B R A I N R E S E A R C H 1 2 2 9 ( 2 0 0 8 ) 1 1 8 – 1 2 4

ava i l ab l e a t www.sc i enced i r ec t . com

www.e l sev i e r. com/ loca te /b ra in res

Research Report

Spinal α3β2* nicotinic acetylcholine receptors tonically inhibitthe transmission of nociceptive mechanical stimuli

Tracey Younga, Shannon Wittenauera, J. Michael McIntoshb,c, Michelle Vinclera,⁎aDepartment of Anesthesiology, Wake Forest University Health Sciences, Medical Center Boulevard, Winston-Salem, NC 27157, USAbDepartment of Psychiatry, University of Utah, Salt Lake City, UT 84112, USAcDepartment of Biology, University of Utah, Salt Lake City, UT 84112, USA

A R T I C L E I N F O

⁎ Corresponding author. Fax: +1 336 716 6744.E-mail address: mvincler@wfubmc.edu (MAbbreviations: α-CTX MII, α-Conotoxin MII

SNL, Spinal nerve ligation; i.t., Intrathecal; RT

0006-8993/$ – see front matter © 2008 Elsevidoi:10.1016/j.brainres.2008.06.086

A B S T R A C T

Article history:Accepted 25 June 2008Available online 2 July 2008

The presence of non-α4β2, non-α7 nicotinic acetylcholine receptors (nAChR) in the ratspinal cord has been suggested previously, but the identity of these nAChRs had not beenshown. Intrathecal administration of the α3β2⁎/α6β2⁎ selective α-conotoxin MII (α-CTX MII)dose- and time-dependently reduced pawwithdrawal thresholds to mechanical pressure innormal rats. The pronociceptive effect of α-CTX MII was partially blocked by NMDA receptorantagonism and lost completely following ablation of C-fibers. The effect of spinal nerveligation on α-CTX MII-induced mechanical hypersensitivity was also assessed. Sensitivitywas lost in the hind paw ipsilateral to spinal nerve ligation, but maintained in thecontralateral hind paw at control levels. Radioligand binding in spinal cord membranesrevealed high and low affinity α-CTX MII binding sites. Spinal nerve ligation did notsignificantly alter α-CTX MII binding ipsilateral to ligation. Finally, no evidence for thepresence of α6-containing nAChRs was identified. The results of these studies show thepresence of 2 populations of α-CTX MII-sensitive nAChRs containing the α3 and β2, but notthe α6, subunits in the rat spinal cord that function to inhibit the transmission of nociceptivemechanical stimuli via inhibiting the release of glutamate from C-fibers. Spinal nerveligation produces a unilateral loss of α-CTX MII-induced mechanical hypersensitivitywithout altering α-CTX MII binding sites. Our data support a peripheral injury-induced lossof a cholinergic inhibitory tone at spinal α3β2⁎ nAChRs, without the loss of the receptorsthemselves, which may contribute to mechanical hypersensitivity following spinal nerveligation.

© 2008 Elsevier B.V. All rights reserved.

Keywords:PainHyperalgesiaGlutamateC-fiberα-Conotoxin

1. Introduction

Multiple populations of nicotinic acetylcholine receptors(nAChRs) within the spinal cord modulate the transmissionof nociceptive stimuli. Exogenous stimulation of spinalnAChRs with intrathecal nAChR agonists produces both

. Vincler).; nAChR, Nicotinic acetylcX, Resiniferatoxin

er B.V. All rights reserved

nociceptive and antinociceptive behaviors via stimulation ofseparable populations of nAChRs (Khan et al., 2001). Intrathe-cally administered nAChR antagonists themselves also reducenociceptive thresholds supporting a role for an endogenouscholinergic inhibitory tone in the spinal cord (Rashid andUeda, 2002). The identity of the nAChRs responsible for these

holine receptor; ACh, Acetylcholine; NMDA, N-methyl-D-aspartate;

.

Fig. 1 – Dose-response of intrathecal α-CTX MII onmechanical paw withdrawal thresholds. (A) The meanpercent changes in baseline paw withdrawal thresholds tomechanical pressure (% Baseline)±S.E.M. following theintrathecal administration of α-CTX MII (0.01, 0.03, or0.1 pmol/10 μl saline) are shown across time. * and #p<0.05compared to pre-drug baseline for 0.03 and 0.1 pmol,respectively (n=8–10). (B) Mean area under the curve(AUC)±S.E.M of the results in panel A. *p<0.05 compared to0.01 pmol; ^p<0.05 compared to 0.03 pmol.

119B R A I N R E S E A R C H 1 2 2 9 ( 2 0 0 8 ) 1 1 8 – 1 2 4

effects has partially been delineated, but has been compli-cated by a lack of compounds that are highly selective fordistinct nAChR subtypes. In the rat, the pronociceptive effectsof intrathecal nAChR agonists epibatidine and A-85380 aremediated by α7⁎1 nAChRs (Khan et al., 1997; Khan et al., 2001)whereas the antinociceptive effects of these 2 agonists aremediated by different populations of nAChRs (Khan et al.,2001). The antinociceptive effects of intrathecal epibatidineare not blocked by either α7⁎ or α4β2⁎ nAChR antagonists,suggesting the presence of additional nAChR subtypes in thespinal cord (Khan et al., 2001).

Peripheral nerve injury alters the pharmacology of spinalnAChRs in both rats and mice (Abdin et al., 2006; Rashid andUeda, 2002). Intrathecal nicotine and epibatidine completelyreverse thermal and mechanical hypersensitivity in partialsciatic nerve-ligatedmice at doses that have no effect in shamanimals (Rashid and Ueda, 2002). In tibial nerve-transectedrats, a prolonged antinociceptive response is observed follow-ing the intrathecal administration of nAChR agonists in theabsence of any algogenic behaviors (Abdin et al., 2006). Thisantinociceptive response to intrathecal nAChR agonists in thepresence of neuropathic pain is thought to result from aperipheral injury-induced loss of spinal cholinergic inhibitorytone (Rashid et al., 2006; Rashid and Ueda, 2002), althoughperipheral injury-induced changes in nAChR expression alsooccur (Vincler and Eisenach, 2004; Yang et al., 2004).

The α3 nAChR subunit has been identified previously in therat spinal cord using immunohistochemistry (Khan et al.,2003; Khan et al., 2008; Vincler and Eisenach, 2004). However,the function and identity of α3-containing nAChRs in thespinal cord is unknown. Using the α3β2⁎/α6β2⁎ nAChR-selective α-conotoxin MII (α-CTX MII), the current series ofstudies delineates the role of this receptor in the rat spinalcord in the transmission of nociceptive mechanical stimuli. Inaddition, injury-induced changes in spinal α3β2⁎/α6β2⁎nAChRs are defined using radioligand binding. The identityof α-CTX MII-sensitive sites in the rat spinal cord is furtherinvestigated using a recently described α-CTX MII analogselective for α6-containing nAChRs (Azam et al., 2008).

2. Results

2.1. Intrathecal α-CTX MII administration in normal rats

Baseline paw withdrawal thresholds to mechanical pressurewere measured in normal rats with mean paw withdrawalthresholds being 148±4 g. As shown in Fig. 1, intrathecal (i.t.)administration of the α3β2⁎/α6β2⁎ nAChR antagonist, α-CTXMII, dose- [F(2,131)=13.8, p<0.001] and time-dependently[F(5,131)=27.9, p<0.001] reduced baseline paw withdrawalthresholds. Both 0.03 and 0.1 pmol α-CTX MII significantlyreduced paw withdrawal thresholds by 31±3% and 37±4%,respectively, with a peak effect of 15 min (Fig. 1A). Pawwithdrawal thresholds remained significantly reduced even at60 min following the i.t. administration of 0.03 and 0.1 pmol α-

1 The asterisk “⁎” indicates that the exact subunit compositionof the nAChR is not known according to the International Unionof Pharmacology (Lukas et al., 1999).

CTX MII. Area under the curve analysis revealed significantdifferences in α-CTX MII doses across time (Fig. 1B).

Previous studies have shown that the pronociceptiveeffects of nAChR agonists are dependent upon glutamaterelease and the activation of spinal NMDA receptors (Khan etal., 2001). However, the role of NMDA receptors in thedisinhibition observed with nAChR antagonists has not beeninvestigated. Therefore, we examined the role of spinal NMDAreceptors in α-CTX MII-induced mechanical hypersensitivity.The NMDA receptor antagonist, MK801 (10 μg/10 μl, i.t.), wasadministered 5 min prior to 0.1 pmol α-CTX MII and pawwithdrawal thresholds to mechanical pressure were mea-sured 15 min post-α-CTX MII. Intrathecal pretreatment withMK801 significantly reduced the pronociceptive effects ofα-CTX MII on mechanical paw withdrawal thresholds (Fig. 2).

The ablation of C-fibers also has been shown previously toreduce the algogenic responses to i.t nAChR agonist admin-istration (Khan et al., 2004). However, the role of C-fibers in thedisinhibition of the transmission of nociceptive mechanicalstimuli observed with nAChR antagonists has not beenexamined. Primary afferent C-fibers were destroyed in adult

Fig. 2 – Effect of spinal NMDA receptor blockade on α-CTXMII-induced mechanical hypersensitivity. Rats wereadministered MK801 (10 μg/10 μl, black box) or saline(white box), 5 min prior to α-CTX MII (0.1 pmol/10 μl) andpaw withdrawal thresholds to mechanical pressure weremeasured 15 min post-α-CTX MII. Mean percent changes inpaw withdrawal thresholds (% Baseline)±S.E.M. are shown.^p<0.05 compared to pre-drug baselines; #p<0.05 comparedto MII.

Fig. 3 – Effect of C-fiber destruction on α-CTX MII-inducedmechanical hypersensitivity. Rats were treated withresiniferatoxin (RTX; 0.2 mg/kg, i.p.) and behavioral testingwas performed 3–5 days post-RTX. (A) Mean pawwithdrawallatencies (PWL) in seconds±S.E.M. of RTX-treated rats usedprior to (pre-RTX) and 3–5 days following (post-RTX) RTXadministration. #p<0.05 compared to Pre-RTX. (B) Mean pawwithdrawal thresholds (PWT) in grams (g)±S.E.M. of thesame group of RTX-treated rats prior to (pre-RTX) and3–5 days following (post-RTX) RTX administration. (n=8–10).(C)α-CTXMII (0.1 pmol/10μl) was administered intrathecallyin normal (α-Ctx MII) and RTX-treated (α-Ctx MII – RTX) ratsand pawwithdrawal thresholds tomechanical pressureweremeasured 15 min post-α-CTX MII administration. The meanpercent changes in paw withdrawal thresholds(% Baseline)±S.E.M. are shown. ^p<0.05 compared toBaseline; #p<0.05 for RTX treatment.

120 B R A I N R E S E A R C H 1 2 2 9 ( 2 0 0 8 ) 1 1 8 – 1 2 4

rats using the intraperitoneal administration of 0.2 mg/kgresiniferatoxin (RTX) as described previously (Pan et al., 2003).Loss of C-fiber functioning was confirmed with an increasedthermal pawwithdrawal latency to radiant heat 3–5 days post-RTX (Fig. 3B). Destruction of C-fibers did not alter baselinemechanical paw withdrawal thresholds (PWTs) in the sametime frame (Fig. 3C). However, RTX treatment resulted in acomplete loss of the pronociceptive effects of i.t. 0.1 pmolα-CTX MII (Fig. 3A).

2.2. Intrathecal α-CTX MII in spinal nerve-ligated rats

In the mouse, peripheral nerve injury results in a loss ofantinociceptive cholinergic tone at α4β2⁎ nAChRs in the spinalcord (Rashid et al., 2006). Because our data suggest that spinalα3β2⁎/α6β2⁎ nAChRs are also under an antinociceptive choli-nergic tone in the normal rat spinal cord, we examined theeffects of spinal nerve ligation (SNL) on MII pharmacology. Ratunderwent SNL as described previously (Kim and Chung, 1992)which significantly reduced mechanical paw withdrawalthresholds in the ipsilateral hind paw (92±12 g ipsilateral vs.142±7 g contralateral) 14 days post-ligation. As shown in Fig. 4,i.t administration of 0.1 pmol α-CTX MII did not alter pawwithdrawal thresholds ipsilateral to ligation. However, sig-nificant mechanical hypersensitivity was observed in thecontralateral hind paw [F(5,29)=2.7, p<0.0001]. The reductionin paw withdrawal thresholds following i.t. α-CTX MII did notdiffer between the contralateral hind paw in SNL rats andnormal rats.

2.3. Radioligand binding in spinal cord membranes

The α-CTX MII-sensitive nAChRs in the rat spinal cord werefurther characterized using radioligand binding. Spinal cordmembranes were prepared from the dorsal half (normal) oripsilateral dorsal quadrant (SNL) of the L4-L6 spinal cord and

incubated with 0.8 nM [3H]epibatidine in the presence of0.3 pM–3 μM α-CTX MII as described in Experimentalprocedures. High and low affinity α-CTX MII binding sites

Fig. 4 – Impact of spinal nerve ligation onα-CTXMII-inducedmechanical hypersensitivity. Rats were administered α-CTXMII (0.1 pmol/10 μl, i.t.) in normal rats or in spinalnerve-ligated (SNL) rats 14 days post-ligation. Mean pawwithdrawal thresholds (PWT) in grams±S.E.M. are shown forhind paws in normal rats and ipsilateral (SNL-Ipsi) andcontralateral (SNL-Contra) hind paws in SNL rats prior to(Baseline) and 5–60 min post i.t. α-CTX MII. (n=10).

121B R A I N R E S E A R C H 1 2 2 9 ( 2 0 0 8 ) 1 1 8 – 1 2 4

were identified in spinal cord membranes from normal andSNL rats (Table 1). Comparison of α-CTX MII binding curvesusing global curve fitting revealed that SNL did not alter spinalα-CTX MII-sensitive sites ipsilateral to ligation (Fig. 5A).

Because α-CTX MII cannot differentiate between α3β2⁎ andα6β2⁎ nAChRs, the identities of the two α-CTX MII bindingsites in the rat spinal cord were investigated further using arecently described α6-selective antagonist (Azam et al., 2008).As shown in Fig. 5B, α-CTX MII[S4A;E11A;L15A] did notdisplace 0.8 nM [3H]epibatidine from normal rat spinal cordmembranes, suggesting that the α6 nAChR subunit does notcontribute to α-CTX MII binding sites in the dorsal horn of theL4–L6 rat spinal cord.

3. Discussion

The results of these studies characterize a previously uni-dentified population of α3β2⁎ nAChRs in the rat spinal cordthat are under tonic cholinergic regulation and function toinhibit the transmission of nociceptive mechanical stimuli.Interestingly, the pronociceptive effects of α3β2⁎ nAChRblockade rely upon the release of glutamate from C-fibers.Following peripheral nerve injury, the inhibitory cholinergictone at the receptor is lost, although no apparent loss of thereceptors themselves is observed. Two α-CTXMII binding siteswere identified in the spinal cord dorsal horn of normal andspinal nerve-ligated rats; neither of which contains the α6subunit. Spinal nerve ligation did not alter α-CTX MII bindingin spinal cord membranes from the dorsal horn ipsilateralto injury.

Intrathecal administration of the α3β2⁎/α6β2⁎ nAChRantagonist, α-CTX MII, reduces paw withdrawal thresholdsto mechanical pressure suggesting that endogenous AChwithin the spinal cord acts to inhibit the transmission ofnociceptive mechanical stimuli. These data are consistentwith the inhibitory cholinergic tone identified previously in

the mouse spinal cord (Rashid and Ueda, 2002). However, inthe mouse spinal cord, the inhibitory cholinergic tone hasbeen definitively shown to be mediated by α4β2⁎ nAChRsusing targeted knockdown of the α4 subunit (Rashid et al.,2006). In the rat spinal cord, the cholinergic inhibitory tone[2682] and the antinociceptive effects of some intrathecalnAChR agonists (Khan et al., 2001) have been attributed to theα4β2⁎ nAChR subtype using only dihydro-β-erythroidine(DHβE). Although previous studies identify the DHβE-sensitivesites as α4β2⁎ nAChRs, the specificity of this antagonist forα4β2⁎ nAChRs following intrathecal administration is ques-tionable. At sub-micromolar concentrations, DHβE also blocksα3β2 nAChRs and displays only a 10–50 fold reduced potencyat α3β4 and α7 nAChRs (Jensen et al., 2005). Based on the highpercentage of displacement of [3H]epibatidine by α-CTX MII inour studies, it seems plausible that previous behavioralstudies using DHβE to implicate a role for spinal α4β2⁎nAChRs, may have been targeting α3β2⁎ nAChRs to a largerdegree.

Intrathecal administration of nAChR agonists elicits exci-tatory amino acid release and produces nociceptive behaviorsthat are blocked by NMDA receptor antagonists and theablation of primary afferent C-fibers (Khan et al., 1996; Khanet al., 2004). Our data show that blockade of spinal α3β2⁎nAChRs also induces pronociceptive behaviors that aremediated by the release of glutamate. Despite the dependenceon the presence of C-fibers, the nociceptive behaviors elicitedby nAChR agonists and antagonists must be occurring viadifferent mechanisms. Nicotinic receptors are localized onprimary afferent C-fibers within the spinal cord and intrathe-cal nAChR agonists likely stimulate these receptors directly toincrease spinal glutamate release (Khan et al., 2003; Khan etal., 1996). However, spinal α3β2⁎ nAChRs function to inhibitthe release of glutamate, an effect that could not be mediatedby the localization of these nAChRs on primary afferents. Theα3β2⁎ nAChRs are likely localized on another spinal structureand are endogenously activated by spinal ACh to release aninhibitory intermediary neuropeptide which then functions toinhibit glutamate release from C-fibers. The α3 subunit islocalized on spinal cord interneurons (Khan et al., 2003;Vincler and Eisenach, 2004) and an α3-containing nAChR isthought to be expressed at presynaptic sites on inhibitoryneurons in the substantia gelatinosa in the adult rat spinalcord (Takeda et al., 2003).

Radioligand binding revealed 2 populations of α-CTX MII-sensitive sites within the dorsal horn of the rat spinal cord, butthe exact nAChR subunit composition of these sites cannot bedetermined from the current data. α-CTX MII cannot differ-entiate between α3-and α6-containing nAChRs (McIntosh etal., 2004), but the absence of the α6 subunit in spinal α-CTXMII-sensitive nAChRs was confirmed using α-CTX MII[S4A;E11A;L15A] which displays N1000-fold preference for α6/α3β2β3 nAChRs over α3β2 nAChRs (Azam et al., 2008). There-fore, α-CTX MII nAChRs in the dorsal horn of the rat spinalcord contain at least the α3 and β2 subunits, but not the α6subunit at the ligand binding site. The β3 and α4 nAChRsubunits are expressed in rat spinal cord and may participatein spinal α-CTX MII-sensitive nAChRs; mice lacking thesesubunits display reduced [125I]-α-CTX MII binding in brain(striatum) (Vincler and Eisenach, 2004; Salminen et al., 2005).

Table 1 – Displacement of [3H]epibatidine binding tospinal cord membranes by α-conotoxin MII

Ki (nM) % Total binding

Highaffinity

Lowaffinity

Highaffinity

Lowaffinity

Normal 2.7±1.6 83.5±3.2 59±11% 41±11%SNL 3.2±1.5 67.5±2.1 54±11% 46±11%

Mean Ki values±S.E.M. of α-CTX MII displacement of 0.8 nM[3H]epibatidine to spinal cord membranes from normal and spinalnerve-ligated (SNL) rats are shown. Ki values were calculated usingthe previously reported KD=0.12 nM of epibatidine to α3β2 nAChRs(Wang et al., 1996). The mean percent contribution±S.E.M. of highand low affinity sites to total α-CTX MII binding in spinal cordmembranes from normal and SNL rats is shown.

122 B R A I N R E S E A R C H 1 2 2 9 ( 2 0 0 8 ) 1 1 8 – 1 2 4

The α5 subunit may also participate in α-CTX MII-sensitivespinal nAChRs, but the absence of this subunit does notreduce α-CTX MII binding to mouse striatal membranes(Salminen et al., 2005). Although the number of [125I]α-CTXMII sites was not reduced in α5 KO mice, the impact of the α5subunit on the affinity of α-CTX MII for α3β2 nAChRs is notknown as mouse striatum contains little α3 (Champtiaux etal., 2002). The α5 subunit is expressed in rat spinal cord (Khanet al., 2003; Vincler and Eisenach, 2004) and is upregulatedfollowing spinal nerve ligation (Vincler and Eisenach, 2004).The inclusion of the α5 subunit with α3 and β2 subunits doesnot alter epibatidine binding, but does reduce the EC50 of AChand nicotine (Wang et al., 1996).

These data identify the presence of α3β2⁎ nAChRs in thelower lumbar rat spinal cord dorsal horn that indirectly inhibitthe release of glutamate from primary afferent C-fibers andreduce the sensitivity to noxious mechanical stimuli. Thelocalization of these nAChRs within the spinal cord is notknown, but they likely facilitate the release of inhibitoryneurotransmitters. Although the presence of α3 and β2subunits in α-CTX MII nAChRs is strongly implicated, thepresence of high and low affinity α-CTX MII nAChRs suggeststhe presence of additional nAChR subunits.

Fig. 5 – α-CTXMII displacement of [3H]epibatidine binding torat spinal cord membranes. (A) Spinal cord membranesprepared from the dorsal half (Normal) or dorsal quadrantipsilateral to spinal nerve ligation (SNL) were incubated with0.8 nM [3H]epibatidine in the presence of 0.3 pM–3 μM ofunlabeled α-CTX MII as described in the Experimentalprocedures section. (B) α-CTX MII[S4A;E11A;L15A]displacement of [3H]epibatidine binding to rat spinal cordmembranes. Membranes prepared from the dorsal half of thelower lumbar (L4–L6) spinal cord were incubated with 0.8 nM[3H]epibatidine in the presence of 0.3 pM–3 μM unlabeledα-CTX MII[S4A;E11A;L15A] as described in the Experimentalprocedures section. Non-specific binding was determined inthe presence of 100 μM unlabeled nicotine. Each pointrepresents the mean±S.E.M. of four separate experiments.Data were best-fitted to a two-site Hill inhibition equation.

4. Experimental procedures

4.1. Animals

All animals used in this study weremale Sprague–Dawley rats(200–250 g; Harlan, IN), housed in pairs prior to surgery andindividually post-catheter implantation with free access tofood and water. Protocols and procedures were approved bythe Animal Care and Use Committee (Wake Forest UniversityHealth Sciences, Winston-Salem, NC).

4.2. Surgical preparations

4.2.1. Intrathecal catheter implantationLumbosacral intrathecal catheters were implanted asdescribed previously (Storkson et al., 1996), with slightmodifications (Milligan et al., 1999). Catheters consisted ofPE-10 tubing stretched to reduce the overall diameter. Briefly,under halothane anesthesia, an incision wasmade in the skin

of the lower back and a sterile 20 G needle was used as a guidecannula and was inserted between the L5 and L6 vertebrae. Atail flick confirmed entry into the intrathecal space. Thestretched PE10 catheter containing a guide wire was gently fedthrough the needle until the catheter extended 3 cm beyondthe tip of the needle to reach the lumbar enlargement. Theneedle and guide wire were gently removed. A loosely tiedknot was made in the catheter and three sutures were used tohold the catheter in place. A small fistula (a modified 1 cm3

syringe hub (Milligan et al., 1999)) was sutured to the musclesurface and the catheter was fed through the fistula. Theremaining externalized catheter was coiled into the fistulaand the rubber plug sealed with a small amount of siliconsealant. The dead space of the catheter ranged from 7–10 μl

123B R A I N R E S E A R C H 1 2 2 9 ( 2 0 0 8 ) 1 1 8 – 1 2 4

and, therefore, all drug administrations were followed by a10 μl saline flush.

4.2.2. Spinal nerve ligationRats underwent spinal nerve ligation (SNL) as describedpreviously (Kim and Chung, 1992). Under halothane anesthe-sia (2–3% halothane in 100% oxygen), the left L5 and L6 spinalnerves were isolated adjacent to the vertebral column andtightly ligatedwith 6.0 silk suture. The incisionwas closed andthe animals returned to their home cages for 12–14 days post-ligation to allow for the development of mechanical allodynia.

4.3. Behavioral testing

All behavioral testing was conducted 12–14 days post-surgerybetween the hours of 9:00 AM and 4:00 PM. Paw withdrawalthresholds (PWT)were determined for left and right hind pawsusing the Randall–Selitto paw pressure technique (Randalland Selitto, 1957). The Analgesy-meter (Ugo Basile, Italy) usesa conical Teflon applicator to apply a constant rate ofincreasing pressure (16 g per second) to the hind paws. Thecut-off pressure was set at 250 g. Prior to experimental testing,animals were first subjected to 4 training sessions prior to SNLto stabilize baseline responses (Taiwo et al., 1989). Each hindpaw was tested 2 times with a 5 min intertrial interval. In SNLrats, the mean PWT for the ipsilateral and contralateral hindpaws was compared to determine the presence of mechanicalhypersensitivity. Mechanical hypersensitivity was defined asthe presence of at least a 40% decrease in PWT for theipsilateral hind paw.

For pharmacological testing, all drugs were dissolved insterile 0.9% saline and administered intrathecally in a volumeof 10 μl. Pawwithdrawal thresholdsweremeasured at 5, 10, 15,30, and 60 min following i.t. administration. When multipledrugs were administered in combination, the first drug wasadministered 10 min prior the second. Data are expressed asthemean pawwithdrawal thresholds (PWTs)±S.E.M. in gramsor as a percentage of pre-drug baseline responses (% Base-line=post-drug PWT/pre-drug PWT PWT×100).

C-fibers were destroyed with an intraperitoneal (i.p.)administration of 0.2 mg/kg resiniferatoxin (RTX) underisoflurane anesthesia as described previously (Pan et al.,2003). RTX was dissolved in a mixture of 10% Tween 80 and10% ethanol in normal saline (Pan et al., 2003). The loss of C-fibers was confirmed using thermal paw withdrawal to anoxious radiant heat source using a commercially availabledevice (Anesthesiology Research Laboratory, Department ofAnesthesiology, UCSD). The mean withdrawal latency of 4applications (2 per hind paw) was calculated 3–5 days post-RTX administration. A cut-off latency of 30 s was employed toavoid tissue damage.

4.4. Radioligand binding

4.4.1. Spinal cord tissue preparationThedorsal half (normal rats) or ipsilateral dorsal quadrant (SNLrats) of the L4-L6 spinal cord tissue was placed in 10 volumes(w/v) of ice-cold hypotonic buffer (14.4 mM NaCl, 0.2 mM KCl,0.2 mM CaCl2, 0.1 mM MgSO4, 2.0 mM HEPES, pH 7.5) andhomogenized using a Kinematica polytron. Homogenizedsamples were centrifuged at 25,000 g for 15 min. The pellet

was resuspended in hypotonic buffer and again centrifuged.The resuspension/centrifugation cycle was repeated twomoretimes. The resulting pellet was stored frozen at −80 °C underfresh hypotonic buffer until ready for use.

4.4.2. Radioligand bindingAt the time of assay, the pellet was thawed and resuspendedwith Tris–HCl buffer (50 mM Tris–HCl, 120 mM NaCl, 5 mMKCl, 1 mM MgCl2, 2 mM CaCl2, pH 7.5) supplemented with0.1 mM PMSF and 5 mM iodoacetamide. The assay mixtureconsisted of 200 μg of membrane protein in a final incubationvolume of 60 μl. Incubations were carried out in a cold roomon a gentle shaker for 60 min. Assays were initiated with theaddition of the membrane suspension with rapid mixing tothe [3H] polypropylene tubes. Incubations were terminated bythe addition of 3 ml of ice-cold assay buffer followed by rapidfiltration through Whatman GF/B filter papers previouslyequilibrated with 0.5% polyethyleneimine at 4 °C using aBrandel cell harvester. Samples were then washed four timeswith 4 ml of ice-cold assay buffer (Tris–HCl buffer supple-mented with 10 μM atropine sulfate). The filters were thenplaced in counting vials, mixed vigorously with scintillationfluid and counted the next day in a Beckman Coulter LS 6500liquid scintillation counter. All assays were done in triplicateand protein was assayed by the Bradford protein assay. Aconcentration of 0.8 nM (±)-[3H]epibatidine was used incompetitive binding assays. Protease inhibitors (10 μg/mleach of aprotinin, leupeptin trifluoroacetate, and pepstatin A)were added for competitive binding assays using MII[S4A;E11A;L15A] as described previously (Whiteaker et al., 2000).

4.5. Materials

MK-801, resiniferatoxin, and all chemical components ofbuffers were purchased from Sigma. [3H]epibatidine waspurchased from Perkin-Elmer. α-Conotoxin MII and α-conotoxin MII[S4A;E11A;L15A] were synthesized as pre-viously described (Cartier et al., 1996).

4.6. Statistics

Behavioral pharmacology was analyzed using one-way ortwo-way repeated measures ANOVA where appropriate.Radioligand binding curves were calculated using a fiveparameter Hill equation in Prism 4 (GraphPad) and comparedbetween groups using global curve fitting and an F test.

Acknowledgments

This work was supported by NIH grants R01 NS48158 (M.V.)and MH53631 and GM48677 (J.M.M.).

R E F E R E N C E S

Abdin, M.J., Morioka, N., Morita, K., Kitayama, T., Kitayama, S.,Nakashima, T., Dohi, T., 2006. Analgesic action of nicotine ontibial nerve transection (TNT)-induced mechanical allodyniathrough enhancement of the glycinergic inhibitory system inspinal cord. Life Sci. 80, 9–16.

124 B R A I N R E S E A R C H 1 2 2 9 ( 2 0 0 8 ) 1 1 8 – 1 2 4

Azam, L., Yoshikami, D., McIntosh, J.M., 2008. Amino acid residuesthat confer high selectivity of the alpha 6 nicotinic acetylcholinereceptor subunit to alpha-conotoxin MII[S4A;E11A;L15A].J. Biol. Chem. 283, 11625–11632.

Cartier, G.E., Yoshikami, D., Gray, W.R., Luo, S., Olivera, B.M.,McIntosh, J.M., 1996. A new alpha-conotoxin which targetsalpha3beta2 nicotinic acetylcholine receptors. J. Biol. Chem.271, 7522–7528.

Champtiaux, N., Han, Z.Y., Bessis, A., Rossi, F.M., Zoli, M., Marubio,L., McIntosh, J.M., Changeux, J.P., 2002. Distribution andPharmacology of alpha 6-Containing Nicotinic AcetylcholineReceptors Analyzed with Mutant Mice. J. Neurosci. 22,1208–1217.

Jensen, A.A., Frolund, B., Liljefors, T., Krogsgaard-Larsen, P., 2005.Neuronal nicotinic acetylcholine receptors: structuralrevelations, target identifications, and therapeuticinspirations. J. Med. Chem. 48, 4705–4745.

Khan, I., Osaka, H., Stanislaus, S., Calvo, R.M., Deerinck, T., Yaksh,T.L., Taylor, P., 2003. Nicotinic acetylcholine receptordistribution in relation to spinal neurotransmission pathways.J. Comp. Neurol. 467, 44–59.

Khan, I.M., Marsala, M., Printz, M.P., Taylor, P., Yaksh, T.L., 1996.Intrathecal nicotinic agonist-elicited release of excitatoryamino acids as measured by in vivo spinal microdialysis inrats. J. Pharmacol. Exp. Ther. 278, 97–106.

Khan, I.M., Stanislaus, S., Zhang, L., Taylor, P., Yaksh, T.L., 2001.A-85380 and epibatidine each interact with disparate spinalnicotinic receptor subtypes to achieve analgesia andnociception. J. Pharmacol. Exp. Ther. 297, 230–239.

Khan, I.M., Wennerholm, M., Singletary, E., Polston, K., Zhang, L.,Deerinck, T., Yaksh, T.L., Taylor, P., 2004. Ablation of primaryafferent terminals reduces nicotinic receptor expression andthe nociceptive responses to nicotinic agonists in thespinal cord. J. Neurocytol. 33, 543–556.

Khan, I.M., Yaksh, T.L., Taylor, P., 1997. Epibatidine binding sitesand activity in the spinal cord. Brain Res. 753, 269–282.

Khan, I.M., Wart, C.V., Singletary, E.A., Stanislaus, S., Deerinck, T.,Yaksh, T.L., Printz, M.P., 2008. Elimination of rat spinalsubstance P receptor bearing neurons dissociatescardiovascular and nocifensive responses to nicotinic agonists.Neuropharmacology 54, 269–279.

Kim, S.H., Chung, J.M., 1992. An experimental model for peripheralneuropathy produced by segmental spinal nerve ligation in therat. Pain 50, 355–363.

Lukas, R.J., Changeux, J.P., Le Novere, N., Albuquerque, E.X.,Balfour, D.J., Berg, D.K., Bertrand, D., Chiappinelli, V.A., Clarke,P.B., Collins, A.C., Dani, J.A., Grady, S.R., Kellar, K.J., Lindstrom,J.M., Marks, M.J., Quik, M., Taylor, P.W., Wonnacott, S., 1999.International Union of Pharmacology. XX. Current status of thenomenclature for nicotinic acetylcholine receptors and theirsubunits. Pharmacol. Rev. 51, 397–401.

McIntosh, J.M., Azam, L., Staheli, S., Dowell, C., Lindstrom, J.M.,Kuryatov, A., Garrett, J.E., Marks, M.J., Whiteaker, P., 2004.

Analogs of α-conotoxin MII are selective for α6-containingnicotinic acetylcholine receptors. Mol. Pharmacol.65, 944–952.

Milligan, E.D., Hinde, J.L., Mehmert, K.K., Maier, S.F., Watkins,L.R., 1999. A method for increasing the viability of theexternal portion of lumbar catheters placed in the spinalsubarachnoid space of rats. Journal of NeuroscienceMethods 90, 81–86.

Pan, H.L., Khan, G.M., Alloway, K.D., Chen, S.R., 2003.Resiniferatoxin induces paradoxical changes in thermal andmechanical sensitivities in rats: mechanism of action. J.Neurosci. 23, 2911–2919.

Randall, L.O., Selitto, J.J., 1957. A method for measurement ofanalgesic activity on inflamed tissueArch. Int. Pharmacodyn.CXI 409–419.

Rashid, M.H., Furue, H., Yoshimura, M., Ueda, H., 2006. Tonicinhibitory role of alpha4beta2 subtype of nicotinicacetylcholine receptors on nociceptive transmission in thespinal cord in mice. Pain 125, 125–135.

Rashid, M.H., Ueda, H., 2002. Neuropathy-specific analgesic actionof intrathecal nicotinic agonists and its spinal GABA-mediatedmechanism. Brain Res. 953, 53–62.

Salminen, O., Whiteaker, P., Grady, S.R., Collins, A.C., McIntosh,J.M., Marks, M.J., 2005. The subunit composition andpharmacology of [alpha]-conotoxin MII-binding nicotinicacetylcholine receptors studied by a novel membrane-bindingassay. Neuropharmacology 48, 696–705.

Storkson, R.V., Kjorsvik, A., Tjolsen, A., Hole, K., 1996. Lumbarcatheterization of the spinal subarachnoid space in the rat.J. Neurosci. Methods 65, 167–172.

Taiwo, Y.O., Coderre, T.J., Levine, J.D., 1989. The contribution oftraining to sensitivity in the nociceptive paw-withdrawal test.Brain Res. 487, 148–151.

Takeda, D., Nakatsuka, T., Papke, R., Gu, J.G., 2003. Modulation ofinhibitory synaptic activity by a non-alpha4beta2, non-alpha7subtype of nicotinic receptors in the substantia gelatinosa ofadult rat spinal cord. Pain 101, 13–23.

Vincler, M., Eisenach, J.C., 2004. Plasticity of spinal nicotinicacetylcholine receptors following spinal nerve ligation.Neurosci. Res. 48, 139–145.

Wang, F., Gerzanich, V., Wells, G.B., Anand, R., Peng, X., Keyser, K.,Lindstrom, J., 1996. Assembly of human neuronal nicotinicreceptor alpha5 subunits with alpha3, beta2, and beta4subunits. J. Biol. Chem. 271, 17656–17665.

Whiteaker, P., McIntosh, J.M., Luo, S., Collins, A.C., Marks, M.J.,2000. 125I-a-conotoxin MII identifies a novel nicotinicacetylcholine receptor population in mouse brain. Mol.Pharmacol. 57, 913–925.

Yang, L., Zhang, F.X., Huang, F., Lu, Y.J., Li, G.D., Bao, L., Xiao, H.S.,Zhang, X., 2004. Peripheral nerve injury inducestrans-synaptic modification of channels, receptors and signalpathways in rat dorsal spinal cord. Eur. J. Neurosci. 19,871–883.