Effects of Adjuvant on Neuropeptide-LikeImmunoreactivity in the Temporomandibular Joint andTrigeminal Ganglia

Joakim Carleson. DDSGraduate StudentDepartment of Physiology and

PharmacologyKarolinska Institute, Stockholm, SwedenCenter for Clinical Oral ScienceSchool of DentistryKarolinska Institute, Huddinge, Sweden

Indre Bileviciute, MDGraduate StudentDepartment of Physiology and

PharmacologyKarolinska Institute, Stockholm, Sweden

Elvar Theodorsson. MD, PhDProfessorDepartment of Clinical ChemistryUniversity HospitalLind ko ping, Svi/eden

Bjorn Appelgren, DDS, PhD, MAAssociate ProfessorDepartment of Physiology and

PharmacologyStockholm, Sweden

Anna Appelgren, DDSGraduate Student

Nancy Yousef, DDSGraduate Student

Sigvard Kopp, DDS, PhDProfessor

Center for Clinical Oral ScienceSchool of DentistryKarolinska InstituteHuddinge, Svueden

Thomas Lundeherg, MD, PhDAssociate ProfessorDepartment of Physiology and

PharmacologyKaroljnska InstituteDepartment of Rehabilitation and

Physical MedicineKarolinska HospitalStockholm, Sweden

Correspondence to:Dr Joakim CarlesonDepartment of Physiology and

PharmacologyKarolinska InstituteS-171 77 StockholmSweden

To study the role of the nervous system in temporomandibularjoint arthritis, substance F-, calcitonin gene-related peptide-, andneuropeptide Y-like immunoreactivity in the trigeminal gangliaand temporomandibular joint of rats was examined. Arthritis wasinduced in female Lewis rats through bilateral injection of a sus-pension of heat-killed Mycobaccerium butyricum in paraffin olíinto the temporomandibular joint. Control rats received paraffinoil via the same route. Tissues were collected for neuropeptide ex-traction 28 days after injection and analyzed hy radioimmuno-assay and reverse-phase high-performance liquid chromatography.Calcitonin gene-related peptide was significantly increased in thearthritic trigeminal ganglia. Substance F, calcitonin gene-relatedpeptide, and neuropeptide Y in the arthritic temporomandibidarjoint were significantly increased as compared to controls. The re-sults of this study show that sensory and sympathetic neuropep-tides may possibly be associated with the development of arthritisin the temporomandibular joint of rats.J OROFACIAL PAIN 1997;11:195-199,

key words: adjuvant arthritis, calcitonin gene-related peptide,immunoreactivity, neuropeptide Y, radioimmuno-assay, substance P, temporomandibular joint, trigemi-nal ganglia

Tbe significance of assessing neuropeptides in joint tissue re-lates to numerous observations suggesting an involvementof cbe nervous system in rbe parhophysiology of inflamma-

tory joint disease. In fact, botb tbe sensory''^ and sympatbetic^nervous systems bave been implicated. Many sensory neurons syn-tbesize both substance P (SP) and calcitonin gene-related peptide(CGRP].''-' In sympathetic neurons, neuropeptide Y (NPY) is syn-tbesized and coreleased with norepinephrine.^ It bas been previ-ously suggested tbat NPY is involved in tbe regulation of inflam-matory processes in tbe buman tempotomandibular joint (TMJ) asit correlates witb TMJ pain and dysfunction.^-^ It bas also beenshown that tbe increased NPY concentrations in tbe TMJ ofpatients suffering from rbeumatoid arthritis wete associated withreduced intraarticulat temperature, suggesting that NPY mediatesthe reduction of tbe microcirculation in the artbritic TMJ, 'Increased concentrations of SP bave also been detected in tbeartbritic TMJ, wbich vLJould mean tbat sympathetic fibers, as wellas sensory fibers, are activated.^

Joumal of Orofacial Pain 195

Carleson et al

The presetit authors have recently reported thatacute experimentally induced TMJ arthritis inrats'^"-''' results in pronounced changes in neu-ropeptide-like immunoreactivity (-LI) in the syn-ovial fluid. Considering that nenropcptide concen-trations in the blood are very low, the contributionof neuropeptides from the hlood circulation of theTMJ synovial fluid is negligible,^ Therefore, themajor portion of the neuropeprides in the synovialfluid is most likely of neuronal origin,"-^^ How-ever, the possibility of a nonticuronal origin can-not be excluded," The aim of the present studywas to elucidate the contribution of the nervoussystem by measuring the concentration of CGRP-LI, SP-LI, and NPY-LI in the TG and TMJ affectedby adjuvant-induced arthritis.

Materials and Methods

The study was performed on 24 female Lewis ratsweighing between 230 and 250 g. The animalswere housed five per cage at 21°C in a 12-hourlight-dark cycle and were given water and pelletsad libitum. Arthritis was induced in 12 of the ratsthrough bilateral intraatticular injection (0,01 mL)of a suspension of heat-killed mycobacteria inparaffin oil (10 mg/mL) (adjuvant arthritis [AA])into the TMJ, The 12 intact control rats received0.01 mL of paraffin oil in the same manner. Therats were killed by decapitation under ether anes-thesia 28 days after inoculation. The TMJ, includ-ing the capsule and the synovial membrane, wasdissected bilaterally separately, and the TG wasexcised, yielding two samples per rat. Each frozentissue sample was weighed before extraction. Theneuropeptides were extracted and quantified in themanner recently described for bone and joint tis-sue,'-^ Prior to extraction, the tissues were cut intosmall pieces, boiled for 10 minutes in 2 mol/Lacetic acid in 4% ethylenediaminetetraacetic acid(EDTA), homogenized in a Polytron (15 seconds),sonicated (30 seconds), and centrifuged at 3000rpm X 9 G for 15 minutes. The supernatants wetelyophilized and diluted in a 2-mL RiA buffer.These samples were further diluted 1:10 and keptat -2O''C until analysis.

Analysis

Calcitonin gene-related peptide was analyzed usingantisernm CGRP8 raised in a rabbit against conju-gated rat CORP. High-petformance liquid chro-matography- (HPLC) purified ^^^I-Histidyl ratCGRP was used as radioligand and rat CGRP as

standard. The detection limit of the assay '°f | ,CGRP was 1 fmol/mL, and crossreactiviry »! _"assay to SP, neurokinin A, neurokinin B. " 'ropeptide K, gastrin, neurotensin, bombesm. ' ';,'ropeptide Y, and calcitonin was less than O.''' "_Crossreactivity with human CGRP a and ß • •'93% and 24%, respectively, and with rat CGKI aand ß 100% and 120%, respectively. Intra- andinterassay coefficients of variation were 8 /a and14%, respectively, ,

Substance P was assessed using antiserum SP2raised in a rabbit against bovine serum albumin-(BSA) conjugated rat SP, The antiserum reactswith SP ,ind SP snlfoxide, but not with othertachykinins in the concentration range 7.8fmol/mL to 3200 fmol/mL. High-performance liq-uid chromatography-purified rat '-^I-Tyrosine SPwas used as radioligand, and rat SP was used asstandard.'^' The detection limit was 1 fmol/mL.Intra- and interassay coefficients of variation were7% and 11%, respectively,

Neuropeptide Y was analyzed using antiserumNI, which crossreacts 0,1% with avian pancreaticpolypeptide but not with other peptides.'-' The de-tection limit of all the assays was 11 fmol/mL.Intra- and interassay coefficients of variation were7% and 12%, respectively.

Reverse-phase HPLC was applied to extracts ofnormai as well as arthritic TMJs to characterizethe neutopeptides studied. Extraction in 2 mol/Lacetic acid in 4% EDTA vi'as found to provide op-timum yield of both sensory and autonomie neu-ropeptides. A Waters Delta Pak C18 300 A 3,9-mm X 150-mm column (Wallac Sverige AB,Upplands Väsby, Sweden) was used, and elutionwas performed with a 40-minute linear gradient ofacetonitrile in water containing 0.1% trifluo-roacetic acid. Two Pharmacia P3500 HPLC pumpswere controlled by a Pharmacia GP250 gradientprogrammer (Pharmacia Biotech Norden,Sollentna, Sweden). A gradient of 20% to 40%acetonitrile was used for SP, whereas a gradient of20% to 50% was used for GGRP and NPY, Thesesamples were passed through Millipore GS filters(0,45 )xm) (Millipore AB, Sundbyberg, Sweden)before chromatography to remove particulate mat-ter, and 200 |i.L of each sample was injected intothe column, Eractions of 0,5 mL were collected atan elution rate of 1,0 mL/minute. Each fractionwas lyophilized and reconstituted in 100 |j,L in dis-tilled water before analysis. The fractions weteassayed for immunoreactivity in the same tubes inwhich they were collected. High-performance liq-uid chromatography analysis of the immunoreac-tive material from ¡oint samples with regard to SP

196 Volume 11, Number 3, 1997

Carleson et al

30-

1 -•

10-

0 -

ns

• ' I I

im 1! • • mSP CGRP NPY

• Arthritic

D Control

i\

Fig 1 Substance P- (SP), calcitonin gene-related peptide-¡CGRP), and neuropeptide Y- (NPY) like immuno-reactivity in the arthritic and in the control trigeminalganglia (N = 8; mean ± SEM, 'P < ,05, ''T < ,01, ' " P< ,005; ns - nonsignificant).

SP CGRP NPY

Fig 2 Substance P- (SP), calcitonin gene-relared peptide-(CGRP), and nenropeptide Y- (NPY) like immunoreac-riviry in the arthritic and in che control cemporo-mandibular joint ¡N - 8; mean ± SEM. *P < .05, °°P <.01, '"^-P<:,005),

CGRP, and NPY consistently resulted in a jnainpeak etured in the position of the correspondingsynthetic peptide. Thus, no evidence of multipleimmunoreactivities was noted for SP, CGRP, orNPY,

Dara were tested for normalcy, and evidence wasfound rhar they were non-Gaussian in most in-stances. Therefore, the nonparametric Wilcoxon'smatched ranked test and the Mann-Whitney U testwere used to analyze differences between the twogroups, and Spearman's rank correlation coefficientwas used to analyze correlation between variables.P values < .05 were considered significant.

Results

Effects of Adjuvant Arthritis on Neuropeptide-LikeImmunoreactivity in the Trigeminai Ganglia

In the TG of the arthritic rats, CGRP-LI was in-creased by 54% as compared to controls. Sub-Stance P-like immunoreactivity and NPY-LIshowed no significant differences as compared tocontrols {Fig 1).

Effects of Adjuvant Arthritis on Neuropeptide-LikeImmunoreactivity in the Temporomandibular Joint

In the TMJ of the arthritic rats, significant in-creases of SP-LI, CGRP-LI, and NPY-Ll concentra-tions, as compared to controls, were detected.

Substance P-like immunoreactivity was increasedby 61%,, CGRP-LI by 72%, and NPY-LI by 38%(Fig 2).

Correiation Between Neuropeptide-LikeImmunoreactivity in the Temporomandibular Jointand in the Trigeminai Ganglia

In the arthritic group, SP correlated by r = -.15 (P= ,72), CGRP by r = -.01 (P = .98), and NPY by r= .03 (P = .95] between the TMJ and the TG. Inthe control group, SP correlated by r = .96 {P =.002), CCRP by r = .2 IP = .63). and NPY by r =.097 (P = ,82) between TMJ and TG.

Discussion

In the presenr study, CGRP-LI was increased inboth the arthritic TMJ and TC, Substance P-hkeimmunoreactivity and NPY-LI increased only inthe TG, but there was a tendency towards anincrease in the TMJ, Adjuvant arthritis, broughtabout by rhe injection of mycobactericum into therat, is used as a chronic animal model in the studyof human rheumatoid arthritis.'^''^ It has beendemonstrated that adjuvant-induced arthritis isrelated to the spread of ad|uvant to regional lymphnodes. ^ No local signs of inflammation were seenin che joints of the control group injected onlywith paraffin oil. However, the animals that weregiven an lntraarticular injection of AA developed

Journal of OrofacisI Pain 197

Csrieson et al

observable sweilitig, redness, and curaneouswarmrh, Cakitonin getie-related peptide was in-creased in both the TMJ and che TG in thearthritic rats; the reasoti for this may be local (neu-rona!) or sysremic- (blood) related factors. How-ever, a previous study-" has shown that the concri-hution from systemic factors is negligible.Substance P-like immunoreactivity and NPY-LIwas increased in the TMJ of the arthritic rats ascompared with the control rats. No significant in-crease in SP-Ll and NPY-LI was detected in theTG of the arthritic rats. These results favor thehypothesis that local peripheral SP and NPY mech-anisms may be involved in the arthritic pro-cess.-'"'^' It is not possible to determine the originof the SP-LI and NPY-LI found in the TMJ basedon the experimental data of the present study.Perhaps some of the SP-LI and NPY-LI are of neu-ronal ongm, stnce both pepttdcs have been foundtn peripheral nerves,''•— However, local factorsmay also contribute. The iocal increase of NPYmay be the result of an increased synthesis ofNerve Growth Factor (NGF) at the site of inflam-mation.''''^ This suggestion is supporred by astudy showing that NGF may induce a productionatid release of NPY by T-lymphocytes,'' Injection-induced adjuvant arthritis triggers an increase ofSP-LI, GGRP-LI, and NPY-LI m TG and TMJ.

The results of the present study show that thealteration in neuropeptide-Hke concentrations inTG and TMJ may reflect neuronal-indueed arthri-tis or arthriris-induced neural activation.

Acknowledgments

The prebeni itudy was supporred by grants from the Anna GretaCraafoords foundation, the Karolinska Insricute Foundation, theKing Gustav Vth 80-Year Anniversary Fund, the ProfessorNanna Svartz foundation, the Swedish Society againstRheumatism, the Magnus Bergvalls Foundation, and the ClasGrnschinsky's Memory Fund,

References

1, Kurai5hi Y, Nanayama T, Ohno H, Fujii N, Otaka A,Yajima H, Satoh M, Gakitonin gene-related peptideiticreases in the dorsal root ganglia of adjuvant arthriticrat, Peptides 1989;ll}i447-452.

2, Levine JD, Oark R, Devor M, Helms G, Moskowitz MA,Basbaum AI, Intrancuronal substance P contributes to theseverity of experimental arthritis. Science 1984;226:547-549,

3, I,eviiie JD, Dardick SJ, Roizen MF, Helms G, Basbaum Al.Gontributicn of sensory afférents and sympathetic efferentsto |Oint injury in experimental arthritis. J Neurosci1986^6:3423-3429,

4. Lee Y, Takami K, Kawai Y, Girgis S, Hillyard CJ;Maclntyre I, et al. Distribution of calcitonin gene-relatepeptide in the rat peripheral nervous system with refereiittto its coexistence with substance P Neurosci iyit5;l5:1227-1237.

5, Wicsenfeld-Hallin Z, Hokfelt T, Lundberg JM, ForssmannWG, Reinecke M, Tschopp FA, Fischer JA, Imnmno-reactive calcironin gene-relaled peptide and subst:ir(-e Pcoexist in sensory neurons to the spinal cord and interactin spinal behavioral responses of the rai, NeurosLi Lett1984;52:199-204,

6. Gray TS, Morley JE, Neuropeptide Y: anatomical distribu-tion and possible function in mamtnalian nervous system|minireview|. Life Sei 1986;3B:389-401,

7. Alstergren P, Appelgrcn A, Appelgren B, Kopp S,Lundeberg T. Theodursson E, Tbe effect on ]oint fluid con-centration of ne tiro peptide Y by intra.articular injection ofglucocorticoid in temporomandihular ¡oint arthritis. ActaOdonrol Scand l996iS4:l-7,

8, Alstergren P, Appelgren A, Appelgren B, Kopp S,Lundeberg T, Tbeodorsson E, Go-variation of neuropep-tide Y, calcitonin gene-related peptide, substance ? andneurokinin A in joint fluid from patients witb temporo-mandibular joinc arthritis. Arch Oral Bioi Í995;4Ü;127-135,

9, Appelgren A, Appelgren B, Kopp S, Lundeberg T,Theodorsson E, Relation between tbe mcra-articular tem-perature of the remporomandibular ]oint and the presenceof neti rope pride Y-like imm uno reactivity in the joint fluid.Acta Odontoi Scand 1993;51:l-8,

10, Garleson J, Alstergren P, Appelgren A, Appelgren B, KoppS, Theodoisson E, Lundeberg T, A model for the study ofeiiperimencally induced temporomandibular arthritis inrats. Effect of human recombinant interleukin-la on neu-ropeptide-like immunoreactiviiy. J Orofacial Painl99é;10:9-14,

11, Garleson J, Alstergren P, Appeigren A, Appelgren B, KoppS, Theodorsson E, Lundeberg T. A model for experimen-tally induced temporomandibulat arthritis in rats. Effectsof adjuvant on neuropeptide-I ike immunoreactivity. ArchOral Biol I996;41:7O5-712.

12, Garleson J, Alstergren P, Appelgren A, Appelgren B, KoppS, Theodorsson E, Lundeberg T. Effects of capsaicin intemporomandibular joint arthritis in rats. Arch Oral Biol1997 (in press),

13, Carleson J, Alstergren P, Appelgren A, Appelgren B, KoppS, Theodorsson E, Lundeberg T, A model for experimentalinduction of acute temporomandibular joint inflammationin rats. Effects of substance P (SP) on neuropoptide-likeimmunoreactivity. Life Sei 1996;59;1193-1201,

14, Lundeberg T, Alstergren P, Appelgren A, Appelgren B,Carleson J, Kopp S, Tbeodoríson E. A model for experi-mentally induced temporomandibular joint arthritis inra ts . Effects of carragenan on neuropeptide-l ikeimmunoreactivity, Neuropeptides 1996;30:37-4].

15, Mabmood A, Neuropeptides in adjuvant artbritis [thesis],Karolinska Institute, Stockholm, Sweden, 1995.

16, Bileviciute I, Lundeberg T, Ekblom A, Tlieodorsson E,Bilateral changes of substance P-, neurokinin A-, calci-tonin gene-related pepcide- and neuropeptide Y-likeimmunoreactivily in rat knee joint synovial fluid duringacute monoarthritis, Neurosci Lett 1993;! •) 3:3 7.40

17, Bracci-Laudiero L, Aloe L, Stenfors G, Tirassa PTheodorsson E, Lundeberg T, Nerve growth factor stimu-lâtes production of neuropeptide Y in human lympho-cytes. Neuro Report 1996;7:4S5-4S8.

198 Volutne 1 1. Number3. 1997

Carleson et al

18, Ward JR, Iones RS, Studies on adjuvant-induced pol-yarthritis in rats, I. Adjuvant composition, roitte of ¡rtjec-tion, and removal of depot site. Arthritis Rheum 1962;5:5,57-564,

19, Freund J, The effect of p iraffin oil and mycob^cteria onatitibody formation and sensiti7iuion. Am J Clin Pathol1951;2h645-656.

20, Appelgren A, Appelgren B, Eriksson S, Ktipp S,Lundtberg T, Nylander M, Theodorsson E, Neuro-peptides in tertiporomandibular joints with rheumatoidarthritis: A dinicai study, Scand J Detii Res 1991; 99;519-521,

21, Fried G, Lundberg JM, Tbcodorsson-Norheim E, Siib-tellular storage and axonal transport iif neuropeptide Y(NPYl iti relation to catccbolamines in the cat. ActaPhysiolSc^nd 19S5;125:145-154,

22, Johansson AS, Isacsson G, Isberg A, Granholm AC,Distribution of substance P-like immunoreactive nervefibers in temporomandibuiar joint soft tissues of monkey,Scand J Dem Res 19S6;94:225-230,

23, Levi-Montalcini R, Sliaper SD, Dal Toso R, Petrelli L,Leon A, Nerve growth factor: From neurotrophin to neu-rokine. Trends Neurosci 199Ê;19:514-520.

Resumen

Los Efectos de un Facilitador sobre la Inmunoreactividadde un Péptido Parecido al Neuropéptido en la Articula-ción Temporomandibuiar y los Ganglios Trigéminos

Se estudió la interacoión del sistema nervioso examinando laSubstanoia P, el Péptido Relaoionado al Gen de la Caloitonina, yla inmunoreactividad de un péptido parecido al Neuropéptido Y,en ios gangiios trigéminos y ias artiouiacicnes teniporoman,dibulares CATWs) de ratas con artritis experimental. La artritisfue inducida en ratas tiembras tipo Lewis por medio de unainyeooión biiateral (en las ATMs de las ratas) de una suspensiónde Uycobaoterium butyrioum muerto a base de calor en aoeitede parafina. Las ratas de controi reoibieron aoeite de parafina através de la misma ruta. Se recoieotaron los tejidos para ia ex-tracción dei neuropéptido y anaiizados por medio de radioinmu-noensayos y de cromatografía liquida de aito rendimiento y defase inversa. Ei Péptido Relacionado ai Gen de la Calcitonmaaumento en ios gangiios trigéminos artritioos, y la Substancia P,el Péptido dei Gen reiacionado a la Calcitonina, y el Neuro-péptido Y aumentaron en ia ATM artritica en oomparaoión oonlos controies. Los fesuitados de este estudio demuestran quehay una interacoión cercana entre ei sistema nervioso y eldesarrollo de la arttitis en la ATM de la rata.

Zusammenfassung

Helfende Auswirkungen auf neuropeptidähnlicherImmunreaktivitÉt im Kiefergelenk und in den Trigeminus-ganglien

Die Interaktion des Nervensystems wurde anhand derUntersuchung der Substanz P, Caicitonin Gene-Related Peptidund Neuropeptid Y-ähnlicher immunreaktivitat in den Trigem-inusgangiien und Kiefergeienken (TMJs) von Ratten wahrendexperimenteller Arthntis studiert. Eine Arthntis wurde in weib-liche Lewisratten durch eine biiaterale Injektion einerSuspension hitieabgetöteter Myobacterium butyricum inParaffinól in die Kiefergeienke der Ralten ausgelost. Kontroll-fatten erhieiten Paraffinoi durch denseiben Weg Gewebe wur-den gesammelt für die Extaktiori von Neuropeptiden und wurdenanaiysiert mittels Radioimmunoassay und Umkehrphaseri-iHochleisturigsflüssigkeitschromatcgraphie. Caloitonin Gene-Reiated Peptd nahm m den arthritischen Tngeminusganglien zu.und Substanz P, Caicitonin Gene-Related Peptid, sowie Neuro-peptid Y nahmen zu in den arthnthischeri Kiefergeienken ver-glichen mit den Kontroiien, Die Resultate dieser Studie zeigen,dass es eine nahe Interaktion zwischen dem Neivensystem undder Entstehung einer Arthritis in den Kiefergeienken der Rattegibt

Journal of Orofacial Pain 199