Vasoactive intestinal peptide induces regulatory dendritic cells with therapeutic effects on autoimmune disorders

Vasoac i e in es inal pep ide induces egula o y dend i ic cells wi h he apeu ic e ec s on au oimmune diso de s Alejo Cho ny*, Elena Gonzalez-Rey*, Amelia Fe nandez-Ma in*, Da id Pozo † , Doina Ganea ‡ , and Ma io Delgado* § *Ins i u e o Pa asi ology and Biomedicine, Consejo Supe io de In es igaciones Cien i icas, 18100 G anada, Spain; †Depa men o Biochemis y and Molecula Biology, Se ille Uni e si y, 41013 Se ille, Spain; and ‡Depa men o Biological Sciences, Ru ge s he S a e Uni e si y o New Je sey, Newa k, NJ 07102 Edi ed by Ralph M. S einman, The Rocke elle Uni e si y, New Yo k, NY, and app o ed July 29, 2005 ( ecei ed o e iew May 31, 2005) The induc ion o an igen-speci ic ole ance is c i ical o he p e- en ion o au oimmuni y and main enance o immune ole ance. In addi ion o hei classical ole as sen inels o he immune- esponse-inducing T cell eac i i y, dend i ic cells (DCs) play an impo an ole in main aining pe iphe al ole ance h ough he induc ion兾ac i a ion o egula o y T cells (T ). The possibili y o gene a e ole ogenic DCs opens new he apeu ic pe spec i es in au oimmune兾in lamma o y diseases. The e o e, he cha ac e iza- ion o he endogenous ac o s ha con ibu e o he de elopmen o ole ogenic DCs is highly ele an . In his s udy, we epo on he use o he known immunosupp essi e neu opep ide, he asoac- i e in es inal pep ide, as a new app oach o induce ole ogenic DCs wi h capaci y o gene a e T cells, o es o e ole ance in i o, and o educe he p og ession o heuma oid a h i is and expe - imen al au oimmune encephalomyeli is. au oimmuni y 兩 egula o y T cell 兩 ole ance The immune sys em is aced wi h he daun ing job o p o ec ing he hos om an a ay o pa hogens, while main aining ole - ance o sel -an igens (Ags). The induc ion o Ag-speci ic ole ance is essen ial o main ain immune homeos asis, o con ol au o eac- i e T cells, p e en ing he onse o au oimmune diseases, and o achie e ole ance owa d ansplan s. Bo h hymic and pe iphe al mechanisms accoun o he abili y o he immune sys em o induce ole ance. A en ion has been ocused ecen ly on induc ion o ac i e supp ession by egula o y T cells (T ) (1), and dend i ic cells (DCs) ha e been shown o con ibu e o T cell ole ance (2, 3). The ma u a ion兾ac i a ion s a e o DCs migh be he con ol poin o he induc ion o pe iphe al ole ance, by p omo ing T di e en i- a ion. Thus, whe eas ma u e DCs (mDCs) a e po en Ag- p esen ing cells enhancing T cell immuni y, imma u e DCs (iDCs) a e in ol ed in he induc ion o pe iphe al T cell ole ance unde s eady-s a e condi ions (2–6). Howe e , he clinical use o iDCs may no be sui able o he ea men o au oimmune diseases, because iDCs a e likely o ma u e in in lamma o y condi ions (6), emphasizing he need o de elop ole ogenic DCs wi h a s ong po en ial o induce T . Immunosupp essi e he apy, adi ionally ocused on lymphocy es, has been e olu ionized by a ge ing he de elopmen and key unc ions o DC, and he gene a ion o ole ogenic DCs in he labo a o y has become he ocus o new he apies (7). Vasoac i e in es inal pep ide (VIP) is a neu opep ide eleased by bo h inne a ion and immune cells, pa icula ly T helpe (Th)2 cells, in esponse o Ag s imula ion and unde in lamma o y兾 au oimmune condi ions (8). VIP elici s a b oad spec um o bio- logical unc ions, including immunomodula ion, p edominan ly ac - ing as a po en an iin lamma o y ac o and a supp essi e agen o Th1 esponses (9). The e o e, VIP has eme ged as a p omising he apeu ic ac o o he ea men o au oimmune兾in lamma o y diseases, including heuma oid a h i is (RA), ulce a i e coli is, u eo e ini is, and expe imen al au oimmune encephalomyeli is (EAE) (10–12). In his s udy, we in es iga ed whe he he p esence o VIP du ing he ea ly phases o DC di e en ia ion induces he gene a ion o egula o y DCs wi h he capaci y o induce T and o p e en au oimmuni y. Ma e ials and Me hods Cell Isola ion and Cul u es. Bone ma ow (BM)-de i ed DCs (BM-DCs) we e gene a ed as desc ibed in e . 13. B ie ly, BM cells (2 ⫻10 6 ) ob ained om BALB兾c (H-2 d ), C57BL兾6 (H-2 b ), o DBA兾1 (H-2 q ) mice we e incuba ed in comple e medium (RPMI medium 1640 supplemen ed wi h 100 uni s/ml penicillin兾 s ec omycin, 2 mM L-glu amine, 50 ␮ M 2-me cap oe hanol, and 10% hea -inac i a ed FCS) con aining 20 ng兾ml g anulocy e mac ophage colony-s imula ing ac o in he p esence o absence o VIP (10 ⫺8 M). A day 6, nonadhe en cells we e collec ed ( ou inely con aining 80–90% CD11c ⫹ cells) and s imula ed o 48 h wi h LPS (1 ␮ g/ml) o induce ac i a ion兾ma u a ion. In some expe imen s, DCs we e pulsed wi h o oalbumin (OVA), collagen II (CII), o myelin oligodend ocy e glycop o ein (MOG) (20 ␮ g兾ml) o 12 h. Allogeneic naı¨ e CD4 T cells we e pu i ied om C57BL兾6 mice by posi i e immunomagne ic se- lec ion (MACS, Mil enyi Bio ec, Aubu n, CA). Flow Cy ome y. Cells we e incuba ed wi h a ious pe idinin– chlo ophyll–p o ein complex (Pe CP)-, FITC- and phycoe y h in (PE)-labeled mAbs (BD Pha mingen), dilu ed a op imal concen- a ion o immunos aining, ixed in 1% pa a o maldehyde, and analyzed on a FACSCalibu low cy ome e (Bec on Dickinson). We used iso ype-ma ched Abs as con ols and IgG block (Sigma) o a oid he nonspeci ic binding o Fc- ecep o s. Cy okine Assays. Cy okine con en s in he cul u e supe na an s we e de e mined by speci ic sandwich ELISAs by using cap u e兾bio in- yla ed de ec ion Abs om BD Pha mingen. Fo in acellula anal- ysis o cy okines in es imula ed CD4 T cells, 10 6 cells pe ml we e collec ed and s imula ed wi h pho bol 12-my is a e 13-ace a e (1 ng兾ml) plus ionomycin (20 ng兾ml) o 8hin hep esence o monensin. Cells we e s ained wi h Pe CP-an i-CD4 mAbs o 30 min a 4°C, washed, ixed兾saponin-pe meabilized wi h Cy o ix兾 Cy ope m, s ained wi h 0.5 ␮ g pe sample FITC- and PE- conjuga ed an icy okine-speci ic mAbs, and analyzed by low cy- ome y. To dis inguish be ween DC and T cell sou ces, This pape was submi ed di ec ly (T ack II) o he PNAS o ice. Abb e ia ions: Ag, an igen; BM, bone ma ow; DC, ded i ic cell; BM-DC, BM-de i ed DC; CIA, collagen-induced a h i is; CII, collagen II; DLN, d aining lymph nodes; DTH, delayed- ype hype sensi i i y; EAE, expe imen al au oimmune encephalomyeli is; iDC, imma u e DC; mDC, ma u e DC; MOG, myelin oligodend ocy e glycop o ein; OVA, o albumin; RA, heuma oid a h i is; T , egula o y T cells; Th, T helpe ; VIP, asoac i e in es inal pep ide §To whom co espondence should be add essed a : Ins i u o de Pa asi ologia y Biomedi- cina, Consejo Supe io de In es igaciones Cien i icas, A enida Conocimien o, PT Ciencias de la Salud, G anada 18100, Spain. E-mail: [email p o ec ed]. © 2005 by The Na ional Academy o Sciences o he USA 13562–13567 兩 PNAS 兩 Sep embe 20, 2005 兩 ol. 102 兩 no. 38 www.pnas.o g兾cgi兾doi兾10.1073兾pnas.0504484102 in acellula cy okine analysis was done exclusi ely in he Pe CP- labeled CD4 T cell popula ion. mRNA Analysis. To al RNA was isola ed om CD4 T cells, and eal- ime PCR was used o de e mine Foxp3 and neu opilin mRNA exp ession, as desc ibed in e s. 14 and 15. Mixed Leukocy e Reac ion and Analysis o T Cell Func ion. Naı¨ e CD4 T cells (2 ⫻10 5 ) we e cul u ed wi h allogeneic DC con ol o DC VIP a a ious T:DC a ios in he p esence o IL-2 (100 uni s兾ml) o 3d. Cell p oli e a ion was e alua ed by using a cell-p oli e a ion assay (B dU d) om Roche Diagnos ics (Mannheim, Ge many), and in acellula cy okine con en was de e mined as desc ibed abo e. In some expe imen s, DCs (10 5 ) we e cul u ed wi h pu i ied allogeneic CD4 T cells (5 ⫻10 5 ). One week la e , CD4 T cells we e eco e ed by immuodeple ion o CD11c ⫹ DCs and cul u ed in di e en numbe s wi h syngeneic CD4 T cells (5 ⫻10 5 )in he p esence o allogeneic mDC (10 5 ), and he p oli e a i e esponse was de e mined. Some cul u es we e pe o med in he p esence o blocking an i-IL-10 (10 ␮ g兾ml) and兾o an i-TGF ␤ 1 (40 ␮ g兾ml) mAbs. To de e mine he cell-con ac -dependence o he egula o y esponse, we placed esponde CD4 T cells (5 ⫻10 5 ) wi h LPS- ma u ed DC (10 5 ) in he bo om well o a T answell sys em (Millipo e) and syngeneic T VIP (2 ⫻10 5 ) wi h allogeneic mDC (10 5 ) in he uppe T answell chambe . A e 72 h, we measu ed he p oli e a i e esponse o he bys ande eac i e CD4 T cells in he bo om well. To gene a e CII- and MOG-speci ic T cells, DBA1兾J and C57BL兾6 DCs (10 5 ) pulsed wi h CII o MOG, espec i ely, we e cul u ed wi h syngeneic CD4 T cells (5 ⫻10 5 ) o 1week in he p esence o CII o MOG (20 ␮ g兾ml). Immuniza ion Model. BALB兾c mice we e injec ed s.c. wi h di e en numbe s ( om 50 o 5 ⫻10 5 ) o cells o me hyla ed BSA (mBSA)- pulsed DC con ol o DC VIP , ollowed a week la e by s.c. immuni- za ion wi h he Ags mBSA o OVA (60 ␮ g) in comple e F eund’s adju an . Fi e days a e Ag immuniza ion, se um Ag-speci ic Ab, d aining lymph nodes (DLN) T cell p oli e a i e esponses, and delayed ype hype sensi i i y (DTH) esponses we e measu ed. Fo he DTH esponses, mice we e injec ed i.d. wi h Ag (5 ␮ g) o saline in o he ea s, and ea swelling was measu ed 24 h la e by using a calipe . Ag-speci ic T cell p oli e a i e esponses we e measu ed a e ex i o s imula ion o DLN cells (4 ⫻10 5 ) wi h 10 ␮ M Ag. Le els o mBSA-speci ic IgG in se um we e de e mined by ELISA, as desc ibed in e . 16. Model o RA and EAE. RA was induced in DBA1兾Jmicebys.c. injec ion o CII, as desc ibed in e . 10. Ch onic EAE was induced in C57BL兾6 mice by s.c. immuniza ion wi h MOG 35–55 , as desc ibed in e . 17. Mice wi h es ablished a h i is (wi h a clinical sco e o 2) we e injec ed i. . wi h di e en numbe s o syngeneic CII-pulsed DC con ol o DC VIP o wi h CII-speci ic T con ol o T VIP . Mice wi h es ablished EAE (wi h a clinical sco e o 1) we e injec ed i. . wi h di e en numbe s o syngeneic MOG 35–55 -pulsed DC con ol o DC VIP o wi h MOG 35–55 -speci ic T con ol o T VIP . The clinical sco e was de e mined daily, based on join in lamma ion o RA and ail兾leg pa alysis o EAE, as desc ibed in e . 17. DLN cells we e isola ed a he peak o he diseases, s imula ed wi h CII o MOG 35–55 (20 ␮ g兾ml), and assayed o p oli e a ion and cy okine p oduc ion, as desc ibed abo e. The con en o se um an i-CII o an i-MOG 35–55 IgG an ibodies was de e mined by ELISA, as desc ibed in e s. 10 and 18. To assess Ag-speci ici y, a h i ic mice we e injec ed wi h unpulsed, OVA-pulsed, o CII- o MOG 35–55 - pulsed DC con ol o DC VIP and immunized s.c. wi h OVA, CII, o MOG 35–55 (150 ␮ g o Ag in comple e F eund’s adju an ) one week la e . A e 5 d, mice ecei ed 5 ␮ g o Ag i.d. in he ea pinna, and he DTH esponse was de e mined, as desc ibed abo e. In some expe imen s, collagen-induced a h i is (CIA) and EAE mice e- cei ed i. . injec ions o neu alizing an i-IL-10 polyclonal Ab, neu alizing an i-TGF ␤ mAb, o p eimmune a IgG used as con ol Ig (500 ␮ g o Ab pe mouse) on al e na e days up o8da e onse o disease. Resul s and Discussion The induc ion o Ag-speci ic ole ance is c i ical o he p e en ion o au oimmuni y and main enance o immune ole ance. In addi- ion o hei classical ole as sen inels o he immune esponse inducing T cell eac i i y, inc easing e idence now indica es ha DCs can induce speci ic T cell ole ance. Al hough unde lying mechanisms a e no ully elucida ed, he capaci y o induce T cells is an impo an p ope y o ole ogenic兾 egula o y DCs. The gene a ion o ‘‘designe ’’ DCs wi h ole ogenic p ope ies in he labo a o y by using speci ic cy okines o immunologic and pha - macologic eagen s is a desi able goal and ep esen s he subjec o in ensi e in es iga ions. Because o i s immunosupp essi e ac ion, VIP is a candida e o he induc ion o egula o y DCs wi h capaci y o gene a e T . In a p e ious s udy, we showed ha VIP ea men o ac i a ed DCs educes hei capaci y o ac i a e allogeneic and syngeneic T cells, an e ec associa ed wi h he p e en ion o CD80兾CD86 up- egula ion (19). VIP ea men o iDC in he absence o ac i a ion esul ed in DCs wi h inc eased capaci y o induce Th2 esponses (19). Howe e , o he immunomodula o y ac o s wi h capaci y o induce ole ogenic DCs ha e been ound o be e ec i e when adminis e ed du ing he di e en ia ion o DCs (6, 7). The e o e, we de e mined whe he exposu e o VIP du ing DC di e en ia ion esul s in DC pheno ypic and unc ional changes. BM-DC Di e en ia ed wi h VIP Induce Regula o y T 1-Like Cells and Tole ance in Vi o .We i s compa ed mu ine BM-de i ed DCs gene a ed in he p esence o absence o VIP in e ms o su ace ma ke s and cy okine p oduc ion. As p e iously desc ibed, BM cells cul u ed wi h g anulocy e mac ophage colony-s imula ing ac o o 6ddi e en ia e in o iDCs (da a no shown). Upon LPS s imula ion, iDCs ma u e o DCs exp essing high le els o DC ma ke s (CD11c), MHC molecules (class I and class II), and cos imula o y molecules (CD40, CD80, and CD86) (Fig. 1a, DC con ol ). Howe e , DCs gene a ed in he p esence o VIP (DC VIP ) we e esis an o he LPS-induced up- egula ion o he cos imula o y molecules (CD40, CD80, and CD86) (Fig. 1a). Upon Fig. 1. VIP induces a s able ‘‘semima u e’’ pheno ype in BM-DCs. DCs we e gene a ed om mouse BM cells in he absence (DCcon ol) o p esence (DCVIP) o VIP and ac i a ed wi h LPS o induce DC ma u a ion. (a)DC con ol and DCVIP we e double-labeled o di e en ma ke s and analyzed by low cy ome y. Numbe s ep esen he pe cen age o posi i e cells (n⫽4). (b) Cy okine con en in he DC supe na an s was de e mined by ELISA (n⫽4). Cho ny e al. PNAS 兩 Sep embe 20, 2005 兩 ol. 102 兩 no. 38 兩 13563 IMMUNOLOGY oll-like ecep o ac i a ion, iDCs ma u e in o cells capable o p oducing high le els o in lamma o y cy okines. In con as o DC con ol , which p oduce TNF and IL-12, and low le els o IL-10, DC VIP p oduce e y low le els o p oin lamma o y cy okines (TNF and IL-12) bu sec e e signi ican le els o he an iin lamma o y cy okine IL-10 (Fig. 1b). Taken oge he , hese esul s indica e ha he DCs gene a ed in he p esence o VIP a e esis an o LPS- induced up- egula ion o cos imula o y molecules and p oduce IL-10. These cha ac e is ics a e qui e simila o hose epo ed o ole ogenic DCs gene a ed wi h o he immunomodula o y ac o s, such as IL-10 o he ac i a ed o m o i amin D 1,25(OH) 2 D 3 (2–4, 20–24). Tole ogenic DCs a e poo s imula o s o T cell p oli e a ion and cy okine p oduc ion (20, 25–28). To examine he capaci y o he DC VIP o s imula e T cells, we cocul u ed DC con ol o DC VIP wi h allo eac i e CD4 T cells. P iming wi h DC con ol esul s in a s ong p oli e a ion o allogeneic CD4 T cells, whe eas DC VIP induce only weak p oli e a ion (Fig. 2a). In addi ion, CD4 T cells p imed wi h DC VIP eexposed o esh LPS-s imula ed allogeneic DCs (mDC) did no p oli e a e (da a no shown), indica ing ha DC VIP induces ane gic T cells and兾o T . Al hough T gene a ed by exposu e o egula o y兾 ole ogenic DCs do no p oli e a e in esponse o he Ag, hey can elease an iin lamma o y cy okines, such as IL-10 and TGF ␤ . The e o e, we assessed he cy okine p o ile o T cells cocul u ed wi h DC VIP . In con as o T cells exposed o DC con ol , which show a p edominan Th1 cy okine p o ile, wi h high le els o IFN ␥ and IL-2, CD4 T cells p imed wi h allogeneic DC VIP exhibi a T 1-like pheno ype, cha ac e ized by IL-10 and TGF ␤ bu no IL-2 and IFN ␥ p oduc ion (Fig. 2b). A e TCR s imula ion, T cells supp ess he p oli e a ion and IL-2 p oduc ion o Ag-speci ic e ec o T cells. To de e mine whe he T cells exposed o DC VIP become unc ional T , we es imula ed CD4 T cells wi h allogeneic mDCs in he p esence o syngeneic CD4 T cells p e iously exposed o allogeneic DC con ol (T con ol )o DC VIP (T VIP ). T VIP inhibi he p oli e a ion o syngeneic esponde CD4 cells in esponse o allogeneic mDCs in a dose-dependen manne , whe eas T con ol a e no supp essi e (Fig. 2c). Simila esul s we e ob ained in espec o IL-2 p oduc ion (da a no shown). The e o e, he pheno ype o T VIP co ela es wi h hei egula o y T cell ac i i y. The obse a ion ha T VIP p oduce high le els o he immuno- supp essi e cy okines IL-10 and TGF ␤ sugges s ha he inhibi o y Fig. 2. Mu ine DCs di e en ia ed in he p esence o VIP induce egula o y T 1 cells and ole ance in i o. DCs we e gene a ed om mouse BM cells in he ab- sence (DCcon ol) o p esence (DCVIP) o VIP and ac i- a ed wi h LPS o induce DC ma u a ion. (a)DC con ol o DCVIP was added o allogeneic CD4 T cells (5 ⫻105), and he p oli e a i e esponse was de e mined. DCcon ol o DCVIP wi hou T cells did no p oli e a e. Each esul is he mean ⫾SD o h ee expe imen s pe o med in duplica e. (b) Pu i ied CD4 T cells we e exposed o allogeneic DCcon ol o DCVIP and ac i a ed wi h pho - bol 12-my is a e 13-ace a e plus ionomycin. In acel- lula cy okines we e de e mined in CD4-ga ed cells by low cy ome y. Numbe s ep esen pe cen age o posi i e cells (n⫽5). (c) Pu i ied CD4 T cells we e s imula ed o 1 week wi h allogeneic DCcon ol o DCVIP. The esul ing egula o y CD4 T cells (T ) we e incuba ed wi h syngeneic esponde CD4 T cells ( CD4) in he p esence o allogeneic mDCs, and he p oli e - a i e esponse was de e mined (n⫽4). (d) Isola ed CD4 T cells we e cocul u ed wi h syngeneic T VIP and allogeneic mDCs in he p esence o absence o block- ing an i-IL10 and兾o an i-TGF ␤ . Addi ionally, CD4⫹mDCs we e sepa a ed om T VIP⫹mDC in a T answell sys em. The p oli e a i e esponse o e- sponde CD4 T cells was de e mined (n⫽4). (e) So ed CD4 T cells gene a ed wi h DCcon ol o DCVIP we e analyzed o neu opilin 1 and Foxp3 mRNA exp ession by eal- ime RT-PCR and o su ace CD103 and glu- coco icoid-induced TNF ecep o (GITR) exp ession by low cy ome y. Open his og ams and dashed lines ep esen iso ype con ols. One ep esen a i e expe - imen o wo is shown. ( ) Mice we e injec ed s.c. wi h inc easing numbe s ( om 50 o 5 ⫻105cells) o Ag- pulsed DCcon ol o DCVIP 1 week be o e p iming wi h Ag. Fi e days la e , mice we e es ed o DLN Ag- speci ic T cell p oli e a ion, se um an ibody le els, and DTH esponses. Mice injec ed wi h Ag alone (None) we e used as con ols. Resul s a e he mean ⫾SD o each g oup (n⫽4) es ed sepa a ely and a e ep e- sen a i e o h ee expe imen s. 13564 兩 www.pnas.o g兾cgi兾doi兾10.1073兾pnas.0504484102 Cho ny e al. e ec o T VIP on esponde CD4 T cell p oli e a ion migh be media ed h ough soluble ac o s p oduced. When T VIP and esponde CD4 T cells we e sepa a ed in answell expe imen s by a semipe meable memb ane ha allows he ee exchange o soluble ac o s bu excludes di ec cell con ac o esponde CD4 T cells and T VIP , he p oli e a ion o e ec o CD4 cells was s ill inhibi ed, indica ing ha soluble ac o s media e he inhibi o y e ec (Fig. 2d). In egula cocul u es, he addi ion o an i-TGF ␤ ,o an i-IL-10 Abs e e sed inhibi ion modes ly. Howe e , he addi ion o bo h an i-IL-10 and an i-TGF ␤ Abs e e ses he inhibi o y e ec almos comple ely (Fig. 2d). Se e al popula ions o CD4 T ha e been desc ibed and cha - ac e ized, including he na u ally occu ing hymic-bo n CD4 ⫹ CD25 ⫹ T and he induced pe iphe al T , consis ing o IL-10-p oducing T 1 and TGF ␤ -sec e ing Th3兾T 2 (29). Regula- o y DCs do no pa icipa e in he gene a ion o na u ally occu ing CD4 ⫹ CD25 ⫹ T ; howe e , hey play an impo an ole in he di e en ia ion o pe iphe ally induced T 1 and Th3兾T 2 T (30– 32). Al hough he CD4 ⫹ CD25 ⫹ popula ion is sligh ly inc eased in T VIP , he ac ha T VIP did no exp ess signi ican le els o he CD4 ⫹ CD25 ⫹ T ma ke s Foxp3, neu opilin-1, glucoco icoid- induced TNF- ecep o - amily- ela ed gene, and CD103 (Fig. 2e), a gues agains he possibili y ha DC VIP induce he gene a ion o CD4 ⫹ CD25 ⫹ T cells. The e a e no epo s on he exp ession o neu opilin-1 in IL-10-induced T 1 cells. Howe e , in con as o CD4 ⫹ CD25 ⫹ T , and in ag eemen wi h ou esul s, T 1 cells gene a ed by epe i i e s imula ion wi h IL-10-sec e ing egula o y DCs ha e been shown o exp ess low le els o CD25 and Foxp3 (33). Al hough he p ecise mechanisms emain unknown, se e al possibili ies may accoun o he gene a ion o T cells by DC VIP . The ac i a ion o naı¨ e CD4 T lymphocy es equi es se e al signals deli e ed by mDCs and media ed h ough Ag兾MHCII–TCR, CD80兾CD86–CD28, and CD40–CD40L in e ac ions. Cos imula- o y molecules, especially CD40, appea o be key de e minan s o he decision be ween ole ance and immuni y (34). The cha ac e - is ic pheno ype o DC VIP , i.e., high le els o MHC plus poo exp ession o cos imula o y molecules, which will deli e s imula- o y bu no cos imula o y signals, is in ag eemen wi h DC VIP ’s ole ance-inducing abili y. In addi ion, he obse a ion ha DC VIP sec e e IL-10 may be linked o he s abili y o DC VIP ’s ole ogenic- Fig. 3. The apeu ic e ec o DC di e en ia ed wi h VIP in RA and EAE. (a) DBA1兾J mice (H-2q) wi h es ablished CIA o C57BL兾6 mice (H-2b) wi h es ablished EAE we e ea ed (a ows) wi h syngeneic CII-pulsed DCs o MOG-pulsed DCs, espec i ely, gene a ed in he absence (DCcon ol, E ) o p esence (DCVIP,o VIP. Un ea ed CIA and EAE mice (none, F ) we e used as con ols. Clinical sco e was moni o ed (n⫽12). (b) CII- and MOG-pulsed DCVIP we e injec ed a di e en doses. (c) CII-induced p oli e a ion and IFN ␥ p oduc ion by spleen T cells, and he le els o an i-CII IgG in se a we e de e mined in CIA mice injec ed wi h DCcon ol o DCVIP (n⫽5). (d) The e ec o DCVIP is Ag-speci ic. A h i ic mice we e ea ed wi h unpulsed, CII-pulsed, o OVA-pulsed DCcon ol o DCVIP a e disease onse . One week la e , mice we e immunized s.c. wi h OVA o CII and challenged i.d. in he ea pinna wi h he espec i e Ag 5 d la e . Clinical sco e and DTH esponses we e de e mined 24 h la e (n⫽5). (e) Un ea ed CIA o EAE mice o animals injec ed wi h DCVIP and ea ed wi h con ol Ig, an i-IL10, an i-TGF ␤ , o a combina ion o bo h mAbs (10 mice pe g oup). Cho ny e al. PNAS 兩 Sep embe 20, 2005 兩 ol. 102 兩 no. 38 兩 13565 IMMUNOLOGY like pheno ype (20, 35–37). P e iously, VIP has been epo ed o inhibi NF- ␬ Bp65 nuclea ansloca ion, DNA-binding, and ans- ac i a ing ac i i y in mac ophages (9), and we ha e ecen ly ound ha bo h NF- ␬ Bp65 nuclea ansloca ion and IkB phospho yla ion a e inhibi ed in DC VIP (M.D., E.G.-R., and D.G., unpublished da a). The connec ion among NF- ␬ B ansac i a ing ac i i y, CD40 exp ession, and DC unc ion (including TNF- ␣ and IL-12 p oduc- ion) has been es ablished in a numbe o ecen s udies. The associa ion be ween ole ance, pa icula ly ole ogenic DCs, and lack o CD40 exp ession o signaling has been demons a ed bo h in i o and in i o (28). Exp ession o CD40 depends on NF- ␬ Bp65 (38), and he inhibi ion o NF- ␬ B in DCs leads o ailu e o CD40, CD80, and CD86 exp ession upon LPS-s imula ion and o he gene a ion o ole ogenic DCs (39). In addi ion, a ecen s udy sugges s ha VIP ea men induced a dec ease o oll-like ecep- o s (TLR-2兾4) exp essions in DCs in a mu ine model o C ohn’s disease by a mechanism ha would in ol e a dec ease o NF- ␬ B ac i a ion (40). The e o e, we would like o p opose ha he mechanism by which VIP induces ole ogenic DCs in ol es he cAMP兾PKA-media ed inhibi ion o I ␬ B phospho yla ion and NF- ␬ Bp65 nuclea ansloca ion, leading o lack o CD40 exp ession, TLR-2兾4 signaling, and in lamma o y cy okine p oduc ion. Because DC VIP appea o ha e a p edominan ly nega i e e ec on Th1 cells, we de e mined he e ec o DC VIP in an in i o model o DTH. Ag (me hyla ed-BSA)-pulsed DC VIP and DC con ol we e adminis e ed i. ., ollowed a week la e by s.c. an igenic immuni- za ion. We de e mined T cell p oli e a ion in esponse o ex i o es imula ion, Ab p oduc ion, and DTH a e a seconda y s.c. Ag adminis a ion (Fig. 2 ). Mice ha ecei ed DC con ol de eloped DTH eac ions highe han con ols (no DCs), whe eas hose ecei ing DC VIP exhibi ed educed DTH. In addi ion, DLN T cells om mice inocula ed wi h DC con ol p oli e a ed a highe le els han con ols (no DCs), and, again, inocula ion o DC VIP esul ed in a subs an ial educ ion in T cell p oli e a ion a e ex i o es imula ion wi h he Ag. Simila ly, mice inocula ed wi h DC con ol p oduced high le els o an i-mBSA Abs, whe eas hose inocula ed wi h DC VIP had an i-mBSA Ab le els below con ol (no DCs) (Fig. 2 ). These esul s indica e ha DC VIP induce ole ance in i o. The induc ion o ole ance is es ic ed o he Ag p esen ed by he inocula ed DC, because we did no obse e educ ion in DTH in mice injec ed wi h mBSA-pulsed DC VIP when we used an un ela ed Ag (OVA) o immuniza ion and ex i o T cell es imula ion (da a no shown). These expe imen s sugges he possibili y o using a VIP in i o sys em o gene a e Ag-speci ic ole ogenic DCs, ollowed by in i o adminis a ion o hese cells o pa ien s wi h au oimmune diseases. The apeu ic E ec o DC VIP in Au oimmuni y. Se e al epo s ha e ecen ly p oposed he possibili y o using egula o y兾 ole ogenic DCs gene a ed ex i o as a he apeu ic ool o p e en o gan- speci ic au oimmune diseases (3, 7, 21). In e es ingly, DC VIP e- ained hei T cell egula o y capaci y in i o and in i o unde in lamma o y condi ions. This obse a ion is pa icula ly ele an o condi ions in which ongoing Ag p esen a ion is associa ed wi h ch onic in lamma ion, including au oimmune diseases. The e o e, we es ed he he apeu ic e ec o DC VIP in wo mu ine models o RA and mul iple scle osis (MS). Fo RA, we used he CIA, an expe imen al disease model induced by immuniza ion wi h CII, which sha es a numbe o clinical, his ologic, and immunological ea u es wi h RA. Fo MS, we used he EAE model induced by MOG 35–55 in C57BL兾6 mice ha mi o di e en clinical cha ac- e is ics o MS. Inocula ion o DC con ol does no amelio a e a h i is (i.e., join in lamma ion, ca ilage des uc ion, and bone e osion) o EAE (i.e., ail and leg pa alysis) (Fig. 3a). In con as , adminis a- ion o syngeneic DC VIP a e he onse o disease ab oga es a h i is and EAE p og ession in a dose-dependen manne (Fig. 3 aand b). The he apeu ic e ec o DC VIP was associa ed wi h he down- egula ion o he au oimmune componen o bo h diseases, because DLN T cells om DC VIP - ea ed mice showed weak p oli e a ion and IFN ␥ p oduc ion in esponse o he au oan igen (Fig. 3c). Fu he mo e, his inhibi ion o he Th1- ype au o eac i e esponse co ela es wi h dec eased le els o CII- and MOG-speci ic au o- an ibodies (Fig. 3c). The e ec o DC VIP was Ag-speci ic. Unpulsed o OVA-pulsed DC VIP showed a weak he apeu ic e ec on a - h i is, while educing OVA-speci ic, bu no CII-speci ic, DTH esponses. In con as , CII-pulsed DC VIP inhibi a h i is and DTH esponse owa d CII bu no owa d OVA (Fig. 3d). Simila Ag-dependence was obse ed in he EAE model (da a no shown). These esul s indica e ha DC VIP gene a ed ex i o could p e en o gan-speci ic au oimmune diso de s in ma ched subjec s, p obably by inducing Ag-speci ic T cells, which supp ess he ongoing au o- eac i e兾in lamma o y esponse. The pa icipa ion o T cells in he he apeu ic e ec o DC VIP co ela ed wi h he ac ha DC VIP induce in i o he gene a ion o IL-10兾TGF ␤ -p oducing egula o y CD4 T cells (Fig. 2). The e o e, we u he examined he ole o T in he he apeu ic e ec o DC VIP on bo h CIA and EAE. In i o blockade expe imen s showed ha ea men wi h an i-IL-10 o an i-TGF ␤ Abs signi ican ly dec ease disease amelio a ion, and ea men wi h bo h Abs ab oga es he bene icial e ec s exe ed by DC VIP (Fig. 3e), sugges ing he pa ial in ol emen o newly gen- e a ed T cells in such ac ion. DC VIP -Induced T Amelio a e Au oimmuni y. In ce ain ci cum- s ances, success ul supp ession o an au oimmune esponse migh equi e high numbe s o T , and he in i o adminis a ion o DC VIP migh no be su icien o a comple e and apid supp ession. The e o e, we decided o gene a e in i o Ag- speci ic DC VIP -induced T cells and o subsequen ly de e mine hei supp essi e capaci y in i o in bo h he CIA and EAE models. We gene a ed CII- o MOG-speci ic T VIP h ough s imula ions o CD4 T cells wi h syngeneic CII- o MOG-pulsed DC VIP .T con ol we e gene a ed in he same manne wi h DC con ol . T ea men wi h T VIP , bu no T con ol , o mice wi h es ablished CIA o EAE p e en ed disease p og ession in a dose-dependen manne (Fig. 4a). This e ec was mainly me- dia ed h ough TGF ␤ and IL-10, because in i o adminis a ion Fig. 4. DCVIP-induced T p e en au oimmuni y. (a) The apeu ic e ec on a h i is. CIA (H-2q) o EAE mice (H-2b) wi h es ablished disease we e ea ed wi h syngeneic CII- and MOG-speci ic T con ol ( E ) o wi h di e en doses o CII- o MOG-speci ic T VIP (106cells, ;5⫻105cells, ƒ;5⫻104cells, ■ ). Un ea ed mice ( F ) we e used as CIA and EAE con ols. Clinical sco e was de e mined (n⫽ 10). (b) Un ea ed CIA兾EAE mice o CIA兾EAE mice injec ed wi h T VIP and ea ed wi h con ol Ig, an i-IL10, an i-TGF ␤ , o an i-IL10 plus an i-TGF ␤ Abs. Clinical sco e was measu ed a he peak o he disease (n⫽10). 13566 兩 www.pnas.o g兾cgi兾doi兾10.1073兾pnas.0504484102 Cho ny e al. o an i-IL10 and兾o an i-TGF ␤ Abs ab oga ed he p o ec i e e ec (Fig. 4b). In bo h models, he p o ec i e e ec o T VIP was Ag-speci ic, because OVA-speci ic T VIP did no e icien ly amelio a e a h i is o pa alysis (da a no shown). These esul s indica e ha Ag-speci ic T 1-like cells gene a ed in i o wi h DC VIP can e icien ly modula e pa hogenic immune esponses in i o. VIP has been p e iously ound o amelio a e CIA and EAE, mainly by down- egula ing he wo componen s o bo h diseases, in lamma ion and Th1-media ed au oimmuni y ( e . 10 and E.G.-R., A.F.-M., A.C., D.P., D.G., and M.D., unpublished esul s). The in ol emen o T cells in he he apeu ic e ec o VIP was demons a ed by he ac ha CD4 T cells isola ed by VIP- ea ed CIA o EAE mice showed an inc eased egula o y兾 supp essi e ac i i y agains sel - eac i e Th1 cells. Pheno ypic analysis o hese T cells indica ed ha hey consis o a mix o Foxp3 ⫹ CD4 ⫹ CD25 ⫹ and IL-10 ⫹ T 1-like cells (E.G.-R., A.C., A.F.-M., D.G., and M.D., unpublished esul s). In addi ion, by using a ansgenic TCR mu ine model, we ound ha VIP induces he in i o gene a ion o Ag-speci ic ole ogenic IL-10- p oducing DCs wi h capaci y o gene a e兾ac i a e T 1-like cells (M.D., E.G.-R., and D.G., unpublished esul s). These indings alida e he da a ob ained in his s udy, demons a ing ha he pha macological use o VIP in he ea men o au oimmuni y is exe ed pa ially h ough he induc ion o ole ogenic DCs and T 1-like cells. I has been p oposed ha ole ance induc ion by DCs equi es ma u a ion signals di e en om mic obial o in lamma o y s imuli. In s eady-s a e condi ions, VIP could ep esen one o he endogenous ma u a ion signals d i ing he di e en ia ion o ole ogenic DCs wi h a egula o y pheno ype. VIP is sec e ed in he lymphoid mic oen i onmen , mainly by Th2 cells, a e Ag s imula ion, and VIP le els a e inc eased in immunopa hologic condi ions, such as au oimmuni y and in lamma ion (8, 9). The e o e, DC VIP may ep esen a popula ion o DCs ha ha e ma u ed o display a s able ole ogenic pheno ype. Unde s eady-s a e condi ions, DC VIP could be loaded wi h sel - and commonly encoun e ed Ags, and, a e mig a ion o he lym- phoid o gans, hey could induce T 1 di e en ia ion and ole - ance. In e es ingly, in subjec s wi h a ious au oimmune diso - de s, educed se um VIP le els and inc eased VIP-speci ic au oan ibodies ha e been epo ed (41). Nume ous s a egies based on immunosupp essi e agen s, such as i amin-D3, IL-10, TGF ␤ , glucoco icoids, and N-ace yl- L-cys eine, alone o in combina ions, ha e been used o induce ole ogenic DCs (7). Howe e , in he case o egula o y DCs induced wi h i amin D analogs, i looks as i hese egula o y DCs induce CD4 ⫹ CD25 ⫹ T cells a he han T 1-like cells (7). Ou da a demons a e ha VIP is e y e icien a he induc ion o egula o y DCs, in compa ison wi h cu en s a egies, and we p opose ha he addi ion o VIP o cock ails o immunomodu- la o y agen s will inc ease hei e ec i eness. In conclusion, he possibili y o gene a ing ole ogenic DC VIP opens he apeu ic pe spec i es o he ea men o au oim- mune兾in lamma o y diseases and in allogeneic ansplan a ion. In i o pulsing o ole ogenic DC VIP wi h sel -Ags, ollowed by in i o injec ion, leads o he di e en ia ion o Ag-speci ic T cells. The e o e, he inclusion o ole ogenic DC VIP in u u e he apeu ic egimens may minimize he dependence on nonspe- ci ic immunosupp essi e d ugs used cu en ly o au oimmune diso de s. This wo k was suppo ed by Spanish Minis y o Heal h G an PI04兾0674 ( o M.D.), Na ional Ins i u es o Heal h G an 2R01A047325 ( o D.G. and M.D.), a g an om he Ramon A eces Founda ion ( o M.D.), and by ellowships om Jun a de Andalucia ( o M.D. and E.G.-R.) and he Spanish Minis y o Educa ion and Science ( o M.D.). 1. Jonulei , H. & Schmi , E. (2003) J. Immunol. 171, 6323–6327. 2. S einman, R. M., Hawinge , D. & Nussenzweig, M. C. (2003) Annu. Re . Immunol. 21, 685–711. 3. Ru ella, S. & Lemoli, R. M. (2004) Immunol. Le . 94, 11–26. 4. Banche eau, J., B ie e, F., Caux, C., Da ous , J., Lebecque, S., Liu, Y. J., Pulend an, B. & Palucka, K. (2000) Annu. Re . Immunol. 18, 767–811. 5. Hawige , D., Inaba, K., Do se , Y., Guo, M., Mahnke, K., Ri e a, M., Ra e ch, J. V., S einman, R. M. & Nussenzweig, M. C. (2001) J. Exp. Med. 194, 769–779. 6. Ronca olo, M. C., Le ings, M. K. & T a e a i, C. (2001) J. Exp. Med. 193, F5–F9. 7. Hacks ein, H. & Thomson, A. W. (2004) Na . Re . Immunol. 4, 24–34. 8. Pozo, D. & Delgado, M. (2004) FASEB J. 18, 1325–1334. 9. Delgado, M., Pozo, D. & Ganea, D. (2004) Pha macol. Re . 56, 249–290. 10. Delgado, M., Abad, C., Ma inez, C., Lece a, J. & Goma iz, R. P. (2001) Na . Med. 7, 563–568. 11. Abad, C., Ma inez, C., Jua anz, M. G., A anz, A., Lece a, J., Delgado, M. & Goma iz, R. P. (2003) Gas oen e ology 124, 961–971. 12. Keino, H., Kezuka, T., Takeuchi, M., Yamakawa, N., Ha o i, T. & Usui, M. (2004) A ch. Oph halmol. 122, 1179–1184. 13. Inaba, K., Inaba, M., Romani, N., Aya, H., Deguchi, M., Ikeha a, S., Mu a- ma su, S. & S einman, R. M. (1992) J. Exp. Med. 176, 1693–1702. 14. Ga in, M. A., Cla ke, S. R., Neg ou, E., Gallegos, E. & Rudensky, A. (2002) Na . Immunol. 3, 33–39. 15. B ude , D., P obs -Keppe , M., Wes endo , A. M., Ge e s, R., Beisse , S., Lose , K., on Boehme , H., Bue , J. & Hansen, W. (2004) Eu . J. Immunol. 34, 623–630. 16. Moseman, E. A., Liang, X., Dawson, A. J., Panoskal sis-Mo a i, A., K ieg, A. M., Liu, Y. J., Blaza , B. R. & Chen, W. (2004) J. Immunol. 173, 4433–4442. 17. Kohm, A. P., Ca pen ie , P. A., Ange , H. A. & Mille , S. D. (2002) J. Immunol. 169, 4712–4716. 18. Sub amanian, S., Ma ejuk, A., Zamo a, A., Vande ba k, A. A. & O ne , H. (2003) J. Immunol. 170, 1548–1555. 19. Delgado, M., Redu a, A., Sha ma, V. & Ganea, D. (2004) J. Leukocy e Biol. 75, 1122–1130. 20. Wakkach, A., Fou nie , N., B un, V., B ei maye , J. P., Co ez, F. & G oux, H. (2003) Immuni y 18, 605–617. 21. Mo elli, A. E. & Thomson, A. W. (2003) Immunol. Re . 196, 125–146. 22. Piemon i, L., Mon i, P., Si oni, M., F a icelli, P., Leone, B. E., Dal Cin, E. All e a, P. & Di Ca lo, V. (2000) J. Immunol. 164, 4443–4451. 23. G i in, M. D., Lu z, W. H., Phan, V. A., Bachman, L. A., McKean, D. J. & Kuma , R. (2000) Biochem. Biophys. Res. Commun. 270, 701–708. 24. G ego i, S., Caso a i, M., Amuchas egui, S., Smi oldo, S., Da alli, A. M. & Ado ini, L. (2001) J. Immunol. 167, 1945–1953. 25. Jonulei , H., Schmi , E., Schule , G., Knop, J. & Enk, A. H. (2000) J. Exp. Med. 192, 1213–1222. 26. Sa o, K., Yamashi a, N. & Ma suyama, T. (2002) Cell. Immunol. 215, 186–194. 27. Sa o, K., Yamashi a, N., Yamashi a, N., Baba, M. & Ma suyama, T. (2003) Immuni y 18, 367–379. 28. Ma in, E., O⬘Sulli an, B., Low, P. & Thomas, R. (2003) Immuni y 18, 155–167. 29. Fehe a i, Z. & Sakaguchi, S. (2004) Cu . Opin. Immunol. 16, 203–208. 30. Akba i, O., DeK uy , R. H. & Ume su, D. T. (2001) Na . Immunol. 2, 725–731. 31. Tang, Q., Boden, E. K., Hen iksen, K. J., Bou -Jo dan, H., Bi, M. & Blues one, J. A. (2004) Eu . J. Immunol. 34, 2996–3005. 32. Mills, K. H. & McGui k, P. (2004) Semin. Immunol. 16, 107–117. 33. Le ings, M. K., G ego i, S., T esoldi, E., Cazzaniga, S., Bonini, C. & Ronca olo, M. G. (2005) Blood 105, 1162–1169. 34. Diehl, L., Den Boe , A. T., an de Voo , E. I., Melie , C. J., O inga, R. & Toes, R. E. (2000) J. Mol. Med. 78, 363–371. 35. Koppelman, B., Nee jes, J. J., de V ies, J. E. & de Waal Male y , T. (1997) Immuni y 7, 861–871. 36. Ding, L., Linsley, P. S., Huang, L. Y., Ge main, R. N. & She ach, E. M. (1993) J. Immunol. 151, 1224–1234. 37. G oux, H., O’Ga a, A., Bigle , M., Rouleau, M., An onenko, S., de V ies, J. E. & Ronca olo, M. G. (1997) Na u e 389, 737–742. 38. Ouaaz, F., Li, M. & Beg, A. A. (1999) J. Exp. Med. 189, 999–1004. 39. Yang, J., Be nie , S. M., Ichim, T. E., Li, M., Xia, X., Zhou, D., Huang, X., S ejan, G. H., Whi e, D. J., Zhong, R. & Min, W. P. (2003) J. Leukocy e Biol. 74, 438–447. 40. Goma iz, R. P., A anz, A., Abad, C., To oba, M., Ma inez, C., Rosignoli, F., Ga cia-Gomez, M., Lece a, J. & Jua anz, Y. (2005) J. Leukocy e Biol. 78, 491–502. 41. Bangale, Y., Ka le, S., Planque, S., Zhou, Y. X., Taguchi, H., Nishiyama, Y., Li, L., Kalaga, R. & Paul, S. (2003) FASEB J. 17, 628–635. Cho ny e al. PNAS 兩 Sep embe 20, 2005 兩 ol. 102 兩 no. 38 兩 13567 IMMUNOLOGY