Get 20M+ Full-Text Papers For Less Than $1.50/day. Start a 7-Day Trial for You or Your Team.

Learn More →

A rush to judgment on Th17

A rush to judgment on Th17 COMMENTARY Lawrence Steinman Some immunologists have characterized T helper (Th)17 T cells as the master In this issue, Kroenke et al. ( 8 ) show that adoptive transfer of either Th1 or mediators of tissue damage in a variety of pathological conditions. New data Th17 cells can induce EAE and clinical now demonstrate that Th1 and Th17 T cells are independently capable of paralysis in mice, but the pathology in- inducing disease in two established models of autoimmunity. Thus, the role of duced by Th17 cells diff ers from that Th17 cytokines as the central mediators of pathological tissue damage seems induced by Th1 cells. Thus Th17 cells to require clarifi cation. are unlikely to be the sole players in Immunologists have famously catego- reversed disease, and blocking IFN-  driving tissue damage in these classical models of autoimmunity. rized diff erent subsets of T cells and at- worsened disease ( 4 – 6 ). These fi ndings tributed diff erent biological functions to seem to contradict the idea that Th1 these cell populations. Nearly 40 years responses drive EAE and suggest that Non – IL-17 culprits in tissue damage In our rush to embrace Th17 cells as the ago, T cells were divided into helper, IFN-  may play diverse roles depending cytotoxic, and suppressor (now “ regu- on the stage of disease, or that certain purveyors of tissue damage, we should latory ” ) cell types. 20 years later, the Th EAE models may not accurately refl ect not forget that cytokines produced by cells were further divided into Th1 and the human disease. For years, the impli- Th1 cells and other cell types are criti- Th2 subsets. Th2 T cells were associated cations of these contradictory data went cal in promoting various forms of in- with allergy, whereas Th1 cells were as- largely unchallenged, as the complexities fl ammation. Administration of IFN-  , sociated with various organ-specifi c au- of the Th1/Th2 axis in this model of T for example, worsened disease in pa- tients with MS ( 13 ). And blocking tu- toimmune diseases. More recently, a new cell – mediated autoimmune disease were subset of Th cells was described and not fully grasped. mor necrosis factor (TNF), which can named Th17 based on production of the The identifi cation of the Th17 sub- be produced by various cell types, is a gold standard for treatment of diseases cytokine interleukin (IL)-17 ( 1 – 3 ). set has now broadened our understand- Although such tidy categorization ing of infl ammatory processes in human now thought to be driven largely by may be attractive in its simplicity, it has disease and has helped to explain some Th17 cells, including RA, Crohn ’ s dis- become apparent that the original Th1/ of the anomalies seen in the Th1/Th2 ease, and various forms of psoriasis ( 1 ). Th2 paradigm is much more compli- axis. However, we may now be facing Furthermore, type I IFNs, which are cated than originally appreciated. Hu- similar pitfalls by invoking Th17 cells to therapeutic in MS ( 14, 15 ), are patho- man diseases such as multiple sclerosis explain disease processes — in particular, genic in systemic lupus erythematosus ( 16 ). It is worth noting that the role of (MS) and rheumatoid arthritis (RA), for immune-mediated tissue damage — with- example, were commonly considered to out considering many as yet unexplained IL-17 in these major human diseases is be Th1 mediated, but we now realize inconsistencies in the experimental much less well understood than TNF, IFN-  , or type I IFNs. that such generalizations were inaccurate data. Immunologists are repeating many and oversimplifi ed. For over a decade, of the intellectual mistakes that were Ex vivo studies have also suggested various anomalies that contradicted the made for Th1/Th2 a decade earlier, as that cytokines of the Th1/Th2 axis Th1/Th2 paradigm went unexplained we confront the new concept of Th17. are critical determinants in mycobacte- ( 1 ). One example was the well-known Two papers in the Journal of Experimen- rial diseases ranging from tuberculoid fi nding that in one version of the “ Th1- tal Medicine , one by Luger et al. in a re- leprosy, which is primarily driven by driven ” disease experimental autoim- cent issue ( 7 ) and another by Kroenke IL-12 and Th1 cells, to lepromatous leprosy, which is mediated by Th2 cells mune encephalitis (EAE), a mouse model et al. ( 8 ) on page 1535 of this issue, as of MS, treating mice with the prototype well as other recent work ( 9 – 12 ), help ( 17 ). And Th2 responses drive many Th1 cytokine interferon (IFN)-  actually provide a more balanced view of the aspects of allergic responses ( 3 ). Al- though Th17 is a welcome addition to role of Th17 cells in autoimmune disease and immune-mediated tissue damage. our understanding of immune-medi- L.S. is at Department of Neurology and Neurological Using a model of experimental au- ated tissue damage, we still need the Sciences and the Interdepartmental Program in Immunology, Beckman Center for Molecular toimmune uveitis (EAU), Luger et al. Th1/Th2 axis and other infl ammatory Medicine, Stanford University, Stanford, CA 94305. ( 7 ) showed that either Th1 or Th17 mediators to explain many aspects of cells can drive tissue damage depending human autoimmune, allergic, and in- CORRESPONDENCE L.S.: [email protected] on the methods used to initiate disease. fectious diseases. © 2008 Steinman The Rockefeller University Press $30.00 J. Exp. Med. Vol. 205 No. 7 1517-1522 www.jem.org/cgi/doi/10.1084/jem.20072066 The Journal of Experimental Medicine Th17 cells as disease inducers comparable in wild-type and IL-17 – de- Th17-polarized myelin-reactive T cells In a recent issue of the JEM , Luger et al. fi cient mice ( 7 ). It should be noted that induces EAE ( 8 ). Ascending paralysis ( 7 ) demonstrated that IL-17 and IL-23, many investigators equate the loss of ensued in both models, but the cellular a cytokine that drives expansion of Th17 IL-17A in the IL-17 mice ( 7 ) with a infi ltrate induced by Th17 cells was rich cells, are important in the pathogenesis complete loss of Th17 cells, but these in neutrophils, whereas macrophages of EAU, a model that refl ects many as- cells also produce other cytokines, in- predominated the Th1-induced infi l- pects of both infectious and autoimmune cluding IL-17F, IL-21, and IL-22. Al- trate — a pathological picture similar to uveitis in man. In that study, adminis- though Luger et al. conclude that the MS, in which neutrophilic infi ltration is tration of antibodies against IL-17 in- development of EAU in IL-17 mice rare. The pathology seen with IL-17 – in- hibited the development of EAU after required an IFN-  – producing eff ector duced EAE was reminiscent of that seen immunization with the retinal antigen T cell, they did not formally rule out a in acute disseminated encephalomyelitis, intra-retinal binding protein (IRBP) in role for other Th17 cytokines ( 7 ). and other rare demyelinating conditions complete Freund ’ s adjuvant (CFA), and These results are diffi cult to interpret such as Marburg ’ s disease, and also re- also reversed established disease ( 7 ). Anti- given other studies showing a protective sembles certain aspects of neuromyelitis bodies against IL-23 also aborted the role for IFN-  in EAU. Neutralization optica — all conditions in which neutro- development of EAU in this model, but of IFN-  has been shown to exacerbate phils are common. Neuromyelitis optica only when given early in the disease EAU, just as it does EAE ( 4 – 6 ). Further- is also similar to Th17 – induced EAE in ( Fig. 1 A ). Blocking IL-23 was no lon- more, EAU develops effi ciently in mice that the pathology is most intense in the ger eff ective by day 7, a time when T depleted of IFN-  , and, perplexingly, spinal cord and optic nerves. Consis- cells capable of inducing uveitis were recombinant IFN-  has been shown to tent with this, Stromnes et al. ( 9 ) showed already present. The authors thus con- protect against disease ( 19, 20 ). How, that EAE is most intense in the brain cluded that IL-23 is important for the then, could an IFN-  – producing eff ec- when Th17/Th1 ratios are high, whereas priming of uveitis-inducing T cells, but tor cell be important for the pathogenesis infl ammation is focused on the spinal not for their eff ector function. Support- of EAU, if recombinant IFN-  is itself cord when Th17/Th1 ratios are more ing a role for IL-23 in EAU induction, protective? These results must be con- widely varied. mice lacking the IL-23 subunit p19 or sidered paradoxical and problematic at The overall picture that emerges p40, which is also shared by IL-12, were the least. Akin to the anomalies and par- suggests that EAE has a variety of his- protected against disease. Mice lacking adoxes seen with Th1 cells in EAE, these tological and anatomical manifestations the IL-12 – only subunit p35, by contrast, results are probably telling us something depending on the cytokines produced were more susceptible to EAU than were highly important, although we do not by the disease-inducing T cells ( 8, 9 ). wild-type mice ( Fig. 1 A ). Finally, trans- yet have enough knowledge to explain Thus, one cannot conclude at all that fer of in vitro – polarized, IRBP-specifi c it. But the data certainly confl ict with EAE is driven by Th17 cells. The data Th17 cells into recipient mice triggered the notion of IL-17 as a master mediator from Kroenke et al. ( 8 ) confi rm that EAE EAU, even in the complete absence of of tissue damage, at least in EAU. can be induced by either Th1 or Th17 IFN-  ( Fig. 1 B ). Luger et al. went on to induce EAU T cells. Finally, earlier work from Lafaille These fi ndings in EAU are similar to by immunizing mice with IRBP-pulsed et al. ( 21 ) indicated that even Th2 cells studies in EAE, in which chronic dis- dendritic cells, a model that is indepen- can induce EAE in RAG-defi cient mice. ease is reduced in IL-17 mice, although dent of CFA ( Fig. 1 C ). In this model, In that study, the cellular infi ltrate was these mice still develop acute paralysis the disease was characterized by an infl ux rich in eosinophils. Such pathology re- of their hind limbs ( 18 ). Adoptive trans- of T cells that produce large amounts plete with eosinophils, typical of allergic fer of myelin-reactive T cells from IL- of IFN-  but little IL-17. And IFN-  – disorders, would be worthy of the origi- 17 mice into wild-type mice also leads defi cient mice immunized with IRBP- nal term for EAE, “ experimental allergic to much milder EAE, as measured by pulsed dendritic cells were protected encephalomyelitis. ” incidence and severity of paralysis, com- against disease despite the robust pro- Other cytokines like osteopontin pared with transfer of myelin-reactive T duction of IL-17 ( 7 ). It is possible that drive the production of IFN-  ( 22, 23 ), cells from wild-type mice ( 18 ). the use of adjuvants in certain animal and osteopontin has also recently been models of autoimmune disease, such as shown to drive the production of IL-17 Th17 cells as innocent bystanders EAE, EAU, and collagen-induced ar- ( 24 ). Osteopontin is critical in trigger- Other data from Luger et al., however, thritis, might skew the importance of ing relapses of EAE and is elevated in suggested that EAU can also develop in IL-17. Indeed, the presence of adjuvants the plasma of patients suff ering MS re- the absence of Th17 cells. Transfer of a that, like CFA, contain killed myco- lapses ( 25 ). Is it thus possible that osteo- Th1 clone specifi c for IRBP, for exam- bacteria might provide a critical clue to pontin is the main driver of tissue damage ple, caused severe disease in the absence explain the discrepancies between dif- and MS relapses, and not IFN-  or IL-17? of IL-17 production ( Fig. 1 B ). More- ferent models. It can hardly be concluded at this point over, disease severity in response to im- In this issue, Kroenke et al. show that Th17 is the sole driver of tissue dam- munization with IRBP plus CFA was that adoptive transfer of either Th1- or age in MS or EAE. 1518 CAVEATS FOR TH17 | Steinman COMMENTARY Figure 1. Th17 or Th1 cells can induce EAU depending on the method of disease induction . (A) Direct vaccination model. Disease was induced by immunizing mice with IRBP plus CFA. There was a signifi cant reduction of EAU in p19 and p40 knockout mice, with a signifi cant increase in EAU in p35 knockout mice compared with wild type. Antibody blockade of IL-17 reduced disease when given either during disease induction or after disease onset, whereas antibody blockade of IL-23 reduced disease only when administered during disease induction. (B) T cell transfer model. Transfer of Th1-polarized, IRBP-specifi c CD4 T cells induced EAU (top). In this model, blocking IL-17 had no effect on disease and blocking IFN-  ameliorated disease. Transfer of Th17-polarized, IRBP-specifi c CD4 T cells induces EAU (bottom), and, in this model, disease was equivalent when cells were transferred into IFN-  – defi - cient mice. (C) Dendritic cell immunization protocol. IRBP-pulsed mature dendritic cells were injected into mice, followed by pertussis toxin at day 2. In this model, disease was decreased in IFN-  – defi cient mice, despite an increased production of IL-17 in the central nervous system as compared with wild-type mice. JEM VOL. 205, July 7, 2008 1519 Th17 cells as protectors TGF-  , were critical for the priming of that its signature function as the mediator Another issue that must be considered is Th17 responses ( 29, 30 ). In one of those of organ-specifi c tissue damage in auto- that IL-17 would not have evolved if it studies, TGF-  suppressed IL-17 pro- immunity and other forms of pathology only did harm. The protective side of duction ( 29 ). More recent data, however, should be refi ned. Exceptions always IL-17 was demonstrated in a recent study revealed that TGF-  , IL-1  and IL-6, teach us something important, as we have of mycobacteria tuberculosis infection, IL-21, or IL-23 (in serum-free condi- learned from our evolving understand- in which IL-17 was required to recruit tions) could induce IL-17 production ing of the Th1/Th2 paradigm ( 1 ). The protective IFN-  – producing CD4 T from naive human CD4 T cells from Th17 pathway has many divergent roles cells into the lung ( 26 ). In an asthma umbilical cord blood ( 31 ). TGF-  sup- in models of autoimmune, infectious, model, neutralization of IL-17 increased pressed ROR- t – induced IL-17 ex- and allergic disease. eosinophilic infi ltration during the eff ec- pression, but this suppression was relieved Indeed, IL-17 is not even a purely tor phase of disease, and administration in the presence of infl ammatory cyto- Th17 cytokine because it is also made by of recombinant IL-17 diminished airway kines. Another study showed that TGF-  macrophages ( 40 ), astrocytes ( 41 ), oligo- hyperreactivity and reduced the num- was essential, along with IL-23, IL-1  , dendroglia ( 41 ), uterine fi broids ( 42 ), bers of eosinophils and lymphocytes in and IL-6 for Th17 diff erentiation ( 32 ). and corneal epithelial cells on the surface bronchial lavage ( 27 ). In this model of Yet another showed that TGF-  en- of the eye ( 43 ). In fact, IL-17 was fi rst asthma, IL-17 thus appears to be a nega- hanced IL-17 in peripheral T cells ( 33 ). cloned from mouse NKT cells ( 44 ). Fi- tive regulator of established disease ( 27 ). At this point it appears that IL-1  and nally, what drives Th17 is not altogether How can we continue to call the Th17 IL-6 drive IL-17A production from certain. Emerging data indicate that mol- pathway the critical mediator of immune- central memory CD4 T cells, whereas ecules such as osteopontin drive both mediated tissue damage with such counter TGF-  and a constellation of other in- Th1- and Th17-mediated tissue damage examples from autoimmunity, infectious fl ammatory cytokines promote the dif- ( 22 – 25 ). We might thus be heading to- disease, and allergic disease? ferentiation of naive CD4 T cells into ward the conclusion that no single mol- Th17 cells ( 34 ). ecule or Th pathway dominates and that Th17 in mouse and in man The role of TGF-  is highly pleio- there is no hierarchical scheme at all. The ultimate signifi cance of the Th17 tropic ( 35, 36 ). Local induction of TGF- It is diffi cult to understand how so pathway in human disease remains un-  in mouse brain, for example, leads to many immunologists developed such clear, and we are only now clarifying the induction of EAE ( 36 ), and its pharma- exuberant enthusiasm for Th17 in the details of the Th17 pathway in humans. cologic suppression reverses paralytic fi rst place, placing it in a starring and The fi rst analyses of the Th17 pathway disease ( 37 ). TGF-  signaling in the paramount role for all immune-medi- came from studies of experimental dis- brain leads to increased production of ated tissue damage. One of the groups eases in mice, and the fi rst studies on the IL-6, which then enhances infl amma- who was instrumental in the discovery Th17 in man revealed signifi cant diff er- tion ( 37 ). TGF-  is elevated in chronic and elucidation of this pathway recently ences. As Natalie Angier wrote in her MS lesions, in which there is also intense reviewed the subject with a balanced elegant book The Canon , “ Whether siz- production of IL-17 from astrocytes and perspective ( 2 ). They enumerated the ing up new acquaintances or seizing on oligodendroglia ( 38 ). Thus, TGF-  multiplicity of “ inconvenient truths ” novel ideas, we remain forever at the might promote IL-17 production during that befuddled this role for Th17 when mercy of our fi rst impressions ” ( 28 ). In the chronic active phase of MS ( 9, 38 ). they asked, “ But are Th17 cells the only immunology (at least these days), our Studies in humans have also revealed eff ector cells capable of inducing organ- fi rst impressions often come from mice. the existence of T cell clones in the in- specifi c autoimmunity? Mice defi cient Again to quote Angier, we should ana- testines of patients with infl ammatory in T-bet and STAT-4, and thus lacking lyze the source of our misconceptions, bowel disease that produce both IL-17 Th1 cells ( 45, 46 ), have overwhelm- and then we “ have a chance of amend- and IFN-  , and are thus designated Th1/ ingly large numbers of Th17 cells ( 47, ing, remodeling, or blowtorching them Th17 cells. These dual producers can 48 ) and yet are resistant to EAE ( 47 – as needed, and replacing them with a also be found in EAE lesions in mice 49 ). Do those data suggest that Th1 closer approximation of science ’ s approxi- ( 39 ) and express both the Th17-inducing cells are pathogenic and Th17 cells are mate truths ” ( 28 ). transcription factor ROR-  t and the not? The truth is probably somewhere The parallels between mouse and Th1-inducing transcription factor T-bet. in the middle. For many reasons, we man fi rst diverged in defi ning a role for The biological importance of these hy- support the idea that both Th1 and transforming growth factor (TGF)-  in brid Th1/Th17 clones is as yet unclear. Th17 cells are capable of inducing au- the diff erentiation of Th17 cells. In mice, toimmunity ” ( 2 ). IL-6 and TGF-  are required for the Concluding remarks An emerging conclusion is that tis- production of IL-17 ( 2 ). Two initial hu- Immunologists ought to be restrained in sue damage and protection are nuanced man studies, however, showed that IL-6 attributing too much to the Th17 path- and are governed by multiple/redun- and IL-1  (produced by monocytes way at this stage. The plentiful excep- dant molecular interactions that involve or conventional dendritic cells), but not tions outlined in this commentary suggest many cytokines, including the type I 1520 CAVEATS FOR TH17 | Steinman COMMENTARY 12 . Kullberg , M.C. , D. Jankovic , C.G. Feng , induced relapse and progression of autoim- IFNs, and other molecular cascades in- S. Hue , P.L. Gorelikc , B.S. McKenzie , D.J. mune brain disease via enhanced survival of cluding, for example, the coagulation Cua , F. Powrie , A.W. Cheever , K.J. Maloy , activated T cells. Nat. Immunol. 8 : 77 – 86 . pathway ( 50 ). Thus, we should reserve and A. Sher . 2006 . IL-23 plays a key role in 26 . Khader , S.A. , G.K. Bell , J.E. Pearl , J.J. our judgment about Th17 and not jump helicobacter hepaticus – induced T cell – depen- Fountain , J. Rangel-Moreno , G. Cilley, F. dent colitis. J. Exp. Med. 203 : 2485 – 2494 . Shen, S.M. Eaton, S.L. Gaff en, S.L. Swain, to conclusions as we did with the Th1/ 13 . Panitch, H.S. , R.L. Hirsch , J. Schindler , and et al . 2007 . IL-23 and IL-17 in the estab- Th2 paradigm ( 1 ). K.P. Johnson . 1987. Treatment of multiple lishment of protective pulmonary CD4+ T sclerosis with gamma interferon: exacerbations cell responses after vaccination and during REFERENCES associated with activation of the immune sys- Mycobacterium tuberculosis challenge. Nat. 1 . Steinman , L. 2007 . A brief history of TH17, tem. Neurology . 37:1097 – 1102. Immunol. 8 : 369 – 377 . the fi rst major revision in the TH1/TH2 14 . Axtell , R.C. , and L. Steinman . 2008 . Type I 27 . Schnyder-Candrian , S. , D. Togbe , I. Couillin , hypothesis of T cell-mediated tissue damage. interferons cool the infl amed brain. Immunity . I. Mercier , F. Brombacher , V. Quesniaux , Nat. Med. 13 : 139 – 145 . 28 : 600 – 602 . F. Fossiez , B. Ryff el , and B. Schnyder . 2 . Bettelli , E. , M. Oukka , and V.K. Kuchroo . 15 . Arnason , B.G. 1999 . Immunological ther- 2006 . Interleukin-17 is a negative regulator 2007 . T(H)-17 cells in the circle of im- apy of multiple sclerosis. Annu. Rev. Med. of established allergic asthma. J. Exp. Med. munity and autoimmunity. Nat. Immunol. 50 : 291 – 302 . 203 : 2715 – 2725 . 8 : 345 – 350 . 16 . Pascual , V. , L. Farkas , and J. Banchereau . 28 . Angier , N. The Canon. 2007 . Houghton 3 . Coff man , R.L. 2006 . Origins of the T(H)1- 2006 . Systemic lupus erythematosus: all roads Miffl in, Boston. 293 pp. T(H)2 model: a personal perspective. Nat. lead to type I interferons. Curr. Opin. Immunol. 29 . Acosta-Rodriguez , E.V. , G. Napolitani , A. Immunol. 7 : 539 – 541 . 18 : 676 – 682 . Lanzavecchia , and F. Sallusto . Interleukins 4 . Billiau , A. , H. Heremans , F. Vandekerckhove , 17 . Kim , J. , K. Uyemura , M.K. Van Dyke , 1beta and 6 but not transforming growth fac- R. Dijkmans , H. Sobis , E. Meulepas , and H. A.J. Legaspi , T.H. Rea , K. Shuai , and R.L. tor-beta are essential for the diff erentiation of Carton . 1988 . Enhancement of experimental Modlin . 2001 . A role for IL-12 receptor ex- interleukin 17-producing human T helper allergic encephalomyelitis in mice by anti- pression and signal transduction in host de- cells. 2007 . Nat. Immunol. 8 : 942 - 949 . bodies against IFN-gamma. J. Immunol. 140 : fense in leprosy. J. Immunol. 167 : 779 – 786 . 30 . Wilson , N.J , K. Boniface , J.R. Chan , B.S. 1506 – 1510 . 18 . Komiyama , Y. , S. Nakae , A. Matsuki , H. McKenzie , W.M. Blumenschein , J.D. Mattson , 5 . Ferber , I.A. , S. Brocke , C. Taylor-Edwards , Nambu , S. Ishigame , K. Kakuta , and Y. B. Basham , K. Smith , T. Chen , F. Morel , et al. W. Ridgway , C. Dinisco , L. Steinman , D. Iwakura . 2006 . IL-17 plays an important role in 2007 . Development, cytokine profi le and func- Dalton , and C.G. Fathman . 1996 . Mice with a the development of experimental autoimmune tion of human interleukin 17-producing helper disrupted interferon-gamma gene are suscep- encephalomyelitis. J. Immunol. 1767 : 566 – 573 . T cells. Nat. Immunol. 8 : 950 – 957 . tible to the induction of experimental auto- 19 . Caspi , R.R. , C.C. Chan , B.G. Grubbs , P.B. 31 . Manel , N. , D. Unutmaz , and D.R. Littman . immune encephalomyelitis (EAE). J. Immunol. Silver , B. Wiggert , C.F. Parsa , S. Bahmanyar , 2008 . The diff erentiation of human Th17 cells 156 : 5 – 7 . A. Billiau , and H. Heremans . 1994 . Endog- requires transforming growth factor-beta and 6 . Voorthuis , J.A. , B.M. Uitdehaag , C.J. De enous systemic IFN-gamma has a protective induction of the nuclear receptor RORgammat. Groot , P.H. Goede , P.H. van der Meide , role against ocular autoimmunity in mice. Nat. Immunol. 9:641 – 649. and C.D. Dijkstra . 1990 . Suppression of J. Immunol. 152 : 890 – 899 . 32 . Volpe , E. , N. Servant , R. Zollinger , S.I. experimental allergic encephalomyelitis by 20 . Jones , L.S. , L.V. Rizzo , R.K. Agarwal , T.K. Bogiatzi , P. Hupe , E. Barillot , and V. Soumelis . intraventricular administration of interferon- Tarrant , C.C. Chan , B. Wiggert , and R.R. 2008 . A critical function for transforming gamma in Lewis rats. Clin. Exp. Immunol. 81 : Caspi . 1997 . IFN-gamma-defi cient mice de- growth factor-beta, interleukin 23 and pro- 183 – 188 . velop experimental autoimmune uveitis in infl ammatory cytokines in driving and mod- 7 . Luger , D. , P. Silver , J. Tang , D. Cua , Z. Chen , the context of a deviant eff ector response. ulating TH-17 responses. Nat. Immunol. 9 : Y. Iwakura , E.P. Bowman , N. Sgambellone , J. Immunol. 158 : 5997 – 6005 . 650 – 657 . C.C. Chan , and R.R. Caspi . 2008 . Either 21 . Lafaille , J.J. , F.V. Keere , A.L. Hsu , J.L. Baron , 33 . Kattah , M.G. , M.T. Wong , M.D. Yocum , a Th17 or a Th1 eff ector response can drive W. Haas , C.S. Raine , and S. Tonegawa . and P.J. Utz . 2008 . Cytokines secreted in autoimmunity: conditions of disease induction 1997 . Myelin basic protein – specifi c T helper response to Toll-like receptor ligand stimula- aff ect dominant eff ector category. J. Exp. Med. 2 (Th2) cells cause experimental autoim- tion modulate diff erentiation in human Th17 205 : 799 – 810 . mune encephalomyelitis in immunodefi cient cells. Arthritis Rheum. 58:1619 – 1629 . 8 . Kroenke , M.A. , T.J. Carlson , A.V. Adnjelkovic , hosts rather than protect them from the dis- 34 . Yang , L. , D.E. Anderson , C. Baecher-Allan , and B.M. Segal . 2008 . IL-12 – and IL-23 – mod- ease. J. Exp. Med. 186 : 307 – 312 . W.D. Hastings , E. Betelli , M. Oukka , V.K. ulated T cells induce distinct types of EAE 22 . Chabas , D. , S. Baranzini , D. Mitchell , C.C.A. Kuchroo , and D.A. Hafl er . 2008 . IL-21 and based on histology, CNS chemokine profi le, Bernard , S. Rittling , D. Denhardt , R. Sobel , TGF-beta are required for diff erentiation of and response to immunomodulatory interven- C. Lock , M. Karpuj , R. Pedotti , et al . 2001 . human Th17 cells. Nature . In press . tion. J. Exp. Med. 205:1535 – 1541. The infl uence of the pro-infl ammatory cyto- 35 . Veldhoen , M. , R.J. Hocking , R.A. Flavell , 9 . Stromnes , I.M. , L.M. Cerretti , D. Liggitt , kine, osteopontin, on autoimmune demye- and B. Stockinger . 2006 . Signals mediated by R.A. Harris , and J.M. Goverman . 2008 . linating disease. Science . 294 : 1731 – 1735 . transforming growth factor beta initiate auto- Diff erential regulation of central nervous sys- 23 . Jansson , M. , V. Panoustsakopoulou , J. Baker , immune encephalomyelitis, but chronic in- tem autoimmunity by T(H)1 and T(H)17 L. Klein , and H.C. Cantor . 2002 . Attenuated fl ammation is needed to sustain disease. Nat. cells. Nat. Med. 14 : 337 – 342 . experimental autoimmune encephalomyeli- Immunol. 7 : 1151 – 1156 . 10 . Cox , C.A. , G. Shi , H. Yin , B.P. Vistica , E.F. tis in Eta-1/osteopontin-defi cient mice. J. 36 . Ming , O.L. , Y.Y. Wan , and R.A.T. Flavell . Wawrousek , C.C. Chan , and I. Gery . 2008 . Immunol. 168 : 2096 – 2099 . 2007 . Cell-produced transforming growth fac- Both Th1 and th17 are immunopathogenic 24 . Shinohara , M.L. , H.J. Kim , J.-H. Kim , V.A. tor-b1 controls T cell tolerance and regulates but diff er in other key biological activities. Garcia , and H. Cantor . 2008 . Alternative Th1- and Th17-cell diff erentiation. Immunity . J. Immunol. 180 : 7414 – 7422 . translation of osteopontin generates intra- 26 : 579 – 591 . 11 . Hue , S. , P. Ahern , S. Buonocore , M.C. cellular and secreted isoforms that mediate 37 . Luo , J. , P.P. Ho , M.S. Buckwalter , T. Hsu , Kullberg , D.J. Cua , B.S. McKenzie , F. distinct biological activities in dendritic cells . L.Y. Lee , H. Zhang , D. Kim , S. Kim , S.S. Powrie , and K. Maloy . 2006 . Interleukin-23 Proc. Natl. Acad. Sci. USA . 105 : 7235 – 7239 . Gambhir , L. Steinman , and T. Wyss-Cory . drives innate and T cell – mediated intestinal 25 . Hur , E. , S. Youssef , M. Haws , S. Zhang , R. 2007 . Glia-dependent TGF-beta signaling, infl ammation. J. Exp. Med. 203 : 2473 – 2483 . Sobel , and L. Steinman . 2007 . Osteopontin independent of the Th17 pathway, is critical JEM VOL. 205, July 7, 2008 1521 for initiation of autoimmune encephalomy- glial cells is associated with active disease novel transcription factor, T-bet, directs Th1 elitis. J. Clin. Invest. 117 : 3306 – 3315 . in multiple sclerosis. Am. J. Pathol. 172 : lineage commitment. Cell . 100 : 655 – 669 . 38 . Lock , C. , G. Hermans , R. Pedotti , A. 146 – 155 . 47 . Chitnis , T ., N. Najafi an, C. Benou, A.D. Brendolan , E. Schadt , H. Garren , A. Langer- 42 . Hever , A. , R.B. Roth , P.A. Hevezi , J. Lee , Salama, M.J. Grusby, M.H. Sayegh, and S.J. Gould , S. Strober , B. Cannella , J. Allard , et al . D. Willhite , E.C. White , E.M. Marin , R. Khoury. 2001 . Eff ect of targeted disruption 2002 . Gene microarray analysis of multiple Herrera , H.M. Acosta , A.J. Acosta , and A. of STAT4 and STAT6 on the induction of sclerosis lesions yields new targets validated Zlotnik . 2006 . Molecular characterization experimental autoimmune encephalomyeli- in autoimmune encephalomyelitis. Nat. Med. of human adenomyosis. Mol. Hum. Reprod. tis. J. Clin. Invest. 108 : 739 – 747 . 8 : 500 – 508 . 12 : 737 – 748 . 48 . Bettelli , E., B. Sullivan, S.J. Szabo, R.A. Sobel, 39 . Ivanov , I. , B.S. McKenzie , Z. Liang , C.E. 43 . Molesworth-Kenyon , S.J. , R. Yin , J.E. L.H. Glimcher, and V.K. Kuchroo. 2004 . Tadokoro , A. Lepelley , J.J. Lafaille , D.J. Cua , Oakes , and R.N. Lausch . 2008 . IL-17 recep- Loss of T-bet, but not STAT1, prevents the and D.R. Littman . 2006 . The orphan nuclear tor signaling infl uences viral induced corneal development of experimental autoimmune receptor RORgammaT directs the diff eren- infl ammation. J. Leukoc. Biol. 83 : 401 – 408 . encephalomyelitis. J. Exp. Med. 200 : 79 – 87 . tiation program of proinfl ammatory IL-17+ 44 . Kennedy, J. , D.L. Rossi , S.M. Zurawski , F. 49 . Lovett-Racke , A.E., A.E. Rocchini, J. Choy, T helper cells. Cell . 126 : 1121 – 1133 . Vega Jr , R.A. Kastelein , J.L. Wagner , C.H. S.C. Northrop, R.Z. Hussain, R.B. Ratts, 40 . Gu , Y. , J. Yang , X. Ouyang , W. Liu , H. Hannum , and A. Zlotnik . 1996 . Mouse IL- D. Sikder, and M.K. Racke. 2004 . Silencing Li , J. Yang , J. Bromberg , S.H. Chen , L. 17: a cytokine preferentially expressed by T-bet defi nes a critical role in the diff erentia- Mayer , J.C. Unkeless , and H. Xiong . 2008 . alpha beta TCR + CD4-CD8-T cells. J. tion of autoreactive T lymphocytes. Immunity. Interleukin 10 suppresses Th17 cytokines se- Interferon Cytokine Res. 16:611 – 617. 21 : 719 – 731 . creted by macrophages and T cells. Eur. J. 45 . Kaplan , M.H. , Y.L. Sun , T. Hoey , and M.J. 50 . Han, M.H. , S. Hwang , D.B. Roy , D.H. Immunol. In print. Grusby . 1996 . Impaired IL-12 responses and Lundgren , J.V. Price , S. Ousman , G. Fernald , 41 . Tzartos , J.S. , M.A. Friese , M.J. Craner , J. enhanced development of T 2 cells in Stat4- B. Gerlitz , W.H. Robinson , S.E. Baranzini , Palace , J. Newcombe , M.M. Esiri , and L. defi cient mice. Nature . 382 : 174 – 177 . et al. 2008 . Proteomic analysis of active mul- Fugger . 2008 . Interleukin-17 production in 46 . Szabo , S. J., S.T. Kim, G.L. Costa, X. Zhang, tiple sclerosis lesions reveals therapeutic tar- central nervous system-infi ltrating T cells and C.G. Fathman, and L.H. Glimcher. 2000 . A gets. Nature. 451 : 1076 – 1081 . 1522 CAVEATS FOR TH17 | Steinman http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png The Journal of Experimental Medicine Pubmed Central

A rush to judgment on Th17

Steinman, Lawrence
The Journal of Experimental Medicine , Volume 205 (7) – Jul 7, 2008
Download PDF
6 pages

Loading next page...
 
/lp/pubmed-central/a-rush-to-judgment-on-th17-5nUam6BjD8

References (55)

Publisher
Pubmed Central
Copyright
© 2008 Steinman
ISSN
0022-1007
eISSN
1540-9538
DOI
10.1084/jem.20072066
Publisher site
See Article on Publisher Site

Abstract

COMMENTARY Lawrence Steinman Some immunologists have characterized T helper (Th)17 T cells as the master In this issue, Kroenke et al. ( 8 ) show that adoptive transfer of either Th1 or mediators of tissue damage in a variety of pathological conditions. New data Th17 cells can induce EAE and clinical now demonstrate that Th1 and Th17 T cells are independently capable of paralysis in mice, but the pathology in- inducing disease in two established models of autoimmunity. Thus, the role of duced by Th17 cells diff ers from that Th17 cytokines as the central mediators of pathological tissue damage seems induced by Th1 cells. Thus Th17 cells to require clarifi cation. are unlikely to be the sole players in Immunologists have famously catego- reversed disease, and blocking IFN-  driving tissue damage in these classical models of autoimmunity. rized diff erent subsets of T cells and at- worsened disease ( 4 – 6 ). These fi ndings tributed diff erent biological functions to seem to contradict the idea that Th1 these cell populations. Nearly 40 years responses drive EAE and suggest that Non – IL-17 culprits in tissue damage In our rush to embrace Th17 cells as the ago, T cells were divided into helper, IFN-  may play diverse roles depending cytotoxic, and suppressor (now “ regu- on the stage of disease, or that certain purveyors of tissue damage, we should latory ” ) cell types. 20 years later, the Th EAE models may not accurately refl ect not forget that cytokines produced by cells were further divided into Th1 and the human disease. For years, the impli- Th1 cells and other cell types are criti- Th2 subsets. Th2 T cells were associated cations of these contradictory data went cal in promoting various forms of in- with allergy, whereas Th1 cells were as- largely unchallenged, as the complexities fl ammation. Administration of IFN-  , sociated with various organ-specifi c au- of the Th1/Th2 axis in this model of T for example, worsened disease in pa- tients with MS ( 13 ). And blocking tu- toimmune diseases. More recently, a new cell – mediated autoimmune disease were subset of Th cells was described and not fully grasped. mor necrosis factor (TNF), which can named Th17 based on production of the The identifi cation of the Th17 sub- be produced by various cell types, is a gold standard for treatment of diseases cytokine interleukin (IL)-17 ( 1 – 3 ). set has now broadened our understand- Although such tidy categorization ing of infl ammatory processes in human now thought to be driven largely by may be attractive in its simplicity, it has disease and has helped to explain some Th17 cells, including RA, Crohn ’ s dis- become apparent that the original Th1/ of the anomalies seen in the Th1/Th2 ease, and various forms of psoriasis ( 1 ). Th2 paradigm is much more compli- axis. However, we may now be facing Furthermore, type I IFNs, which are cated than originally appreciated. Hu- similar pitfalls by invoking Th17 cells to therapeutic in MS ( 14, 15 ), are patho- man diseases such as multiple sclerosis explain disease processes — in particular, genic in systemic lupus erythematosus ( 16 ). It is worth noting that the role of (MS) and rheumatoid arthritis (RA), for immune-mediated tissue damage — with- example, were commonly considered to out considering many as yet unexplained IL-17 in these major human diseases is be Th1 mediated, but we now realize inconsistencies in the experimental much less well understood than TNF, IFN-  , or type I IFNs. that such generalizations were inaccurate data. Immunologists are repeating many and oversimplifi ed. For over a decade, of the intellectual mistakes that were Ex vivo studies have also suggested various anomalies that contradicted the made for Th1/Th2 a decade earlier, as that cytokines of the Th1/Th2 axis Th1/Th2 paradigm went unexplained we confront the new concept of Th17. are critical determinants in mycobacte- ( 1 ). One example was the well-known Two papers in the Journal of Experimen- rial diseases ranging from tuberculoid fi nding that in one version of the “ Th1- tal Medicine , one by Luger et al. in a re- leprosy, which is primarily driven by driven ” disease experimental autoim- cent issue ( 7 ) and another by Kroenke IL-12 and Th1 cells, to lepromatous leprosy, which is mediated by Th2 cells mune encephalitis (EAE), a mouse model et al. ( 8 ) on page 1535 of this issue, as of MS, treating mice with the prototype well as other recent work ( 9 – 12 ), help ( 17 ). And Th2 responses drive many Th1 cytokine interferon (IFN)-  actually provide a more balanced view of the aspects of allergic responses ( 3 ). Al- though Th17 is a welcome addition to role of Th17 cells in autoimmune disease and immune-mediated tissue damage. our understanding of immune-medi- L.S. is at Department of Neurology and Neurological Using a model of experimental au- ated tissue damage, we still need the Sciences and the Interdepartmental Program in Immunology, Beckman Center for Molecular toimmune uveitis (EAU), Luger et al. Th1/Th2 axis and other infl ammatory Medicine, Stanford University, Stanford, CA 94305. ( 7 ) showed that either Th1 or Th17 mediators to explain many aspects of cells can drive tissue damage depending human autoimmune, allergic, and in- CORRESPONDENCE L.S.: [email protected] on the methods used to initiate disease. fectious diseases. © 2008 Steinman The Rockefeller University Press $30.00 J. Exp. Med. Vol. 205 No. 7 1517-1522 www.jem.org/cgi/doi/10.1084/jem.20072066 The Journal of Experimental Medicine Th17 cells as disease inducers comparable in wild-type and IL-17 – de- Th17-polarized myelin-reactive T cells In a recent issue of the JEM , Luger et al. fi cient mice ( 7 ). It should be noted that induces EAE ( 8 ). Ascending paralysis ( 7 ) demonstrated that IL-17 and IL-23, many investigators equate the loss of ensued in both models, but the cellular a cytokine that drives expansion of Th17 IL-17A in the IL-17 mice ( 7 ) with a infi ltrate induced by Th17 cells was rich cells, are important in the pathogenesis complete loss of Th17 cells, but these in neutrophils, whereas macrophages of EAU, a model that refl ects many as- cells also produce other cytokines, in- predominated the Th1-induced infi l- pects of both infectious and autoimmune cluding IL-17F, IL-21, and IL-22. Al- trate — a pathological picture similar to uveitis in man. In that study, adminis- though Luger et al. conclude that the MS, in which neutrophilic infi ltration is tration of antibodies against IL-17 in- development of EAU in IL-17 mice rare. The pathology seen with IL-17 – in- hibited the development of EAU after required an IFN-  – producing eff ector duced EAE was reminiscent of that seen immunization with the retinal antigen T cell, they did not formally rule out a in acute disseminated encephalomyelitis, intra-retinal binding protein (IRBP) in role for other Th17 cytokines ( 7 ). and other rare demyelinating conditions complete Freund ’ s adjuvant (CFA), and These results are diffi cult to interpret such as Marburg ’ s disease, and also re- also reversed established disease ( 7 ). Anti- given other studies showing a protective sembles certain aspects of neuromyelitis bodies against IL-23 also aborted the role for IFN-  in EAU. Neutralization optica — all conditions in which neutro- development of EAU in this model, but of IFN-  has been shown to exacerbate phils are common. Neuromyelitis optica only when given early in the disease EAU, just as it does EAE ( 4 – 6 ). Further- is also similar to Th17 – induced EAE in ( Fig. 1 A ). Blocking IL-23 was no lon- more, EAU develops effi ciently in mice that the pathology is most intense in the ger eff ective by day 7, a time when T depleted of IFN-  , and, perplexingly, spinal cord and optic nerves. Consis- cells capable of inducing uveitis were recombinant IFN-  has been shown to tent with this, Stromnes et al. ( 9 ) showed already present. The authors thus con- protect against disease ( 19, 20 ). How, that EAE is most intense in the brain cluded that IL-23 is important for the then, could an IFN-  – producing eff ec- when Th17/Th1 ratios are high, whereas priming of uveitis-inducing T cells, but tor cell be important for the pathogenesis infl ammation is focused on the spinal not for their eff ector function. Support- of EAU, if recombinant IFN-  is itself cord when Th17/Th1 ratios are more ing a role for IL-23 in EAU induction, protective? These results must be con- widely varied. mice lacking the IL-23 subunit p19 or sidered paradoxical and problematic at The overall picture that emerges p40, which is also shared by IL-12, were the least. Akin to the anomalies and par- suggests that EAE has a variety of his- protected against disease. Mice lacking adoxes seen with Th1 cells in EAE, these tological and anatomical manifestations the IL-12 – only subunit p35, by contrast, results are probably telling us something depending on the cytokines produced were more susceptible to EAU than were highly important, although we do not by the disease-inducing T cells ( 8, 9 ). wild-type mice ( Fig. 1 A ). Finally, trans- yet have enough knowledge to explain Thus, one cannot conclude at all that fer of in vitro – polarized, IRBP-specifi c it. But the data certainly confl ict with EAE is driven by Th17 cells. The data Th17 cells into recipient mice triggered the notion of IL-17 as a master mediator from Kroenke et al. ( 8 ) confi rm that EAE EAU, even in the complete absence of of tissue damage, at least in EAU. can be induced by either Th1 or Th17 IFN-  ( Fig. 1 B ). Luger et al. went on to induce EAU T cells. Finally, earlier work from Lafaille These fi ndings in EAU are similar to by immunizing mice with IRBP-pulsed et al. ( 21 ) indicated that even Th2 cells studies in EAE, in which chronic dis- dendritic cells, a model that is indepen- can induce EAE in RAG-defi cient mice. ease is reduced in IL-17 mice, although dent of CFA ( Fig. 1 C ). In this model, In that study, the cellular infi ltrate was these mice still develop acute paralysis the disease was characterized by an infl ux rich in eosinophils. Such pathology re- of their hind limbs ( 18 ). Adoptive trans- of T cells that produce large amounts plete with eosinophils, typical of allergic fer of myelin-reactive T cells from IL- of IFN-  but little IL-17. And IFN-  – disorders, would be worthy of the origi- 17 mice into wild-type mice also leads defi cient mice immunized with IRBP- nal term for EAE, “ experimental allergic to much milder EAE, as measured by pulsed dendritic cells were protected encephalomyelitis. ” incidence and severity of paralysis, com- against disease despite the robust pro- Other cytokines like osteopontin pared with transfer of myelin-reactive T duction of IL-17 ( 7 ). It is possible that drive the production of IFN-  ( 22, 23 ), cells from wild-type mice ( 18 ). the use of adjuvants in certain animal and osteopontin has also recently been models of autoimmune disease, such as shown to drive the production of IL-17 Th17 cells as innocent bystanders EAE, EAU, and collagen-induced ar- ( 24 ). Osteopontin is critical in trigger- Other data from Luger et al., however, thritis, might skew the importance of ing relapses of EAE and is elevated in suggested that EAU can also develop in IL-17. Indeed, the presence of adjuvants the plasma of patients suff ering MS re- the absence of Th17 cells. Transfer of a that, like CFA, contain killed myco- lapses ( 25 ). Is it thus possible that osteo- Th1 clone specifi c for IRBP, for exam- bacteria might provide a critical clue to pontin is the main driver of tissue damage ple, caused severe disease in the absence explain the discrepancies between dif- and MS relapses, and not IFN-  or IL-17? of IL-17 production ( Fig. 1 B ). More- ferent models. It can hardly be concluded at this point over, disease severity in response to im- In this issue, Kroenke et al. show that Th17 is the sole driver of tissue dam- munization with IRBP plus CFA was that adoptive transfer of either Th1- or age in MS or EAE. 1518 CAVEATS FOR TH17 | Steinman COMMENTARY Figure 1. Th17 or Th1 cells can induce EAU depending on the method of disease induction . (A) Direct vaccination model. Disease was induced by immunizing mice with IRBP plus CFA. There was a signifi cant reduction of EAU in p19 and p40 knockout mice, with a signifi cant increase in EAU in p35 knockout mice compared with wild type. Antibody blockade of IL-17 reduced disease when given either during disease induction or after disease onset, whereas antibody blockade of IL-23 reduced disease only when administered during disease induction. (B) T cell transfer model. Transfer of Th1-polarized, IRBP-specifi c CD4 T cells induced EAU (top). In this model, blocking IL-17 had no effect on disease and blocking IFN-  ameliorated disease. Transfer of Th17-polarized, IRBP-specifi c CD4 T cells induces EAU (bottom), and, in this model, disease was equivalent when cells were transferred into IFN-  – defi - cient mice. (C) Dendritic cell immunization protocol. IRBP-pulsed mature dendritic cells were injected into mice, followed by pertussis toxin at day 2. In this model, disease was decreased in IFN-  – defi cient mice, despite an increased production of IL-17 in the central nervous system as compared with wild-type mice. JEM VOL. 205, July 7, 2008 1519 Th17 cells as protectors TGF-  , were critical for the priming of that its signature function as the mediator Another issue that must be considered is Th17 responses ( 29, 30 ). In one of those of organ-specifi c tissue damage in auto- that IL-17 would not have evolved if it studies, TGF-  suppressed IL-17 pro- immunity and other forms of pathology only did harm. The protective side of duction ( 29 ). More recent data, however, should be refi ned. Exceptions always IL-17 was demonstrated in a recent study revealed that TGF-  , IL-1  and IL-6, teach us something important, as we have of mycobacteria tuberculosis infection, IL-21, or IL-23 (in serum-free condi- learned from our evolving understand- in which IL-17 was required to recruit tions) could induce IL-17 production ing of the Th1/Th2 paradigm ( 1 ). The protective IFN-  – producing CD4 T from naive human CD4 T cells from Th17 pathway has many divergent roles cells into the lung ( 26 ). In an asthma umbilical cord blood ( 31 ). TGF-  sup- in models of autoimmune, infectious, model, neutralization of IL-17 increased pressed ROR- t – induced IL-17 ex- and allergic disease. eosinophilic infi ltration during the eff ec- pression, but this suppression was relieved Indeed, IL-17 is not even a purely tor phase of disease, and administration in the presence of infl ammatory cyto- Th17 cytokine because it is also made by of recombinant IL-17 diminished airway kines. Another study showed that TGF-  macrophages ( 40 ), astrocytes ( 41 ), oligo- hyperreactivity and reduced the num- was essential, along with IL-23, IL-1  , dendroglia ( 41 ), uterine fi broids ( 42 ), bers of eosinophils and lymphocytes in and IL-6 for Th17 diff erentiation ( 32 ). and corneal epithelial cells on the surface bronchial lavage ( 27 ). In this model of Yet another showed that TGF-  en- of the eye ( 43 ). In fact, IL-17 was fi rst asthma, IL-17 thus appears to be a nega- hanced IL-17 in peripheral T cells ( 33 ). cloned from mouse NKT cells ( 44 ). Fi- tive regulator of established disease ( 27 ). At this point it appears that IL-1  and nally, what drives Th17 is not altogether How can we continue to call the Th17 IL-6 drive IL-17A production from certain. Emerging data indicate that mol- pathway the critical mediator of immune- central memory CD4 T cells, whereas ecules such as osteopontin drive both mediated tissue damage with such counter TGF-  and a constellation of other in- Th1- and Th17-mediated tissue damage examples from autoimmunity, infectious fl ammatory cytokines promote the dif- ( 22 – 25 ). We might thus be heading to- disease, and allergic disease? ferentiation of naive CD4 T cells into ward the conclusion that no single mol- Th17 cells ( 34 ). ecule or Th pathway dominates and that Th17 in mouse and in man The role of TGF-  is highly pleio- there is no hierarchical scheme at all. The ultimate signifi cance of the Th17 tropic ( 35, 36 ). Local induction of TGF- It is diffi cult to understand how so pathway in human disease remains un-  in mouse brain, for example, leads to many immunologists developed such clear, and we are only now clarifying the induction of EAE ( 36 ), and its pharma- exuberant enthusiasm for Th17 in the details of the Th17 pathway in humans. cologic suppression reverses paralytic fi rst place, placing it in a starring and The fi rst analyses of the Th17 pathway disease ( 37 ). TGF-  signaling in the paramount role for all immune-medi- came from studies of experimental dis- brain leads to increased production of ated tissue damage. One of the groups eases in mice, and the fi rst studies on the IL-6, which then enhances infl amma- who was instrumental in the discovery Th17 in man revealed signifi cant diff er- tion ( 37 ). TGF-  is elevated in chronic and elucidation of this pathway recently ences. As Natalie Angier wrote in her MS lesions, in which there is also intense reviewed the subject with a balanced elegant book The Canon , “ Whether siz- production of IL-17 from astrocytes and perspective ( 2 ). They enumerated the ing up new acquaintances or seizing on oligodendroglia ( 38 ). Thus, TGF-  multiplicity of “ inconvenient truths ” novel ideas, we remain forever at the might promote IL-17 production during that befuddled this role for Th17 when mercy of our fi rst impressions ” ( 28 ). In the chronic active phase of MS ( 9, 38 ). they asked, “ But are Th17 cells the only immunology (at least these days), our Studies in humans have also revealed eff ector cells capable of inducing organ- fi rst impressions often come from mice. the existence of T cell clones in the in- specifi c autoimmunity? Mice defi cient Again to quote Angier, we should ana- testines of patients with infl ammatory in T-bet and STAT-4, and thus lacking lyze the source of our misconceptions, bowel disease that produce both IL-17 Th1 cells ( 45, 46 ), have overwhelm- and then we “ have a chance of amend- and IFN-  , and are thus designated Th1/ ingly large numbers of Th17 cells ( 47, ing, remodeling, or blowtorching them Th17 cells. These dual producers can 48 ) and yet are resistant to EAE ( 47 – as needed, and replacing them with a also be found in EAE lesions in mice 49 ). Do those data suggest that Th1 closer approximation of science ’ s approxi- ( 39 ) and express both the Th17-inducing cells are pathogenic and Th17 cells are mate truths ” ( 28 ). transcription factor ROR-  t and the not? The truth is probably somewhere The parallels between mouse and Th1-inducing transcription factor T-bet. in the middle. For many reasons, we man fi rst diverged in defi ning a role for The biological importance of these hy- support the idea that both Th1 and transforming growth factor (TGF)-  in brid Th1/Th17 clones is as yet unclear. Th17 cells are capable of inducing au- the diff erentiation of Th17 cells. In mice, toimmunity ” ( 2 ). IL-6 and TGF-  are required for the Concluding remarks An emerging conclusion is that tis- production of IL-17 ( 2 ). Two initial hu- Immunologists ought to be restrained in sue damage and protection are nuanced man studies, however, showed that IL-6 attributing too much to the Th17 path- and are governed by multiple/redun- and IL-1  (produced by monocytes way at this stage. The plentiful excep- dant molecular interactions that involve or conventional dendritic cells), but not tions outlined in this commentary suggest many cytokines, including the type I 1520 CAVEATS FOR TH17 | Steinman COMMENTARY 12 . Kullberg , M.C. , D. Jankovic , C.G. Feng , induced relapse and progression of autoim- IFNs, and other molecular cascades in- S. Hue , P.L. Gorelikc , B.S. McKenzie , D.J. mune brain disease via enhanced survival of cluding, for example, the coagulation Cua , F. Powrie , A.W. Cheever , K.J. Maloy , activated T cells. Nat. Immunol. 8 : 77 – 86 . pathway ( 50 ). Thus, we should reserve and A. Sher . 2006 . IL-23 plays a key role in 26 . Khader , S.A. , G.K. Bell , J.E. Pearl , J.J. our judgment about Th17 and not jump helicobacter hepaticus – induced T cell – depen- Fountain , J. Rangel-Moreno , G. Cilley, F. dent colitis. J. Exp. Med. 203 : 2485 – 2494 . Shen, S.M. Eaton, S.L. Gaff en, S.L. Swain, to conclusions as we did with the Th1/ 13 . Panitch, H.S. , R.L. Hirsch , J. Schindler , and et al . 2007 . IL-23 and IL-17 in the estab- Th2 paradigm ( 1 ). K.P. Johnson . 1987. Treatment of multiple lishment of protective pulmonary CD4+ T sclerosis with gamma interferon: exacerbations cell responses after vaccination and during REFERENCES associated with activation of the immune sys- Mycobacterium tuberculosis challenge. Nat. 1 . Steinman , L. 2007 . A brief history of TH17, tem. Neurology . 37:1097 – 1102. Immunol. 8 : 369 – 377 . the fi rst major revision in the TH1/TH2 14 . Axtell , R.C. , and L. Steinman . 2008 . Type I 27 . Schnyder-Candrian , S. , D. Togbe , I. Couillin , hypothesis of T cell-mediated tissue damage. interferons cool the infl amed brain. Immunity . I. Mercier , F. Brombacher , V. Quesniaux , Nat. Med. 13 : 139 – 145 . 28 : 600 – 602 . F. Fossiez , B. Ryff el , and B. Schnyder . 2 . Bettelli , E. , M. Oukka , and V.K. Kuchroo . 15 . Arnason , B.G. 1999 . Immunological ther- 2006 . Interleukin-17 is a negative regulator 2007 . T(H)-17 cells in the circle of im- apy of multiple sclerosis. Annu. Rev. Med. of established allergic asthma. J. Exp. Med. munity and autoimmunity. Nat. Immunol. 50 : 291 – 302 . 203 : 2715 – 2725 . 8 : 345 – 350 . 16 . Pascual , V. , L. Farkas , and J. Banchereau . 28 . Angier , N. The Canon. 2007 . Houghton 3 . Coff man , R.L. 2006 . Origins of the T(H)1- 2006 . Systemic lupus erythematosus: all roads Miffl in, Boston. 293 pp. T(H)2 model: a personal perspective. Nat. lead to type I interferons. Curr. Opin. Immunol. 29 . Acosta-Rodriguez , E.V. , G. Napolitani , A. Immunol. 7 : 539 – 541 . 18 : 676 – 682 . Lanzavecchia , and F. Sallusto . Interleukins 4 . Billiau , A. , H. Heremans , F. Vandekerckhove , 17 . Kim , J. , K. Uyemura , M.K. Van Dyke , 1beta and 6 but not transforming growth fac- R. Dijkmans , H. Sobis , E. Meulepas , and H. A.J. Legaspi , T.H. Rea , K. Shuai , and R.L. tor-beta are essential for the diff erentiation of Carton . 1988 . Enhancement of experimental Modlin . 2001 . A role for IL-12 receptor ex- interleukin 17-producing human T helper allergic encephalomyelitis in mice by anti- pression and signal transduction in host de- cells. 2007 . Nat. Immunol. 8 : 942 - 949 . bodies against IFN-gamma. J. Immunol. 140 : fense in leprosy. J. Immunol. 167 : 779 – 786 . 30 . Wilson , N.J , K. Boniface , J.R. Chan , B.S. 1506 – 1510 . 18 . Komiyama , Y. , S. Nakae , A. Matsuki , H. McKenzie , W.M. Blumenschein , J.D. Mattson , 5 . Ferber , I.A. , S. Brocke , C. Taylor-Edwards , Nambu , S. Ishigame , K. Kakuta , and Y. B. Basham , K. Smith , T. Chen , F. Morel , et al. W. Ridgway , C. Dinisco , L. Steinman , D. Iwakura . 2006 . IL-17 plays an important role in 2007 . Development, cytokine profi le and func- Dalton , and C.G. Fathman . 1996 . Mice with a the development of experimental autoimmune tion of human interleukin 17-producing helper disrupted interferon-gamma gene are suscep- encephalomyelitis. J. Immunol. 1767 : 566 – 573 . T cells. Nat. Immunol. 8 : 950 – 957 . tible to the induction of experimental auto- 19 . Caspi , R.R. , C.C. Chan , B.G. Grubbs , P.B. 31 . Manel , N. , D. Unutmaz , and D.R. Littman . immune encephalomyelitis (EAE). J. Immunol. Silver , B. Wiggert , C.F. Parsa , S. Bahmanyar , 2008 . The diff erentiation of human Th17 cells 156 : 5 – 7 . A. Billiau , and H. Heremans . 1994 . Endog- requires transforming growth factor-beta and 6 . Voorthuis , J.A. , B.M. Uitdehaag , C.J. De enous systemic IFN-gamma has a protective induction of the nuclear receptor RORgammat. Groot , P.H. Goede , P.H. van der Meide , role against ocular autoimmunity in mice. Nat. Immunol. 9:641 – 649. and C.D. Dijkstra . 1990 . Suppression of J. Immunol. 152 : 890 – 899 . 32 . Volpe , E. , N. Servant , R. Zollinger , S.I. experimental allergic encephalomyelitis by 20 . Jones , L.S. , L.V. Rizzo , R.K. Agarwal , T.K. Bogiatzi , P. Hupe , E. Barillot , and V. Soumelis . intraventricular administration of interferon- Tarrant , C.C. Chan , B. Wiggert , and R.R. 2008 . A critical function for transforming gamma in Lewis rats. Clin. Exp. Immunol. 81 : Caspi . 1997 . IFN-gamma-defi cient mice de- growth factor-beta, interleukin 23 and pro- 183 – 188 . velop experimental autoimmune uveitis in infl ammatory cytokines in driving and mod- 7 . Luger , D. , P. Silver , J. Tang , D. Cua , Z. Chen , the context of a deviant eff ector response. ulating TH-17 responses. Nat. Immunol. 9 : Y. Iwakura , E.P. Bowman , N. Sgambellone , J. Immunol. 158 : 5997 – 6005 . 650 – 657 . C.C. Chan , and R.R. Caspi . 2008 . Either 21 . Lafaille , J.J. , F.V. Keere , A.L. Hsu , J.L. Baron , 33 . Kattah , M.G. , M.T. Wong , M.D. Yocum , a Th17 or a Th1 eff ector response can drive W. Haas , C.S. Raine , and S. Tonegawa . and P.J. Utz . 2008 . Cytokines secreted in autoimmunity: conditions of disease induction 1997 . Myelin basic protein – specifi c T helper response to Toll-like receptor ligand stimula- aff ect dominant eff ector category. J. Exp. Med. 2 (Th2) cells cause experimental autoim- tion modulate diff erentiation in human Th17 205 : 799 – 810 . mune encephalomyelitis in immunodefi cient cells. Arthritis Rheum. 58:1619 – 1629 . 8 . Kroenke , M.A. , T.J. Carlson , A.V. Adnjelkovic , hosts rather than protect them from the dis- 34 . Yang , L. , D.E. Anderson , C. Baecher-Allan , and B.M. Segal . 2008 . IL-12 – and IL-23 – mod- ease. J. Exp. Med. 186 : 307 – 312 . W.D. Hastings , E. Betelli , M. Oukka , V.K. ulated T cells induce distinct types of EAE 22 . Chabas , D. , S. Baranzini , D. Mitchell , C.C.A. Kuchroo , and D.A. Hafl er . 2008 . IL-21 and based on histology, CNS chemokine profi le, Bernard , S. Rittling , D. Denhardt , R. Sobel , TGF-beta are required for diff erentiation of and response to immunomodulatory interven- C. Lock , M. Karpuj , R. Pedotti , et al . 2001 . human Th17 cells. Nature . In press . tion. J. Exp. Med. 205:1535 – 1541. The infl uence of the pro-infl ammatory cyto- 35 . Veldhoen , M. , R.J. Hocking , R.A. Flavell , 9 . Stromnes , I.M. , L.M. Cerretti , D. Liggitt , kine, osteopontin, on autoimmune demye- and B. Stockinger . 2006 . Signals mediated by R.A. Harris , and J.M. Goverman . 2008 . linating disease. Science . 294 : 1731 – 1735 . transforming growth factor beta initiate auto- Diff erential regulation of central nervous sys- 23 . Jansson , M. , V. Panoustsakopoulou , J. Baker , immune encephalomyelitis, but chronic in- tem autoimmunity by T(H)1 and T(H)17 L. Klein , and H.C. Cantor . 2002 . Attenuated fl ammation is needed to sustain disease. Nat. cells. Nat. Med. 14 : 337 – 342 . experimental autoimmune encephalomyeli- Immunol. 7 : 1151 – 1156 . 10 . Cox , C.A. , G. Shi , H. Yin , B.P. Vistica , E.F. tis in Eta-1/osteopontin-defi cient mice. J. 36 . Ming , O.L. , Y.Y. Wan , and R.A.T. Flavell . Wawrousek , C.C. Chan , and I. Gery . 2008 . Immunol. 168 : 2096 – 2099 . 2007 . Cell-produced transforming growth fac- Both Th1 and th17 are immunopathogenic 24 . Shinohara , M.L. , H.J. Kim , J.-H. Kim , V.A. tor-b1 controls T cell tolerance and regulates but diff er in other key biological activities. Garcia , and H. Cantor . 2008 . Alternative Th1- and Th17-cell diff erentiation. Immunity . J. Immunol. 180 : 7414 – 7422 . translation of osteopontin generates intra- 26 : 579 – 591 . 11 . Hue , S. , P. Ahern , S. Buonocore , M.C. cellular and secreted isoforms that mediate 37 . Luo , J. , P.P. Ho , M.S. Buckwalter , T. Hsu , Kullberg , D.J. Cua , B.S. McKenzie , F. distinct biological activities in dendritic cells . L.Y. Lee , H. Zhang , D. Kim , S. Kim , S.S. Powrie , and K. Maloy . 2006 . Interleukin-23 Proc. Natl. Acad. Sci. USA . 105 : 7235 – 7239 . Gambhir , L. Steinman , and T. Wyss-Cory . drives innate and T cell – mediated intestinal 25 . Hur , E. , S. Youssef , M. Haws , S. Zhang , R. 2007 . Glia-dependent TGF-beta signaling, infl ammation. J. Exp. Med. 203 : 2473 – 2483 . Sobel , and L. Steinman . 2007 . Osteopontin independent of the Th17 pathway, is critical JEM VOL. 205, July 7, 2008 1521 for initiation of autoimmune encephalomy- glial cells is associated with active disease novel transcription factor, T-bet, directs Th1 elitis. J. Clin. Invest. 117 : 3306 – 3315 . in multiple sclerosis. Am. J. Pathol. 172 : lineage commitment. Cell . 100 : 655 – 669 . 38 . Lock , C. , G. Hermans , R. Pedotti , A. 146 – 155 . 47 . Chitnis , T ., N. Najafi an, C. Benou, A.D. Brendolan , E. Schadt , H. Garren , A. Langer- 42 . Hever , A. , R.B. Roth , P.A. Hevezi , J. Lee , Salama, M.J. Grusby, M.H. Sayegh, and S.J. Gould , S. Strober , B. Cannella , J. Allard , et al . D. Willhite , E.C. White , E.M. Marin , R. Khoury. 2001 . Eff ect of targeted disruption 2002 . Gene microarray analysis of multiple Herrera , H.M. Acosta , A.J. Acosta , and A. of STAT4 and STAT6 on the induction of sclerosis lesions yields new targets validated Zlotnik . 2006 . Molecular characterization experimental autoimmune encephalomyeli- in autoimmune encephalomyelitis. Nat. Med. of human adenomyosis. Mol. Hum. Reprod. tis. J. Clin. Invest. 108 : 739 – 747 . 8 : 500 – 508 . 12 : 737 – 748 . 48 . Bettelli , E., B. Sullivan, S.J. Szabo, R.A. Sobel, 39 . Ivanov , I. , B.S. McKenzie , Z. Liang , C.E. 43 . Molesworth-Kenyon , S.J. , R. Yin , J.E. L.H. Glimcher, and V.K. Kuchroo. 2004 . Tadokoro , A. Lepelley , J.J. Lafaille , D.J. Cua , Oakes , and R.N. Lausch . 2008 . IL-17 recep- Loss of T-bet, but not STAT1, prevents the and D.R. Littman . 2006 . The orphan nuclear tor signaling infl uences viral induced corneal development of experimental autoimmune receptor RORgammaT directs the diff eren- infl ammation. J. Leukoc. Biol. 83 : 401 – 408 . encephalomyelitis. J. Exp. Med. 200 : 79 – 87 . tiation program of proinfl ammatory IL-17+ 44 . Kennedy, J. , D.L. Rossi , S.M. Zurawski , F. 49 . Lovett-Racke , A.E., A.E. Rocchini, J. Choy, T helper cells. Cell . 126 : 1121 – 1133 . Vega Jr , R.A. Kastelein , J.L. Wagner , C.H. S.C. Northrop, R.Z. Hussain, R.B. Ratts, 40 . Gu , Y. , J. Yang , X. Ouyang , W. Liu , H. Hannum , and A. Zlotnik . 1996 . Mouse IL- D. Sikder, and M.K. Racke. 2004 . Silencing Li , J. Yang , J. Bromberg , S.H. Chen , L. 17: a cytokine preferentially expressed by T-bet defi nes a critical role in the diff erentia- Mayer , J.C. Unkeless , and H. Xiong . 2008 . alpha beta TCR + CD4-CD8-T cells. J. tion of autoreactive T lymphocytes. Immunity. Interleukin 10 suppresses Th17 cytokines se- Interferon Cytokine Res. 16:611 – 617. 21 : 719 – 731 . creted by macrophages and T cells. Eur. J. 45 . Kaplan , M.H. , Y.L. Sun , T. Hoey , and M.J. 50 . Han, M.H. , S. Hwang , D.B. Roy , D.H. Immunol. In print. Grusby . 1996 . Impaired IL-12 responses and Lundgren , J.V. Price , S. Ousman , G. Fernald , 41 . Tzartos , J.S. , M.A. Friese , M.J. Craner , J. enhanced development of T 2 cells in Stat4- B. Gerlitz , W.H. Robinson , S.E. Baranzini , Palace , J. Newcombe , M.M. Esiri , and L. defi cient mice. Nature . 382 : 174 – 177 . et al. 2008 . Proteomic analysis of active mul- Fugger . 2008 . Interleukin-17 production in 46 . Szabo , S. J., S.T. Kim, G.L. Costa, X. Zhang, tiple sclerosis lesions reveals therapeutic tar- central nervous system-infi ltrating T cells and C.G. Fathman, and L.H. Glimcher. 2000 . A gets. Nature. 451 : 1076 – 1081 . 1522 CAVEATS FOR TH17 | Steinman

Journal

The Journal of Experimental Medicine – Pubmed Central

Published: Jul 7, 2008

There are no references for this article.