Antiinflammatory and antinociceptive effects of strontium ranelate on the zymosaninduced joint inflammatory in rats depend on TNF a inhibition

Original

article

Anti-in

fl

ammatory

and

anti-nociceptive

effects

of

strontium

ranelate

on

the

zymosan-induced

temporomandibular

joint

in

fl

ammatory

hypernociception

in

rats

depend

on

TNF-

a

inhibition

Sheila

Moreira

Alves

a

_,

_Susana

_Capistrano

_Abreu

a

_,

_Jonas

_Cavalcante

_Lemos

a

_,

Francisco

Isaac

Fernandes

Gomes

a

_,

_Shirley

_Moreira

_Alves

a

_,

_Danielle

_Rocha

_do

_Val

b

_,

Raul

Sousa

Freitas

a

_,

_Karuza

_Maria

_Alves

_Pereira

a

_,

_Vicente

_de

_Paulo

_Teixeira

_Pinto

a

_,

Gerly

Anne

de

Castro

Brito

c

_,

_Mirna

_Marques

_Bezerra

a

_,

_Gerardo

_{Cristino-Filho}

a

_,

Hellíada

Vasconcelos

Chaves

a,

_*

a_{FederalUniversityofCeará,AvenidaComandanteMaurocélioRochaPontes,100Derby,CEP:62.042-280,Sobral,Ceará,Brazil}

b_{NortheastBiotechnologyNetwork(Renorbio),FederalUniversityofPernambuco,Av.Prof.MoraesRego,1235CidadeUniversitária,CEP:50670-901,Recife,}

Pernambuco,Brazil

c_{DepartmentofMorphology,FederalUniversityofCeará,RuaDelmirodeFarias,Porangabussu,CEP:60440-261,Fortaleza,Ceará,Brazil}

ARTICLE INFO

Articlehistory:

Received2July2016

Receivedinrevisedform24January2017 Accepted10March2017

Availableonline12March2017

Keywords:

Temporomandibularjoint Arthritis

Strontiumranelate TNF-a

ABSTRACT

Background: Temporomandibular joint (TMJ) disorders show inflammatory components, heavily impacting on quality of life. Strontium ranelate has previously shown anti-inflammatory and antinociceptiveeffectsonotherexperimentalinflammatorypainmodels.Thus,weaimtoinvestigate thestrontiumranelateefficacyinreducingthezymosan-inducedinflammatoryhypernociceptioninthe TMJofratsbyevaluatingtheTNF-a,IL-1b,andhemeoxygenase-1(HO-1)involvement.

Methods:Wistarratsweretreatedwithstrontiumranelate(0.5,5or50mg/kg,peros)1hbeforezymosan injection(iart).MechanicalthresholdwasassessedbyVonFreytestandsynoviallavagewascollectedfor leukocyte countingand myeloperoxidase measurement, jointtissue and trigeminal ganglionwere excisedforhistopathologicalanalysis(H&E)andTNF-a/IL-1blevelsdosage(ELISA).Moreover,ratswere pre-treated with ZnPP-IX (3mg/kg, sc), a specific HO-1 inhibitor, before strontium ranelate administration (0.5mg/kg,peros),andEvansBlue(5mg/kg,iv)wasadministered toassessplasma extravasation.Pre-treatmentwithindomethacin(5mg/kg,sc)wasusedaspositivecontrolwhilethe shamgroupreceived0.9%sterilesaline(perosandiart).

Results:Strontiumranelatedidnotreduceleukocytecounting,myeloperoxidaseactivity,EvansBlue extravasation,IL-1blevels,andTNF-a/IL-1bimmunolabeling;butitincreasedthenociceptivethreshold andreducedTNF-alevels.Additionally,HO-1inhibitiondidnotchangethestrontiumranelateeffects. Conclusion:StrontiumranelatemayachieveitsantinociceptiveeffectsthroughthereductionofTNF-a

levelsinthetrigeminalganglion,butnotsuppressingIL-1bexpressionnorinducingtheHO-1pathway. ©2017InstituteofPharmacology,PolishAcademyofSciences.PublishedbyElsevierSp.zo.o.Allrights reserved.

Introduction

Thepathogenesisoftemporomandibularjoint(TMJ)disorders stillremainsunclearevenwiththeremarkableprogressalready madetowardstheelucidationoftheirpathogenesis.Experimental models that allow the investigation of the inflammatory pain relatedtothesedisordersareofgreatclinicalrelevanceandour

group developed an experimental model of zymosan-induced

arthritisinrats.Zymosanisayeastcell-derivedpolysaccharidethat *Correspondingauthor.

E-mailaddresses:eishyla2010@hotmail.com(S.M.Alves),

suzanact@hotmail.com(S.C.Abreu),jonas_cavalcante_@hotmail.com(J.C.Lemos), isaac.odonto@hotmail.com(F.I.F.Gomes),shirley_sma31@yahoo.com.br (S.M.Alves),danielleval@hotmail.com(D.R.doVal),raul.sf2@gmail.com (R.S.Freitas),karuzaalves@yahoo.com.br(K.M.A.Pereira),pintovicente@gmail.com (V.dePauloTeixeiraPinto),gerlybrito@hotmail.com(G.A.deCastroBrito), mirnabrayner@gmail.com(M.M.Bezerra),gerardocristino@uol.com.br (G.Cristino-Filho),helliadachaves@yahoo.com.br(H.V.Chaves).

http://dx.doi.org/10.1016/j.pharep.2017.03.007

1734-1140/©2017InstituteofPharmacology,PolishAcademyofSciences.PublishedbyElsevierSp.zo.o.Allrightsreserved. –

ContentslistsavailableatScienceDirect

Pharmacological

Reports

producesasevereanderosivesynovitisalongwithinflammatory paininanimalmodelsofkneearthritis[1–3].

Inflammatorystimulicausemechanicalhypernociceptionbya definedsequentialreleaseofcytokineswithininflamedjointsby manycellssuchastumornecrosisfactor-

a

(TNF-

a

)and interleu-kin-1

b

[4]. TNF-

a

and IL-1

b

arehighly expressedin thejoints affected by TMJ disorders. Many studies showed considerable levelsofbothinthesynovialfluidofpatientssufferingfromthis condition [5,6]. Furthermore, studies demonstrated that heme oxygenase–1(HO-1)exertsanantioxidantroleanditsinduction wouldleadtonegativefeedbackforcellactivationandproduction ofinflammatorymediators[7–9].

Strontiumranelate(Sran){5-[bis(carboxy-methyl)

amino]-2-carboxy-4-cyano-3-thiophen-acetic acid distrontium salt} is a compoundwithtwostablestrontiumatomsand ranelicacid. It affectstheboneturnoveranditisanorallyactivetreatmentwhich

decreasestherisk ofvertebraland hipfracturesinosteoporotic women [10,11]. Although its mechanism of action is not fully understood,this drugmaypossessanalgesiceffects[12].Recent studiesshowedthatstrontiumranelatehadaprotectiveeffectin experimental osteoarthritis [13] and it promoted analgesia in arthriticrats,whichwasassociatedtoinhibitionofthereleaseof inflammatorycytokinesintoinflamedjoints[14].

Thus,thepresentstudyattemptstoinvestigatetheunexplored anti-nociceptive and anti-inflammatory effects of strontium ranelateonthezymosan-inducedinflammatoryhypernociception intheTMJofratsbyevaluatingtheIL-1

b

andTNF-

a

levelsafter strontium ranelate treatment. Further, we aim to determine whetherstrontiumranelateeffectsonthisexperimentalmodelof arthritiswoulddependonHO-1pathwayintegrityasourprevious results showedthatHO-1pathwayinhibitionisassociatedwith increasedinflammatoryresponses[8].

Fig.1.ExperimentaldesigntoassessstrontiumranelateeffectsontheTMJarthritis.(a)Evaluationoftheanti-inflammatoryandantinociceptiveeffectsofstrontiumranelate onthezymosan-inducedmechanicalallodyniaintherattemporomandibularjoint.(b)Investigationoftheinvolvementofhemeoxygenase-1intheanti-inflammatoryand antinociceptiveeffectsofstrontiumranelateonthezymosan-inducedmechanicalallodyniaintherattemporomandibularjoint.

Materialsandmethods

Animals

MaleWistarrats(n=6pergroup;160–220g)werehousedin standardplasticcages,theyhadaccesstofoodandwateradlibitum andweremaintainedinatemperature-controlledroom(232_C) with a 12/12-h light-dark cycle. This study was conducted in accordancewiththelocalInstitutionalAnimalCareandwiththe approvalofthelocalethicalcommittee(registrationnumber54/ 12)andwedesignedittoreduceanimalsufferingandthenumber ofanimals.

Zymosan-inducedinflammatoryhypernociception

To induce the inflammatory hypernociception, rats were

anesthetized with inhaled isoflurane (4%) and received intra-articular(iart) injectionof zymosan (2mg;40

m

L)dissolved in sterilesaline into theleft TMJ using a 30-gauge needle. Sham animalsreceivedsalinesolution(peros)beforezymosanorsaline solutioninjections.TheTMJskinwasshaved,thepostero-inferior borderof the zygomaticarch was located,and the needle was inserted in a position inferior to this point until the needle contactedthecondyle,whichwasverifiedbythemovementofthe mandibleandthepunctureoftheneedleintothejointspacewas confirmedbythelossofresistance. Gentleaspiration ruledout intravascular injection, after which zymosan or salinesolution wereinjected.Aspreviouslyshownbyourgroup[1]the zymosan-induced inflammatory hypernociception is maximal at 4h of arthritiswhilepolymorphonuclearcellinfluxpeaksafter6h.Thus, weused thesetime points toassess thefollowing parameters: headwithdrawalthreshold,totalcellcounting,and myeloperox-idaseactivity.

Mechanicalthresholdevaluation

Inflammatoryhypernociceptionin theTMJ was evaluatedby measuringthethresholdofforceneededtobeappliedtotheTMJ regionuntil the head withdrawaloccurred. The measurements wereperformedbyablindedexaminerwhousedadigitaldevice (Insight, Brazil) that consisted of a rigid filament linkedto an electronicdevice–automaticVonFreyanesthesiometer,whichin turn measures the response threshold in grams (g) when the filament isappliedtothesurfaceofthetestedregion[15].The facialareastobetestedaroundtheTMJwereshavedbeforethe experimentalprocedureandtheanimalswereplacedinindividual plastic cages 45min before the tests. The animals underwent conditioningsessionsinthetestingroomfor4consecutivedays. Ondayfive,thebasalforcethresholdvaluewasrecordedthree timesbeforeand4haftertheintra-articularinjectionsofeither

zymosan or vehicle. Then, we measured the inflammatory

hypernociception of the zymosan-group animals and treated controls. The applied mechanical stimuli were innocuous in controlanimals.

Pharmacologicalmodulation

Strontium ranelate (PROTOS1 _{2g, Les Laboratoires Servier} Industry,45 520Gidy, France) (0.5,5 or 50mg/kg, per os) was administeredonehourpriortozymosaninjection(iart).Foodwas removed1hbeforethetreatmentinordertoavoidanychangesin thepharmacokineticprofileofstrontiumranelate(Fig.1a).Data validationwasachievedbyusingapositivecontrolgroupthatwas pre-treatedwithindomethacin(5mg/kg,sc)1hbeforezymosan injection.Shamgroupreceived(perosandiart)0.9%sterilesaline.

ToanalyzethepossibleeffectofHO-1pathwayonanti-nociceptive and anti-inflammatory efficacy of strontium ranelate, animals were pre-treated (sc) with ZnPP IX (3mg/kg), a specific HO-1 inhibitor,followedbyaninjection(peros)ofstrontiumranelate (0.5mg/kg) 30min later (Fig. 1b). After 1h, intra-articular zymosan-injectionwasperformedandatthe4thhour, inflamma-toryhypernociceptionintheTMJwasevaluated.

SynovialLavageCollection,CellCountingandMyeloperoxidase ActivityAssessment

Six hours after zymosan injections, the rats weresacrificed underanesthesiaandexsanguinated.Thesuperficialtissueswere dissectedandtheTMJcavitywaswashedtwotimestocollectthe synovial fluid by the pumping and aspiration technique using 0.05mLofEDTA(1.77mg EDTA/1mLPBS).Thetotal numberof whitecellsinthesynoviallavagewascountedusingaNeubauer

chamber. MPO activity assay measurement was described by

Bradleyetal.anditwasconductedonthecollectedsynoviallavage whichwascentrifugedat4500rpmfor12minat4_{C.MPOactivity} was assayedbymeasuringthechangein absorbanceat450nm using o-dianisidine dihydrochloride and 1% hydrogenperoxide. TheresultsarereportedastheMPOunits/jointfluidandaunitof MPOactivitywasdefinedastheconversionofa

m

molofhydrogen peroxidetowaterin1minat22_C.

Evansblueextravasationassay

Strontium ranelate(0.5mg/kg, peros) was administered 1h prior to zymosan injection. Thirty minutes before euthanasia, Evans Blue (5mg/kg, iv) was then injected to assess plasma extravasation. After its excision, the periarticular tissue was weighed,immersedintoformamide solution(1mL), andplaced intowaterbath(60_{C)overnight.Theresultingsupernatantwas}

collected and the absorbance was read at 620nm using a

spectrophotometer.The concentrationwas determinedby com-parisonwithastandardcurveofEvansbluedyeintheextraction solution and the amount of Evans blue dye (

m

g) was then calculatedpermLofexudate[16].

Histopathologicalanalysis

The TMJ was excised six hours after the induction of the inflammatoryhypernociception.Thespecimenswerefixedin10% neutralbufferedformalinfor24h,demineralizedin10%EDTAfor 7days,embeddedinparaffin,andsectionedalongthelongaxisof the TMJ. Sectionsof 5

m

m, includingthe condyle,the articular cartilage, the articular disc, the synovial membrane, the peri-articulartissue,andtheskeletalmusclewereevaluatedunderlight microscopy (400). The specimens were prepared for routine

hematoxylin-eosin(H&E)stainingandhistologicalanalysis con-sidereda0–4scorebasedonthefollowingparameters:cellinflux intothesynovialmembrane,cellinfluxintotheconnectivetissue andtheskeletalmuscleoftheperiarticulartissue, andsynovial membranethickness.

Immunohistochemistry

Immunohistochemistryfor TNF-

a

and IL-1

b

was performed usingthestreptavidin-biotin(LabeledStreptavidinBiotin –LSAB) method in formalin-fixed, paraffin-embedded tissue sections (5

m

m thickness), mounted on glass slides prepared with an organosilane-basedadhesive(3-aminopropyltriethoxysilane, Sig-maChemicalCoJ_{,StLouis,MO,USA).Thesectionsunderwent2} bathsinxylolfortenminuteseachone.Theywereafterimmersed

three timesinto alcohol at 100%concentration,and washedin distilledwater.

Antigenrecoverywasperformedwithcitrate(pH6.0;30min; 99_{C).Afterreturningtoambienttemperature,thesectionswere} immersedintoa3%hydrogenperoxideblockingsolutionfor10min. Thesectionswerethenincubatedovernight(4_{C)withaprimary} rabbitanti-TNF-

a

and anti- IL-1

b

antibody (ABCAMJ_{, England,} UK),atthedilutionof1:200,andwashedwithPBSsolution.

ThesampleswereincubatedwiththesecondaryantibodyLSAB Kit for 10min at ambient temperature. Next, incubation was performedinachromogensolutionpreparedwith3,30 diamino-benzidine(DAB) (DAKOJ_{, Carpentaria,CA,USA),for 10minina}

dark chamber. Afterwards, the specimens were washed into

running water and then into distilled water. Counter-staining wasperformedwithhematoxylin,andafterwardsthespecimens weredehydratedinalcoholanddiaphanizedinxylol.Finally,they weremountedonglassslides.Thenegativecontrolsectionswere performed excluding the application of the primary antibody.

The parameter of positivity for the immunohistochemical

marking of the antigen in all the specimens included in the sampleconsistedofthecellsthatexhibitedbrownstainingintheir cytoplasmirrespectiveoftheintensityoftheimmunomarking.

TNF-

a

andIL-1

b

ELISAassays

TheTMJ tissueand thetrigeminalganglionwereexcised6h afterthezymosan-injectioninratsandwerehomogenizedina solutionof RIPALysis BufferSystem (SantaCruz Biotechnology, USA).Thesampleswerecentrifugedat10000rpmfor15minat 4_{C.Thesupernatantswerestoredat80Cforposterioranalysisto} evaluatetheproteinlevelsofTNF-

a

andIL-1

b

intheTMJtissueand thetrigeminalganglion.Thecytokinelevelswerequantifiedbythe following kits:TNF-

a

–RatTNF-alpha/TNFSF1AQuantikineELISA Kit (R&DSystems, catalog number RTA00); and IL-1

b

–RatIL-1 beta/IL-1F2QuantikineELISAKit(R&DSystems,catalognumber DY501).Theabsorbancewasmeasuredat450nm.IL-1

b

andTNF-

a

concentrationswereexpressedaspg/mL.

Statisticalanalysis

ThedataarepresentedasthemeanSEMormedianswhere

appropriate. Differencesbetween means were compared using one-way ANOVA followed by theBonferroni test. The Kruskal-WallistestfollowedbyDunn’stestwasusedtocomparemedians. Avalueofp<0.05indicatedsignificantdifferences.

Fig.2.Effectsofstrontiumranelateonthezymosan-inducedTMJinflammatoryhypernociception.(a)Headwithdrawalthresholdinstrontiumranelate-treatedrats(b) Leukocytecountinginstrontiumranelate-treatedrats(c)MPOactivityfromTMJsynoviallavageinstrontiumranelate-treatedrats.(d)Plasmaextravasationinstrontium ranelate-treatedrats.DataareexpressedasthemeanSEMof6miceforeachgroup;*p<0.05indicatesasignificantdifferencefromtheshamgroup,+p<0.05indicatesa

significantdifferencefromthezymosangroup(ANOVA,Bonferroni).

Results

Effectsofstrontiumranelateonthezymosan-inducedinflammatory hypernociception(Fig.2)

Theintra-articularinjectionofzymosancausedinflammatory hypernociceptionthatisobservedasadecreaseinthemechanical thresholdofheadwithdrawal(Fig.2a).Itresultedinasignificant increaseinthenumberofpolymorphonuclearcells(Fig.2b)which inturnwascertifiedbytheincreaseofMPOactivityintheTMJ synoviallavageafterzymosaninjection(Fig.2c). Thesechanges were followed by plasma extravasation into the TMJ after 6h

(Fig. 2d). Sham animals showed no significant changes in

withdrawalthreshold,polymorphonuclearcells count,and MPO activity(Fig. 2 a, b, c). Strontium ranelate(0.5, 5 or 50mg/kg) injected (per os) 1h prior to zymosan injection significantly (p<0.05)increasedthenociceptivethreshold(Fig.2a).However, strontiumranelatefailedtodecreasethenumberof polymorpho-nuclearcells(Fig.2b),MPOactivity(Fig.2c),andEvansbluedye extravasationinthesynoviallavage(Fig.2d).

EffectsofzincprotoporphyrinIX(ZnPPIX)onthestrontiumranelate efficacy(Fig.3)

ToinvestigatetheroleofHO-1activityintheantinociceptive effectofstrontium ranelate,theanimalswerepre-treated with ZnPPIX (3mg/kg; sc), a specific HO-1 inhibitor. The effects of strontiumranelate(0.5mg/kg)onthezymosan-inducedinfl am-matoryhypernociception(Fig.3)werenotchangedinthepresence ofZnPP-IX(3mg/kg).

JointtissueandtrigeminalganglionTNF-

a

andIL-1

b

ELISAassays (Fig.4)

Theintra-articularinjectionofzymosanresultedinasignificant increaseinTNF-

a

(Fig.4aand4b)andIL-1

b

(Fig.4candd)levelsin bothjointtissue andtrigeminalganglionafter.Albeitstrontium ranelatetreatmentwasnotabletosignificantlyreduceIL-1

b

levels

when compared with the zymosan group (Fig. 4c and 4d),

strontiumranelatereducedTNF-

a

levelsinbothjointtissueand trigeminalganglion(Fig.4aand4b).

Histopathologicalanalysis(Fig.5)

Inflammatory cell influx was observed into the synovial

membrane(Fig.5b)6hafterzymosan-injectioncomparedwith the sham group (Fig. 5a). The predominant cell types were neutrophils,whichcharacterizedacuteinflammation.Edemawas alsoobservedinthesynovium(Fig.5b).Table1showsthescores attributed to TMJ histopathological analysis and compares the

values between the sham and zymosan groups. A significant

(p<0.05)increaseintheinflammatoryparameterswasobserved inthezymosangroup.Table1alsoshowsthescoresattributedto the TMJ histopathological analysis and compares the values betweenthezymosan andstrontium ranelate(0.5,5 or 50mg/ kg)groups.Strontiumranelate(0.5,5or50mg/kg)didnotreduce theinflammatoryparameters.Figs.5cand5dshowtheTMJofrats pre-treatedwithstrontiumranelate(0.5mg/kg,peros).

Immunohistochemicalanalysis(Fig.6)

TheimmunohistochemicalanalysisofTNF-

a

andIL-1

b

showed increasedimmunolabelingforbothTNF-

a

andIL-1

b

in synovio-cytesand neutrophlisafterzymosanchallengethatwas charac-terizedbybrown-coloredcellsinthesynovialmembrane(Fig.6). Thesynovialcellsinthesynovialmembraneofthezymosanand strontiumranelate-treatedanimalsalsoshowedbothTNF-

a

and IL-1

b

expression (Fig. 6). However, in the conjunctive tissue, strontiumranelate(0.5mg/kg)treatmentreducedTNF-

a

expres-sion.Thenegativecontrolgroupsectionsconsistedof zymosan-inducedTMJinflammatoryhypernociceptionthatwerenottreated with anti-TNF-

a

or anti-IL-1

b

antibody. None of the negative controlsshowedTNF-

a

orIL-1

b

immunoreactivity.

Discussion

Wedemonstratedthattheeffectsofstrontiumranelateonthe zymosan-inducedTMJinflammatoryhypernociceptioninratsmay occurviaTNF-

a

suppressionaswellasitsmechanismofactionin this disease model is IL-1

b

/HO-1 independent. Experimental animalmodelsofTMJinflammatoryhypernociceptionhavebeen usedtostudyinflammatoryconditionsandweperformedthefirst demonstration of TMJ arthritis induced by zymosan, through

which we showed that zymosan caused a time-dependent

leucocyte migration, plasma extravasation, mechanical hyper-nociception,andneutrophilaccumulation[1].Thisshowsthatthe zymosan-induced TMJ arthritis is a reproducible experimental modelthatcanbeusedtoexplorethemechanismsunderlyingTMJ inflammationandpotentialtherapies.

Strontiumranelatewasoriginallydesignedtotreat osteoporo-sis[17–21],butitmayexerteffectsonosteoarthritisandithasled

topositiveoutcomesinaphaseIIIclinicalstudy[13,22].Albeitthe mechanismofactionofstrontiumranelateisnotfullyunderstood, itappearstostimulatethedifferentiationofosteoblastsbyeliciting thecalciumsensor receptor,inhibitingosteoclastdifferentiation byinhibitingRANKL production,andincreasingosteoprotegerin (OPG)activity[23,24].

In addition, patients treated with strontium ranelate had a greaterreductioninthetotalscoreandpainsubscorecompared withtheplacebogroup[25].Sinceosteoporosisandosteoarthritis areassociatedwithavarietyofsymptoms,includingpain,itcould behypothesizedthatstrontiumranelatemayalsobeeffectivein reducingthetemporomandibularjointinflammatory hypernoci-ception.Ourresultsdemonstratedthattheintra-articularzymosan injectiondiminishedthemechanicalnociceptivethreshold,which Fig.3. EffectofzincprotoporphyrinIX(ZnPPIX),aspecificHO-1inhibitor,onthe

strontiumranelateefficacyonzymosan-inducedTMJinflammatory hypernocicep-tion.DataareexpressedasthemeanSEMof6ratsforeachgroup;*p<0.05

indicates a significant difference from the sham group,+_{p<0.05 indicates a} significantdifferencefromthezymosangroup(ANOVA,Bonferroni).

in turn was increased by the strontium ranelate treatment. However, the inflammatory parameters cell influx and MPO activitywerenotreduced afterstrontiumranelatetreatment.

Evansblueextravasationmeasurementintothesynoviallavagedid notchangefollowingthedrugadministration.TheTMJ histopath-ological analysisafter zymosan injection showed inflammatory cellinfluxintothesynovialmembrane,periarticular tissue,and musculoskeletal tissue associated with thickness of synovial membrane, being the drug treatment unable to reverse these findingstoa normalstatus. Likewise,itwas demonstratedthat strontiumranelatetreatmenthadnoparticulareffectonsynovitis indogs[13].

Studies correlate the HO-1 activity with oxidative damage inhibitionandreductioninproinflammatorycytokinesproduction [7].Itwasreportedthatthisenzymehadantinociceptiveeffectson acetic acid-evoked nociception and positive outcomes after inductionofHO-1inazymosan-inducedairpouchinflammation model[8,26].Consideringthesedata,weevaluatedthe involve-mentofHO-1inthestrontiumranelateantinociceptiveeffectsand weobservedthattheywerenotchangedafterthepre-treatment withZnPP-IX,suggestingthatHO-1activityisnotinvolvedinits antinociceptiveeffects.

Manycelltypesproducecytokinesinresponsetoavarietyof stimuli,whichisalinkbetweencellularinjuryandthe develop-ment of localsigns and symptoms of inflammation. There is a cascadeofreleaseofcytokineslinkinginjuriesandthereleaseof

nociceptive mediators in rats: a concept that allows us to understand why the inhibition of cytokines causes analgesia [27].ManystudieshavedemonstratedthecontributionofTNF-

a

to inflammatoryhyperalgesiaandtheclinicalsuccessofthe

anti-TNF-a

therapyof rheumatoid arthritis alsoexemplifies this concept [4,28]. During the inflammatory response, TNF-

a

is the first released cytokine and IL-1

b

is a potent pleiotropic mediator involvedininflammatoryresponses[29].Hence,TNF-

a

andIL-1

b

arerecognizedcontributorstothepathogenesisofjointdiseases, leadingtosynovialfibroblasthyperplasiaandtothedestructionof theextracellularmatrix[30,31].

In the present study, the zymosan injection resulted in a significantincrease in both TNF-

a

and IL-1

b

levelssothatour findingsareinaccordancewithotherones,suggestingthatTNF-

a

is asdrivingcytokineofthenociceptiveprocess.TNF-

a

playsa crucial role in the development of inflammatory hyperalgesia duringtheinflammatoryresponseinrats,beinghighlyexpressed in the synovial fluid of patients with TMJ disorders [5–7]. Additionally,wealsodemonstratedthattreatmentwithstrontium ranelatereducedTNF-

a

levelsinbothjointtissueandtrigeminal ganglion.

TNF-

a

is thefirst cytokinereleased duringan inflammatory response, triggering the release of IL-1

b

known to activate inflammatoryanddegradativepathwaysinsynovialcells.Studies suggested high IL-1

b

levels in the synovial fluid of patients suffering from TMJ disorders [5,29,32]. Nunes et al. [14] Fig.4.JointtissueandtrigeminalganglionTNF-a (a/b)andIL-1b(c/d)levelsfromstudyratseithersubjectedornottoonzymosan-inducedTMJinflammatory hypernociceptionandassayedonthe6thhourpostchallenge.DataareexpressedasthemeanSEMof6ratsforeachgroup;*p<0.05indicatesasignificantdifferencefrom

theshamgroup(ANOVA,Bonferroni).

investigatedtheanti-inflammatoryactivityofstrontiumranelate inthearticularincapacitationtest,inthepaw-pressuretest,andin the anterior cruciate ligament transection model. Unlike our protocol,theanimalsreceivedstrontiumranelateathigherdoses: 30–300mg/kg per os. They find that strontium ranelate dose-dependentlyinhibitedjointpaininbothtypesofarthritismodels, butitdidnotaltercellinfluxwhichisasimilarresultobtained here.Contrarytothem,ourfindings suggestthatthestrontium ranelate treatment is not capable of reducing IL-1

b

levels comparedwiththezymosangroup.

Thismightberelatedtothedifferencesinthedesignperformed byNunesetal.[14].Thestrontiumranelatedoseadministeredin theirprotocolisfarhigher thanthe oneweused, which could potentiate the analgesic effects of this drug. Other essential difference is the experimental model of diseases that could

culminateindistinctresultsasdifferentdiseasemodelsmayimply incontrastingpathogenesismechanisms,thus,leadingtodifferent results.AspointedoutbyNunesetal.[14],naloxoneabolishedthe strontiumranelateanalgesiceffectwhichisvaluableinformation onthestrontiumranelateunspecificmechanismofaction.

Furthermore,the immunohistochemicalanalysis showed in-creasedTNF-

a

and IL-1

b

immunolabeling in the synovialcells after the intra-articular zymosan injection. Albeit strontium ranelate slightly reduced the TNF-

a

immunolabeling in the conjunctive tissue, it was not able to diminish the IL-1

b

immunolabeling one. Contrary to this result, in the synovial membraneofdogsundergoingsectioningoftheanteriorcruciate ligament,thegeneticexpressionofIL-1

b

wassignificantlyreduced bystrontiumranelatetreatmentatthedosesof50or75mg/kgper

day for 16 weeks [13]. A possible explanation for these

Fig.5. Photomicrographsofthehistopathologicalanalysisoftemporomandibularjoints(TMJ).(a)shamgroupTMJ(100);(b)zymosan2mggroup(400)showing

inflammatorycellinfluxinthesynovialmembrane;(c)and(d)TMJofratspretreated(peros)withstrontiumranelate(0.5mg/kg)andinjected(i.art.)withzymosan2mg(100 and400x,respectively).C:condyle;AC:articularcartilage;AD:articulardisc;SM:synovialmembrane;PAT:periarticulartissue.Hematoxylinandeosin(H&E)staining.

Table1

Histopathologicalanalysisbyhematoxylin-eosinstaining(HE)ofthetemporomandibularjointofratsafterintra-articularzymosaninjectionandstrontiumranelate(Sran) treatment.

Groups Cellinfluxinthe synovialmembrane

Periarticularcellinflux Cellinfluxinthe Musculartissue

Sham 1(0–1) 1(0–2) 0(0–0)

Zy 4(2–4)* _3(3–4)* _2.5(2–4)*

Indo 2(1–2)** _2(1–2)** _1.5(1–2)**

Sran0.5 3.5(2–4) 3.5(3–4) 3.5(2–4)

Sran5 3(2–3) 3.5(3–4) 3.5(2–4)

Sran50 3(2–4) 4(2–4) 4(2–4)

*_{p<0.05versusSham;}**_{p<0.05versusZymosan(Kruskal-Wallis,Dunn’s).}

contradictory results could be the different animal model of diseasealongwiththeadoptedposology,leadingtocontradictory resultsfromwhatweobtained.

Therefore, albeit the mechanisms of action through which strontiumranelateexertsantinociceptiveeffectsremainrelatively elusive,thisstudyprovidesnovelinformationonitseffectsonthe zymosan-inducedTMJ inflammatory hypernociceptionas stron-tiumranelateprimarilysuppressedTNF-

a

levelsandreducedthe nociceptivethreshold.Theinflammatorystimuliortissueinjuries stimulatetherelease of characteristiccytokine cascades,which ultimatelytriggertherelease of final mediatorsresponsiblefor inflammatorypain.Thesefinalmediators,suchasprostanoidsor sympathetic amines, act directly on the nociceptors to cause hypernociception,which resultsfromtheloweringofthreshold duetomodulationofspecificvoltage-dependentsodiumchannels. As reported here,strontium ranelate could decrease hyper-nociceptionthresholdsbyreducingTNF-

a

levelsinthe periartic-ulartissuesandtrigeminalganglion.Giventheimportanceofthese

structures to the temporomandibular joint pain onset and

progression,we suggest thatTNF-

a

functionsasa target point forstrontiumranelate.Thissuggeststhatstrontiumranelatemight beapotentialcandidatefor thetreatmentof TMJpainthrough TNF-

a

inhibition and more studies of longer duration are necessarytovalidate theuse ofstrontium ranelatein thepain management.

Conflictofinterests

The authors declare that they have no conflict of interest regardingthepublicationofthisstudy.

Funding

Thisworkwassupported byBrazilian grantsfromFundação CearensedeApoioao DesenvolvimentoCientíficoeTecnológico (FUNCAP), Conselho Nacional de Desenvolvimento Científico e Tecnológico(CNPq),CoordenaçãodeAperfeiçoamentodePessoal

deNívelSuperior(CAPES),andInstitutodeBiomedicinado Semi-ÁridoBrasileiro(INCT).

Acknowledgment

The authorsthank AdalbertoNascimentode LimaJúnior for technicalassistance.

References

[1]ChavesHV, RibeiroRA, SouzaAM,Silva AAR,GomesAS,Vale ML,etal. Experimentalmodelofzymosan-inducedarthritisintherat

temporomandibularjoint:roleofnitricoxideandneutrophils.JBiomed Biotechnol2011;200:1–2011.

[2]Bezerra MM, Brain SD, Girão VCC, Greenacre S, Keeble J, Rocha FAC. Neutrophils-derivedperoxynitritecontributestoacutehyperalgesiaand cellinfluxinzymosanarthritis.NaunynSchmiedebergsArchPharmacol 2007;374:265–73.

[3]GegoutP,GilletP,ChevrierD,GuingampC,TerlainB,NetterP.Characterization ofzymosan-inducedarthritisintherat:effectsonjointinflammationand cartilagemetabolism.LifeSci1994;55:321–6.

[4]CunhaFQ,PooleS,LorenzettiBB,FerreiraSH.Thepivotalroleoftumour necrosisfactoralphainthedevelopmentofinflammatoryhyperalgesia.BrJ Pharmacol1992;107:660–4.

[5]AkutsuM,OguraN,Ito K,KawashimaM,KishidaT,KondohT. Effectsof interleukin-1bandtumornecrosisfactor-aonmacrophageinflammatory protein-31productioninsynovialfibroblast-likecellsfromhuman temporomandibularjoints.JOralPatholMed2013;42:491–8.

[6]Gulen H, Ataoglu H, Haliloglu S, Isik K. Proinflammatorycytokines in temporomandibularjointsynovialfluidbeforeandafterarthrocentesis. OralSurgOralMedOralPatholOralRadiolEndod2009;107:1–4.

[7]AlcarazMJ,FernándezP,GuillénMI.Anti-inflammatoryactionsoftheheme oxygenase-apathway.CurrPharmDes2003;9:2541–51.

[8]GrangeiroNMG,AguiarJA,ChavesHV,SilvaAAR,LimaV,BenevidesNMB,etal. Hemeoxygenase/carbonmonoxide-biliverdinpathwaymaybeinvolvedinthe antinociceptiveactivityofetoricoxib,aselectiveCOX-2inhibitor.Pharmacol Rep2011;63:112–9.

[9]VanderleiESO,AraújoIWF,QuinderéALG,PontesBP,EloyYRG,RodriguesJAG, etal.TheinvolvementoftheHO-1pathwayintheanti-inflammatoryactionof asulfatedpolysaccharideisolatedfromtheredseaweedGracilariabirdiae. InflammRes2011;60:1121–30.

[10]CianferottiL,D’AstaF,BrandiLM.Areviewonstrontiumranelatelong-term antifractureefficacyinthetreatmentofpostmenopausalosteoporosis.Her AdvMusculoskelDis2013;0:1–13.

[11]ReginsterJ,BadurskiJ,BellamyN,BensenW,ChapurlatR,ChevalierX.Efficacy andsafetyofstrontiumranelateinthetreatmentofkneeosteoarthritis: Fig.6.RepresentativeimmunohistochemistryofTMJtissuesforIL-1b(upperpanel),andTNF-a(lowerpanel)fromratsatthesixthhourafterzymosaninjection.Theupper panelshowsincreasedIL-1bimmunolabelingofsynoviocytesandneutrophlisafterzymosan-challenge(400).ThebottompanelshowsincreasedTNF-aimmunolabelingof

synoviocytesandneutrophlis afterzymosan-challenge(400).Control: negativecontrol(sectionsin theabsenceofanti-IL-1band anti-TNF-a antibody);Sham:

resultsofadouble-blind,randomised,placebo-controlledtrial.AnnRheum Dis2013;72:179–86.

[12]KarsdalMA,Bay-JensenAC,LoriesRJ,AbramsonS,SpectorT,PastoureauP, etal.Thecouplingofboneandcartilageturnoverinosteoarthritis: opportunitiesforboneantiresorptivesandanabolicsaspotentialtreatments? AnnRheumDis2014;73(2):336–48.

[13]PelletierJP,KapoorM,FahmiH.Strontiumranelatereducestheprogressionof experimentaldogosteoarthritisbyinhibitingtheexpressionofkeyproteases incartilageandofIL-1betainthesynovium.AnnRheumDis2013;72:250–7. [14]NunesRM,MartinsMR,SilvaJuniorFS,MeloLeiteAC,GirãoVC,CunhaFQ,etal. Strontiumranelateanalgesiainarthritismodelsisassociatedtodecreased cytokinereleaseandopioid-dependentmechanisms.InflammRes 2015;64:781–7.

[15]Denadai-SouzaA,CamargoL,de L,RibelaMT,KeebleJE,CostaSK,etal. ParticipationofperipheraltachykininNKreceptorsinthecarrageenan– inducedinflammationoftherattemporomandibularjoint.EurJPain 2010;13:812–9.

[16]KwanCL, HuJW,SessleBJ.Neuroplasticeffectsofneonatalcapsaicin on neuronsinadultrattrigeminalnucleusprincipalisandsubnucleusoralis.J Neurophysiol1996;75:210–98.

[17]AmmannP,ShenV,RobinB,MaurasY,BonjourJP,RizzoliR.Strontiumranelate improvesboneresistancebyincreasingbonemassandimprovingarchitecture inintactfemalerats.JBoneMinerRes2004;19:2012–20.

[18]BuehlerJ,ChappuisP,SaffarJL,TsouderosY,VigneryA.Strontiumranelate inhibitsboneresorptionwhilemaintainingboneformationinalveolarbonein monkeys.Bone2001;29:176–9.

[19]DelannoyP,BazotD,MarieP.Long-termtreatmentwithstrontiumranelate increasesvertebralbonemasswithoutdeleteriouseffectinmice.Metabolism 2002;51:906–11.

[20]ParkSY,KimYH,KimEK,RyuEY,LeeSJ.Hemeoxygenase-asignalsareinvolved inpreferentialinhibitionofpro-inflammatorycytokinereleasebysurfactinin cellsactivatedwithporphyromonasgingivalislipopolysaccharide.ChemBiol Interact2010;3:437–45.

[21]ReginsterJ,DeroisyR,NeuprezA,HiligsmannA,ZegelsB,BruyereO.Strontium ranelate:newdataonfracturepreventionandmechanismsofaction.Cur OsteoporosRep2009;7:96–102.

[22]Alexandersen P, Karsdal MA, Qvist P. Strontium ranelate reduces the urinarylevelofcartilagedegradationbiomarkerCTX-IIinpostmenopausal women.Bone2007;40:218–22.

[23]AtkinsGJ,WelldonKJ,HalboutP.Strontiumranelatetreatmentofhuman primaryosteoblastspromotesanosteocyte-likephenotypewhileelicitingan osteoprotegerinresponse.OsteoporosInt2009;20:653–64.

[24]CoulombeJ,FaureH,RobinB.Invitroeffectsofstrontiumranelateonthe extracellularcalcium-sensingreceptor.BiochemBiophysResCommun 2004;323:1184–90.

[25]BruyereO,DelferriereD,RouxC. Effectsofstrontiumranelateonspinal osteoarthritisprogression.AnnRheumDis2008;67:335–9.

[26]VicenteAM,Guillen MI,HabibA,AlcarazMJ.Beneficialeffects ofheme oxygenase-aup-regulationinthedevelopmentofexperimentalinflammation inducedbyzymosan.JPharmacolExpTher2003;307:1030–7.

[27]CunhaTM,VerriWA.JrSilvaJS,PooleS,CunhaFQ:Acascadeofcytokines mediatesmechanicalinflammatoryhyperalgesiainmice.ProcNatlAcadSciU SA2005;102(5):1755–60.

[28]RankinEC,ChoyEH,KassimosD,KingsleyGH,SopwithAM,InsenbergDA,etal. Thetherapeuticeffectsofanengineeredhumananti-tumournecrosisfactor alphaantibody(CDP571)inrheumatoidarthritis.Rheumatol1995;34:334–42. [29]CunhaTM,VerriWA,SilvaJrJS,PooleS,CunhaFQ.Acascadeofcytokines mediatesmechanicalinflammatoryhyperalgesiainmice.ProcNatlAcadSciU SA2005;102(5):1755–60.

[30]KayJ,CalabreseL.Theroleofinterleukin-ainthepathogenesisofrheumatoid arthritis.Rheumatology2004;43:2–9.

[31]FieldM.Tumournecrosisfactorpolymorphismsinrheumaticdiseases.QJ Med2001;94:237–46.

[32]KubotaE, KubotaT, Matsumoto J,Shibata T,Murakami K.Synovialfluid cytokinesandproteinasesasmarkersoftemporomandibularjointdisease.J OralMaxillofacSurg1998;56:192–8.