Calcitonin gene-related peptide content, basal outflow and electrically-evoked release from monoarthritic rat spinal cord in vitro

P 66a(1996) i351-358 n , Associationfor the Studyof Pain.All rightsreserved0304-3959/96/$ 15.(XI @ 1996 International

PAIN3036

Calcitonin gene-related peptide content, basal outflow and electrically-evokedrelease from monoarthritic rat spinal cord in vitro Marzia Malcangio* and Norman G. Bowery Department of Pharmacology, The School of Pharmacy, 29/39 Brunswick Square, London WCIN IAX, (UK)

(Received1 June1995,accepted16November1995)

In this study, Freund’sadjuvant-inducedmonoarthritisin the rat hind paw was used to inducechronic pain and inflammation.In order to compare the basal outflow,electrically-evokedreleaseand total content of calcitonin gene-relatedpeptidelikeimrnunoreactivity(CGRP-LI) with previouslyreportedchangesin substanceP (SP-LI), the lumbar enlargement of monoarthitic (complete Freund’s adjuvant-treated, CFA rat) and control (incomplete Freund’sadjuvant-treated,IFA rat) spinalcords were used. During the 4-wk period after injection,neither the basal nor the evokedreleaseof CGRP-LI from CFA cords differedfrom controls. Bycontrast, we have previouslyreported that SP-LI releasefrom CFA rat spinal cords was significantlyhigher than from controls, 21 days after inoculation with Freund’sadjuvant. Electrically-evokedCGRP-LI releasefrom 21-dayCFA rat spinalcord sliceswas not modified by superfusionwith a GABA~ antagonist, CGP 36742(100 KM) which could greatly increase SP-LI release. However,the release of both peptideswas significantlyincreasedto the same extent in IFA and normal tissue but to a lesserextentin CFA cords, by superfusionwith the opioidantagonistnaloxone(1 PM). In conclusion,CGRP-LI, unlikeSP-LI, did not appear to be susceptibleto any changesin the lumbar enlargementof the rat spinalcord during inflammationof the hind paw. In addition, CGRP-LI releasewas increasedby antagonismof opiate but not GABAB receptors,suggestingthat during chronic inflammationof one hind paw, the GABABergicsystem,unlike the opioid system,might be activated to selectivelyinhibit the enhanced SP-LI release but not CGRP-LI releasewhich is not changed.

Summary

Key words: CGRP-LI; Chronic pain; Chronic inflammation;Spinal cord; GABABreceptor; Naloxone

Introduction The changesoccurring within the spinal cord during developmentof chronic inflammationin peripheral tissues, when acute pain turns into chronic pain, have recentlyreceivedmuch attention. The focushas been on fastneurotransmitters,suchas glutamate(Schaibleet al. 1991;Sluka and Westlund 1993)and on neuropeptides (Oku et al. 1987;Schaible et al. 1990;Malcangio and Bowery 1994)which are contained in terminals of primary afferent fibres (De Biasi and Rustioni 1988; Tuchschererand Seybold 1989)responsiblefor conveying noxious stimuli from the periphery to the spinal cord.

“ Corresponding author: MarziaMalcangio;Presentaddress: Departmentof Pharmacology,Universityof London, QueenMaryand WestfieldCollege,MileEndRoad,London El 4NS, UK. Tel.: (44) 171 9826351; Fax: (44) 181 9830470. PII 0304-3959 (96)03036-9

We have recentlyused a model of peripheral inflammation in the rat, Freund’s adjuvant-induced monoarthritis in the hindpaw, which is characterized by chronicsegmentalhyperalgesiaand loss of body weight (Malcangioand Bowery1994).The spinalcords of these monoarthriticrats, 21 daysafter inoculationof the paw, releasedsignificantlylarger amounts of the nociceptive peptidesubstanceP (SP). This releasecould be greatly enhanced by the GABAB antagonist CGP 36742 whereasnaloxoneonlyenhancedreleaseto the sameextent as in naive animals. CGP 36742also potentiated monoarthritis-inducedhyperalgesia.We have thus suggested that in monoarthritic rats the GABABergicsystem withinthe spinalcord is activatedto counteract the sensationof pain by inhibiting the release of SP from primary afferent fibres (Malcangioand Bowery 1994). SP is not the only peptide thought to play a role in nociceptionand whichmay be susceptibleto changesin its apparent content (Minarni et al. 1989;Smith et al. 1992)and release (Oku et al. 1987; Malcangio and

352

Bowery1994)during the establishmentof peripheralinflammation. For example, galanin (Hope et al. 1994), neuropeptide Y (Ji et al. 1994)and calcitonin generelated peptide(CGRP) (Weiheet al. 1988;Smith et al. 1992)haveall beenconsidered.CGRP is found in dorsal root ganglia (DRG) neurones colocalizedwith SP and their absoluteconcentrationis higherat the lumbar than at the cervical level (Pohl et al. 1990).Terminals immunoreactiveto CGRP innervate the supedlcial levels of the dorsal horn (Conrath et al. 1989)and the peptide coexistswith SP in primary afferent varicositieswhich can be depletedfollowingrhizotomy(Gibsonet al. 1984; Tuchscherer and Seybold 1989) and by capsaicin (Franco-Cerecedaet al. 1987).CGRP-likeimmunoreactivity (CGRP-LI) can be released from rat spinal cord slicesin responseto capsaicinand high K+ (Saria et al. 1986),electricalfield stimulation(Santicioliet al. 1992), electricalstimulationof the dorsal roots (Andreevaand Rang 1993)and noxious stimulation of the skin (Morton and Hutchison 1990).CGRP has a direct biphasic hyperpolarizingand depolarizing action on the membrane of rat spinal dorsal horn neurones (Ryu et al. 1988)and facilitatesSP-inducedexcitationof theseneurons (Biellaet al. 1991).Intrathecal injectionof CGRP has no effecton hindpawwithdrawallatencyto thermal and mechanicalstimuli, howeverthe CGRP antagonist CGRPs-37prevents SP-inducedhyperalgesia(Yu et al. 1994) and intrathecal injection of antisemm against CGRP induces antinociceptionin rats (Kuraishi et al. 1988).CGRP potentates SP-induced aversive behaviour (Wiesenfeld-Hallinet al, 1984)most probably because it inhibits SP breakdown (U Greves et al. 1985) and the level of CGRP increases ipsilaterally in the DRG and dorsalhorn of monoarthriticrats (Smithet al. 1992;Weihe et al. 1988). The aim of this study was to evaluate in the monomthritis model whether electrically-evokedCGRP-LI release from the spinal cord was altered during the establishment of chronic inflrtmmation. Results were interpreted in relation to the previously reported changesin SP (Malcangioand Bowery 1994).

Iing(score O-2) and low thresholdto thermalstimuli(score 3-4) (Malcangioand Bowery1994),and reductionof Imomotoractivity (score4) (Caatro-Lope.s et al. 1992). Rot sptkzl cord slice preparation and superfuion

Rat spinal cord slices were obtained as previouslydescribed (MalcangioandBowery1993).Briefly,hernisected dorsalhunbosacral slices (300 pm) with attacheddorsal roots were preparedusing a Vibratome(Carnpden Instruments,London)andmountedin a threecompartment chamber.Thetissuewaspositionedin thecentralcompmtrnentwhereit wascontinuouslyperfusedwithKrebs’solutionat 1rnlhninandthedorsalrootswereplacedacrosstwopairsof bipolar electrodesandimmersedin mineraloil in the lateralcompartments. Leak-proofpartitionbarriersof perspexandparafh greaseensured electricalisolation. After 1 h wash, normal Krebs’ solution (m rnillirnolar: NaCl, 11S;KCI,4; MgS04,1.2;KH#04, 1.2;NaHC03, 25;CaC12,2.5 andglucose,11)was substitutedwith Krebs’solution containing0.17. bovine serumalbumin,100 AM captopril,1 PM phosphorarnidon, 20@nl bacitracinanddithiothreitol(6PM)to preventmetabolicbreakdownof peptides.Eight-or tive-rninfractions werecollectedin the foUowingorder:3 fractionsto measurebasal outflow,one fractionto evaluatereleasefollowingelectricalstimulationat20V, 0.5msec,10Hzeitherin theabsenceorpresenceof CGP 36742 (3-aminopropyl-n-butyl-phosphinic acid) or naloxone HC1 (SIGMA)and3 fractionsto evaluaterecoveryto basallevels.At the completionof each experiment,the slice was blotted,weighedand CGRP-LIextractedin 1ml glacialaceticacid.Samplesweredesalted by using 100mg SEP-PAKC18reversephase silica gel cartridges (Millipore).Cartridgea wereconditionedwith5ml acetonitrilefollowedby3 MItritluoroacetic acidO.l%(TFA)andNaCl(3.5gl). Samples were then loadedinto the cartridgefollowedby 3 MITFA-NaCl. CGRP-LIwasthenelutedby using2 ml TFA-NaClandacetonitrile (2:S).Theeluatesweredriedby evaporationat 55°Cundernitrogen andstoredat -SWCuntiltheycouldbe assayedforCGRP-LIcontent byRIAusingthescintillationproximityassaybeadtechnique(ArnershamInternational,UK) (Malcangioand Bowery1993). CGRP radwitnrnunoassays

(RIA)

In each experiment,RIA standardcurves were constructedfor CGRPbyusing known amountsofsyntheticpcptidesrangingbetween 1and200fmolkube.Thedriedsampleswereresuspendedin 300PIof 50mMphosphatebutTer(pH 7.4)containingbovineserumalbumin and Triton X-1OO(0.1’?40) and assayedin amountsof CGRP immunoreactivity stdlkientto placethemwithinthe rangeof the standardcurve.CGRPcontentwasmeasuredbyusinga polyclonalrabbit antiserum specifwfor humanCGRP(PeninsulaLaboratoriesUSA). Theincubationmixturesof eaehassaycontainedsampleor standard (100AI),tracercontaining-6000 countshnirr of the isotope(lIMI@, dilutedantiserum(100jd) and 50 PIof scintillationproximityassay proteinA reagent.All sampleswereassayedin duplicate.Thereaction mixturewasvortexedandincubatedat roomtemperature on a shaker for 4S h beforebeing placedin the scintillation-counter.

Results Methods Induction of monoarthritis Thepresentstudywascarriedout on mate Wistarrats(250g) and Lewisratsweighing1S0-200g at the timeexperimentswerestarted. Theywerehousedfourto a cageandgivenfoodandwateradIibiturn. Inductionof monoorthritis

As previouslydescribed(Matcan@oand Bowery1994),monoarthritiswasinducedin Lewisratsby singleintraderrnal injection,into the left hind paw, of 5W pg heat-killedand dried Mycobacterium tuberculosis(H37 Ra, Difco laboratories,UK) in 0.1 MImineraloil (completeFreund’sadjuvant)(CFA rats)underhalothaneanesthesia. Controlanimalswere injectedwith either 0.1 ml mineraloil (incompleteFreund’sadjuvant)(IFArats)(Difcolaboratories,UK)or 100~ phosphatebtiered saline(PBSAts).Arthriticratswerebedded wasassessedby on softpapershavings.Theseventyof inflarmnation subjectivescoring(O-4)takinginto considerationpawredness,swel-

The hind paw of rats inoculated with complete Freund’sadjuvant (CFA rats) showedclear signsof inflammationsuchas rednessand swellingwithin4 h after inoculation.By 3 wks after treatment most of the animals reached the higher scores(Fig. 1) and did not use their paw to walk but tended to protect it. CGRP-LI releasefrom the dorsal horn of the spinal cord

The basal outflow of CGRP-LI from rat spinal cord slicesappears to vary between preparations. This may

353 40-

F .Q30 z s ::20

*

● Smin

o8min

= qlo L % V.

0,

# & 1234567

14

21

28

Fractions

Days post-injection Fig. 1.Assessmentof the severityof inflammation of the injectedpaw of CFA rats (closedcircles,n = 12)and IFA rats(opencircles,n = 12) duringthe4-wkperiodafterinjection.Arthriticscorewasdetermined byanabsolutescaleof O-4(Castro-Lopes et al., 1992;Malcangioand Bowery1994).ErrorbarsindicatedS.E.of mean.Differencebetween the two groups was significant (F’ < 0.05) at all time-points (Mann-Whitney C/-test).

Fig.2. Time-dependencyofelectrically-evoked(20V, 0.5msec, 10Hz, horizontal bar) CGRP-LI release from spinal cord slices. Closed circles:lackof significantreleaseafter5-reinperiodof electricalstimulation of the dorsalroots (n= 5). Open circles: Highly significant releaseof CGRP-LIfollowing8-reinperiodof stimulation(n = 8).The fractionvolumewas 5 ml in the case of 5-reinperiod stimulationand 8 ml when 8-reinperiod stimulationwas applied.

have been due, in part, to the relativelylow peptide recovery(- 60Yo)after loading samplesthrough the SEPPack cartridges. Electrical stimulation of the dorsal roots at 20 V, 0.5 msec, 10 Hz induced a significantrelease of CGRP-LI from spinal cord sliceswhen an 8tnin, but not 5-rein period of stimulation,was applied (Fig. 2). The peptide content was still high in the fraction followingstimulation and then declined to reach basal values(Fig. 2). These conditionswere identicalto those necessary to stimulate SP-LI release in a Ca2+and tetrodotoxin-dependent manner (Malcangio and Bowery 1993).

PBS and IFA rat spinal cord release at 7, 14, 21 or 28 daysafter inoculationof the hind paw (Fig. 3).Table III comparesCGRP-LI and SP-LI release from the spinal cord of monoarthritic rats. Effect of superfusion with a GABA~ antagonist and naloxone

Superfusionwith the GABA~antagonist CGP 36742 (100AM)of PBS, IFA and CFA rat spinal cord slices, prepared 21 days after inoculationwith Freund’s adjuvant, modified neither the basal outflow (data not shown)nor the amount of CGRP-LI releasedafter electrical stimulationof the dorsal roots (Fig. 4A). On the other hand, naloxone(1 PM) superfusedthrough PBS, IFA and CFA rat spinalcord slicessignificantlyenhanced electrically-evoked CGRP-LI release (Fig. 4B) without altering the basal outflow (data not shown). However, the apparent increase in release from CFA cords was significantlylessthan that which occurred in PBSor IFA cords(Fig.4B).CGRP-LI total content was modifiedneitherby superfusionwith CGP 36742nor by naloxone(data not shown).

CGRP-LI release from the spinal cord of monoarthritic rats

Comparablebasal outflowand total content (Table I and II) of CGRP-LI were found in naive,IFA and CFA rat spinal cords. However, 14 days after treatment the CGRP-LI total content was found to be higher in CFA rat spinal cord. Electricallyevoked CGRP-LI release from the spinalcord of CFA rats was not differentfrom

TABLE I EFFECT OF HIND PAW MONOARTHRITIS ON CGRP-LI OUTFLOW FROM DORSAL HORN SLICES OBTAINED FROM THE LUMBAR ENLARGEMENT OF THE RAT SPINAL CORD Treatment

Naive IFA CFA

CGRP-LI basal outflow (fmol/fraction); days after inoculation

o

7

14

21

28

14.9 + 2.9 —

— 9.7 * 3.5 11.2 + 2.7

14.7 + 2.8 9.9 * 2.1

— 17.7 * 2.4 15.0 ● 3.2

11.0 ● 1.3 11.7 + 1.8

Values are means + S.E. of mean (n= 4 for each group). The fraction volume was 8 ml. IFA and CFA rats were inoculated with incomplete and complete Freunds’ adjuvant, respectively.

354 TABLE II EFFECT OF HIND PAW MONOARTHRITIS ON CGRP-LI TOTAL CONTENT IN THE DORSAL HORN OF THE LUMBAR ENLARGEMENT OF THE RAT SPINAL CORD Treatment

CGRP-LI total content (fmol/mg tissue); days after inoculation

Naive IFA CFA

o

7

37 & 6.9 —

— 26.9 & 6.8 30.9 * 5.4

14

21

28

27.9 & 6.1 49.4 & 8.2*

— 27.2 & 3.9 29.4 & 9.5

33.2 A 3.3 40.6 + 6.9

Spinal cord sliceswereblotted, weighedand CGRP-LI extractedin glacialaceticacid. The sliceweightswere25 + 3.4mg (n = 32). Valuesare means + S.E. of mean (n= 3-4 for each group). IFA and CFA, see legendTable I; *P < 0.05, Student’st-test. TABLE 111 COMPARISON OF ELECTRICALLY-EVOKED CGRP-LI AND SP-LI RELEASE FROM THE SPINAL CORD OF MONOARTHRITIC RATS AT VARIOUS DAYS AFTER INDUCTION OF INFLAMMATION Peptide release from CFA rat spinal cord (?Ao IFA rat spinal cord release);days after inoculation

CGRP-LI SP-LI=

7

14

21

28

93.0 + 31.1 91.1 * 15.5

105.0● 50.5 172.7+ 39.6

98.6 & 54.2 258.0+ 41.3*

118.6+ 38.4 26.3 & 25.4”

Valuesare means + S.E. of mean calculatedas percentageof the amount of peptide released(fmol/fraction)from the spinal cord of IFA rats; n >4 for each group; *Student’st-test performedon original values (fmol/8tnl/fraction). ‘Data modified from Malcangio and Bowery, 1994.

301T20

Discussion 6 In

In this study we have determined the total content, basal outflow and electrically-evokedrelease of the neuropeptide CGRP from the isolated spinal cord of monoarthriticrats at 7, 14, 21, 28 days after induction of hind paw monoarthritis. This method is known to produce a local acute reaction with the duration of the inflammatorylesionsand paw hyperalgesiarestrictedto

~lo a) W m w “ 50 L -cJ

.+ \ None

CGP 36742 (100 uM)

80.

“-

0

7

14

21

28

Dayspost-injection Fig. 3. Electrically-evokedrelease(20 V, 0.5 msec, 10Hz for 8 miss) of CGRP-LIfrom spinalcord slicesof PBSrats (whitecolumn),IFA rats (hatched column) and CFA rats (heavily hatched column) at variousdaysafter inoculationof the hindpawwith Freund’sadjuvant. Basaloutflowof CGRP-LIwas 13.3+ 2.3fmol/8rnllfraction(n = 30, pooled means). Values are means + S.E. of mean, n z 3 for each group at each time interval.

Fig. 4. (A) Lack of effect of the GABABantagonist CGP 36742on electrically-evoked (20V, 0.5msec, 10Hz for 8 rnin)CGRP-LIrelease from the spinal cord of monoarthritic rats. Basal outflow was 10.4+ 2.1 fmol/8rtWfraction(n= 13).CGP 36742was superfused5 tin prior to and during 8-rnin pxiod stimulation. Values are means * S.E.of mean. PBSrats (whiteand shadedcolumns)wereinoculatedinto the hindpawwithphosphatebuffersaline;IFA (hatched cohsmn)and CFA (heavilyhatchedcohrmn)rats wereinoculatedwith incomplete and complete Freund’s adjuvant, respectively. (B) Naloxone-indueedincrease in electrically-evokedCGRP-LI release from monoarthritic rat spinal cord slices. Basal outflow was 12.2+ 1.4(n = 14).Naloxonewas superfu~d 5 min prior to and during8-orinperiodstimulation.Valuesare means + S.E. of mean. PBS, IFA and CFA, see above; *P < 0.05; **P < 0.01,Student’sr-test.

355

between2 and 4 wks (Donaldsonet al. 1993;Malcangio and Bowery 1994).Electrical stimulation of the dorsal roots for 8 min induced a significantreleaseof the peptide from the spinal cord. The same conditions were necessaryfor releasingSP (Malcangioand Bowery1993, 1994)but not excitatory amino acids which only required a 3-reinstimulationfor significantrelease(Teoh et al. 1995). The data indicate that CGRP-LI total content, with the exception of values at day 14, and basal outflowfrom dorsal horn sliceswere not modified by inductionof chronicinflammation.Theseresultsalso showed that, even though CGRP-LI could be released from dorsal horn slicesfollowingelectricalstimulation of the attached roots, this releasewas not susceptibleto any change during the developmentof either mild inflammation(IFA rats vs. PBS rats) or severeinflammation (CFA rats vs.PBSrats). Theseresultsweredifferent from those obtained with the neuropeptideSP-LI under the same experimental conditions (Malcangio and Bowery 1994).Although the total content and basal outflowof SP-LI and CGRP-LI did not change during the development of monoarthritis, the electricallyevokedSP-LI releasefrom isolated spinalcord of CFA rats wassignificantlyhigherthan in control rats, 21 days after inoculation with Freunds’ adjuvant (Malcangio and Bowery 1994).By contrast, CGRP-LI release was unaltered. One more differencebetweenCGRP-LI and SP-LIelectrically-evokedreleasefrom monoarthriticrat spinalcord was noted in the effectof GABAergicdrugs. Perfusionwith the GABABantagonist CGP 36742failed to alter CGRP-LI releasein this study whereasit increasedSP-LI release(Malcangioand Bowery 1994).In normal rats the electrically-evokedrelease of both CGRP-LI and SP-LI could be inhibitedby the GABAB agonist (–)baclofen (Santicioli et al. 1992;Malcangio and Bowery 1993, 1995).Thus, GABABreceptor activation can suppressthe releaseof SP-LI and CGRP-LI but the apparent increasein GABAergictone within the spinalcord which occurs in CFA rats (Castro-Lopeset al. 1992;Malcangioand Bowery 1994)only appears to influenceSP-LI but not CGRP-LI release. CGRP-LI release from the spinal cord of naive, as well as IFA and CFA rats, was increased by perfusion with the opioid antagonist, naloxone.These data are in agreement with the naloxone-induced increase of CGRP-LI in polyarthriticrat spinalcord in vivo(Collin et al. 1993).Naloxone also induced the same effect on SP-LIrelease(Malcangioand Bowery1994),supporting the existenceof opioid receptorson primary afferent fibres containing both peptides (Pohl et al. 1990).However,thesedata do not support the previouslysuggested increasein opioid control in arthritic rats (Collin et al. 1993)and the greater analgesiceffect of mu-agonistsin arthritic animals(Kayser and Guilbaud 1983;Millan et al. 1986)becausenaloxone induced an increasein peptide releasefrom isolatedspinalcords in naive,IFA and

CFA rats. This incongruencemightbe ascribedto lower severityof the monoarthritis model compared to polyarthritis.The total number of bindingsitesfor naloxone is reported to be either unchanged(Cesselinet al. 1980) or enhanced (Millan et al. 1986)in the spinal cord of polyarthriticrats. Furthermore, rats with unilateral inflammationof the paw showedno changein the density of mu, delta and kappa binding sites in the lumbar enlargementof the spinalcord ipsilateralto the inflammation (Iadarola et al. 1988). To summarize, CGRP-LI release from the spinal cord, unlike SP-LI release, was not altered during development of monoarthritis. CGP 36742 failed to modify CGRP-LI release, but increased SP-LI release (Malcangioand Bowery 1994).Naloxone increased the release of both peptides in normal and inflamed rats (thisstudy;Malcangioand Bowery 1994).However,the increase in CGRP-LI release produced in CFA cords was significantlyless than control tissue. The explanation(s)for this observationare not clear particularly as there was no differencebetween the increase in SP-LI release from control and CFA cords (Malcangio and Bowery 1994). SP and CGRP have been reported to coexist in the superficiallayers of the spinal cord (Tuchscherer and Seybold 1989).CGRP has been claimed to be a good marker for primary afferent fibres being only localized in theirterminals(Gibsonet al. 1984;Chunget al. 1988). However, this is not certain because it has also been shown that some CGRP is still present in the dorsal horn after rhizotomy and capsaicin (Wiesenfeld-Hallin et al. 1984;Franco-Cereceda et al. 1987; Pohl et al. 1990). Although SP is present in many spinal intemeurones(Hokfeltet al. 1975),we showedthat the release of the neuropeptide, but not basal outflow and total content, measured under our experimentalconditions is unlikely to involve this pool (Malcangio and Bowery 1993). However, the possibility that the observed difference in SP and CGRP evoked release might be ascribed to intrinsic versus primary afferent mechanismsshouldalso be considered.Even though the two peptidescooperativelymediate the transmissionof noxious stimuli within the spinal cord (WiesenfeldHallin et al. 1984;Woolf and Wiesenfeld-Hallin1986), CGRP does not seem to play a role on its own in the transmissionof noxious stimuli but it has only been shown to potentate SP-inducedaversivebehaviour. In the monoarthriticmodel of chronic pain and inflammation CGRP does not behaveas SP as it appeared not to be influencedby peripheral inflammation.The lack of increasein CGRP-LI releaseconcomitant to SP during the development of monoarthritis, excludes the possibilityof metabolicprotection of SP by CGRP (Le Greves et al. 1985;Schaible et al. 1992)which might have accounted for the enhanced SP-LI release .(Malcangio and Bowery 1994).These data do not sup-

port the previously reported increase in SP-LI and CGRP-LI content (Kar et al. 1991;Smith et al. 1992; Kuraishi et al. 1989;Marlier et al. 1991)and synthesis (Donaldson et al. 1992;Hanesch et al. 1993)in DRG and dorsal horn of mono- and polyarthritic rat spinal cord, the increase of CGRP-LI release from the spinal cord of polyarthritic rats (Collin et al. 1993) and increasein SP-LI and CGRP-LI in synovialfluid in a model of acute arthritis (Bileviciuteet al. 1993).However, they do support the lack of enhancement of CGRP-LI levelsin the dorsal zone of the spinalcord of rats bearingan inflamedhind paw (Kuraishiet al. 1989), the loss of CGRP-containingnerve21 days after induction of monoarthritisin thejoint (Mapp et al. 1993)and CGRP decrease in the lumbar versus cervical dorsal horn at 8 h after inductionof acute inflammationin the monkey(Slukaet al. 1992).An explanationfor this discrepancy may stem from the differentmethods used to induce arthritis and on the severity of inflammation which, in the case of monoarthritis, does not appear to be enough to alter CGRP, but only SP release. We concludethat CGRP, unlike SP, does not appear to play a role in the changesoccurringwithin the dorsal horn of the lumbar enlargementof the spinalcord during the establishmentof chronic peripheral inflammation in the hind paw (monoarthritis). In addition CGRP-LI release,unlikeSP-LI, appeared to be increased by blockade of opiate but not GABAB receptors suggesting that the GABABergicsystem within the spinalcord mightbe selectivelyactivatedduringchronic inflammationto inhibit the enhanced release of SP-LI but not CGRP-LI release which is not changed. The opioid system,however,appears to be tonicallyactivated in normal animals and inflamed rats to inhibit the releaseof both SP-LI and CGRP-LI to the sameextent. Maybeit is only when polyarthritisis induced,that rats show a supersensitivityto opioids (Kaiser et al. 1991; Collin et al. 1993). Acknowledgement

We wish to thank CIBA-Geigy for providing CGP 36742.This work was supported by the Arthritis and RheumatismCouncil (B0196). Refereneea Andreeva,L. and Rang H.P., Effectof bradykininand prostaglandins on the releaseof calcitoningene-relatedpeptide-likeimmunoreactivity from the rat spinal cord in vitro, Br. J. Pharmacol., 108 (1993)185-190. Bileviciute,L, Lundeberg,T., Ekblom,A. and TheodorssonE., Bilateral changes of substance P-, neurokinin A-, calcitonin generelated peptide-,and neuropeptideY-1ikeimmunoreactivityin rat knee joint synovial fluid during acute monoarthritis, Neuroaci. Lett., 153(1993)37-40.

Biella,G., Panara, C., Pecile,A. and Sotgiu, M.L., Facilitator role of calcitoningene-relatedpeptide (CGRP) on excitation induced by substanceP (SP)and noxiousstimuliin rat spinal dorsal horn neurons.An iontophoreticstudy in vivo, Brain Res., 559(1991) 352-356. Castro-Lopes,J.M., Tavares, I., Tone, T.R., Coito, A. and Coimbra, A., Increase in GABAergiccells and GABA levels in the spinal cord in unilateralinflammationof the hindlimbof the rat, Eur. J. Neuroaci.,4 (1992)763-773. Cesselin,F., Montastruc,C., Gros, C., Bourgoin,S. and Hamon, M., Met-enkephalinlevelsand opiate receptors in the spinal cord of chronicallysufferingrats, Brain Res., 191(1980)289–293. Collin,E., Mantelet,S., Frechilla,D., Pohl, M., Bourgoin,S., Hamon, M. and Cesselin,F., Increasedin vivo release of calcitonin generelated peptide-likematerial from the spinalcord in arthritic rats, Pain, 54 (1993)203-211. Conrath, M., Taquet, H., Pohl, M. and Carayon, A., Immunocytochemicalevidenceof calcitoningene-relatedpcptide-likeneurons in the dorsal horn and lateral spinal nucleus of the rat cervical spinal cord, J. Chem. Neuroanat., 2 (1989)335-347. Chung, K., Lee, W.T. and Carlton, S.M., The effects of dorsal rhizotomyand spinal cord isolation on calcitonin gene-related peptide-laheledterminals in the lumbar dorsal horn, Neurosci. Lett., 90 (1988)27-32. De Biasi,S. and Rustioni,A., Glutamate and substance P coexistin primaryafferentterminalsin the superficialIaminaeof spinalcord, Proc. Natl. Acad. Sci. USA, 85 (1988)7820-7824. Donaldson,L.F., Harrnar, A.J., McQueen,D.S. and Seckl,J.R., Increased expression of preprotachykinin,calcitonin gene-related peptide, but not vasoactiveintestinal peptide messengerRNA in dorsal root ganglia during the developmentof adjuvant monoarthritis in the rat, Mol. Brain Res., 16(1992)143-149. Donaldson,L.F., Seckl,J.R. and McQueen,D.S.,A discreteadjuvantinduced monoarthritis in the rat: effects of adjuvant dose, J. Neurosci.Methods,49 (1993)5-10. Franco-Cereceda,A., Henke, H., Lundberg,J.M., Petermann, J.B., Hokfelt, T. and Fischer, J.A., Calcitonin gene-related peptide (CGRP) in capsaicin-sensitivesubstance P-immunoreactiveacnsoryneuronsin animalsand man: distributionand releaseby capsaicin, Peptides,8 (1987)399-410. Gibson,S.J., Polak,J.M., Bloom,S.R., Sabate, I.M., Mrdderry,P.M., Ghatel, M.A., McGregor, G.P., Morrison, J.F.B., Kelly, J.S., Evans,R.M. and Rosenfeld,M.G., Calcitoningene-relatedpeptide irnmunoreactivityin the spinalcord of man and eightother species, J. Neurosci.,4 (1984)3101-3111. Hanesch,U., Pfromrner,U., Grubb, B.D., Heppelmann,B.and Schaible, H-G.,Acuteand chronicphasesof unilateral inflammationin rat’s ankle are associated with an increase in the proportion of calcitonin gene-related peptide-immunoreactive dorsal root ganglioncells, Eur. J. Neurosci., 5 (1993)154-161. Hokfelt,T., Kellerth,J.O., Nilsson,G. and Pemow, B., SubstanceP: localizationin the central nervous system and in some primary sensoryneurons, Science,190(1975)889-890. Hope,P.J., Lang,C.W., Grubb, B.D. and Duggan,A.W., Releaseof irnmunoreactivegalanin in the spinal cord of rats with ankle inflammation:studies with antibody rnicroprobes,Neuroscience, 60 (1994)801-807. Iadarola, M.J., Ruda, M.A., Cohen,L.V., Flores,C.M. and Naranjo, J.R., Enhanceddynorphingene expressionin spinal cord dorsal horn neurons during peripheral inflammation: behavioral, neuropeptide,immunocytochernicaland mRNA studies. In: R.

357 Dubner,G.F. Gebhart, and M.R. Bond(Eds.),Proceedingsof the Vth World Congress on Pain, Elsevier, Amsterdam, 1988,pp. 61-71. Ji, R-R., Zhang,X., Wiesenfeld-Hallin,Z. and Hokfelt,T., Expression of neuropeptideY and neuropeptideY (Yl) receptormRNAin rat spinal cord and dorsal root ganglia followingperipheral tissue inflammation,J. Neurosci., 14(1994)6423-6434. Kaiser, V. and Guilbaud, G. The analgesiceffects of morphine,but not those of enkephalinasethiorphan, are enhanced in arthritic rats, Brain Res., 267(1983)131-138. Kaiser, V., Besson,J.M. and Guilbald, G., Effects of the analgesic agent tramadol in normal and arthritic rats: comparisonwith the effectsof differentopioids,includingtoleranceand cross-tolerance to morphine,Eur. J. Pharmacol., 195(1991)37-45. Kar, S., Gibson, S.J., Rees, R.G., Jura, W.G.Z.O., Brewerton,D.A. and Polak, J.M., Increased calcitonin gene-related peptide (CGRP),substancsP, and enkephalinimmunoreactivitiesin dorsal spinal cord and loss of CGRP-immunoreactivemotoneuronesin arthritic rats depend on intact peripheral nerve supply, J. Mol. Neurosci.,3 (1991)7-18. Kuraishi,Y., Nanayama,T., Ohno, H., Minami, M. and Satoh, M., Antinociceptioninduced in rats by intratheeal administrationof antiserumagainstcalcitoningene-relatedpeptide,Neurosci.Lett., 92 (1988)325-329. Kuraishi,Y., Nanayama,T., Ohno, H., Fujii, N., Otaka, A., Yajima, H. and Satoh, M., Calcitoningene-relatedpcptideincreasesin the dorsal root ganglia of adjuvant arthritic rat, Peptides, 10(1989) 447-452. Le Greves, P., Nyberg, F., Terenius, L. and Hokfelt, T., Calcitonin gene-relatedpeptideis a potent inhibitor of substanceP degradation, Eur. J. PharmacoL,115(1985)309-311. Malcangio,M. and Bowery,N.G., y-Aminobtrtyricacida, but not ~aminobutyric acid~, receptor activation, inhibits electncallyevokedsubstanceP-likeimmunoreactivityfrom rat spinal cord in vitro, J. Pharmacol.Exp. Ther., 266(1993)1490-1496. Malcangio, M. and Bowery, N.G., Spinal cord SP release and hyperalgesiain monoarthritic rats: involvementof the GABAB receptor system,Br. J. Pharmacol., 113(1994)1561-1566. Malcangio,M. and Bowery,N.G., Possibletherapeuticapplicationof GABABreceptor agonists and antagonists, Clin. Neuropharmacol., 18(1995)285-305. Mapp, P.I., Terenghi, G., Walsh, D.A., Chen, S.T., Cruwys, S.C., Garrett, N., Kidd, B.L., Polak, J.M. and Blake, D.R., Monoarthritis in the rat knee induces bilateral and time-dependent changes in substance P and calcitonin gene-related peptide immunoreactivityin the spinal cord, Neuroscience, 4 (1993) 1091-1096. Marlier,L., Pordat,P., Rajaofetra, N and Privat,A., Modificationsof serotonin-, substance P- and calcitonin gene-relatedpeptide-like immunoreactivitiesin the dorsal horn of the spinalcord of arthritic rats: a quantitative immunocytochemicalstudy, Exp. Brain Res., 85 (1991)482-490. Millan, M.J., Millan, M.H., Czlonkowski,A., Hollt, V., Pilcher, C.W.T., Herz, A. and Colpaert, F.C., A modelof chronic pain in the rat: responseof multiple opioid systemsto adjuvant-induced arthritis, J. Neurosci.,7 (1986)899–906. Minami,M., Kuraishi,Y., Kawamura,M., Yamaguchi,T., Masu, Y., Nakanishi,S. and Satoh, M., Enhancementof preprotachykininin the rat spinalcord: possibleinvolvementof substanceP-containing spinalneuronsin nociception,Neurosci.Lett., 98 (1989)105-110.

Morton, C.R. and Hutchison,W.D., Morphinedoes not reduce the intraspinal release of calcitonin gene-relatedpeptide in the cat, Neurosci.Lett., 117(1990)319-324. Oku, R., Satoh, M. and Takagi, H., Releaseof substanceP from the spinal dorsal horn is enhanced in polyarthritic rats, Neurosci. Lett., 74 (1987)315-319. Pohl, M., Benoliel,J.J., Bourgoin,S., Lombard, M.C., Mauborgne, A., Taquet, H., Carayon, A., Besson, J.M., Cesselin, F. and Hamon, M., Regional distribution of calcitonin gene-related peptide-, substance P-, cholecystokinin-,Mets-enkephalin-,and dynorphinA (l-8)-like materialsin the spinalcord and dorsal root gangliaof adult rats: effectof dorsal rhizotomyand neonatal capsaicin, J. Neurochem.,55 (1990)1122-1130. Ryu, P.D., Gerber, G., Murase, K. and Randic M., Actionsof calcitoningene-relatedpeptideon rat spinaldorsal horn neurons,Brain Res., 441(1988)357-361. Santicioli,P., Del Bianco,E., Tramontana, M. and Maggi,C.A., Adenosine inhibits action potential-dependentrelease of calcitonill gene-related peptide- and substance P-1ike immunoreactivities from primary afferents in rat spinal cord, Neurosci. Lett., 144 (1992)211-214. Saria, A., Gamse, R., Petermann, J., Fischer, J.A., TheodorssonNorhein,E. and Lundberg,J.M., Simultaneousreleaseof several tachykininsand CGRP fromrat spinalcord slices,Neurosci.Lett., 63 (1986)310-314. Schaible,H.-G.,Jarrott, B., Hope,P.J. and Duggan,A.W., Releaseof immrsnoreactive substancePin the spinalcord duringdevelopment of acutearthritis in the kneejoint of the cat: a studywith antibody microprobe, Brain Res., 529(1990)214-223. Schaible,H.-G., Grubb, B.D., Neugebauer, V. and Oppmann, M., The effects of NMDA antagonists on neuronal activity in cat spinalcord evokedby acute inflammationin the kneejoint, Eur. J. Neurmci., 3 (1991)981-991. Schaible,H.-G., Hope, P.J., Lang, C.W. and Duggan,A.W., Calcitonin gene-relatedpeptide causes intraspinal spreading of substance P released by peripheral stimulation, Eur. J. Neurosci.,4 (19S2)750-757. Sluka, K.A., Dougherty, P.M., Okin, L.S., Willis, W.D. and Westlund,K.N., Neural changesin acute arthritis in monkeysIII. Changesin substanceP, CGRP and glutamatein the dorsal horn of the spinal cord, Brain Res. Rev., 17(1992)29–38. Shska,K.A and Westhrnd,K.N., An experimentalarthritis model in rats: the effectsof NMDA and non-NMDAantagonistson aspartate and ghstamatereleasein the dorsal horn, Neurosci.Lett., 149 (1993)99-102. Smith,G.D., Harmar, A.J., MeQueen,D.S. and Seckl,J.R., Increase in substanceP and CGRP, but not somatostatincontent of innervating dorsal root ganglia in adjuvant monoarthritis in the rat, Neurosci.Lett., 137(1992)257-260. Teoh, H., Fowler,L. and Bowery,N.G., Effectof Iamotngineon the electrically-evokedreleaseof endogenousamino acids from slices of dorsal horn of the rat spinal cord. Neuropharmiacology34 (1995)1273-1278. Tuchscherer,M.M. and Seybold,V.S., A quantitative study of the coexistence of peptides in varicosities within the supetilcial Iaminaeof the dorsal horn of the rat spinal cord, J. Neurosci., 9 (1989)195-205. Weihe,E., Nohr, D., Millan,M.J., Stein,C., Muller,S., Gramsch,C. and Herz,A., Peptideneuroanatomyof adjuvant-inducedarthritic inflammationin rat, AgentsActions, 25 (1988)255-259.

358 Wiesenfeld-Hallin,Z., Hokfek, T., Lundberg, J.M., Forssmann, W.G., Reinecke,M., Tschopp, F.A. and Fischer, J.A., hnrnunoreactivecalcitoningene-relatedpeptideand substanceP coexistin sensoryneurons to the spinal cord and interact in spinal behavioral responsesof the rat, Neurosci. Lett., 52 (1984)199-204. Woolf, C.J. and Wiesenfeld-Hallin,Z., Substance P and calcitonin

gene-related Wptide synergisticallymodulate the gain of the nociceptiveflexor reflex in the rat, Neurosci. Lett., 66 (1986) 226-230. Yu,L.-C.,Hansson,P. and Lundeberg,T., Thecdcitonin gene-related peptideantagonistCGRP5-37increasesthe latency to withdrawal responsesin rats, Brain Res., 653(1994)223-230.