectopic activity within neuromas (Burchiel, 1988; Devor, et al., 1992;Omana-Zapata, et al., 1997; Yaari and Devor, 1985).

Nine isoforms of sodium channels have been identified, each withdistinct voltage-dependence, kinetic and pharmacological properties(see (Catterall, et al., 2005). In normal adult rats, five sodium channelisoforms (NaV1.1, NaV1.6, NaV1.7, NaV1.8 and NaV1.9) are expressed inneurons within dorsal root ganglion (DRG) neurons, includingnociceptive neurons (Dib-Hajj and Waxman, 2010), while NaV1.3 isup-regulated in a subpopulation of these neurons following injury toperipheral nerves (Black, et al., 1999; Lindia and Abbadie, 2003;Waxman, et al., 1994). Previous studies have demonstrated accumu-lation of sodium channels within experimental and human neuromas.Initial descriptions utilized antibodies that did not distinguishdiffering sodium channel isoforms (Devor, et al., 1993; England, etal., 1996). However, subsequent studies have shown accumulations ofNaV1.3, NaV1.7 and NaV1.8 in human neuromas (Bird, et al., 2007;Black, et al., 2008; Kretschmer, et al., 2002) and NaV1.3 inexperimental neuromas (Black, et al., 1999; Shah, et al., 2004).

The accumulation of NaV1.7 in axons reported here has importantimplications for the genesis and persistence of chronic pain withinneuromas. NaV1.7 is expressed in ~70% of functionally-identified

ntrol (upper panel) and neuroma (middle and lower panel) were doubled-labeled forper and middle panel. Montage of low magnification images from control and neuromacalization (yellow) of PERK1/2 (green) and NaV1.7 (red) within individual axons. Scale

Fig. 4. Phosphorylated p38 is not increased in axons within neuromas. Neuroma sections do not display activated p38 (Pp38, red) in peripherin-immunopositive nerve fibers (blue).However, macrophages within neuromas exhibit robust phosphorylated p38 immunolabeling (red; e.g. arrow). Inset: CD68-positive macrophage (green) displays activated p38labeling (red). Scale bar, 50 μm.

278 A.-K. Persson et al. / Experimental Neurology 230 (2011) 273–279

nociceptive DRG neurons (Djouhri, et al., 2003) where its biophysicalproperties poise it to amplify generator potentials (Cummins, et al.,1998; Herzog, et al., 2003; Klugbauer, et al., 1995). NaV1.7 is a fast-activating, fast-inactivating and slow-repriming (recovery frominactivation) channel, that produces substantial ramp currents inresponse to small, slow depolarizations (Cummins, et al., 1998;Herzog, et al., 2003; Klugbauer, et al., 1995). Thereby, NaV1.7 can actas a threshold channel for firing action potentials (Rush, et al., 2007)and thus sets the gain for nociceptive neurons (Waxman, 2007). A keyrole for NaV1.7 in pain signaling is supported by recent studiesdemonstrating that two severe pain conditions, inherited erythrome-lalgia (IEM) and paroxysmal extreme pain syndrome, are produced bydominant gain-of-function mutations in SCN9A, the gene encodingNaV1.7 (Dib-Hajj, et al., 2007; Drenth and Waxman, 2007). Inaddition, recessive loss-of-function mutations in NaV1.7 are linkedto congenital insensitivity to pain (Ahmad, et al., 2007; Cox, et al.,2006). These observations support the suggestion that accumulation/modulation of NaV1.7 sodium channels in axons within neuromas cancontribute to spontaneous ectopic firing of these fibers, leading topain.

MAP kinases transduce extracellular signals into cellular re-sponses, including post-translational and transcriptional modifica-tions (Chang and Karin, 2001; Seger and Krebs, 1995). The MAPkinases p38, ERK1/2 and c-Jun N-terminal kinase (JNK) have receivedconsiderable attention for their contribution to the onset andmaintenance of neuropathic pain (Ji, 2004; Ji, et al., 2009; Obata andNoguchi, 2004). While much of this work has focused on differentialactivation of MAP kinases in microglia and astrocytes, and a resultingrelease of proinflammatory and pronociceptive mediators by theseglia that can modulate neuronal responsiveness (Ji and Suter, 2007;Scholz and Woolf, 2007), recent work has demonstrated that directphosphorylation of sodium channels byMAP kinases can directly alterthe activities of these channels. Thus, activation of MAP kinases mayinitiate dual converging pathways to affect neuronal excitability—indirectly through the release of bioactive molecules by glia anddirectly by phosphorylation of sodium channels.

The NaV1.8 current density in DRG neurons is significantlyincreased following phosphorylation by activated p38, withoutaffecting the gating properties of the channel (Hudmon, et al.,2008). To date, however, a modulatory role for Pp38 on NaV1.7channels has not been studied. In the present study, phosphorylatedp38 was detected at high levels within infiltrating macrophages, butnot in axons within neuromas, thus eliminating p38 as a modulator ofNaV1.7 activity in axons within neuromas. In contrast, we demon-strated significantly increased levels of activated ERK1/2 within axonsin neuromas compared to control nerves, and phosphorylated ERK1/2was co-localized with NaV1.7 in individual axons. ERK1/2 has recentlybeen shown to phosphorylate NaV1.7 and to modulate the excitabilityof the channel by causing a hyperpolarizing shift in activation andfast-inactivation (Stamboulian, et al., 2010). In the context of ectopic

firing within neuromas, the co-localization of NaV1.7 and ERK1/2within individual axons would be expected to facilitate channelopening in response to weak stimuli, and thus would supportspontaneous firing and hypersensitivity of the neuroma.

Conclusions

The present results demonstrate enhanced levels of NaV1.7 andphosphorylated ERK1/2 in individual axons of acute experimental ratneuromas. These observations extend previous studies and identifyERK1/2 and NaV1.7 as targets for therapeutic intervention in theattenuation of chronic pain associated with neuromas.

Acknowledgments

Acknowledgements: A-KP was supported by a fellowship from theSwedish Research Council (K2010-78PK-21636-01-2). This work wassupported in part by grants from the Rehabilitation Research Serviceand Medical Research Service, Department of Veterans Affairs. TheCenter for Neuroscience and Regeneration Research is a Collaborationof the Paralyzed Veterans of America and the United SpinalAssociation with Yale University.

References

Ahmad, S., Dahllund, L., Eriksson, A.B., Hellgren, D., Karlsson, U., Lund, P.E., Meijer, I.A.,Meury, L., Mills, T., Moody, A., Morinville, A., Morten, J., O'Donnell, D., Raynoschek,C., Salter, H., Rouleau, G.A., Krupp, J.J., 2007. A stop codon mutation in SCN9A causeslack of pain sensation. Hum. Mol. Genet. 16, 2114–2121.

Baker, M.D., Wood, J.N., 2001. Involvement of Na+ channels in pain pathways. TrendsPharmacol. Sci. 22, 27–31.

Bird, E.V., Robinson, P.P., Boissonade, F.M., 2007. Na(v)1.7 sodium channel expressionin human lingual nerve neuromas. Arch. Oral Biol. 52, 494–502.

Black, J.A., Cummins, T.R., Plumpton, C., Chen, Y.H., Hormuzdiar, W., Clare, J.J., Waxman,S.G., 1999. Upregulation of a silent sodium channel after peripheral, but not central,nerve injury in DRG neurons. J. Neurophysiol. 82, 2776–2785.

Black, J.A., Nikolajsen, L., Kroner, K., Jensen, T.S., Waxman, S.G., 2008. Multiple sodiumchannel isoforms and mitogen-activated protein kinases are present in painfulhuman neuromas. Ann. Neurol. 64, 644–653.

Burchiel, K.J., 1984. Effects of electrical and mechanical stimulation on two foci ofspontaneous activity which develop in primary afferent neurons after peripheralaxotomy. Pain 18, 249–265.

Burchiel, K.J., 1988. Carbamazepine inhibits spontaneous activity in experimentalneuromas. Exp. Neurol. 102, 249–253.

Cantrell, A.R., Catterall, W.A., 2001. Neuromodulation of Na+ channels: an unexpectedform of cellular plasticity. Nat. Rev. Neurosci. 2, 397–407.

Catterall, W.A., Goldin, A.L., Waxman, S.G., 2005. International Union of Pharmacology.XLVII. Nomenclature and structure–function relationships of voltage-gated sodiumchannels. Pharmacol. Rev. 57, 397–409.

Chahine, M., Ziane, R., Vijayaragavan, K., Okamura, Y., 2005. Regulation of Na v channelsin sensory neurons. Trends Pharmacol. Sci. 26, 496–502.

Chang, L., Karin, M., 2001. Mammalian MAP kinase signalling cascades. Nature 410,37–40.

Cox, J.J., Reimann, F., Nicholas, A.K., Thornton, G., Roberts, E., Springell, K., Karbani, G.,Jafri, H., Mannan, J., Raashid, Y., Al-Gazali, L., Hamamy, H., Valente, E.M., Gorman, S.,Williams, R., McHale, D.P., Wood, J.N., Gribble, F.M., Woods, C.G., 2006. An SCN9Achannelopathy causes congenital inability to experience pain. Nature 444,894–898.

279A.-K. Persson et al. / Experimental Neurology 230 (2011) 273–279

Cravioto, H., Battista, A., 1981. Clinical and ultrastructural study of painful neuroma.Neurosurgery 8, 181–190.

Cummins, T.R., Howe, J.R., Waxman, S.G., 1998. Slow closed-state inactivation: a novelmechanism underlying ramp currents in cells expressing the hNE/PN1 sodiumchannel. J. Neurosci. 18, 9607–9619.

Devor, M., Govrin-Lippmann, R., 1983. Axoplasmic transport block reduces ectopicimpulse generation in injured peripheral nerves. Pain 16, 73–85.

Devor, M., Wall, P.D., Catalan, N., 1992. Systemic lidocaine silences ectopic neuroma andDRG discharge without blocking nerve conduction. Pain 48, 261–268.

Devor, M., Govrin-Lippmann, R., Angelides, K., 1993. Na+ channel immunolocalizationin peripheral mammalian axons and changes following nerve injury and neuromaformation. J. Neurosci. 13, 1976–1992.

Dib-Hajj, S.D., Waxman, S.G., 2010. Isoform-specific and pan-channel partners regulatetrafficking and plasma membrane stability; and alter sodium channel gatingproperties. Neurosci. Lett. 486, 84–91.

Dib-Hajj, S.D., Cummins, T.R., Black, J.A., Waxman, S.G., 2007. From genes to pain: Na v1.7 and human pain disorders. Trends Neurosci. 30, 555–563.

Dib-Hajj, S.D., Black, J.A., Waxman, S.G., 2009. Voltage-gated sodium channels:therapeutic targets for pain. Pain Med. 10, 1260–1269.

Djouhri, L., Newton, R., Levinson, S.R., Berry, C.M., Carruthers, B., Lawson, S.N., 2003.Sensory and electrophysiological properties of guinea-pig sensory neuronesexpressing Nav 1.7 (PN1) Na+ channel alpha subunit protein. J. Physiol. 546,565–576.

Drenth, J.P., Waxman, S.G., 2007. Mutations in sodium-channel gene SCN9A cause aspectrum of human genetic pain disorders. J. Clin. Invest. 117, 3603–3609.

England, J.D., Happel, L.T., Kline, D.G., Gamboni, F., Thouron, C.L., Liu, Z.P., Levinson, S.R.,1996. Sodium channel accumulation in humans with painful neuromas. Neurology47, 272–276.

Fjell, J., Hjelmstrom, P., Hormuzdiar, W., Milenkovic, M., Aglieco, F., Tyrrell, L., Dib-Hajj,S., Waxman, S.G., Black, J.A., 2000. Localization of the tetrodotoxin-resistant sodiumchannel NaN in nociceptors. Neuroreport 11, 199–202.

Fried, K., Govrin-Lippmann, R., Rosenthal, F., Ellisman, M.H., Devor, M., 1991.Ultrastructure of afferent axon endings in a neuroma. J. Neurocytol. 20, 682–701.

Frisen, J., Risling, M., Fried, K., 1993. Distribution and axonal relations of macrophages ina neuroma. Neuroscience 55, 1003–1013.

Herzog, R.I., Cummins, T.R., Ghassemi, F., Dib-Hajj, S.D., Waxman, S.G., 2003. Distinctrepriming and closed-state inactivation kinetics of Nav1.6 and Nav1.7 sodiumchannels in mouse spinal sensory neurons. J. Physiol. 551, 741–750.

Hudmon, A., Choi, J.S., Tyrrell, L., Black, J.A., Rush, A.M., Waxman, S.G., Dib-Hajj, S.D.,2008. Phosphorylation of sodium channel Na(v)1.8 by p38 mitogen-activatedprotein kinase increases current density in dorsal root ganglion neurons. J.Neurosci. 28, 3190–3201.

Ji, R.R., 2004. Mitogen-activated protein kinases as potential targets for pain killers.Curr. Opin. Investig. Drugs 5, 71–75.

Ji, R.R., Suter, M.R., 2007. p38 MAPK, microglial signaling, and neuropathic pain. Mol.Pain 3, 33.

Ji, R.R., Gereau, R.W.T., Malcangio, M., Strichartz, G.R., 2009. MAP kinase and pain. BrainRes. Rev. 60, 135–148.

Klugbauer, N., Lacinova, L., Flockerzi, V., Hofmann, F., 1995. Structure and functionalexpression of a new member of the tetrodotoxin-sensitive voltage-activatedsodium channel family from human neuroendocrine cells. EMBO J. 14, 1084–1090.

Kretschmer, T., Happel, L.T., England, J.D., Nguyen, D.H., Tiel, R.L., Beuerman, R.W., Kline,D.G., 2002. Accumulation of PN1 and PN3 sodium channels in painful humanneuroma-evidence from immunocytochemistry. Acta Neurochir. (Wien) 144,803–810 discussion 810.

Lindia, J.A., Abbadie, C., 2003. Distribution of the voltage gated sodiumchannelNa(v)1.3-likeimmunoreactivity in the adult rat central nervous system. Brain Res. 960, 132–141.

Nikolajsen, L., Black, J.A., Kroner, K., Jensen, T.S., Waxman, S.G., 2010. Neuroma removalfor neuropathic pain: efficacy and predictive value of lidocaine infusion. Clin. J. Pain26, 788–793.

Nordin, M., Nystrom, B., Wallin, U., Hagbarth, K.E., 1984. Ectopic sensory discharges andparesthesiae in patients with disorders of peripheral nerves, dorsal roots and dorsalcolumns. Pain 20, 231–245.

Nystrom, B., Hagbarth, K.E., 1981. Microelectrode recordings from transected nerves inamputees with phantom limb pain. Neurosci. Lett. 27, 211–216.

Obata, K., Noguchi, K., 2004. MAPK activation in nociceptive neurons and painhypersensitivity. Life Sci. 74, 2643–2653.

Ochoa, J.L., Torebjork, H.E., 1980. Paraesthesiae from ectopic impulse generation inhuman sensory nerves. Brain 103, 835–853.

Omana-Zapata, I., Khabbaz, M.A., Hunter, J.C., Clarke, D.E., Bley, K.R., 1997. Tetrodotoxininhibits neuropathic ectopic activity in neuromas, dorsal root ganglia and dorsalhorn neurons. Pain 72, 41–49.

Patino, G.A., Isom, L.L., 2010. Electrophysiology and beyond: multiple roles of Na+channel beta subunits in development and disease. Neurosci. Lett. 486, 53–59.

Rivera, L., Gallar, J., Pozo, M.A., Belmonte, C., 2000. Responses of nerve fibres of the ratsaphenous nerve neuroma to mechanical and chemical stimulation: an in vitrostudy. J. Physiol. 527 (Pt 2), 305–313.

Rush, A.M., Cummins, T.R., Waxman, S.G., 2007. Multiple sodium channels and theirroles in electrogenesis within dorsal root ganglion neurons. J. Physiol. 579, 1–14.

Scholz, J., Woolf, C.J., 2007. The neuropathic pain triad: neurons, immune cells and glia.Nat. Neurosci. 10, 1361–1368.

Seger, R., Krebs, E.G., 1995. The MAPK signaling cascade. FASEB J. 9, 726–735.Seo, T.B., Oh, M.J., You, B.G., Kwon, K.B., Chang, I.A., Yoon, J.H., Lee, C.Y., Namgung, U.,

2009. ERK1/2-mediated Schwann cell proliferation in the regenerating sciaticnerve by treadmill training. J. Neurotrauma 26, 1733–1744.

Shah, B.S., Rush, A.M., Liu, S., Tyrrell, L., Black, J.A., Dib-Hajj, S.D., Waxman, S.G., 2004.Contactin associates with sodium channel Nav1.3 in native tissues and increaseschannel density at the cell surface. J. Neurosci. 24, 7387–7399.

Sheu, J.Y., Kulhanek, D.J., Eckenstein, F.P., 2000. Differential patterns of ERK and STAT3phosphorylation after sciatic nerve transection in the rat. Exp. Neurol. 166,392–402.

Sorkin, L.S., Yaksh, T.L., 2009. Behavioral models of pain states evoked by physical injuryto the peripheral nerve. Neurotherapeutics 6, 609–619.

Stamboulian, S., Choi, J.S., Ahn, H.S., Chang, Y.W., Tyrrell, L., Black, J.A., Waxman, S.G.,Dib-Hajj, S.D., 2010. ERK1/2 mitogen-activated protein kinase phosphorylatessodium channel Na(v)1.7 and alters its gating properties. J. Neurosci. 30,1637–1647.

Wall, P.D., Devor, M., 1983. Sensory afferent impulses originate from dorsal root gangliaas well as from the periphery in normal and nerve injured rats. Pain 17, 321–339.

Wall, P.D., Gutnick, M., 1974. Ongoing activity in peripheral nerves: the physiology andpharmacology of impulses originating from a neuroma. Exp. Neurol. 43, 580–593.

Waxman, S.G., 2006. Neurobiology: a channel sets the gain on pain. Nature 444,831–832.

Waxman, S.G., 2007. Nav1.7, its mutations, and the syndromes that they cause.Neurology 69, 505–507.

Waxman, S.G., Dib-Hajj, S., 2005. Erythermalgia: molecular basis for an inherited painsyndrome. Trends Mol. Med. 11, 555–562.

Waxman, S.G., Kocsis, J.D., Black, J.A., 1994. Type III sodium channel mRNA is expressedin embryonic but not adult spinal sensory neurons, and is reexpressed followingaxotomy. J. Neurophysiol. 72, 466–470.

Waxman, S.G., Cummins, T.R., Black, J.A., Dib-Hajj, S., 2002. Diverse functions anddynamic expression of neuronal sodium channels. Novartis Found. Symp. 241,34–51 discussion 51–60.

Welk, E., Leah, J.D., Zimmermann,M., 1990. Characteristics of A- and C-fibers ending in asensory nerve neuroma in the rat. J. Neurophysiol. 63, 759–766.

Wittmack, E.K., Rush, A.M., Hudmon, A., Waxman, S.G., Dib-Hajj, S.D., 2005. Voltage-gated sodium channel Nav1.6 is modulated by p38 mitogen-activated proteinkinase. J. Neurosci. 25, 6621–6630.

Yaari, Y., Devor, M., 1985. Phenytoin suppresses spontaneous ectopic discharge in ratsciatic nerve neuromas. Neurosci. Lett. 58, 117–122.