PA

EAa

Cb

R

Aifwnnipmpnt(3cagtnoTMttmrit(ectastoc

*E(Arssm(msap

Neuroscience 152 (2008) 511–520

0d

RONOCICEPTIVE RESPONSE ELICITED BY TRPA1 RECEPTOR

CTIVATION IN MICE

httA

Kc

MnleiahcaHstp

c(idfic(sas2a

os2dcsi(eameMt

. L. ANDRADE,a A. P. LUIZ,a J. FERREIRAb

ND J. B. CALIXTOa*

Department of Pharmacology, Federal University of Santa Catarina,CB, Campus Universitário Trindade, 88049-900, Florianópolis, SC, Brazil

Department of Chemistry, Federal University of Santa Maria, Santa Maria,S, Brazil

bstract—Ankyrin-repeat transient receptor potential 1 (TRPA1)s a member of the transient receptor potential (TRP) channelamily and it is found in sensory neurons. In the present study,e found that TRPA1 receptor activation with allyl isothiocya-ate or cinnamaldehyde caused dose-dependent spontaneousociception when injected into the mouse hind paw. Very sim-

lar results were obtained when stimulating transient receptorotential vanilloid 1 (TRPV1) receptors with capsaicin. Pretreat-ent with the TRP receptor antagonist Ruthenium Red (1 nmol/aw) inhibited capsaicin-(0.1 nmol/paw) and allyl isothiocya-ate–(1 nmol/paw) induced nociceptive responses. However,he nonselective TRPV1 receptor antagonist capsazepine1 nmol/paw) and the selective TRPV1 receptor antagonist SB66791 (1 nmol/paw) only attenuated capsaicin-induced no-iception. In contrast, the intrathecal treatment with TRPA1ntisense oligodeoxynucleotide (2.5 nmol/site) and the de-eneration of the subset of primary afferent fibers sensitiveo capsaicin significantly reduced allyl isothiocyanate–inducedociception. Consequently to TRPA1 antisense oligodeoxynucle-tide treatment there was a marked decrease of the expression ofRPA1 receptor in both sciatic nervous and spinal cord segments.oreover, capsaicin and allyl isothiocyanate–induced nocicep-

ion were not significantly changed by chemical sympathec-omy produced by guanethidine. The previous degranulation of

ast cells by compound 48/80 and treatment with antagonist H1

eceptor antagonist pyrilamine (400 �g/paw) both significantlynhibited the capsaicin- and allyl isothiocyanate–induced nocicep-ion. The selective NK1 receptor antagonist N2-[(4R)-4-hydroxy-1-1-methyl-1H-indol-3-yl) carbony-1-L-prolyl]-N-methyl-N-phenylm-thyl-3-2-(2-naphtyl)-L-alaninamide (10 nmol/paw) reduced eitherapsaicin- or allyl isothiocyanate–induced nociception. Collec-ively, the present findings demonstrate that the TRPA1 agonistllyl isothiocyanate produces a consistent nociceptive re-ponse when injected into the mouse paw, an effect that seemso be mediated via activation of TRPA1 receptor and dependentn the capsaicin-sensitive fibers, release of histamine by mastells and participation of tachykinins. Thus, the TRPA1 receptor

Corresponding author. Tel: �55-483721-9491; fax: �55-483232-9139.-mail address: calixto@farmaco.ufsc.br or calixto3@terra.com.br

J. B. Calixto).bbreviations: AS-ODN, antisense oligodeoxynucleotide; DRG, dorsal

oot ganglia; ED50, dose of allyl isothiocyanate, cinnamaldehyde or cap-aicin producing nociceptive responses to the order of 50% of the re-ponse relative to the control value; FK 888, N2-[(4R)-4-hydroxy-1-(1-ethyl-1H-indol-3-yl) carbony-1-L-prolyl]-N-methyl-N-phenylmethyl-3-2-

2-naphtyl)-L-alaninamide; i.pl., intraplantarly; i.t., intrathecal; MM-ODN,ismatch oligodeoxynucleotide; PBS, phosphate-buffered solution; SDS,

odium dodecyl sulfate; TRP, transient receptor potential; TRPA1,

dnkyrin-repeat transient receptor potential 1; TRPV1, transient receptorotential vanilloid 1.

306-4522/08$32.00�0.00 © 2008 IBRO. Published by Elsevier Ltd. All rights reseroi:10.1016/j.neuroscience.2007.12.039

511

as an apparently relevant role in nociceptive processes andhe selective TRPA1 antagonist might possess a potential an-inociceptive property. © 2008 IBRO. Published by Elsevier Ltd.ll rights reserved.

ey words: transient potential receptor, allyl isothiocyanate,innamaldehyde, capsaicin, mast cells, histamine.

embers of the transient receptor potential (TRP) family ofonselective cation channels are involved in several physio-

ogical and pathological states (Calixto et al., 2005; Bautistat al., 2006). Transient receptor potential vanilloid 1 (TRPV1)

s expressed in a subset of peripheral pain-sensing neuronsnd is gated by different painful stimuli, including protons,eat, some lipid mediators, and vanilloid compounds such asapsaicin, the pungent principle of chili peppers (Caterina etl., 1997; Tominaga et al., 1998; Zygmunt et al., 1999;wang et al., 2000). An increasing amount of evidence nowuggests that TRPV1 receptor is widely involved in nocicep-ion, contributing to both the detection and integration ofainful chemical and thermal stimuli (Caterina et al., 2000).

Recently, novel members of TRP channels have beenloned. The ankyrin-repeat transient receptor potential 1TRPA1) (formerly named ANKTM1) channel was firstdentified as an ankyrin-like protein, with transmembraneomains, that is lost after oncogenic transformation of humanbroblasts (Jaquemar et al., 1999). TRPA1 has been furtherharacterized as a thermoreceptor activated by noxious coldStory et al., 2003). Similar to TRPV1, some plant-derivedubstances, namely allyl isothiocyanate, cinnamaldehydend allicin isolated from mustard, cinnamon and garlic, re-pectively, are capable of stimulating TRPA1 (Bandell et al.,004; Jordt et al., 2004; Bautista et al., 2005; Macpherson etl., 2005; see for review Calixto et al., 2005).

TRPA1 is selectively expressed by the peptidergic subsetf sensory fibers that also express TRPV1, implying its pos-ible role in pain processing (Bautista et al., 2005; Story et al.,003). In fact, recent findings using knockout mice or knock-own techniques have implicated TRPA1 in the acute andhronic cold nociception processes (Obata et al., 2005; Kat-ura et al., 2006; Kwan et al., 2006; Bautista et al., 2006). It

s also well established that topical application of mustard oilthat is composed mainly of allyl isothiocyanate) producesxcitation of small diameter afferent fibers (Reeh et al., 1986)nd consequently induces pain and hyperalgesia in experi-ental animals and humans (Jiang and Gebhart, 1998; Silvat al., 1997; Laird et al., 2001; Cervero and Laird, 1996).oreover, very recently McNamara et al. (2007) have shown

hat TRPA1 is the principal site through which formalin pro-

uces nociceptive response into the mice paw.

ved.

sstk

A

Atc6wp(aCmdntg

A

Iiil(t0op2snt

cpet

Ri

Tip1pcdC

ii(gd1taii

acatoM

T

twsdl1p1s(fbmtuB

msBSflz((baBsleTcva

Ra

Tn(vdu(tcd0mtd

ct(

E. L. Andrade et al. / Neuroscience 152 (2008) 511–520512

The present study therefore, aimed at investigatingome of the mechanisms underlying the nociceptive re-ponses caused by TRPA1 receptor activation in mice ando compare these effects with that elicited by the well-nown TRPV1 agonist capsaicin.

EXPERIMENTAL PROCEDURES

nimals

dult male Swiss mice weighing 30–35 g were used throughouthe experiments. All animals were housed in a room maintained at aonstant temperature of 22�1 °C under a 12-h light/dark cycle at0–80% humidity with food and water available ad libitum. Animalsere acclimatized to the laboratory for at least 2 h before testing. Allrocedures were approved by our Institutional Ethics Committeeprocess numbers 262/CEUA and 23080.035334/2003-16/UFSC)nd were in accordance with the National Institutes of Health Animalare Guidelines (NIH Publications No. 80-23). The number of ani-als and the nociceptive stimuli were the minimum necessary toemonstrate the consistent effects of drug treatments. To reduce theociceptive intensity in order to minimize the animal suffering, weested the effects of drugs always with sub-maximal doses of algo-enic substances.

lgogen-induced overt nociception in mice

nitially, a time course experiment for the allyl isothiocyanate–nduced nociception was carried out. A volume of 20 �l of allylsothiocyanate (1 nmol/paw) dissolved in phosphate-buffered so-ution (PBS) plus ethanol and Tween was injected intraplantarlyi.pl.) under the surface of the right hind paw. The final concen-rations of ethanol and Tween 80 did not exceed 0.095% and.005%, respectively. Control animals received the same volumef vehicle. Animals were placed individually in chambers (trans-arent glass cylinders of 20 cm in diameter) and were adapted for0 min before algogen or vehicle injection. The amount of timepent licking, raising and shaking the injected paw (indicative ofociception) was timed each minute with a chronometer and theotal over a period of 5 min was considered.

Afterward, complete dose-response curves were plotted forapsaicin (0.01–1 nmol/paw), allyl isothiocyanate (0.1–100 nmol/aw) and cinnamaldehyde (1–100 nmol/paw). After challenge,ach mouse was observed individually for 5 min and the nocicep-ive response was evaluated as described above.

ole of TRP receptor stimulation in allylsothiocyanate–induced nociception

o assess the role of TRP receptors in the nociceptive responsesnduced by allyl isothiocyanate (1 nmol/paw) or capsaicin (0.1 nmol/aw), animals received Ruthenium Red (nonselective TRP inhibitor,nmol/paw), capsazepine (nonselective TRPV1 antagonist, 1 nmol/aw) or SB 366791 (a selective TRPV1 antagonist, 1 nmol/paw). Thehoice of the dose of each drug was selected on the basis of previousata from the literature or on previous experiments (Santos andalixto, 1997a; Ferreira et al., 2005; Claudino et al., 2006).

In order to gain more direct and conclusive evidence that allylsothiocyanate is activating TRPA1 receptor, in a separate exper-ments, animals received the antisense oligodeoxynucleotideAS-ODN; 5= TCTATGCGGTTATGTTGG-3=) or mismatch oli-odeoxynucleotide (MM-ODN; 5=-ACTACTACACTAGACTAC-3=)irected to TRPA1, administrated intrathecally (2.5 nmol/site) each2 h, for 3 days and a last administration 1 h before of nociceptiveest. Several reports have demonstrated that intrathecal (i.t.) ODNccumulates in dorsal root ganglia (DRG) neurons, therefore reach-

ng sufficient concentration (Barclay et al., 2002; Lai et al., 2002). The

.t. injection of ODN was carried out as described previously (Hylden c

nd Wilcox, 1980; Scheidt et al., 2002). Injections were given to fullyonscious mice awake. For this, mice were restrained manually, and30-gauge needle attached to a 25 �l microsyringe, was inserted

hrough the skin and between the vertebras into the subdural spacef the L5–L6 spinal segments. A volume of 5 �l of the AS-ODN orM-ODN was given over a period of 5 s.

RPA1 expression following AS-ODN treatment

Preparation of tissue for Western blot studies. The effect ofhe treatment with AS-ODN on the expression of TRPA1 receptorsas assessed by Western blot analysis. The s.c. tissue of paws,ciatic nerves and portions of spinal cords (site of injection) wereissected, snap-frozen in liquid nitrogen, and homogenized in a

yses buffer containing 10 mM Hepes, pH 7.4; 2 mM MgCl2,0 mM KCl, 1 mM phenylmethylsulphonyl fluoride, 1 �g/ml leu-eptin, 1 �g/ml pepstatin A, 1 �g/ml leupeptin, 1 �g/ml aprotinin,mM sodium orthovanadate, 10 mM �-glycerophosphate, 50 mM

odium fluoride and 0.5 mM dithiothreitol. After centrifugation14,000�g for 45 min), the supernatant containing the cytosolicraction was dispensed. The pellet was reconstituted with lysesuffer added of Triton X. After new centrifugation (14,000�g for 45in), the supernatant containing the solubilized membrane frac-

ion was collected. The protein concentration was determinedsing the Bio-Rad protein assay kit (Bio-Rad, São Paulo, SP,razil).

Western blot analysis. Amounts of proteins (35 �g) wereixed in buffer (Tris 200 mM, glycerol 10%, sodium dodecyl

ulfate (SDS) 2%, �-mercaptoethanol 2.75 mM and Bromophenollue 0.04%) and boiled for 5 min. Proteins were resolved in 7%DS gel by electrophoresis. After transferring to a polyvinylideneuoride (PVDF) membrane (GE Healthcare, São Paulo, SP, Bra-il), the blots were blocked with 5% fat-free dry milk–TBS buffertris-buffered saline) overnight at 4 °C, and then washed with TBSTtris-buffered saline and 5% Tween-20). The membranes were incu-ated overnight at 4 °C with 1:500 dilutions of primary antibodiesnti-ANKTM1 (TRPA1) or Na�,K�-ATPase (� subunity) (Santa Cruziotechnology, Santa Cruz, CA, USA). The expression of the �ubunity of Na�,K�-ATPase was used as a loading control (Gal-agher et al., 2007) and analyzed in same membranes in which wasvaluated the TRPA1 expression. Blots were washed four times withBST for 5 min followed by incubation with adjusted peroxidase-oupled secondary antibody for 1 h. The transferred proteins wereisualized with an enhanced chemiluminescence (ECL) detection kitccording to the manufacturer’s instructions (GE Healthcare).

ole of capsaicin-sensitive fibers, sympathetic fibersnd mast cells in TRPA1 agonist-induced nociception

o explore further the role of capsaicin-sensitive fibers in theociceptive effect induced by allyl isothiocyanate, neonatal miceon day 2 of life) received either capsaicin (50 mg/kg, s.c.) orehicle alone (10% ethanol, 10% Tween-80 and 80% PBS), asescribed previously (Ferreira et al., 2004). The animals weresed 6–7 weeks after the treatment with capsaicin or vehicleused as control). To assess whether a complete degeneration ofhe C sensory fibers had occurred after neonatal treatment withapsaicin, the animals were first submitted to an eye-wiping test, asescribed by Ikeda et al. (2001). For this, 20 �l capsaicin solution.01% (p/v) was instilled in the eye and the number of wiping move-ents that occurred for 1 min was counted. The animals that wiped

heir eyes no more than five times were considered to have beenesensitized by the neonatal capsaicin treatment.

To assess the role of sympathetic fibers in the allyl isothio-yanate or capsaicin-induced nociception, a chemical sympathec-omy was produced by the treatment of mice with guanethidine30 mg/kg, i.p.) 3 days before capsaicin or allyl isothiocyanate

hallenge, as described previously (Ferreira et al., 2005).

tc4e1awaoTtir(p

Rc

Tnirca

D

TctSskspwiabc0mm

S

AvctgdmopbwKwwi

T

Tp

an1it1cdelh8fBTdA

Facl

E. L. Andrade et al. / Neuroscience 152 (2008) 511–520 513

To investigate further the participation of mast cell products inhe allyl isothiocyanate or capsaicin-induced nociception, mastells were degranulated by previous treatment with compound8/80, a mast cell degranulator drug (Mousli et al., 1990). To thisnd, a daily administration for four consecutive days (1, 3, 10,0 �g/paw), before allyl isothiocyanate challenge, was achieveds described previously (Piovezan et al., 2004). To confirmhether or not the degranulation occurred, a separate group ofnimals received the same treatment with compound 48/80, andn the fifth day they received compound 48/80 (10 �g/paw) again.he inhibition of the nociceptive response to compound 48/80 was

aken as being indicative of mast cell degranulation. Also, tonvestigate the role of histamine, a separate group of animalseceived the H1 histaminergic receptor antagonist pyrilamine400 �g/paw) coadministered with allyl isothiocyanate (1 nmol/aw) or capsaicin (0.1 nmol/paw).

ole of tachykinins in allyl isothiocyanate orapsaicin-induced nociception

o assess the participation of tachykinins in the allyl isothiocya-ate- or capsaicin-induced nociception, mice received a coadmin-

stration of these agonists in association with the selective NK1

eceptor antagonist N2-[(4R)-4-hydroxy-1-(1-methyl-1H-indol-3-yl)arbony-1-L-prolyl]-N-methyl-N-phenylmethyl-3-2-(2-naphtyl)-L-laninamide (FK 888) (10 nmol/paw).

rugs

he following drugs were used: capsaicin, allyl isothiocyanate,innamaldehyde, compound 48/80, guanethidine, pyrilamine, ru-henium red (ammoniated ruthenium oxychloride) (all from Sigma,t. Louis, MO, USA), SB 366791 and capsazepine (Tocris Cook-on Inc., Ellisville, USA). The NK1 receptor antagonist FK 888 wasindly supplied by Fujisawa Pharmaceutical (Osaka, Japan). Thetock solutions of capsaicin, FK 888 and SB 366791 were pre-ared in absolute ethanol and put into siliconized plastic tubeshich were kept in a freezer at �18 °C until use. Solutions of allyl

sothiocyanate and cinnamaldehyde were made in PBS plus eth-nol and Tween 80 and diluted on the day of experiment justefore use. Other drugs were dissolved in PBS. The maximal finaloncentrations of ethanol and Tween 80 were 0.0095% and.0005%, respectively and had no effect per se, or in agonist-ediated responses. Antisense and MM-ODN were designed andanufactured by Biognostik, Germany.

tatistical analysis

ll values are expressed as mean�S.E.M., except for the ED50

alues (i.e. the dose of allyl isothiocyanate, cinnamaldehyde orapsaicin producing nociceptive responses to the order of 50% ofhe response relative to the control value), which are given as theeometric means accompanied by their respective 95% confi-ence limits. The percentages of inhibition are reported as theean�S.E.M., calculated for the maximal developed responsesbtained after injection of allyl isothiocyanate or capsaicin in com-arison with vehicle-treated animals. The statistical significanceetween groups was assessed by unpaired Student’s t-test or one-ay analysis of variance followed by Dunnett’s or Student-Newman-euls’ tests when appropriate. P values less than 0.05 (P�0.05)ere considered to be indicative of significance. The ED50 valuesere determined by linear regression analysis from individual exper-

ments using GraphPad Software 4.0 (GraphPad, USA).

RESULTS

RPA1 or TRPV1 agonist-induced nociception

he intraplantar injection of allyl isothiocyanate (1 nmol/

aw) caused a rapid onset pain response, being maximal

sc

s early as 1 min after its administration, decreasing in theext minutes, and disappearing altogether at 5 min (Fig.A). Based on these preliminary results, the further exper-

ments were conducted within 5 min after algogen injec-ion. The intraplantar administration of capsaicin (0.01–

nmol/paw), allyl isothiocyanate (0.1–100 nmol/paw) orinnamaldehyde (1–100 nmol/paw) to mice produced aose-dependent spontaneous nociception (Fig. 1B). Thestimated mean ED50 values (and the 95% confidence

imits) for capsaicin, allyl isothiocyanate and cinnamalde-yde were 0.05 (0.04–0.07), 0.6 (0.4–0.8) and 7.1 (6.1–.2) nmol/paw, and the maximal observed nociceptive ef-ects were 140�8, 111�30 and 129�16 s, respectively.ased on these results, allyl isothiocyanate was theRPA1 agonist chosen to carry out further experimentsue its greater potency compared with cinnamaldehyde.dditionally, we induced nociception further with submaxi-

ig. 1. Temporal course for the nociception caused by i.pl. injection ofllyl isothiocyanate (A). Dose-response curves for the nociceptionaused by i.pl. injection of capsaicin, allyl isothiocyanate or cinnama-dehyde in mice (B). The effects of the drugs are expressed as time

pent licking, raising and shaking the injected paw. Each point on theurve represents the mean�S.E.M. of six to eight mice.

mnl

Rnt

aer1i

n

F(ri

E. L. Andrade et al. / Neuroscience 152 (2008) 511–520514

al doses of capsaicin (0.1 nmol/paw) or allyl isothiocya-ate (1 nmol/paw) in order to avoid supramaximal stimu-

ation and consequent unnecessary animal discomfort.The coadministration of the TRP receptor antagonists

uthenium Red (1 nmol/paw) significantly reduced theociception induced by capsaicin (83.4�9.4%) or allyl iso-

hiocyanate (84.3�9.4%) (Fig. 2A, B). However, the co-

ig. 2. Effect of i.pl. treatment with the TRP receptor antagonist Rut1 nmol/paw, C and D) or SB 366791 (1 nmol/paw, E and F) on caps

epresents the mean�S.E.M. of six to eight mice. The asterisks denote the signsothiocyanate–treated mice # (one-way ANOVA followed by Student-Newman

dministration of the TRPV1 receptor antagonists capsaz-pine (1 nmol/paw) or SB 366791 (1 nmol/paw) significantlyeduced capsaicin-induced nociceptive responses by 72.2�0.6% and 68.2�5.5%, while causing no significant inhibition

n the allyl isothiocyanate–induced nociception (Fig. 2C–F).Of interest, we found that allyl isothiocyanate–induced

ociception, but not the capsaicin-induced nociception,

ed (1 nmol/paw, A and B), TRPV1receptor antagonist capsazepineallyl isothiocyanate–induced nociception, respectively. Each column

henium Raicin- or

ificance levels. * P�0.05, ** P�0.01, compared with capsaicin- or allyl-Keuls’ test).

wTdi

Ts

Tpcep

mwa

Tds

Ttas

F(cToaw

E. L. Andrade et al. / Neuroscience 152 (2008) 511–520 515

as significantly reduced by previous treatment with theRPA1 AS-ODN (78.0�8.5%) (Fig. 3A, B). Thus, knock-own of the TRPA1 gene specifically reduced the allyl

sothiocyanate–induced nociception in mice.

RPA1 protein expression in paw, sciatic nerve andpinal cord following AS-ODN treatment

he treatment with AS-ODN significantly reduced the ex-ression of TRPA1 protein in both sciatic nerve and spinalord in comparison to MM-ODN treatment (Fig. 3C). How-ver, there was no difference in the expression of TRPA1

ig. 3. Effect of i.t. treatment of TRPA1 AS-ODN (2.5 nmol/site) or MMA and B, respectively). Each column represents the mean�S.E.M. oompared with MM-ODN (2.5 nmol/site) -treated mice # (one-way ANORPA1 (1) and Na�,K�-ATPase (2) expression in mouse paw, sciaticr with MM-ODN (2.5 nmol/site) (C). The expression of the � subunitys arbitrary density unit. Each column represents the mean�S.E.M. ofith the respective MM-ODN group # (Student’s unpaired t-test).

rotein between AS-ODN and MM-ODN groups in the c

ouse paw (Fig. 3C). Moreover, Na�,K�-ATPase (whichas used as loading control) expression was not alteredfter AS-ODN treatment (Fig. 3C).

RPA1 agonist-induced nociception afteregeneration of the C sensory fibers, chemicalympathectomy and degranulation of mast cells

o investigate further the role played by capsaicin-sensi-ive fibers in TRPA1 agonist-mediated nociception, thenimals were treated during the neonatal period with cap-aicin (50 mg/kg, s.c.). As previously demonstrated for

.5 nmol/site) on capsaicin- or allyl isothiocyanate–induced nociceptionight mice. The asterisks denote the significance levels. *** P�0.001,wed by Student-Newman-Keuls’ test). Western blot analysis showingd spinal cord after i.t. treatment with TRPA1 AS-ODN (2.5 nmol/site)�-ATPase was used as a loading control. Results were represented

ce. The asterisks denote the significance levels. * P�0.05, compared

-ODN (2f six to eVA follonerve anof Na�,Kthree mi

apsaicin (Ferreira et al., 2004), the neonatal treatment with

c(sda

pde

wtc(4man

FwbnmN

E. L. Andrade et al. / Neuroscience 152 (2008) 511–520516

apsaicin reduced allyl isothiocyanate–induced nociceptionFig. 4A) (95.9�4.1%). On the other hand, the chemicalympathectomy produced by guanethidine (30 mg/kg, i.p., 3ays before) did not significantly change the capsaicin- orllyl isothiocyanate–induced nociception (Fig. 4D, E).

The previous mast cell degranulation caused by com-ound 48/80 (1, 3, 10 and 10 �g/paw) for four consecutiveays resulted in a significant inhibition of the nociceptive

ig. 4. Disruption of the nociceptive effect caused by the i.pl. injectionith capsaicin (50 mg/kg, s.c.). Influence of pretreatment with compouny i.pl. injection of capsaicin (0.1 nmol/paw, B) or allyl isothiocyanaociception induced by capsaicin (0.1 nmol/paw, D) or allyl isothiocyanice. The asterisks denote the significance levels. * P�0.05, ** P�ewman-Keuls’ test).

ffect produced by either capsaicin or allyl isothiocyanate, s

ith inhibitions of 44.9�2.2% and 67.9�10.3%, respec-ively (Fig. 4B, C). Likewise, the nociceptive responseaused by compound 48/80 was significantly inhibited66.3�13.5%) in animals previously treated with compound8/80 (results not shown). The coadministration of the hista-inergic H1 receptor antagonist pyrilamine (400 �g/paw)lso significantly reduced either capsaicin or allyl isothiocya-ate–induced nociception (57.0�9.9% and 75.4�5.9%, re-

othiocyanate (1 nmol/paw, A) produced by treatment of neonate mice1, 3, 10, 10 �g/paw) for four consecutive days on the effects induciblel/paw, C) on the 5th day. Effect of treatment with guanethidine onol/paw, E). Each column represents the mean�S.E.M. of six to eightpared with control group # (one-way ANOVA followed by Student-

of allyl isd 48/80 (

te (1 nmoate (1 nm0.01 com

pectively) (Fig. 5A, B).

Ia

Tm

cm(ai

TTcwtsptgveotib

troabpmJt2daatib

FpscicS

F(s

E. L. Andrade et al. / Neuroscience 152 (2008) 511–520 517

nvolvement of tachykinins in TRPA1 or TRPV1gonist-induced nociception

o investigate further the role played by tachykinins,ainly substance P, through NK1 receptor in allyl isothio-

ig. 5. Influence of i.pl. treatment with the H1 receptor antagonistyrilamine (400 �g/paw) on the nociceptive effects produced by cap-aicin (0.1 nmol/paw, A) or allyl isothiocyanate (1 nmol/paw, B). Eacholumn represents the mean�S.E.M. of six to eight mice. The aster-

sks denote the significance levels. ** P�0.01 compared with controlsapsaicin and allyl isothiocyanate # (one-way ANOVA followed bytudent-Newman-Keuls’ test).

ig. 6. Influence of i.pl. treatment with the NK1 receptor antagonis

0.1 nmol/paw, A) or allyl isothiocyanate (1 nmol/paw, B). Each column represignificance levels. * P�0.05 compared with controls capsaicin and allyl isothioc

yanate–mediated nociception, a separate group of ani-als received the NK1 receptor antagonist FK 888

10 nmol/paw). FK888 decreased in about 45.8�10.5%nd 68.9�10.5% the capsaicin- or allyl isothiocyanate–

nduced nociception, respectively (Fig. 6A, B).

DISCUSSION

he data from the present study demonstrated that theRPA1 agonist, allyl isothiocyanate, like the TRPV1 agonistapsaicin, elicited a dose dependent nociceptive responsehen injected i.pl. into the mouse paw. Our results suggest

hat these nociceptive responses seem to be mediated, re-pectively, by the activation of TRPA1 and TRPV1 receptorsresent in small diameter afferent fibers, an event that seems

o be mediated by release of histamine from mast cells. Ofreat interest, both allyl isothiocyanate and capsaicin hadery similar efficacy, but with different potencies. This differ-nce in the potency observed in the present study reinforcesur previous findings (Andrade et al., 2006) demonstrating

hat capsaicin was about 4000-fold more potent that allylsothiocyanate in inducing contractile response in rat urinaryladder in vitro.

Our present results also confirm and largely extendhose from a previous study of Bandell et al. (2004), whoeported a nociceptive response after intraplantar injectionf cinnamaldehyde in mice. Moreover, other studies havelready shown that mustard oil produces nociception inoth experimental animals and humans. It has been re-orted that topic administration or intra-articular injection ofustard oil causes hyperalgesia in rats (Silva et al., 1997;

iang and Gebhart, 1998), while its intracolonic adminis-ration causes visceral nociception in mice (Laird et al.,001). Recently, it has been demonstrated that TRPA1-eficient mice present a reduced nociceptive responsefter s.c. administration of allyl isothiocyanate (Bautista etl., 2006; Kwan et al., 2006). In sensory neurons of ratrigeminal ganglia, the application of mustard oil evokes anncrease in intracellular calcium and elicits large mem-rane currents (Jordt et al., 2004). Moreover, the topic

(10 nmol/paw) on the nociceptive effects produced by capsaicin

t FK 888 ents the mean�S.E.M. of six to eight mice. The asterisks denote theyanate # (one-way ANOVA followed by Student-Newman-Keuls’ test).

ah

antcnnvtnd

stcm(cSdTdrntcesna2n(tTiaislec

sAanirrgomteao

tts

mtcDsrTste

ratnsbncsdn

pocicrpthorchcTcmsbrsit

tcmstd

E. L. Andrade et al. / Neuroscience 152 (2008) 511–520518

pplication of mustard oil is also capable of causing pain inumans (Cervero and Laird, 1996).

The nociceptive responses produced by capsaicin orllyl isothiocyanate were significantly reduced by Ruthe-ium Red, an inorganic polycationic dye that non-selec-

ively blocks response to several members of the TRPhannel family, suggesting that TRPV1 and TRPA1 ago-ists might be able to activate calcium influx to induceociception. These results are in line with our own pre-ious findings (Andrade et al., 2006) indicating that Ru-henium Red inhibits both capsaicin and allyl isothiocya-ate–induced contractile responses in rat urinary blad-er in vitro.

However, we have found that treatment with the non-elective TRPV1 antagonist capsazepine or with the selec-ive TRPV1 antagonist SB 366791 is effective in antagonizingapsaicin-induced nociception, but not allyl isothiocyanate–ediated nociceptive response. Of interest, Varga et al.

2005) have demonstrated the effect of SB 366791 onapsaicin-evoked responses both in vitro and in vivo in rat.uch results suggest that TRPV1 receptor is probably notirectly involved in the nociceptive response evoked byRPA1 stimulation through allyl isothiocyanate. In accor-ance with our data, mice lacking TRPV1 receptor presenteduced nociceptive response to capsaicin, but normalociception to mustard oil or cinnamaldehyde, indicatinghat TRPV1 and TRPA1 agonists likely interact with noci-eptors through distinct molecular mechanisms (Caterinat al., 2000; Bandell et al., 2004). On the other hand,ensory neurons from TRPA1 knockout mice displayedormal sensitivity to capsaicin, but reduced response tollyl isothiocyanate (Bautista et al., 2006; Kwan et al.,006). Moreover, mice lacking TRPA1 did not presentociceptive response after allyl isothiocyanate injectionBautista et al., 2006; Kwan et al., 2006). Contributing withhe study about a possible interaction between TRPA1 andRPV1 receptors, we demonstrated herein that i.t. admin-

stration of TRPA1 AS-ODN was capable of reducing thellyl isothiocyanate–induced nociception, without interfer-

ng significantly with capsaicin-mediated nociceptive re-ponse. Such finding provided additional support for theack of direct involvement of TRPA1 receptor in nociceptiveffect produced by stimulation of TRPV1 receptor withapsaicin.

The reduction of the TRPA1 protein expression in bothpinal cord and sciatic nerve confirmed the efficiency of theS-ODN treatment, which seems to reach to lumbar DRGnd consequently to reduce TRPA1 protein in sensoryeuron terminals. Interestingly, this is the first report show-

ng TRPA1 protein expression in spinal cord. Our result inelation to TRPA1 expression in spinal cord confirms aecent study carried out by Kosugi et al. (2007) that sug-ested the localization of TRPA1 at presynaptic terminalsn substantia gelatinosa neurons and in terminals of pri-ary afferents innervating onto inhibitory interneurons of

he rat spinal cord. An additional result was the TRPA1xpression in the paw of the mice not to present alterationfter AS-ODN treatment. Our result also suggests that

ther cells than neurons in paw tissue express TRPA1 and r

hat TRPA1 protein in sensory neuron terminals in suchissue has a minor part in the total expression. However,tudies must be carried out to elucidate this point.

It has been well established that the neonatal treat-ent of animals with capsaicin is able to produce a selec-

ive degeneration of C and some A� (Holzer, 1991), asso-iated with decrease of the expression of TRPV1 at theRG (Rashid et al., 2003). Our data clearly show that theame neonatal treatment with capsaicin also significantlyeduced the allyl isothiocyanate–mediated nociception.hese results are in accordance with a recent findinghowing that the expression of TRPA1 and TRPV1 recep-ors is restricted to putative C-fibers neurons (Kobayashit al., 2005).

Besides the sensory neurons, the mRNA for TRPA1eceptor is also found in sympathetic neurons (Smith etl., 2004). Furthermore, sympathetic fibers are thought

o be involved in the development of inflammatory andeuropathic pain, delivering pronociceptive mediatorsuch as noradrenalin and prostaglandins (Safieh-Gara-edian et al., 1997, 2002). Our findings show that theociceptions caused by capsaicin and by allyl isothio-yanate are probably unrelated with the stimulation ofympathetic fibers, since chemical sympathectomy pro-uced by guanethidine was not capable of altering theirociceptive responses.

An interesting and new finding that emerges from theresent study is the fact that the previous degranulationf mast cells by treatment with compound 48/80 effi-iently reduces both allyl isothiocyanate- and capsaicin-

nduced nociception. Moreover, allyl isothiocyanate- andapsaicin-induced nociceptions were also significantlyeduced by treatment with the H1 receptor antagonistyrilamine. Massaad et al. (2004) have shown that in-

raplantar treatment with an H1 receptor antagonist in-ibits the hyperalgesia induced by intraplantar injectionf capsaicin in rat. A possible explanation for theseesults would be the direct stimulation of mast cells byapsaicin and by allyl isothiocyanate. In this respect, itas been reported that TRPV1 mRNA is detected inultured mast cells (Stokes et al., 2004) and that theRPV1 protein has also been found in human skin mastells (Stander et al., 2004). Contrasting with the directast cell degranulation by capsaicin, another study has

hown that capsaicin is able to induce calcium uptake,ut not degranulation in mast cell lines and bone mar-ow– derived mast cells (Biro et al., 1998). So far, it istill undefined whether or not TRPA1 is also expressed

n mast cells. Further studies are required to elucidatehis point.

Besides stimulating mast cells directly through recep-ors TRPV1 and TRPA1, capsaicin and allyl isothiocyanateould stimulate mast cells indirectly via excitation of pri-ary afferent fibers with the consequent release of sub-

tance P. In fact, it is well known that substance P is ableo activate mast cells through tachykinin NK1 receptorependent or independent mechanisms, leading to the

elease of histamine and other mediators (Ansel et al.,

1e

omAtruercstbtbNfBstls

Tnaosica

AC(TldNcs

A

A

B

B

B

B

B

B

C

C

C

C

C

D

F

F

G

H

H

H

I

J

J

J

K

E. L. Andrade et al. / Neuroscience 152 (2008) 511–520 519

993; Dimitriadou et al., 1997; Okada et al., 1999; Suzukit al., 1999).

Our data show that the selective NK1 receptor antag-nist FK888 decreased the nociception caused by sub-aximal doses of both allyl isothiocyanate and capsaicin.ccordingly, capsaicin is well known to produce nocicep-

ion, an event that is probably mediated by tachykinineceptor stimulation when a higher dose of capsaicin issed (Santos and Calixto, 1997b). Interestingly, Bánvölgyit al. (2004) have showed that mustard oil–induced neu-ogenic ear edema is mediated by activation of NK1 re-eptor. Therefore, our findings reinforce the concept thatub-maximal activation of both TRPV1 and TRPA1 recep-ors produces a nociception which seems to be mediatedy tachykinins, especially substance P. However, the par-

icipation of other tachykinins cannot be fully excluded,ecause neurokinin A and neurokinin B can interact withK1 receptor, although NK1 receptor exhibits preference

or substance P (Mussap et al., 1993; Regoli et al., 1994).ased on the above outcomes, it is quite plausible that thetimulation of mast cells by both capsaicin and allyl iso-hiocyanate could be likely mediated indirectly by the re-ease of substance P. However, additional experimentshould be done to confirm this hypothesis.

Collectively, the present findings demonstrate that theRPA1 agonist allyl isothiocyanate caused a consistentociceptive response when injected into the mouse paw,n effect that seems to have been mediated via activationf TRPA1 receptor and dependent on the capsaicin-sen-itive fibers, release of histamine by mast cells and partic-

pation of tachykinins via NK1 receptor. Thus, TRPA1 re-eptor is a potential target for the development of newnalgesics.

cknowledgments—This study was supported by grants from theoordenação de Aperfeiçoamento de Pessoal de Ensino Superior

CAPES), the Conselho Nacional de Desenvolvimento Científico eecnológico (CNPq), the Fundação de Apoio a Ciência e Tecno-

ogia do Estado de Santa Catarina (FAPESC), the Financiadorae Estudos e Projetos (FINEP) and the Programa de Apoio aosúcleos de Excelência (PRONEX), Brazil. We thank the pharma-eutical company for the kind donation of FK 888 used in thistudy. E.L.A. is a CNPq PhD fellowship student in pharmacology.

REFERENCES

ndrade EL, Ferreira J, André E, Calixto JB (2006) Contractile mech-anisms coupled to TRPA1 receptor activation in rat urinary bladder.Biochem Pharmacol 72:104–114.

nsel JC, Brown JR, Payan DG, Brown MA (1993) Substance Pselectively activates TNF-alpha gene expression in murine mastcells. J Immunol 150:4478–4485.

andell M, Story GM, Hwang SW, Viswanath V, Eid SR, Petrus MJ,Earley TJ, Patapoutian A (2004) Noxious cold ion channel TRPA1is activated by pungent compounds and bradykinin. Neuron41:849–857.

ánvölgyi A, Pozsgai G, Brain SD, Helyes ZS, Szolcsányi J, Ghosh M,Melegh B, Pintér E (2004) Mustard oil induces a transient receptorpotential vanilloid 1 receptor-independent neurogenic inflammationand a non-neurogenic cellular inflammatory component in mice.Neuroscience 125:449–459.

arclay J, Patel S, Dorn G, Wotherspoon G, Moffatt S, Eunson L,

Abdel’al S, Natt F, Hall J, Winter J, Bevan S, Wishart W, Fox A,

Ganju P (2002) Functional downregulation of P2X3 receptorsubunit in rat sensory neurons reveals a significant role in chronicneuropathic and inflammatory pain. J Neurosci 22:8139–8147.

autista DM, Jordt SE, Nikai T, Tsuruda PR, Read AJ, Poblete J,Yamoah EN, Basbaum AI, Julius D (2006) TRPA1 mediates theinflammatory actions of environmental irritants and proalgesicagents. Cell 124:1269–1282.

autista DM, Movahed P, Hinman A, Axelsson HE, Sterner O, Hoges-tatt ED, Julius D, Jordt SE, Zygmunt PM (2005) Pungent productsfrom garlic activate the sensory ion channel TRPA1. Proc NatlAcad Sci U S A 102:12248–12252.

iro T, Maurer M, Modarres S, Lewin NE, Brodie C, Acs G, Acs B,Paus R, Blumberg PM (1998) Characterization of functional va-nilloid receptors expressed by mast cells. Blood 91:1332–1340.

alixto JB, Kassuya CA, André E, Ferreira J (2005) Contribution ofnatural products to the discovery of the transient receptor (TRP)channels family and their functions. Pharmacol Ther 106:179–208.

aterina MJ, Leffler A, Malmberg AB, Martin WJ, Trafton J, Petersen-Zeitz KR, Koltzenburg M, Basbaum AI, Julius D (2000) Impairednociception and pain sensation in mice lacking the capsaicin re-ceptor. Science 288:306–313.

aterina MJ, Schumacher MA, Tominaga M, Rosen TA, Levine JD,Julius D (1997) The capsaicin receptor: a heat-activated ion chan-nel in the pain pathway. Nature 389:816–824.

ervero F, Laird JM (1996) Mechanisms of allodynia: interactionsbetween sensitive mechanoreceptors and nociceptors. Neurore-port 7:526–528.

laudino RF, Kassuya CA, Ferreira J, Calixto JB (2006) Pharmaco-logical and molecular characterization of the mechanisms involvedin prostaglandin E2-induced mouse paw edema. J Pharmacol ExpTher 318:611–618.

imitriadou V, Rouleau A, Tuong MDT, Newlands GJF, Miller HRP,Luffau G, Schwartz JC, Garbarg M (1997) Functional relationshipsbetween sensory nerve fibers and mast cells of dura mater innormal and inflammatory conditions. Neuroscience 77:829–839.

erreira J, Da Silva GL, Calixto JB (2004) Contribution of vanilloidreceptors to the spontaneous nociception induced by B2 kininreceptor activation in mice. Br J Pharmacol 141:787–794.

erreira J, Trichês KM, Medeiros R, Calixto JB (2005) Mechanismsinvolved in the nociception produced by peripheral protein kinase cactivation in mice. Pain 117:171–181.

allagher MJ, Ding L, Maheshwari A, Macdonald RL (2007) TheGABAA receptor alpha1 subunit epilepsy mutation A322D inhibitstransmembrane helix formation and causes proteasomal degrada-tion. Proc Natl Acad Sci U S A 104:12999–13004.

olzer P (1991) Capsaicin: cellular targets, mechanisms of action, andselectivity for thin sensory neurons. Pharmacol Rev 43:14–200.

wang SW, Cho H, Kwak J, Lee SY, Kang CJ, Jung J, Cho S, Min KH,Suh YG, Kim D, Oh U (2000) Direct activation of capsaicin receptorby products of lipoxygenases: endogenous capsaicin-like sub-stances. Proc Natl Acad Sci U S A 97:6155–6160.

ylden JLK, Wilcox GL (1980) Intrathecal morphine in mice: A newtechnique. Eur J Pharmacol 67:313–316.

keda Y, Ueno A, Naraba H, Oh-ishi S (2001) Involvement of vanilloidreceptor VR1 and prostanoids in the acid-induced writhing re-sponses of mice. Life Sci 69:2911–2919.

aquemar D, Schenker T, Trueb B (1999) An ankyrin-like protein withtransmembrane domains is specifically lost after oncogenic trans-formation of human fibroblasts. J Biol Chem 274:7325–7333.

iang MC, Gebhart GF (1998) Development of mustard oil-inducedhyperalgesia in rats. Pain 77:305–313.

ordt SE, Bautista DM, Chuang HH, McKemy DD, Zygmunt PM,Hogestatt ED, Meng ID, Julius D (2004) Mustard oils and canna-binoids excite sensory nerve fibres through the TRP channelANKTM1. Nature 427:260–265.

atsura H, Obata K, Mizushima T, Yamanaka H, Kobayashi K, Dai Y,

Fukuoka T, Tokunaga A, Sakagami M, Noguchi K (2006) Antisense

K

K

K

L

L

M

M

M

M

M

O

O

P

R

R

R

S

S

S

S

S

S

S

S

S

S

S

T

V

Z

E. L. Andrade et al. / Neuroscience 152 (2008) 511–520520

knock down of TRPA1, but not TRPM8, alleviates cold hyperalgesiaafter spinal nerve ligation in rats. Exp Neurol 200:112–123.

obayashi K, Fukuoka T, Obata K, Yamanaka H, Dai Y, Tokunaga A,Noguchi K (2005) Distinct expression of TRPM8, TRPA1, and TRPV1mRNAs in rat primary afferent neurons with adelta/c-fibers and colo-calization with trk receptors. J Comp Neurol 493:596–606.

osugi M, Nakatsuka T, Fujita T, Kuroda Y, Kumamoto E (2007)Activation of TRPA1 channel facilitates excitatory synaptic trans-mission in substantia gelatinosa neurons of the adult rat spinalcord. J Neurosci 27:4443–4451.

wan KY, Allchorne AJ, Vollrath MA, Christensen AP, Zhang DS,Woolf CJ, Corey DP (2006) TRPA1 contributes to cold, mechani-cal, and chemical nociception but is not essential for hair-celltransduction. Neuron 50:277–289.

ai J, Gold MS, Kim CS, Bian D, Ossipov MH, Hunter JC, Porreca F(2002) Inhibition of neuropathic pain by decreased expression of thetetrodotoxin-resistant sodium channel, NaV1.8. Pain 95:143–152.

aird JMA, Martinez-Caro L, Garcia-Nicas E, Cervero F (2001) A newmodel of visceral and referred hyperalgesia in the mouse. Pain92:335–342.

acpherson LJ, Geierstanger BH, Viswanath V, Bandell M, Eid SR,Hwang S, Patapoutian A (2005) The pungency of garlic: activationof TRPA1 and TRPV1 in response to allicin. Curr Biol 15:929–934.

assaad CA, Safieh-Garabedian B, Poole S, Atweh SF, Jabbur SJ,Saadé NE (2004) Involvement of substance P, CGRP and histaminein the hyperalgesia and cytokine upregulation induced by intraplantarinjection of capsaicin in rats. J Neuroimmunol 153:171–182.

cNamara CR, Mandel-Brehm J, Bautista DM, Siemens J, Kari L,Deranian KL, Zhao M, Hayward NJ, Chong JA, Julius D, MoranMM, Fanger CM (2007) TRPA1 mediates formalin-induced pain.Proc Natl Acad Sci U S A 104:13525–13530.

ousli M, Bueb JL, Bronner C, Rouot B, Landry Y (1990) G proteinactivation: a receptor-independent mode of action for cationic am-phiphilic neuropeptides and venom peptides. Trends PharmacolSci 11:358–362.

ussap CJ, Geraghty DP, Burcher E (1993) Tachykinin receptors: aradioligand binding perspective. J Neurochem 60:1987–2009.

bata K, Katsura H, Mizushima T, Yamanaka H, Kobayashi K, Dai Y,Fukuoka T, Tokunaga A, Tominaga M, Noguchi K (2005) TRPA1induced in sensory neurons contributes to cold hyperalgesia afterinflammation and nerve injury. J Clin Invest 115:2393–2401.

kada T, Hirayama Y, Kishi S, Miyayasu K, Hiroi J, Fujii T (1999)Functional neurokinin NK-1 receptor expression in rat peritonealmast cells. Inflamm Res 48:274–279.

iovezan AP, D’Orleans-Juste P, Frighetto M, Souza GE, HenriquesMG, Rae GA (2004) Endothelins contribute towards nociceptioninduced by antigen in ovalbumin-sensitised mice. Br J Pharmacol141:755–763.

ashid MH, Inoue M, Kondo S, Kawashima T, Bakoshi S, Ueda H(2003) Novel expression of vanilloid receptor 1 on capsaicin-insen-sitive fibers accounts for the analgesic effect of capsaicin cream inneuropathic pain. J Pharmacol Exp Ther 304:940–948.

eeh PW, Kocher L, Jung S (1986) Does neurogenic inflammationalter the sensitivity of unmyelinated nociceptors in the rat? Brain

Res 384:42–50.

egoli D, Boundon A, Fauchere JL (1994) Receptors and antagonistsfor substance P and related peptides. Pharmacol Rev 46:551–599.

afieh-Garabedian B, Kanaan SA, Haddad JJ, Jaoude PA, Jabbur SJ,Saade NE (1997) Involvement of interleukin-1 beta, nerve growthfactor and prostaglandin E2 in endotoxin-induced localized inflam-matory hyperalgesia. Br J Pharmacol 121:1619–1626.

afieh-Garabedian B, Poole S, Haddad JJ, Massaad CA, Jabbur SJ,Saade NE (2002) The role of the sympathetic efferents in endo-toxin-induced localized inflammatory hyperalgesia and cytokineupregulation. Neuropharmacology 42:864–872.

antos AR, Calixto JB (1997a) Ruthenium red and capsazepine an-tinociceptive effect in formalin and capsaicin models of pain inmice. Neurosci Lett 235:73–76.

antos AR, Calixto JB (1997b) Further evidence for the involvement oftachykinin receptor subtypes in formalin and capsaicin models ofpain in mice. Neuropeptides 31:381–389.

cheidt C, Santos AR, Ferreira J, Malheiros A, Cechinel-Filho V,Yunes RA, Calixto JB (2002) Evidence for the involvement ofglutamatergic receptors in the antinociception caused in mice bythe sesquiterpene drimanial. Neuropharmacology 43:340–347.

ilva E, Cleland CL, Gebhart GF (1997) Contributions of glutamatereceptors to the maintenance of mustard oil-induced hyperalgesiain spinalized rats. Exp Brain Res 117:379–388.

mith MP, Beacham D, Ensor E, Koltzenburg M (2004) Cold-sensitive,menthol-insensitive neurons in the murine sympathetic nervoussystem. Neuroreport 15:1399–1403.

tander S, Moormann C, Schumacher M, Buddenkotte J, Artuc M,Shpacovitch V, Brzoska T, Lippert U, Henz BM, Luger TA, MetzeD, Steinhoff M (2004) Expression of vanilloid receptor subtype 1 incutaneous sensory nerve fibers, mast cells, and epithelial cells ofappendage structures. Exp Dermatol 13:129–139.

tokes AJ, Shimoda LM, Koblan-Huberson M, Adra CN, Turner H(2004) A TRPV2-PKA signaling module for transduction of physicalstimuli in mast cells. J Exp Med 200:137–147.

tory GM, Peier AM, Reeve AJ, Eid SR, Mosbacher J, Hricik T, EarleyT, Hergarden A, Andersson D, Hwang S (2003) ANKTM1, a TRP-like channel expressed in nociceptive neurons, is activated by coldtemperatures. Cell 112:819–829.

uzuki R, Furuno T, McKay DM, Wolvers D, Teshima R, Nakanishi M,Bienenstock J (1999) Direct neurite-mast cell communication invitro occurs via the neuropeptide substance P. J Immunol163:2410–2415.

ominaga M, Caterina MJ, Malmberg AB, Rosen TA, Gilbert H, Skin-ner K, Raumann BE, Basbaum AI, Julius D (1998) The clonedcapsaicin receptor integrates multiple pain-producing stimuli. Neu-ron 21:531–543.

arga A, Németh J, Szabó A, McDougall JJ, Zhang C, Elekes K, PintérE, Szolcsányi J, Helves Z (2005) Effects of the novel TRPV1receptor antagonist SB366791 in vitro and in vivo in the rat. Neu-rosci Lett 385:137–142.

ygmunt PM, Petersson J, Andersson DA, Chuang H, Sorgard M, DiMarzo V, Julius D, Hogestatt ED (1999) Vanilloid receptors onsensory nerves mediate the vasodilator action of anandamide.

Nature 400:452–457.

(Accepted 27 December 2007)(Available online 9 January 2008)