Inhibition of indoleamine 2,3-dioxygenase-mediated tryptophan catabolism accelerates collagen-induced arthritis in mice

Open Access A ailable online h p://a h i is- esea ch.com/con en /9/3/R50 Page 1 o 7 (page numbe no o ci a ion pu poses) Vol 9 No 3 Resea ch a icle Inhibi ion o indoleamine 2,3-dioxygenase-media ed yp ophan ca abolism accele a es collagen-induced a h i is in mice Sándo Szán ó1,2, Tamás Ko eny1, Ka alin Mikecz1, Tibo T Glan 1, Zol án Szekanecz2 and John Va ga3 1Sec ion o Molecula Medicine, Depa men o O hopedic Su ge y, Rush Uni e si y Medical Cen e , Cohn Resea ch Building, Room 708, 1735 W. Ha ison, Chicago, IL 60612, USA 2Ins i u e o Medicine, Di ision o Rheuma ology, Uni e si y o Deb ecen, Medical and Heal h Science Cen e , 22 Mó icz S ee , Deb ecen, H-4012, Hunga y 3Di ision o Rheuma ology, No hwes e n Uni e si y Medical School, 303 Eas Chicago A e., Chicago, IL 60611, USA Co esponding au ho : Sándo Szán ó, szan [email protected] e.hu Recei ed: 13 Dec 2006 Re isions eques ed: 17 Jan 2007 Re isions ecei ed: 24 Ap 2007 Accep ed: 18 May 2007 Published: 18 May 2007 A h i is Resea ch & The apy 2007, 9:R50 (doi:10.1186/a 2205) This a icle is online a : h p://a h i is- esea ch.com/con en /9/3/R50 © 2007 Szán ó e al.; licensee BioMed Cen al L d. This is an open access a icle dis ibu ed unde he e ms o he C ea i e Commons A ibu ion License (h p://c ea i ecommons.o g/licenses/by/2.0), which pe mi s un es ic ed use, dis ibu ion, and ep oduc ion in any medium, p o ided he o iginal wo k is p ope ly ci ed. Abs ac Indoleamine 2,3-dioxygenase (IDO) is one o he ini ial and a e- limi ing enzymes in ol ed in he ca abolism o he essen ial amino acid yp ophan. In cul u ed cells, he induc ion o IDO leads o deple ion o yp ophan and yp ophan s a a ion. Recen s udies sugges ha modula ion o yp ophan concen a ion ia IDO plays a undamen al ole in inna e immune esponses. Induc ion o IDO by in e e on-γ in mac ophages and dend i ic cells esul s in yp ophan deple ion and supp esses he immune-media ed ac i a ion o ib oblas s and T, B, and na u al kille cells. To assess he ole o IDO in collagen-induced a h i is (CIA), a model o heuma oid a h i is cha ac e ized by a p ima ily Th1-like immune esponse, ac i i y o IDO was inhibi ed by 1-me hyl- yp ophan (1-MT) in i o. The esul s showed signi ican ly inc eased incidence and se e i y o CIA in mice ea ed wi h 1-MT. Ac i i y o IDO, as de e mined by measu ing he le els o kynu enine/ yp ophan a io in he se a, was inc eased in he acu e phase o a h i is and was highe in collagen-immunized mice ha did no de elop a h i is. T ea men wi h 1-MT esul ed in an enhanced cellula and humo al immune esponse and a mo e dominan pola iza ion o Th1 in mice wi h a h i is compa ed wi h ehicle- ea ed a h i ic mice. The esul s demons a ed ha de elopmen o CIA was associa ed wi h inc eased IDO ac i i y and enhanced yp ophan ca abolism in mice. Blocking IDO wi h 1-MT agg a a ed he se e i y o a h i is and enhanced he immune esponses. These indings sugges ha IDO may play an impo an and no el ole in he nega i e eedback o CIA and possibly in he pa hogenesis o heuma oid a h i is. In oduc ion Locally p oduced p oin lamma o y cy okines such as umo nec osis ac o -α, in e leukin (IL)-1, and IL-6 play a pi o al ole in he pa hology o heuma oid a h i is (RA). These cy okines, by up egula ion o se e al genes, a e esponsible bo h o he ec ui men and con inuous ac i a ion o he in lamma o y cells and o inducing p oduc ion o he enzymes ha des oy bone and ca ilage. Howe e , hese in lamma o y e en s need o be balanced by he p oduc ion o endogenous inhibi o s, as in lamma o y esponses a e gene ally localized and he conse- quen des uc ion o a ec ed join s is less se e e [1]. The ubiqui ously exp essed heme enzyme indoleamine 2,3- dioxygenase (IDO) ca alyzes he non-hepa ic oxida i e deg a- da ion o yp ophan, he ini ial and a e-limi ing s ep in yp- ophan me abolism, esul ing in deple ion o his leas abundan essen ial amino acid. T yp ophan s a a ion enables he hos o es ic he g ow h o in acellula pa hogens [2]. The exp ession o IDO in heal hy issues is gene ally qui e low bu is ma kedly up egula ed in esponse o in ec ion and in lamma ion. Recen s udies ha e es ablished an en i ely no el impo an biological unc ion o IDO. These s udies indi- ca e ha IDO-media ed yp ophan deple ion in i o esul s in inhibi ion o ma ix-deg ading me allop o einase enzyme p o- duc ion, supp ession o in lamma o y esponses, and 1-MT = 1-me hyl- yp ophan; CFA = comple e F eund's adju an ; CIA = collagen-induced a h i is; CII = ype II collagen; CTLA-4 = cy o oxic T lym- phocy e-associa ed an igen-4; ELISA = enzyme-linked immunoso ben assay; IDO = indoleamine 2,3-dioxygenase; IFN-γ = in e e on-γ; IgGAM = immunoglobulins G, A, and M; IL = in e leukin; i.p. = in ape i oneally; K = kynu enine; NK = na u al kille ; RA = heuma oid a h i is; T = yp ophan. A h i is Resea ch & The apy Vol 9 No 3 Szán ó e al. Page 2 o 7 (page numbe no o ci a ion pu poses) p omo ion o immune ole ance [3,4]. Acco dingly, an eme g- ing immunological pa adigm ega ds IDO as a key egula o y con ol enzyme in bo h inna e and adap i e immune esponses. Indeed, modula ion o cellula IDO exp ession and/ o ac i i y is now inc easingly implica ed in in lamma ion and au oimmuni y, o gan ansplan a ion and e al ejec ion, and e asion o immune su eillance by umo cells. Fu he mo e, agen s ha inhibi IDO a e unde ac i e in es iga ion as po en- ial adju an s o umo immuno he apy [5]. IDO-media ed yp ophan ca abolism esul s in deple ion o yp ophan wi h concomi an gene a ion o kynu enine and ul i- ma ely NAD (nico inamide adenine dinucleo ide). Exp ession o he IDO gene is induced in mos cell ypes in esponse o in ec ion wi h mic obial agen s ia ac i a ion o oll-like ecep- o s. Fu he mo e, in e e on-gamma (IFN-γ), IL-10, and cy o- oxic T lymphocy e-associa ed an igen-4 (CTLA-4) ha e also been shown o s imula e IDO [4,6]. In con as , IL-4, IL-13, and ans o ming g ow h ac o -β a e supp esso s o IDO [7]. In p e ious s udies, we ha e shown ha IFN-γ caused ime- dependen induc ion o IDO gene exp ession and ac i i y in cul u ed no mal ib oblas s [4]. The consequen yp ophan ca abolism and ensuing yp ophan s a a ion in hese cul u es we e associa ed wi h p o ound supp ession o collagenase and s omelysin gene exp ession induced by IL-1β. Ab oga ion o IL-1β-induced ma ix me allop o einase s imula ion upon ac i a ion o IDO in hese ib oblas s was di ec ly due o he educ ion o local yp ophan concen a ions a he han he accumula ion o kynu enine and o he yp ophan me aboli es. These obse a ions led us o p opose he hypo hesis ha , h ough i s abili y o induce ansien yp ophan s a a ion, IDO ep esen ed an impo an endogenous an i-in lamma o y mechanism and ha dis up ion o IDO induc ion o unc ion would be associa ed wi h exagge a ed in lamma o y esponses. Collagen-induced a h i is (CIA) is gene a ed in gene ically suscep ible mouse o a s ains by immuniza ion wi h ype II collagen (CII) dissol ed in comple e F eund's adju an (CFA). CIA highly esembles RA acco ding o o e lapping immun- opa hogenic pa hways and simila his opa hologic ea u es [8]. Among o he easons, he cen al ole o class II majo his- ocompa ibili y complex [9], he in ol emen o CD4+ T cells in he immunopa hogenesis o he disease, he p edominan ly Th1 ype immune esponse o CII in he induc ion phase [10,11], as well as he cy okine p o ile h oughou he e olu- ion o CIA [12] make his expe imen al animal model one o he mos adequa e models o RA. To add ess he ques ion o whe he IDO plays a nega i e egula o y ole in an animal model o a h i is, we assessed IDO ac i i y in he se a o mice a di e en s ages o a h i is and he e ec o 1-me hyl- yp- ophan (1-MT), a speci ic inhibi o o IDO, on a h i is de elop- men and an igen-speci ic immune esponses in CIA. Ma e ials and me hods Immuniza ion and assessmen o collagen-induced a h i is in mice ecei ing ei he 1-me hyl- yp ophan o ehicle Mice we e housed and b ed unde s anda d condi ions a he Compa a i e Resea ch Cen e o Rush Uni e si y (Chicago, IL, USA). The Ins i u ional Animal Ca e and Use Commi ee app o ed all animal expe imen s. DBA/1 male mice, 6 o 8 weeks o age, we e pu chased om The Jackson Labo a o y (Ba Ha bo , ME, USA). Mice we e immunized by a s anda d immuniza ion p o ocol. B ie ly, 100 μg o human CII was emul- si ied in CFA (Di co Labo a o ies Inc., now pa o Bec on Dickinson and Company, F anklin Lakes, NJ, USA) and injec ed in o he p oximal ail o mice. A second injec ion o he same dose and adju an was gi en in ape i oneally (i.p.) on day 21. Mice ha did no de elop a h i is wi hin 3 weeks o he second an igen injec ion we e boos ed wi h a hi d injec- ion adminis e ed in equally di ided doses i.p. and in o he p oximal ail. All mice we e sac i iced 8 weeks a e he i s injec ion. A e he second immuniza ion, all mice we e exam- ined o swelling and e y hema o dis al join s. Paws we e con- side ed o ha e a h i is when swelling and e y hema we e no ed in a leas wo digi s and/o o he join s. The clinical se e i y o a h i is was g aded on a scale o 0 o 4 o each paw, acco ding o swelling and edness. Speci ically, sco ing was pe o med as ollows: 0 = heal hy; 1 = mild swelling and e y hema; 2 = mode a e swelling and e y hema; 3 = mo e in ense e y hema, swelling, and edness a ec ing a g ea e p opo ion o he paw; 4 = se e e e y hema, swelling, and ed- ness a ec ing he en i e paw. A cumula i e sco e anging om 0 o 16, based on indi idual paw sco es o 0 o 4, was assigned o each animal. Table s con aining slow elease o D, L-1-MT (Inno a i e Resea ch o Ame ica, Sa aso a, FL, USA) o ehicle we e su - gically inse ed unde he do sal skin o mice on days 22 and 29 a e he i s immuniza ion. The able s eleased 10 mg/day 1-MT o a pe iod o 7 o 10 days. Measu emen o an ibody p oduc ion and T-cell esponse An ibodies o he immunizing human and mouse (sel ) ca i- lage CII we e de e mined by enzyme-linked immunoso ben assay (ELISA), as desc ibed elsewhe e [13-15]. B ie ly, Max- iso p 96-well pla es (Nalge Nunc, Nape ille, IL, USA) we e coa ed wi h 0.1 μg o human o mouse ca ilage-de i ed CII in 100 μl o coa ing bu e . An ibodies we e de e mined in se ial dilu ions o se a (1:500 o 1:62,500) using pe oxidase-conju- ga ed goa an i-mouse immunoglobulins G, A, and M (IgGAM) seconda y an ibodies (Zymed Labo a o ies Inc., now pa o In i ogen Co po a ion, Ca lsbad, CA, USA). Se um an ibody le els we e exp essed in millig ams pe li e using mouse IgGAM as con ol (In i ogen Co po a ion). A ailable online h p://a h i is- esea ch.com/con en /9/3/R50 Page 3 o 7 (page numbe no o ci a ion pu poses) An igen-speci ic T-cell esponses, including IL-2 p oduc ion and T-cell p oli e a ion, we e measu ed in quad uplica e sam- ples o spleen cells (3 × 105 cells pe well) cul u ed in he p esence o 100 μg o collagen p o ein pe millili e . IL-2 was measu ed in supe na an s ha es ed on day 2 by he p oli e a- ion o he IL-2-dependen CTLL-2 cell line. An igen-speci ic T- cell p oli e a ion was assessed on day 5 by he inco po a ion o 3 [H] hymidine. In bo h cases, he an igen-speci ic esponse was exp essed as s imula ion index, which is a a io o inco - po a ed 3 [H] hymidine (coun s pe minu e) in sil e -s imula ed cul u es ela i e o coun s pe minu e in non-s imula ed cul- u es [13,14]. An igen-speci ic IFN-γ and IL-4 p oduc ion by T cells was de e mined in cul u e condi ions iden ical o hose desc ibed o T-cell p oli e a ion in 4-day-old condi ioned medium (2.5 × 106 mononuclea cells pe millili e ) using cap- u e ELISAs (R&D Sys ems, Inc., Minneapolis, MN, USA). High-pe o mance liquid ch oma og aphy analysis Se um yp ophan (T) and kynu enine (K) concen a ions we e measu ed simul aneously by high-pe o mance liquid ch oma- og aphy, as p e iously desc ibed [16]. As ela i ely high ol- umes o se a a e needed o he de e mina ion o K and T, pooled se a we e used, K and T concen a ions we e de e - mined, and hen K/T a ios we e calcula ed. S a is ical analysis Analyses o he a h i is sco e and disease incidence a di e - en ime poin s we e ca ied ou using he non-pa ame ic Mann-Whi ney U es and chi-squa e con ingency analysis, espec i ely. S uden es was used o s a is ical analysis o all o he da a. Analyses we e pe o med using SPSS e sion 7.5 so wa e package (SPSS Inc., Chicago, IL, USA). Signi i- cance was se a a p alue o less han 0.05. Resul s Clinical and his ological ea u es o a h i is: inc ease o bo h he incidence and se e i y in mice ea ed wi h 1- me hyl- yp ophan Mice we e immunized wi h CII and moni o ed o he de elop- men o a h i is o 54 days a e he p ima y immuniza ion. Clinical signs o a h i is ypically appea ed wi hin 4 o 8 days a e he second and hi d injec ions, which we e adminis e ed on days 21 and 42 a e he p ima y immuniza ion, espec- i ely. T ea men wi h 1-MT adminis e ed by he inse ion o able s con aining 1-MT (o ehicle) on days 22 and 29 a e he i s injec ion esul ed in an inc eased incidence o a h i is (Figu e 1a). In addi ion o he highe disease incidence, he se e i y indica ed by he cumula i e a h i is sco e was signi i- can ly inc eased in mice ecei ing 1-MT as compa ed o hose ecei ing ehicle be ween days 37 and 47 a e he p ima y immuniza ion (Figu es 1b and 2). His ological analysis o he ankle, me a a sophalangeal, and in e phalangeal join s o 1- MT- ea ed and ehicle- ea ed mice showed ypical a h i is cha ac e ized by ex ensi e leukocy e in il a ion, syno ial p oli - e a ion, pannus o ma ion, and e osions. The se e i y o in lam- ma o y cell in il a ion and des uc ion o ca ilage and bone e lec ed he clinical s a e, bu no quali a i e di e ences could be obse ed in mice ecei ing 1-MT compa ed o con ol mice. Cellula and humo al immune esponses Because o he di e ences in incidence and se e i y o a h i is be ween mice ecei ing 1-MT o ehicle, i seemed o be p u- den o de ec immune esponses o CII. Signi ican inc eases we e ound in bo h CII-speci ic au oan ibody and he e oan i- body i e s in 1-MT- ea ed mice compa ed o he ehicle g oup; howe e , hese di e ences disappea ed by he end o he ollow-up pe iod (day 55) (Figu e 3a,b). When assessing T-cell p oli e a ion in he wo g oups a he end o ollow-up, no Figu e 1 Incidence and se e i y o collagen-induced a h i is (CIA) in mice ea ed wi h 1-me hyl- yp ophan (1-MT) o ehicleIncidence and se e i y o collagen-induced a h i is (CIA) in mice ea ed wi h 1-me hyl- yp ophan (1-MT) o ehicle. A h i is was i s de ec ed on day 26 a e immuniza ion wi h ype II collagen on days 0, 21, and 42 (solid a ows) in mice ea ed wi h 1-MT o ehicle (open a ows). (a) Incidence o CIA exp essed as he pe cen age o a h i ic animals. (b) Disease se e i y exp essed as he cumula i e a h i is sco e in a ec ed animals. S a is ically signi ican di e ences in a h i is sco es we e ound be ween days 37 and 47 (p < 0.05). Values a e p e- sen ed as he mean and s anda d e o o he mean o 25 1-MT- ea ed and 20 ehicle- ea ed DBA/1 mice pe g oup and ep esen wo inde- penden expe imen s. A h i is Resea ch & The apy Vol 9 No 3 Szán ó e al. Page 4 o 7 (page numbe no o ci a ion pu poses) di e ences in T-cell esponses o he e ologous CII, and only mode a ely inc eased T-cell p oli e a ion o au ologous CII in he 1-MT g oup, could be de ec ed. Simila ly o T-cell p oli e - a ion, IL-2 concen a ions in he supe na an s o spleen cells s imula ed wi h mouse o human CII we e sligh ly (bu non-sig- ni ican ly) highe in he 1-MT- ea ed g oup han in ehicle- ea ed g oup a he end o he ollow-up (da a no shown). Spleen cell supe na an cy okine le els in mice wi h collagen-induced a h i is ea ed wi h 1-me hyl- yp ophan o ehicle The concen a ions o IFN-γ and IL-4 we e de e mined in he supe na an s o spleen cells om mice wi h CIA and ea ed wi h 1-MT o ehicle, espec i ely, a he end o he ollow-up. Cy okine p oduc ion o cells was assessed ei he wi hou s im- ula ion o a e challenging wi h mouse o human CII. A lowe concen a ion o IL-4 was de ec ed in supe na an s o spleen cells om mice ea ed wi h 1-MT in compa ison o ea ed hose wi h ehicle, bu he di e ence was signi ican only when challenging he cells wi h human CII (p < 0.05) (Figu e 4a). In con as , sligh ly (bu non-signi ican ly) highe concen- a ions o IFN-γ could be measu ed in supe na an s o spleen cells om 1-MT- ea ed mice compa ed o hose om ehicle- ea ed animals (Figu e 4b). Kynu enine/ yp ophan a ios in collagen-induced a h i is mice ea ed wi h 1-me hyl- yp ophan o ehicle To es ima e he enzyma ic ac i i y o IDO in di e en s ages o a h i is, K and T concen a ions we e measu ed and K/T a ios we e calcula ed in he pooled se a o mice ea ed wi h 1-MT o ehicle. Inc eased K and dec eased T concen a ions we e measu ed, and hus inc eased K/T a io was de ec ed in mice wi h acu e a h i is, bu his a io was e en highe in immunized and immedia e p e-a h i ic mice (day 33). Thus, on day 33, animals ha will no subsequen ly de elop a h i is ha e highe K/T a ios in compa ison o hose ha will de elop a h i is. As was expec ed, ea men wi h 1-MT dec eased he K/T a io in he se um o a h i ic and immedia e p e-a h i ic mice com- pa ed o he ehicle- ea ed ones. A he end o ollow-up, he K/T a io o mice was as low as ha o he immunized mice. 1- MT ea men did no ha e any e ec on K/T a io (Figu e 5). Discussion RA and expe imen al in lamma o y join diseases ha e a p o- g essi e cha ac e wi h in ol emen o inc easing numbe s o join s; howe e , he ini ial and agg essi e acu e phase in a ec ed join s slows down o e ime and he in lamma o y p ocesses bu n ou . Se e al lines o e idence indica e ha he e ec o mechanism ha ini ially a acks he join s is T cell- d i en in esponse o he e ec o p oin lamma o y cy okines, bu he mechanisms esponsible o he limi a ion o acu e in lamma o y p ocesses a e much less unde s ood. The no el inding o ou s udy is ha IDO ac i i y is up egula ed in he acu e phase o CIA e lec ed by he inc eased K/T a io in he se um. Fu he mo e, we could also demons a e ha inhibi ion o IDO in his expe imen al model augmen s he incidence and se e i y o he disease and inc eases he immune esponses o he au oan igens and alloan igens. These da a sugges ha IDO plays a cen al ole in he nega i e egula o y eedback o immunological mechanisms in in lamma o y join diseases. CIA, like RA in humans, is cha ac e ized by he accumula ion o T cells, plasma cells, mac ophages, B cells, mas cells, na - u al kille (NK) cells, and dend i ic cells in he syno ial sublin- ing [17,18]. Fu he mo e, in lamma o y cells in il a ing he syno ial issue in RA and in he acu e phase o CIA exhibi a p edominan ly Th1 pa e n o cy okine exp ession [10,11]. By p iming he Th1- ype in lamma o y cell esponses, IFN-γ is one o he mos impo an p oin lamma o y ac o s in he induc ion o T cell-d i en au oimmune a h i is, such as CIA. Howe e , IFN-γ plays an ambiguous ole in au oimmuni y. A e he ac i- a ion o sel - eac i e lymphocy e clones and o bys ande and accesso y cells in he acu e phase, IFN-γ down egula es he au oimmune p ocesses [19]. Indeed, CIA and CFA de eloped mo e eadily in IFN-γ ecep o -de icien mice han in wild- ype Figu e 2 Hind paw images o DBA miceHind paw images o DBA mice. (a) Hind paws o heal hy non-immunized DBA mice. (b) Type II collagen-immunized (a h i ic) DBA mice ea ed wi h 1-me hyl- yp ophan 6 o 7 days a e he onse o collagen-induced a h i is. (c) Type II collagen-immunized (a h i ic) DBA mice ea ed wi h ehicle 6 o 7 days a e he onse o collagen-induced a h i is. Acco ding o ou sco ing sys em (see Ma e ials and me hods), he h ee s ages ha e been assigned sco es o 0 (a), 4 (b), and 3 (c). A ailable online h p://a h i is- esea ch.com/con en /9/3/R50 Page 5 o 7 (page numbe no o ci a ion pu poses) li e ma es [20]. As a possible explana ion, i has eme ged ha CFA elici s s ong myelopoiesis and expansion o Mac-1+ cells, which play a c ucial ole in disease pa hogenesis, and his p ocess is down egula ed by IFN-γ [20]. Howe e , he exac mechanism esponsible o he e ec o IFN-γ in CIA has no been ully elucida ed. One o he mos likely mechanisms o he down egula ion o CIA by IFN-γ is he inc eased exp ession o IDO by non-T cells [21]. In ib oblas s [22], mac ophages [23], and dend i ic cells [3], IFN-γ s imula es he enzyme IDO, which deg ades he amino acid yp ophan o o m kynu enine, esul ing in he inhi- bi ion o au oimmune p ocesses. Acco ding o his assump- ion, we demons a ed an ele a ed K/T a io, indica ing high IDO ac i i y du ing he acu e phase o CIA. Mo eo e , he K/T a io was e en highe in mice ha ecei ed CII and CFA bu de eloped no clinical and his ological signs o a h i is. These da a sugges ha IDO ac s as a nega i e eedback in his model, and he onse and se e i y o expe imen al a h i is a e in e sely p opo ional o IDO ac i i y. To con i m he egula o y ole o IDO in CIA, we used 1-MT, a known compe i i e inhibi o o IDO. 1-MT did no in luence IFN-γ, bu i signi ican ly supp essed IL-4 p oduc ion by spleen cells, esul ing in an inc eased Th1/Th2 esponse. In 1-MT- ea ed mice, we could demons a e a signi ican dec ease o K/T a io in he immedia e p e-a h i ic and acu e phase o a h i is compa ed o ehicle- ea ed animals, sugges ing he high ac i i y o IDO only in hese s ages o in lamma o y p oc- esses. In o he wo ds, he low blocking ac i i y o 1-MT ei he in he p e-a h i ic phase o in he ch onic phase o a h i is deno es he an i-in lamma o y e ec o IDO only in he case o Figu e 3 Humo al immune esponses in DBA mice immunized wi h ype II colla-gen (CII) and comple e F eund's adju an (CFA) and ea ed wi h 1-me hyl- yp ophan (1-MT) o ehicleHumo al immune esponses in DBA mice immunized wi h ype II colla- gen (CII) and comple e F eund's adju an (CFA) and ea ed wi h 1- me hyl- yp ophan (1-MT) o ehicle. (a) Concen a ions o an ibodies ( o al concen a ions o IgGAM [immunoglobulins G, A, and M]) o he - e ologous (human) CII we e de e mined in he se um o DBA/1 mice immunized wi h CII and CFA and ea ed wi h 1-MT o ehicle. (b) Con- cen a ions o an ibodies ( o al concen a ions o IgGAM) o au ologous (mouse) CII we e de e mined in he se um o DBA/1 mice immunized wi h CII and CFA and ea ed wi h 1-MT o ehicle. Se a we e ob ained on days 5, 22, 33, and 54. Values a e p esen ed as he mean and s anda d e o o he mean o 25 1-MT- ea ed and 20 ehicle- ea ed DBA/1 mice pe g oup and ep esen wo independen expe imen s. *p < 0.05 be ween ehicle and 1-MT- ea ed mice on he co esponding days. Figu e 4 Concen a ions o in e leukin-4 (IL-4) and in e e on-γ (IFN-γ) in he supe na an s o spleen cells ha es ed om mice wi h collagen-induced a h i is and ea ed wi h 1-me hyl- yp ophan (1-MT) o ehicleConcen a ions o in e leukin-4 (IL-4) and in e e on-γ (IFN-γ) in he supe na an s o spleen cells ha es ed om mice wi h collagen- induced a h i is and ea ed wi h 1-me hyl- yp ophan (1-MT) o ehi- cle. The elease o IL-4 (a) and IFN-γ (b) in o he supe na an s o spleen cells in esponse o human o mouse ype II collagen (CII) was de e - mined a he end o he expe imen . Values a e p esen ed as he mean and s anda d e o o he mean o 16 animals pe g oup (*p < 0.01). A h i is Resea ch & The apy Vol 9 No 3 Szán ó e al. Page 6 o 7 (page numbe no o ci a ion pu poses) up egula ion o in lamma o y p ocesses, especially Th1 esponses. As a yp ophan-ca abolizing enzyme, IDO can induce he pe iphe al ole ance and educe he pe sis en immune ac i a- ion. On one hand, IDO dec eases he yp ophan concen a- ion in he mic oen i onmen o in lamma o y cells. Al hough yp ophan is an essen ial amino acid indispensable o he biosyn hesis o p o eins, he low yp ophan concen a ion esul s in he a es o cell p oli e a ion in he mid-G1 a es poin . T cells a e speci ically sensi i e o yp ophan dep i a- ion [2,5] and hus IDO ac i i y can block he po en ial ha m ul au oimmune esponse. On he o he hand, Zhu and colleagues [24] p oposed ha syno ial T cells de i ed om RA syno ial luids a e esis an o IDO-media ed yp ophan dep i a ion. This may be one mechanism by which au o eac i e T cells a e sus ained in i o in pa ien s wi h a h i is [24]. In addi ion, selec ed me aboli es on he yp ophan-kynu enine pa hway a e able o supp ess p oli e a ion o allogeneic T cells and, o a lesse ex en , B and NK cells [25]. Mo eo e , some o he kynu enine de i a es can induce in i o he selec i e apop o- sis o Th1 cells, bu no Th2 cells [7]. In acco dance wi h hese esul s, 1-MT ea men in ou s udy esul ed in an inc eased, mainly Th1 cell-media ed immune esponse o he CII and con- sequen ly in he signi ican wo sening o se e i y and inc ease o onse o CIA media ed by Th1 esponse. In his ega d, ecen s udies wi h he CIA model o a h i is using an an ibody o he cos imula o y molecule CD137, a membe o he umo nec osis ac o ecep o supe amily, a e o g ea in e es . These esul s showed ha ea men o mice wi h he an i-CD137 an ibody esul ed in induc ion o IDO in i o, which was associa ed wi h signi ican amelio a ion o he se e i y o CIA [26]. Fu he mo e, pha macological inhibi ion o IDO e e sed he e ec s o he an i-CD137 an ibody and agg a a ed he a h i is in his model. The yp ophan-IDO pa hway may ha e impo an ele ance o he biological he apy o RA. In a s udy o Boasso and col- leagues [27], CTLA-4-Fc ea men o human pe iphe al blood CD4+ T cells esul ed in inc eased IDO exp ession by hese cells. This e ec was no obse ed in CD8+ T cells. Thus, aba acep (CTLA4-Ig) he apy may ac , a leas in pa , by he s imula ion o IDO p oduc ion. Conclusion The impo ance o IDO ac i i y in he egula ion o CIA sup- po s he hypo hesis ha IDO exp ession by an igen-p esen - ing cells is esponsible o supp ession o undesi able Th1 cells in human in lamma o y join diseases, as yp ophan deg- ada ion could be demons a ed in pa ien s wi h RA [28]. These indings heo e ically aise he possibili y ha he local o sys ema ic induc ion o IDO ac i i y could be es ed in in lamma o y join diseases. Compe ing in e es s The au ho s decla e ha hey ha e no compe ing in e es s. Au ho s' con ibu ions SS pe o med he expe imen al wo k and p epa ed he manu- sc ip . TK pe o med he expe imen al wo k. ZS ad ised on he s udy. KM and TTG a e he heads o labo a o y, supe ised he expe imen al wo k, and ad ised on he s udy. JV is he senio esea che and supe iso o he expe imen al wo k and ad ised on he s udy. All au ho s ead and app o ed he inal manusc ip . Acknowledgemen s The au ho s acknowledge Kuniaki Sai o, o he Na ional Ins i u es o Heal h (NIH), o pe o ming he se um kynu enine assays and he con- ibu ions o many colleagues in he Sec ion o Biochemis y and Molec- ula Biology, Rush Uni e si y and Sec ion o Rheuma ology, Uni e si y o Illinois College o Medicine. This wo k was suppo ed by a g an om he NIH (AR047163). Re e ences 1. A end WP: Physiology o cy okine pa hways in heuma oid a h i is. A h i is Rheum 2001, 45:101-106. 2. Taylo MW, Feng GS: Rela ionship be ween in e e on-gamma, indoleamine 2,3-dioxygenase, and yp ophan ca abolism. FASEB J 1991, 5:2516-2522. 3. Va ga J, Yu i T, B own RR: Inhibi ion o collagenase and s omelysin gene exp ession by in e e on-gamma in human de mal ib oblas s is media ed in pa ia induc ion o yp- ophan deg ada ion. J Clin In es 1995, 96:475-481. 4. Mello AL, Munn DH: IDO exp ession by dend i ic cells: ole - ance and yp ophan ca abolism. Na Re Immunol 2004, 4:762-774. 5. King NJ, Thomas SR: Molecules in ocus: indoleamine 2,3-diox- ygenase. In J Biochem Cell Biol in p ess. 2007 Jan 20 Figu e 5 Kynu enine/ ypophan (K/T) a ios in se a o DBA/1 mice immunized wi h ype II collagen (CII) and comple e F eund's adju an (CFA) and ea ed wi h 1-me hyl- yp ophan (1-MT) o ehicleKynu enine/ ypophan (K/T) a ios in se a o DBA/1 mice immunized wi h ype II collagen (CII) and comple e F eund's adju an (CFA) and ea ed wi h 1-me hyl- yp ophan (1-MT) o ehicle. K and T concen a- ions we e de e mined and K/T a ios we e calcula ed using pooled se a o mice immunized wi h CII and CFA. Each sample con ained he se a o e e y mouse o he co esponding g oup in equal olume. Because mos animals de elop a h i is a e day 30 (Figu e 1), day 33 ep esen s p e-symp oma ic animals. As shown he e a day 33, hose animals, which subsequen ly will no de elop a h i is (-), ha e much highe K/T a io han hose which will de elop a h i is (+). A ailable online h p://a h i is- esea ch.com/con en /9/3/R50 Page 7 o 7 (page numbe no o ci a ion pu poses) 6. Munn DH, Mello AL: Indoleamine 2,3-dioxygenase and umo - induced ole ance. J Clin In es 2007, 117:1147-1154. 7. Yuan W, Collado-Hidalgo A, Yu i T, Taylo M, Va ga J: Modula ion o cellula yp ophan me abolism in human ib oblas s by ans o ming g ow h ac o -be a: selec i e inhibi ion o indoleamine 2,3-dioxygenase and yp ophanyl- RNA syn- he ase gene exp ession. J Cell Physiol 1998, 177:174-186. 8. S ua JM, Townes AS, Kang AH: Type II collagen-induced a h i is. Ann N Y Acad Sci 1985, 460:355-362. 9. Yang HT, Ji hol J, S ensson L, Sund all M, Jansson L, Pe e sson U, Holmdahl R: Iden i ica ion o genes con olling collagen- induced a h i is in mice: s iking homology wi h suscep ibili y loci p e iously iden i ied in he a . J Immunol 1999, 163:2916-2921. 10. Miossec P, an den BW: Th1/Th2 cy okine balance in a h i is. A h i is Rheum 1997, 40:2105-2115. 11. Mau i C, Williams RO, Walmsley M, Feldmann M: Rela ionship be ween Th1/Th2 cy okine pa e ns and he a h i ogenic esponse in collagen-induced a h i is. Eu J Immunol 1996, 26:1511-1518. 12. Ma ino a-Mu a chie a L, Williams RO, Mason LJ, Mau i C, Feld- mann M, Maini RN: Dynamics o p oin lamma o y cy okine exp ession in he join s o mice wi h collagen-induced a h i is (CIA). Clin Exp Immunol 1997, 107:507-512. 13. Ada iche VA, Ba dos T, Ch is odoulou S, Phillips T, Mikecz K, Glan TT: Majo his ocompa ibili y complex con ols suscep i- bili y and dominan inhe i ance, bu no he se e i y o he dis- ease in mouse models o heuma oid a h i is. Immunogene ics 2002, 54:184-192. 14. Ada iche VA, Valdez JC, Ba dos T, Finnegan A, Mikecz K, Glan TT: Combined au oimmune models o a h i is e eal sha ed and independen quali a i e (bina y) and quan i a i e ai loci. J Immunol 2003, 170:2283-2292. 15. O o JM, Chand aseke an R, Ve mes C, Mikecz K, Finnegan A, Ricke SE, Ende s JT, Glan TT: A genome scan using a no el gene ic c oss iden i ies new suscep ibili y loci and ai s in a mouse model o heuma oid a h i is. J Immunol 2000, 165:5278-5286. 16. Widne B, We ne ER, Schennach H, Wach e H, Fuchs D: Simul- aneous measu emen o se um yp ophan and kynu enine by HPLC. Clin Chem 1997, 43:2424-2426. 17. Yanni G, Whelan A, Feighe y C, B esnihan B: Analysis o cell popula ions in heuma oid a h i is syno ial issues. Semin A h i is Rheum 1992, 21:393-399. 18. E landsson HH, Liljes om M, Kla eskog L: Cha ac e is ics o syn- o ial luid e usion in collagen-induced a h i is (CIA) in he DA a ; a compa ison o his ology and an ibody eac i i ies in an expe imen al ch onic a h i is model and heuma oid a h i is (RA). Clin Exp Immunol 1997, 107:480-484. 19. Rosloniec EF, La ham K, Guedez YB: Pa adoxical oles o IFN- gamma in models o Th1-media ed au oimmuni y. A h i is Res 2002, 4:333-336. 20. Ma hys P, Ve mei e K, Mi e a T, He emans H, Huang S, Schols D, De Wol -Pee e s C, Billiau A: Enhanced au oimmune a h i is in IFN-gamma ecep o -de icien mice is condi ioned by myco- bac e ia in F eund's adju an and by inc eased expansion o Mac-1+ myeloid cells. J Immunol 1999, 163:3503-3510. 21. By ne GI, Lehmann LK, Ki schbaum JG, Bo den EC, Lee CM, B own RR: Induc ion o yp ophan deg ada ion in i o and in i o: a gamma-in e e on-s imula ed ac i i y. J In e e on Res 1986, 6:389-396. 22. Dai W, Gup a SL: Regula ion o indoleamine 2,3-dioxygenase gene exp ession in human ib oblas s by in e e on-gamma. Ups eam con ol egion disc imina es be ween in e e on- gamma and in e e on-alpha. J Biol Chem 1990, 265:19871-19877. 23. Ca lin JM, Bo den EC, Sondel PM, By ne GI: In e e on-induced indoleamine 2,3-dioxygenase ac i i y in human mononuclea phagocy es. J Leukoc Biol 1989, 45:29-34. 24. Zhu L, Ji F, Wang Y, Zhang Y, Liu Q, Zhang JZ, Ma sushima K, Cao Q, Zhang Y: Syno ial au o eac i e T cells in heuma oid a h i- is esis IDO-media ed inhibi ion. J Immunol 2006, 177:8226-8233. 25. Te ness P, Baue TM, Rose L, Du e C, Wa zlik A, Simon H, Opelz G: Inhibi ion o allogeneic T cell p oli e a ion by indoleamine 2,3-dioxygenase-exp essing dend i ic cells: media ion o sup- p ession by yp ophan me aboli es. J Exp Med 2002, 196:447-457. 26. Seo SK, Choi JH, Kim YH, Kang WJ, Pa k HY, Suh JH, Choi BK, Vinay DS, Kwon BS: 4-1BB-media ed immuno he apy o heu- ma oid a h i is. Na Med 2004, 10:1088-1094. 27. Boasso A, He beu al JP, Ha dy AW, Winkle C, Shea e GM: Reg- ula ion o indoleamine 2,3-dioxygenase and yp ophanyl- RNA-syn he ase by CTLA-4-Fc in human CD4+ T cells. Blood 2005, 105:1574-1581. 28. Sch oecksnadel K, Winkle C, Du ne C, Wi lei ne B, Schi me M, Fuchs D: T yp ophan deg ada ion inc eases wi h s age in pa ien s wi h heuma oid a h i is. Clin Rheuma ol 2006, 25:334-337.