IL-4 During Th17 Maturation Sensitivity and Resistance to Regulation By

Sensitivity and Resistance to Regulation by IL-4 during Th17 Maturation Laura A. Cooney, Keara Towery, Judith Endres and David A. Fox This information is current as of September 30, 2021. J Immunol 2011; 187:4440-4450; Prepublished online 26 September 2011; doi: 10.4049/jimmunol.1002860 http://www.jimmunol.org/content/187/9/4440 Downloaded from

Supplementary http://www.jimmunol.org/content/suppl/2011/09/27/jimmunol.100286 Material 0.DC1

References This article cites 45 articles, 20 of which you can access for free at: http://www.jimmunol.org/ http://www.jimmunol.org/content/187/9/4440.full#ref-list-1

Why The JI? Submit online.

• Rapid Reviews! 30 days* from submission to initial decision

• No Triage! Every submission reviewed by practicing scientists by guest on September 30, 2021

• Fast Publication! 4 weeks from acceptance to publication

*average

Subscription Information about subscribing to The Journal of Immunology is online at: http://jimmunol.org/subscription Permissions Submit copyright permission requests at: http://www.aai.org/About/Publications/JI/copyright.html Email Alerts Receive free email-alerts when new articles cite this article. Sign up at: http://jimmunol.org/alerts

The Journal of Immunology is published twice each month by The American Association of Immunologists, Inc., 1451 Rockville Pike, Suite 650, Rockville, MD 20852 Copyright © 2011 by The American Association of Immunologists, Inc. All rights reserved. Print ISSN: 0022-1767 Online ISSN: 1550-6606. The Journal of Immunology

Sensitivity and Resistance to Regulation by IL-4 during Th17 Maturation

Laura A. Cooney,1 Keara Towery,2 Judith Endres, and David A. Fox

Th17 cells are highly pathogenic in a variety of immune-mediated diseases, and a thorough understanding of the mechanisms of cytokine-mediated suppression of Th17 cells has great therapeutic potential. In this article, we characterize the regulation of both in vitro- and in vivo-derived Th17 cells by IL-4. We demonstrate that IL-4 suppresses reactivation of committed Th17 cells, even in the presence of TGF-b, IL-6, and IL-23. Downregulation of IL-17 by IL-4 is dependent on STAT6 and mediated by inhibition of STAT3 binding at the Il17a promoter. Although Th1 cytokines were shown to induce IFN-g expression by Th17 cells, IL-4 does not induce a Th2 phenotype in Th17 cells. Suppression by IL-4 is stable and long-lived when applied to immature Th17 cells, but cells that have undergone multiple rounds of stimulation, either in vivo during a Th17-mediated inﬂammatory disease, or in vitro,

become resistant to suppression by IL-4 and lose the ability to signal through IL-4R. Thus, although IL-4 is a potent suppressor of Downloaded from the Th17 genetic program at early stages after differentiation, prolonged stimulation renders Th17 cells impervious to regulatory cytokines. The Journal of Immunology, 2011, 187: 4440–4450.

h17 cells, which express the proinﬂammatory cytokine shown that cytokine-regulatory networks can change as T cells IL-17, play an important role in the pathogenesis of a mature. For example, IL-27 suppresses Th17 development from +

variety of autoimmune and inflammatory diseases. Naive naive CD4 T cells, but it fails to suppress reactivation of com- http://www.jimmunol.org/ T+ CD4 T cells differentiate into Th17 cells in response to TGF-b, mitted Th17 cells (15). Understanding the regulation of pre- IL-6, and IL-21 and become more pathogenic in the presence of existing activated T cells may also have therapeutic applications IL-23 (1–7). In addition, Th17 cell development depends on the in the control of chronic T cell-mediated diseases, such as rheu- transcription factors STAT3, IRF4, RORgt, and RORa (8–11). matoid arthritis. In vitro, induction of Th17 differentiation is inhibited by the Th1 Several groups recently demonstrated a high degree of plasticity cytokines IFN-g and IL-12, as well as the Th2 cytokines IL-4 and in Th17 cells, such that stimulation with IL-12 upregulates T-bet IL-13 (12–14). and IFN-g and induces a Th1-like phenotype, whereas stimula- Most of what is known about the cross-regulation of Th17 cells is tion with TGF-b upregulates Foxp3 and induces a regulatory limited to the earliest stages of differentiation, which occur within T cell-like phenotype (16, 17). One report suggested that the Th17 by guest on September 30, 2021 the first few hours to days following the initial T cell activation. phenotype is unstable, and Th17 cells will spontaneously convert However, it is also important to address the role of cytokine- to Th1 cells in lymphopenic hosts (18). However, another group mediated regulation of committed Th17 cells. Little is known demonstrated that in vitro-generated Th17 cells quickly lose IL-17 about the regulation of secondary stimulation, which may be more expression unless IL-23 is added and opposing cytokines are representative of a Th17 cell that has become activated and dif- blocked, whereas in vivo-generated Th17 cells continue to express ferentiated in a lymph node in the presence of TGF-b and IL-6, IL-17, regardless of which cytokines are added (19). Similarly, exited the lymph node, and traveled to a site of inflammation with in vitro-generated Th17 cells could convert to a Th1 or Th2 a distinct cytokine milieu, such as an inflamed joint. It was also phenotype when stimulated under Th1- or Th2-skewing conditions, whereas in vivo-generated Th17 cells were refractory to Division of Rheumatology, Department of Internal Medicine and Rheumatic Disease Th1- and Th2-polarizing signals. Research Core Center, University of Michigan, Ann Arbor, MI 48109 Some Th17-to-Th1 conversion may not be surprising, consid- 1Current address: Malaria Program, Seattle Biomedical Research Institute, Seattle, ering the evidence for the close relationship between these two WA. lineages. Th17 cells and Th1 cells were thought to share a com- 2Current address: Department of Microbiology, Michigan State University, East mon lineage precursor (20), and IL-17/IFN-g double-positive cells Lansing, MI. are quite common in humans (21, 22). Similarly, the shared de- Received for publication August 23, 2010. Accepted for publication August 18, 2011. pendence on TGF-b implies some regulatory T cell–Th17 com- This work was supported by a grant from the Arthritis Foundation and by National monality. However, there is much less evidence of any Th2–Th17 Institutes of Health Grants AR38477 and P30 AR048310. L.A.C. was supported by a National Institutes of Health immunopathology training grant, a National Institutes commonality. Unlike the frequent occurrence of IL-17/IFN-g of Health regenerative sciences training grant, and a University of Michigan Rack- double-positive cells, IL-17/IL-4 double-positive cells have only ham postdoctoral fellowship. rarely been encountered (23), suggesting that the relationship Address correspondence and reprint requests to Dr. David A. Fox, University of between Th1 and Th17 may be very different from the relationship Michigan, 3918 Taubman Center, 1500 East Medical Center Drive, SPC 5358, Ann Arbor, MI 48109-5358. E-mail address: [email protected] between Th2 and Th17. The online version of this article contains supplemental material. Although IL-4 was shown to inhibit Th17 differentiation and IL- Abbreviations used in this article: BM-DC, bone marrow-derived dendritic cell; ChIP, 17 expression, nothing is known about the molecular mechanisms chromatin immunoprecipitation; CIA, collagen-induced arthritis; ICS, intracellular of this suppression or about the effects of IL-4 on other Th17- cytokine staining; i.d., intradermally; IRS, insulin receptor substrate; KLH, keyhole family genes, such as RORgt and IL-22. Similarly, the effects of limpet hemocyanin; SOCS, suppressor of cytokine signaling. IL-4 on developing versus committed Th17 cells have not been Copyright Ó 2011 by The American Association of Immunologists, Inc. 0022-1767/11/$16.00 explored. In this study, we demonstrate that IL-4 suppresses www.jimmunol.org/cgi/doi/10.4049/jimmunol.1002860 The Journal of Immunology 4441 a subset of Th17 genes distinct from IFN-g. Suppression of IL-17 with streptavidin HRP (BioLegend). The plates were developed with depends on STAT6, but not GATA-3, and correlates with a loss of OptEIA TMB substrate (BD) and absorbance at 450 nm was quantitated STAT3 binding at the Il17a promoter. IL-4 induces a stable, ir- with a Bio-Rad (Hercules, CA) plate reader using KC4 software (Biotek, Winooski, VT). reversible loss of IL-17 expression without inducing conversion to a Th2 phenotype. However, repeated stimulation renders both Flow cytometry in vivo- and in vitro-derived Th17 cells refractory to suppression For intracellular cytokine staining (ICS), cells were stimulated for 6 h with by IL-4. 5 ng/ml PMA and 500 ng/ml ionomycin, with 10 mg/ml brefeldin A added for the last 5 h (all chemicals were purchased from Sigma). Cells were then treated with mouse Fc Block anti-CD16/32, stained with FITC- or Materials and Methods PE-conjugated anti-CD4 (clone GK1.5), and fixed overnight. The next Mice day, cells were permeabilized with saponin and stained with fluorescently For in vitro Th17 differentiation and in vivo keyhole limpet hemocyanin labeled anti–IL-17 (clone TC11-18H10.1), anti–IFN-g (clone XMG 1.2), (KLH) immunization, male 6–8-wk-old BALB/c mice were obtained from and anti–IL-4 (clone 11B11) or the appropriate isotype control (all Abs Jackson Laboratories. For cII/CFA immunization, male 8–10 wk-old from BioLegend). Staining was measured with a FACSCalibur, and data DBA1 mice were obtained from Jackson Laboratories. STAT6-deficient were analyzed using Cell Quest software (BD). and IL-4R mutant mice on a BALB/c background were obtained from Real-time PCR Jackson Laboratories. All animals were housed in specific pathogen-free conditions, and all procedures were approved by the University Committee Gene expression at the mRNA level was analyzed by TaqMan-based real- for the Use and Care of Animals, University of Michigan. Single-cell time PCR with specific primers and probes. First, RNA was collected from suspensions from spleens and thymi of CD4-Cre/STAT5a/bflox mice on frozen cell pellets with the RNeasy Mini kit and treated with DNase a C57BL/6 background were a kind gift from the laboratory of Dr. John (Qiagen). cDNA was generated using the High Capacity cDNA archive kit Downloaded from O’Shea (National Institutes of Health, Bethesda, MD) and were shipped (Applied Biosystems). Relative quantification using the comparative cycle overnight on ice. Freshly isolated spleens from GATA3 conditional- threshold method was carried out using TaqMan Universal PCR Master knockout mice were a kind gift from the laboratory of Dr. James Engel Mix or Gene Expression Master Mix (Applied Biosystems) and run on an (University of Michigan). AB7500 machine. The following primer and probe sets were obtained from Applied Biosystems: IL-17A, IL-17F, IL-22, RORg, IL-23R, IL-4, IL-4R, Generation of bone marrow-derived dendritic cells STAT6, GATA3, b-actin, and GAPDH.

Bone marrow was isolated from femurs and tibias, treated with hypotonic Chromatin immunoprecipitation http://www.jimmunol.org/ ammonium chloride buffer to lyse erythrocytes, and cultured for 6 d at 1 3 106 cells/ml with 10 ng/ml recombinant mouse IL-4 and GM-CSF Chromatin immunoprecipitation (ChIP) was carried out according to the EZ (PeproTech) in RPMI 1640 (10% FCS, 2% L-glutamine, 1% penicillin/ ChIP protocol (Upstate). Briefly, Th17 cells were fixed with formaldehyde streptomycin, and 55 mM 2-ME). The cells were then collected using and lysed with SDS. Lysates were sonicated to shear DNA and immuno- a cell scraper, and CD11c+ cells were positively selected by two rounds of precipitated with protein G and Abs to STAT3. Eluted DNA was quantitated MACS (Miltenyi Biotec). by real-time PCR with SYBR Green Master Mix (Applied Biosystems) and the following primers: Il17a promoter forward: 59-AGGGAGAGCTTC- Purification of naive T cells ATCTGTGG-39, Il17a promoter reverse: 59-AGATTCATGGACCCCAA- Spleens were collected, and CD4+ T cells were magnetically isolated by CAG-39, Il17a/f intergenic region forward: 59-CAGACTCCAAGCAC- negative selection using the EasySep kit from Stem Cell Technologies. The ATCATG-39, Il17a/f intergenic region reverse: 59-GACTGACCTACATT- by guest on September 30, 2021 purified CD4+ T cells were then labeled with CD4 FITC, CD25 PE, CD44 GTGGGC-39, Rorc promoter forward: 59-AGGCTCCTGACCTTTGATT- PE-Cy7, and CD62L allophycocyanin (BioLegend). The CD4+CD252 G-39, and Rorc promoter reverse: 59-AGGGGGTGCTGAGTAATCAC-39. CD44loCD62L+ cells were sorted on a FACSVantage, FACSAria, or FACSDiva. Western blot Th17 cells were washed and rested overnight in RPMI 1640 with 2% FCS Th17 differentiation plus cytokine-neutralizing Abs to minimize background levels of STAT Bone marrow-derived dendritic cells (BM-DCs) and naive T cells were activation. The cells were then incubated with 50 ng/ml IL-4. The re- 6 plated in six-well plates in RPMI 1640 at 0.125 3 10 BM-DCs and 0.25 3 action was stopped with cold PBS plus 1 mM Na3VO4, and the cells were 106 naive T cells/ml with 4 mg/ml anti-CD3 (145-2C11), 10 mg/ml anti– lysed with Phosphosafe extraction reagent (Novagen) supplemented with IL-4 (11B11), 10 mg/ml anti–IFN-g (R4-6A2), 1 ng/ml recombinant hu- protease inhibitor mixture (Calbiochem). Lysates were reduced and de- man TGF-b1 (PeproTech), 20 ng/ml recombinant mouse IL-6 (Pepro- natured by boiling with SDS loading buffer with 100 mM DTT. Samples Tech), and 10 ng/ml recombinant mouse IL-23 (eBioscience). For in- were run on a 10% Precise Tris-Hepes-SDS gel (Pierce, Rockford, IL) and hibition of Th17 differentiation, anti–IL-4 was omitted from the culture, transferred to polyvinylidene fluoride membrane (Millipore). Membranes and recombinant mouse IL-4 (PeproTech) was added at 10 ng/ml, unless were stained with the following primary Abs at 1:1000, unless noted stated otherwise. Alternatively, anti–IFN-g was omitted from the culture, otherwise: anti-STAT6 (Cell Signaling), anti–phospho-Tyr641 STAT6 and recombinant mouse IFN-g (PeproTech) was added at 10 ng/ml, unless (Calbiochem), and anti-GAPDH (1:400; BioLegend). Secondary Abs stated otherwise. Cells were stimulated for 6 d and then collected, washed were goat anti-rabbit IgG HRP (1:1000; Cell Signaling) or rabbit anti-goat twice with cold 2% newborn calf serum/PBS, and put back into culture in IgG HRP (1:10,000; Abcam). Chemiluminescence was developed with the same volume without stimulation for 2 d. For inhibition of Th17 Pierce ECL Western blotting substrate and detected on blue autoradiog- restimulation, IL-4 was added to the culture during the 2-d rest period or raphy film (MidSci). Band intensities were quantified using Kodak 1d 3.6 during a 2-d restimulation with anti-CD3 following the rest period. software. Th17 maturation Immunizations Naive T cells underwent 6 d of Th17 differentiation, followed by 2 d of rest, CFA was prepared by mixing heat-inactivated mycobacterial strain H37Ra according to the protocol described above. To induce maturation, the cells in IFA at 4 mg/ml. For KLH immunization, Imject mcKLH subunits (Pierce) were then expanded in a 2-fold culture volume with the addition of fresh were diluted in PBS to 2 mg/ml and mixed at a 1:1 ratio with CFA. Mice BM-DCs, restimulated with the same cytokine and neutralizing Ab mixture were immunized i.p. with 100 mg KLH. For collagen immunization, ly- for 5 d, and then washed and rested for 2 d. This cycle of 5 d of stimulation ophilized chicken collagen (Chondrex, Redmond, WA) was dissolved and 2 d of rest was repeated, for a total of 3 wk of culture. At the end of 3 wk, overnight in acetic acid at 4 mg/ml. CFA and collagen were mixed at a 1:1 the Th17 cells were restimulated for 2 d with anti-CD3 and rIL-4. ratio to form an emulsion, and 100 mg collagen was injected intradermally (i.d.) at the base of the tail. ELISA Statistical analysis IL-17A was measured by ELISA. Plates were coated with purified anti–IL- 17A (clone TC11-18H10.1, BioLegend), blocked, and then loaded with The p values were calculated by the unpaired two-way Student t test, one- tissue culture supernatants or serum. The plates were washed and treated sample t test, one-way ANOVA, or two-way ANOVA with the Dunnett with biotin-conjugated anti–IL-17A (clone TC11-8H4, BioLegend) mixed posttest, using Prism, as described in the figure legends. 4442 EFFECTS OF IL-4 ON Th17 CELLS

Results Th2 cytokine, also inhibited IL-17 production, confirming recent Th1 and Th2 cytokines suppress reactivation of in vivo-derived data from Newcomb et al. (14) and furthering the idea that Th2 Th17 cells cells can suppress Th17 activity through multiple mechanisms. In addition, neutralizing Abs to IL-4, IFN-g, or IL-12 upregulated Previous work showed that IL-4 can inhibit in vitro Th17 differ- + IL-17 production, implying that the collagen-specific Th17 re- entiation from naive CD4 T cells (12), but much less is known sponse is constrained by endogenous cytokines, possibly from about the effects of IL-4 on pre-existing or memory Th17 cells, collagen-specific T cells of opposing lineages (Fig. 1B). To con- such as those that are likely to persist during many inflammatory firm that these observations are not restricted to DBA mice, the diseases. To examine the cytokine-mediated regulation of Th17 collagen-specific response, or an autoimmune disease, we mea- cells generated in vivo in the context of an IL-17–dependent au- sured Th17 responses in BALB/c mice 2 wk after i.p. immuni- toimmune disease, we immunized DBA1/LacJ mice i.d. with zation with KLH. Despite the well-known Th2 bias in BALB/c chick type II collagen emulsified in CFA, following the standard mice, immunization induced significant KLH-specific IL-17 pro- protocol for induction of collagen-induced arthritis (CIA). After 2 duction in the spleen, which was negatively regulated by both wk, spleen cells were restimulated in vitro with collagen, in the endogenous and exogenous Th1 and Th2 cytokines during Ag- presence or absence of rTh1 and rTh2 cytokines or neutralizing specific rechallenge, recapitulating our previous findings in the Abs. After 5 d of culture, IL-17 production was measured by collagen system (Fig. 1C). These data implied that, even after dif- ELISA. The results in Fig. 1A demonstrate that in vivo-derived, ferentiation, Th17 cells are still susceptible to cytokine-mediated committed Th17 cells are susceptible to counter-regulation by counter-regulation, which raises many interesting questions about a number of other Th1 and Th2 cytokines. In particular, IL-4 and the stability of lineage commitment, the mechanisms of suppres- Downloaded from IL-12 were potent suppressors, and IFN-g was a weak suppressor, sion, and the role of these regulatory pathways in chronic in- of collagen-specific IL-17 production. IL-13, another important flammation. http://www.jimmunol.org/ by guest on September 30, 2021

FIGURE 1. Ag-specific IL-17 production is regulated by endogenous and exogenous Th1 and Th2 cytokines. DBA mice were immunized i.d. with cII/CFA, and spleens were collected 2 wk later. Single-cell suspensions were restimulated in vitro with 50 mg/ml heat-denatured collagen in the presence of recombinant cytokines (A)or10mg/ml purified cytokine-neutralizing Abs (B) for 5 d. Supernatants were collected, and IL-17 was measured by ELISA. Error bars represent SEM of triplicate culture samples. *p , 0.05, **p , 0.01, ***p , 0.001, versus 0 ng/ml (A) or isotype control (B)(one-wayANOVA).C, BALB/c mice were immunized i.p. with KLH in CFA, and spleens were collected 2 wk later. Single-cell suspensions were restimulated in vitro with 50 mg/ml heat-denatured KLH in the presence of 10 ng/ml recombinant cytokines or 10 mg/ml purified cytokine-neutralizing Abs for 5 d. Supernatants were collected, and IL-17 was measured by ELISA. Error bars represent SEM of triplicate culture samples. **p , 0.01, ***p , 0.001, versus KLH or isotype (one-way ANOVA). The Journal of Immunology 4443

Suppression by IL-4 and IFN-g dominates over strong Th17 cells to become Ag independent could have important stimulation of Th17 cells implications for the development of chronic inflammation. Although several groups showed that the combination of TGF-b Given that TGF-b, IL-6, and IL-23 can greatly enhance the and IL-6 acts on naive T cells to induce Th17 differentiation (2, activation of pre-existing Th17 cells, we wondered how these 3, 24), it is not clear what effect TGF-b and IL-6 have on mem- positive signals might interact with negative signals coming from Th1 and Th2 cytokines. Thus, we asked whether IL-4, IFN-g, and ory Th17 cells. Therefore, we decided to restimulate collagen- IL-12 would continue to suppress IL-17 production in the pres- immunized spleen cells in the presence of TGF-b and/or IL-6, ence of TGF-b, IL-6, and IL-23. ELISAs of cytokines produced in with the assumption that 2 wk after immunization most collagen- spleen cell cultures showed that the suppressive signals from IL-4 specific T cells have already differentiated; thus, TGF-b and IL-6 overpowered any combination of activating signals, and even very are more likely to be acting on activated or memory Th17 cells elevated production of IL-17 was still potently downregulated rather than inducing new Th17 differentiation in naive T cells. The (Fig. 2B). Interestingly, IFN-g continued to suppress IL-17 in the results showed that IL-6 alone, and, to a much greater extent, presence of TGF-b, IL-6, and IL-23, whereas IL-12 did not (Fig. TGF-b alone, was able to upregulate IL-17, a result consistent 2C). These results point to different mechanisms of suppression with the presence of some endogenous IL-6. However, the com- downstream of the IFN-g and IL-12 receptors and may reflect bination of TGF-b and IL-6 was significantly better at inducing competition for shared signaling pathways by IL-12 and IL-23. IL-17 than was either cytokine alone, and addition of IL-23 to the mixture enhanced IL-17 production further (Fig. 2A). Moreover, IL-4 and IFN-g selectively inhibit unique subsets of Th17 genes TGF-b, IL-6, and IL-23, alone or in combination, induced con- Downloaded from siderable IL-17 production, even in the absence of exogenous To more carefully investigate the regulatory mechanisms at work collagen, suggesting that the right cytokine milieu can stimulate during different stages of Th17 development, we next analyzed pre-existing Th17 cells without the need for concomitant TCR in vitro-derived Th17 cells. Purified naive CD4+ T cells were stimulation. This inference assumes that Ag from the in vivo differentiated into Th17 cells for 5 or 6 d, as described in Mate- immunization does not persist in these cultures. The potential for rials and Methods. The cells were then rested for 2 d, followed by http://www.jimmunol.org/ by guest on September 30, 2021

FIGURE 2. TGF-b, IL-6, and IL-23 synergistically upregulate IL-17 production during collagen restimulation, but IL-4 and IFN-g continue to suppress. Splenocytes from immunized mice were restimulated with or without collagen, IL-23, TGF-b, IL-6, or 10 mg/ml neutralizing Abs to IL-4 and IFN-g (A)orwith collagen, TGF-b, IL-6, IL-23, and increasing concentrations of IL-4, IFN-g,orIL-12(B, C). Supernatants were collected after 5 d, and IL-17 was measured by ELISA. Error bars represent SEM of triplicate culture samples. A,**p , 0.01, ***p , 0.001 versus collagen alone; ##p , 0.01, ###p , 0.001 versus unstimulated (one-way ANOVA). B,**p , 0.01, ***p , 0.001 versus 0 ng/ml IL-4 by two-way ANOVA. C,***p , 0.001 versus 0 ng/ml cytokine (two-way ANOVA). 4444 EFFECTS OF IL-4 ON Th17 CELLS challenge with rTh1 and rTh2 cytokines in the presence or absence that endogenous cytokines also inhibit IL-17 expression via of anti-CD3 for 2 d. This protocol of 5 d of priming/2 d of rest/ STAT6 (Fig. 3D). STAT5 can also be activated by IL-4 and was 2 d of restimulation allows us to specifically focus on the effects of reported to mediate suppression of IL-17 downstream of IL-2 (25, opposing cytokines on activation of pre-existing Th17 cells, rather 26). However, using STAT5-deficient spleen cells stimulated with than the effects on differentiation of naive T cells into Th17. Th17 anti-CD3, we found that STAT5 was dispensable for suppression activity was measured by ICS, ELISA, and real-time PCR. As of IL-17 by IL-4 (Supplemental Fig. 3). Similarly, IL-4R signaling predicted, the results clearly demonstrated that adding IL-4 or was shown to activate insulin receptor substrate (IRS)-2; however, IFN-g during restimulation inhibited the expression of IL-17 when we used mice bearing a point mutation in the IL-4R at the protein, sometimes by as much as 95%, and with a more potent IRS-2 binding site, thus preventing recruitment and activation of inhibition by IL-4 (Fig. 3A). We also found that IL-4 and IFN-g IRS-2 by IL-4 without impacting STAT6 activation, rIL-4 con- inhibited expression of IL-17A, IL-17F, and RORgt mRNA (Fig. tinued to inhibit IL-17 expression, demonstrating that IRS-2 ac- 3C). Interestingly, however, these cytokines selectively inhibited tivation is not required for suppression of IL-17 (Supplemental the expression of a unique subset of Th17 genes: IL-4 suppressed Fig. 1). IL-23R but not IL-22, whereas IFN-g suppressed IL-22 but not Previous data from other groups suggested that suppression of IL-23R. The differential regulation of IL-22 and IL-23R implied Th1 development by IL-4 is mediated by GATA-3, the transcription that the Th17 gene-expression program is not completely reversed factor induced by IL-4 that acts as the master regulator of Th2 in the presence of IFN-g and IL-4, leaving open the possibility differentiation (27–30). In addition, we found that GATA-3 mRNA that these conflicting cytokine milieus yield either a mixed pop- is upregulated in Th17 cells as early as 2 h after the addition of ulation of cells or cells of a mixed phenotype. IL-4, prior to downregulation of IL-17 mRNA (data not shown). Downloaded from Thus, we hypothesized that GATA-3 may play a role in sup- Suppression by IL-4 depends on STAT6 but not GATA3 pression of IL-17 by IL-4. Mice bearing a floxed Gata3 gene Because IL-4R signaling is mediated primarily by STAT6, we with inducible Cre recombinase under control of the IFN-indu- hypothesized that STAT6 was required for suppression of IL-17 by cible Mx1 promoter (Gata3fl/fl:TgMx1cre mice) were treated with IL-4; to address this question, we repeated our previous in vitro polyinosinic-polycytidylic acid to induce excision of the Gata3

experiments using STAT6-deficient Th17 cells. In the restimulation gene, which resulted in a 70% decrease in total splenic levels of http://www.jimmunol.org/ of in vitro-derived Th17 cells, IL-4 had no effect on IL-17 ex- GATA3 mRNA (Supplemental Fig. 2). Although the in vivo con- pression in the absence of STAT6 (Fig. 3B). Similarly, by real-time ditional deletion was less than ideal, published reports using these RT-PCR, IL-4 failed to robustly suppress IL-17A, IL-17F, RORgt, mice demonstrated a near-complete suppression of early T lineage and IL-23R expression in STAT6-deficient Th17 cells (Fig. 3C). progenitor development, suggesting that the incomplete deletion is Spleen cells from STAT6-deficient mice stimulated with anti-CD3 sufficient to significantly alter T cell function (31). Spleen cells in the absence of exogenous IL-4 also had increased IL-17 ex- from these mice and control mice that are unable to delete GATA3 pression by ELISA compared with wild-type T cells, indicating were cultured with anti-CD3 in the presence or absence of Th17- by guest on September 30, 2021

FIGURE 3. IL-4 suppresses a different subset of Th17 genes than does IFN-g, and the effect of IL-4 is mediated by STAT6. Naive CD4+ T cells from wild-type or STAT6-deficient mice were cultured under Th17-skewing conditions for 5 d, followed by 2 d of rest and 2 d of restimulation in the presence or absence of 50 ng/ml IL-4 or IFN-g. A, Cells were treated for 6 h with PMA, ionomycin, and brefeldin A, and IL-17 expression was measured by ICS. B, Cells were treated for 6 h with PMA, ionomycin, and brefeldin A, and IL-17 expression was measured by ICS. Error bars represent SEM of triplicate cultures. ***p , 0.001 versus 0 ng/ml IL-4 (one-way ANOVA). C, RNA was collected, and IL-17A, IL-17F, RORgt, IL-22, and IL-23R expression was measured by real-time PCR with primers and probes from Applied Biosystems. Data were normalized first to b-actin, the internal control, and then to the matched sample without IL-4 or IFN-g. Error bars represent SEM of triplicate PCRs. D, Augmented IL-17 expression in STAT6-deficient mice. Spleen cells from wild-type BALB/c, STAT6 knockout, IL-4 knockout, or IL-4R knockout mice were stimulated in vitro with anti-CD3. Five days later, supernatant was collected, and IL-17 expression was measured by ELISA. Bars represent mean SEM of triplicate samples. ***p , 0.001 versus wild-type (one-way ANOVA). The Journal of Immunology 4445 skewing cytokines, and the effect of IL-4 on IL-17 expression was measured by ICS and real-time PCR. However, contrary to our prediction, the results in Fig. 4 show that deletion of Gata3 had no effect on suppression of IL-17 by IL-4. Suppression by IL-4 results in loss of STAT3 binding at the Il17a promoter Th17 development in mice depends on the transcription factor STAT3, which is activated by IL-6 and IL-23. STAT3 has multiple roles in Th17 development: in activated Th17 cells stimulated with IL-23, it binds directly to the Il17a promoter and induces IL-17 expression, and in naive T cells stimulated with TGF-b and IL-6, it is required for induction of RORgt expression (10, 32–34). Thus, we hypothesized that IL-4 may mediate suppression of Th17 activity by inhibiting STAT3 binding at Th17 gene loci. To test this hypothesis, we restimulated Th17 cells with anti-CD3 and Th17- FIGURE 5. IL-4 inhibits STAT3 binding at the Il17a promoter but not at skewing cytokines in the presence or absence of IL-4 for 6 or 24 h other sites. Th17 cells were generated in vitro, as described. After 1 wk, the and measured STAT3 binding at the Il17a promoter, Il17a/f inter- cells were restimulated for 6 h with anti-CD3 and Th17-polarizing cyto- genic region, and Rorc promoter by ChIP. The results in Fig. 5 kines in the presence or absence of 50 ng/ml IL-4. ChIPs were carried out Downloaded from demonstrate very strong STAT3 binding at all three regions. In- using the EZ-ChIP kit, according to manufacturer’s instructions (Milli- terestingly, IL-4 inhibited STAT3 binding at the Il17a promoter but pore), with Ab specific for STAT3 or isotype control. Eluted DNA was had no effect at the Il17a/f intergenic region or the Rorc promoter. quantitated by SYBR Green real-time PCR with primers specific for the Il17a promoter, Il17a/f intergenic region, or Rorc promoter. Data are Suppression by IL-4 is stable but does not induce Th2 normalized to the corrected cycle threshold values of the 1% input sample. conversion Error bars represent SEM of triplicate PCR reactions. **p , 0.01 versus http://www.jimmunol.org/ Several groups recently demonstrated a high degree of plasticity in Th17 conditions (Student t test). Th17 cells (17, 35); thus, we decided to ask whether prolonged culture with IL-4, the chief Th2-skewing cytokine, would induce in Th17 cells by ICS and found that ,2% of the cells expressed Th17-to-Th2 conversion. Th17 cells were generated in vitro, with IL-4 after 2 d of restimulation in Th0 or Th2 conditions. Fur- 5 d of differentiation followed by 2 d of rest. Following the rest thermore, the IL-4 expression on day 2 was extinguished by day 4 period, the Th17 cells were restimulated in Th0, Th2, or Th17 or 6 of Th2 culture, and no cells coexpressed IL-4 and IL-17, conditions for 2, 4, or 6 d. Th2 conversion was assessed by real- implying that there was no Th17-to-Th2 conversion (Fig. 6B, time PCR for IL-4 and GATA-3, as well as IL-17 and RORgt. The data not shown). Thus, although low levels of IL-4 and GATA-3 results in Fig. 6A show that low levels of mRNA for IL-4 and mRNA persist, IL-4 protein expression is quickly extinguished, by guest on September 30, 2021 GATA-3 were expressed in Th2-stimulated Th17 cultures, but suggesting a role for posttranscriptional regulation. similar levels of IL-4 and GATA-3 message were also expressed in Although we demonstrated that Th17 cells stimulated in the Th17 cells stimulated with anti-CD3 alone. Although IL-4 mRNA presence of IL-4 do not convert to Th2 cells, it was not clear expression persisted longer in cultures restimulated with Th2 con- whether IL-17 expression was only temporarily downregulated, ditions versus anti-CD3, these results may indicate a small number requiring constant IL-4R signaling to maintain suppression, or of contaminating Th0 or Th2 cells in the culture, rather than in- persistently extinguished, even after removal of IL-4. Thus, we duction of the Th17-to-Th2 conversion. In addition, the levels of again generated Th17 cells in vitro (primary culture), which were IL-4 mRNA expressed by Th17 cells restimulated in Th2 con- restimulated in secondary culture with Th2 conditions (anti-CD3, ditions were significantly lower than the levels of IL-4 mRNA 10 ng/ml IL-4, anti–IFN-g), Th0 conditions (anti-CD3, anti–IL-4, expressed by naive T cells differentiated under Th2 conditions anti–IFN-g), or “Th23 conditions” (anti-CD3, 20 ng/ml IL-23, in vitro (Supplemental Fig. 3). We also examined IL-4 expression anti–IL-4, anti–IFN-g) to maintain the existing Th17 cell population without inducing new differentiation. After 2 d of secondary culture, the cells were washed and rested in tertiary culture for 1, 2, or 3 d to allow the Th17 cells to regain IL-17 expression. After each phase of culture, cells were restimulated with PMA, ionomycin, and brefeldin A for ICS. The results in Fig. 7 show that secondary stimulation in Th0 or Th23 conditions imprinted Th17 cells for increasing levels of IL-17 expression after stimulation was removed, whereas secondary stimulation in Th2 conditions resulted in a continual decline in IL-17 expression, even after IL-4 was removed. These results demonstrated that IL-4– mediated suppression of IL-17 is stable and does not require continual exposure to IL-4, suggesting that IL-4 may induce heri- table changes in chromatin structure at the Il17a locus. Cell num- bers and viability were consistent within groups and over time FIGURE 4. GATA3 is not required for suppression of IL-17 by IL-4. during tertiary culture, implying that there was no selective pro- Spleen cells from mice with inducible Gata3 deletion or controls without liferation or cell death. Thus, IL-17 expression is easily sup- deletion were stimulated with anti-CD3 and increasing concentrations of pressed by low doses of IL-4, despite the presence of activating IL-4. After 5 d, cells were treated with PMA, ionomycin, and brefeldin A cytokines, and the suppression by IL-4 is also persistent and not for 6 h, and IL-17 expression was measured by ICS. passively reversed. 4446 EFFECTS OF IL-4 ON Th17 CELLS Downloaded from http://www.jimmunol.org/ by guest on September 30, 2021

FIGURE 6. Th17 cells restimulated in Th2 conditions do not convert to Th2 cells. A, Th17 cells were restimulated for 2, 4, or 6 d in Th0, Th2, or Th17 conditions, and expression of IL-4 and GATA-3 was measured by real time PCR. Results were normalized to b-actin expression. Error bars represent the SEM of triplicate PCRs. B, IL-17A and IL-4 expression was measured by ICS.

Repeated stimulation renders Th17 cells resistant to disease where they may encounter many suppressive cytokines. suppression by IL-4 One explanation may be the process of maturation or stabilization. The results shown thus far suggested that Th17 cells are very Previous reports from other groups demonstrated that developing sensitive to suppression by IL-4, which raises the question of how Th1 cells progress through several stages of maturation, gradually Th17 cells manage to persist in vivo during an inﬂammatory stabilizing their phenotype in response to cytokine stimulation.

FIGURE 7. Suppression of IL-17 expression persists after removal of IL-4. Th17 cells were generated in vitro and treated with IL-4 to suppress IL-17 expression, as described, and then washed and put back into culture for 1, 2, or 3 d to allow cells to regain IL- 17 expression. Primary culture = 5 d of Th17 differentiation and 2 d of rest, as described. Secondary culture = 2 d restimulation in Th0 conditions (anti-CD3, anti–IL-4, anti–IFN-g), Th23 conditions (anti-CD3, anti–IL-4, anti–IFN-g, 20 ng/ml IL-23), or Th2 conditions (anti-CD3, 10 ng/ml IL-4, anti–IFN-g). Tertiary culture = 1, 2, or 3 d resting culture. After each phase of culture, cells were restimulated with PMA, ionomycin, and brefeldin A, and IL-17 expression was assessed by ICS. The Journal of Immunology 4447

Early in the culture, IL-4 suppresses Th1 differentiation and IFN-g with Ag in the presence or absence of IL-4 for 2 d, followed by IL-17 expression, but after prolonged stimulation, Th1 cells lose the ELISA of the supernatants. As the results in Fig. 8C show, we found ability to respond to IL-4 and become resistant to suppression (29, that 3 d of ex vivo restimulation with Ag was sufficient to induce 36, 37). Previous data from our laboratory on the effect of IL-4– Th17 maturation and resistance to suppression by IL-4 in whole- transduced dendritic cells on IL-17 production during CIA sug- spleen cultures from 2-wk immunized mice. Similar results were gested that Th17 cells are less susceptible to regulation by IL-4 observed for KLH-immunized BALB/c mice (data not shown). after the onset of arthritis (38). Thus, we decided to address the The ability of Th17 cells to become resistant to suppression maturation of Th17 cells in vitro and in vivo, as measured by could have important implications for the development of auto- insensitivity to suppression by IL-4. Th17 cells were generated immune disease, and IL-4–transduced dendritic cells were shown in vitro with 5 d of differentiation and 2 d of rest, as in previous to suppress collagen-specific IL-17 production from mice before experiments. To induce maturation, we repeated this process two the onset of arthritis but not after (38). Thus, we asked whether additional times, for a total of 3 wk of culture, and then assayed Th17 maturation correlated with disease progression or severity in the effect of IL-4 on IL-17 expression during a 2-d restimulation CIA. To address this question, we collected spleens from mice with anti-CD3, followed by ICS. Fig. 8A shows representative dot at different time points after immunization with cII/CFA and plots for IL-17 expression in Th17 cells treated with IL-4 after 1 assessed the Th17 sensitivity to suppression by IL-4 on day 0, 1, or 3 wk of culture. The results showed that after three rounds of or 3 of culture. We found that, 6 wk after immunization, splenic stimulation, Th17 cells are largely resistant to suppression by Th17 cells were moderately sensitive to suppression by IL-4 di- IL-4. Interestingly, the results in Fig. 8B demonstrate that Th17 rectly ex vivo. However, Th17 responses of cells from mice that cells cultured for 3 wk become resistant to suppression by IL-4 had undergone prolonged immunizations matured more quickly ex Downloaded from and IFN-g but remain sensitive to suppression by IL-12. vivo than did Th17 responses of cells from very short immuni- We already showed that IL-4 suppresses IL-17 expression by zations (Fig. 8D, Table I). collagen-immunized splenocytes, thus demonstrating that Ag-spe- Mature Th17 cells express IL-4R but lose the ability to cific Th17 cells in the spleen 2 wk after immunization have not phosphorylate STAT6 fully matured. However, we decided to look for ex vivo maturation of splenocytes from immunized mice by stimulating with Ag Our observation of desensitization of Th17 cells to challenge with http://www.jimmunol.org/ for 1 or 3 d to induce maturation and then washing and challenging IL-4 begged the question of whether signaling through IL-4R by guest on September 30, 2021

FIGURE 8. Th17 cells become resistant to suppression by IL-4 both in vitro and ex vivo. A, Th17 cells were generated by 5 d of stimulation and 2 d of rest, as described, followed by 2 d of restimulation in the presence or absence of IL-4 and ICS for IL-17. Alternatively, cells were cultured for two more rounds of 5 d of Th17 stimulation and 2 d of rest, for a total of 3 wk of culture. After 3 wk, the cells were restimulated for 2 d in the presence or absence of IL-4, followed by ICS for IL-17. B, Th17 cells cultured for 3 wk to induce maturation were restimulated for 2 d with increasing concentrations of IL-4, IFN-g, or IL-12, followed by ICS for IL-17. C, DBA mice were immunized i.d. with cII/CFA, and spleens were collected 2 wk later. Total spleen cells were restimulated with heat-denatured collagen for 0, 1, or 3 d and then collected, washed, and replated with collagen and increasing concentrations of IL-4 for 2 d. Supernatants were collected, and IL-17 was measured by ELISA. Bars represent mean 6 SEM of triplicate culture samples, normalized to 0 ng/ml IL-4. *p , 0.05, ***p , 0.001 versus 0 ng/ml IL-4 (one-way ANOVA). D, DBA mice were immunized with cII/CFA, and spleens were collected 4, 15, or 31 d later. Spleen cells were restimulated with collagen, and 50 ng/ml IL-4 was added on day 0, 1, or 3. Supernatants were collected on day 5, and IL-17 was measured by ELISA. IL-17 expression is normalized to the sample with no IL-4. 4448 EFFECTS OF IL-4 ON Th17 CELLS

Table I. In vivo experience correlates with ex vivo development of not signiﬁcantly downregulated in 3-wk Th17 cultures versus 1- or resistance to IL-4 2-wk cultures.

Time Since Immunization (wk) Culture until IL-4 Resistance (d) Discussion 1 .3 The data presented demonstrate a remarkable degree of complexity 22in the cytokine-mediated regulation of Th17 cells. Initial reports 41suggested that Th cell cross-regulation is rather black and white: 6 0–1 TGF-b, IL-6, and IL-23 promote Th17 cells, whereas IL-4, IFN-g, DBA mice were immunized with cII/CFA and spleens were collected after 1, 2, 4, and IL-12 inhibit Th17 cells. However, upon closer inspection, we or 6 wk. Spleen cells were restimulated with collagen and IL-4 was added on d 0, 1, or 3 of culture. IL-17 was measured by ELISA on d 5. A designation of resistance see many shades of gray. For instance, Th17 cells developing in required that IL-17 production in the presence of IL-4 was $70% of IL-17 pro- the presence of IL-4 may continue to express IL-22, whereas Th17 duction in the absence of IL-4. Spleens from 1-wk–immunized mice continued to respond to IL-4 when it was added after 3 d of culture, whereas spleens from 6-wk– cells developing in the presence of IFN-g may continue to express immunized mice failed to respond when IL-4 was added on day 0 or 1. IL-23R, suggesting that there may be an array of Th cell subset phenotypes that include intermediate states that are not fully polarized to one of the well-defined Th subsets. It is also particularly remained intact following maturation. To measure IL-4R signal- interesting that IL-4 and IFN-g continue to suppress IL-17 pro- ing, Th17 cells were rested in low-serum media with cytokine- duction in the presence of TGF-b, IL-6, and IL-23, whereas IL-12 neutralizing Abs overnight to minimize the background level of does not. Another important difference appears in the process of activation; stimulated with IL-4 for 15, 30, or 120 min; and lysed in Th17 maturation, which results in resistance to suppression by Downloaded from the presence of protein phosphatase inhibitors. STAT6 activation IL-4 and IFN-g but not to IL-12. was measured by Western blot with phospho-STAT6 specific Abs. Th1 cytokines can suppress IL-17 expression but also induce The results in Fig. 9A show that IL-4 induced significantly less a Th1-like phenotype in Th17 cells. In contrast, IL-4 induces potent STAT6 activation in mature Th17 cells cultured for 3 wk versus and stable suppression without inducing any Th2 conversion, even immature Th17 cells cultured for 1 wk, regardless of whether in the face of strong Th17 stimuli. This is consistent with previous phospho-STAT6 levels were normalized to total STAT6 or to the work from our laboratory demonstrating that a single injection of http://www.jimmunol.org/ loading control GAPDH. In addition, when total STAT6 expres- IL-4–transduced dendritic cells induced long-lasting protection sion was normalized to GAPDH, we found that STAT6 was ac- from CIA, which is likely mediated by suppression of collagen- tually upregulated in mature Th17 cells (data not shown). Thus, specific Th17 responses (38, 39). These results suggest a funda- the loss of STAT6 activation was not simply due to downregula- mental difference in the mechanisms underlying regulation of tion of STAT6 expression. Th17 cells by Th1 and Th2 cytokines. How Th17 cells integrate One plausible explanation for the loss of STAT6 activation is a complex array of positive and negative signals is an interesting downregulation of IL-4R. Therefore, we decided to measure IL- area for future research and may depend on the ability of key 4Ra expression at different stages of Th17 maturation by flow transcription factors, such as T-bet and GATA-3, to bind to Th17 cytometry. The results in Fig. 9B show that IL-4R expression was gene loci and interact with RORgt. by guest on September 30, 2021

FIGURE 9. Impaired IL-4R signaling in mature Th17 cells. A, Th17 cells were generated in vitro with 1 or 3 wk of stimulation and rested overnight in low-serum media with cytokine-neutralizing Abs to reduce the background level of STAT activation. Cells were then washed and stimulated with 50 ng/ ml IL-4 for 15, 30, or 120 min and lysed with PhosphoSafeLysis Buffer (Nova- gen). Lysates were reduced and run on 10% SDS-PAGE gels and stained with Abs for phospho-STAT6, total STAT6, or GAPDH. Band intensities were quantitated with Kodak software, and phospho- STAT6 intensity was normalized either to total STAT6 intensity or GAPDH intensity; p , 0.05 for the average band intensity between mature and immature Th17 cells (two-way t test). B, IL-4R expression in immature and mature Th17 cells was measured by ﬂow cytometry. Bold lines represent IL-4R; dotted lines represent isotype control. The Journal of Immunology 4449

The molecular mechanisms mediating Th17 suppression and/or 41). In this regard, our in vitro culture conditions may more lineage conversion downstream of Th1 and Th2 cytokines remain closely approximate the natural setting, with respect to Th17 cell largely unknown. We showed that IL-4–mediated suppression maturation, compared with more short-term Th17 cultures. is dependent on STAT6 and independent of STAT5, IRS-2, and Similar to published data on mature Th1 cells (36, 37, 42), we GATA-3. One potential mechanism of suppression downstream were able to demonstrate a loss of STAT6 activation in response to of IL-4R could be direct binding of STAT6 to Th17 gene loci IL-4 in mature Th17 cells, despite normal levels of all of the IL- and inhibition of transcription. However, we performed multiple 4R–signaling components. One potential mechanism for loss of STAT6 ChIPs, scanning wide regions of the Il17a and Rorc pro- STAT6 activation in mature Th17 cells is upregulation of a mem- moters, and found no evidence for direct binding (Supplemental ber of the suppressor of cytokine signaling (SOCS) family of Fig. 4). Given that IL-17 expression is dependent on STAT3, we proteins. Most SOCS family proteins preferentially interact with hypothesized that IL-4R signaling may suppress IL-17 expression one or more specific cytokine receptors, thereby inhibiting acti- by inhibiting STAT3 DNA-binding activity, which we confirmed vation of unique STAT molecules. For example, SOCS3, was by ChIP for STAT3 at the Il17a promoter. Interestingly, STAT3 shown to suppress Th17 differentiation by binding to IL-6R and binding was specifically downregulated at the Il17a promoter, inhibiting STAT3 activation, whereas SOCS5 was shown to bind whereas there was no loss of STAT3 binding at the Rorc promoter to IL-4R and inhibit STAT6 activation in Th1 cells (42). However, or Il17a/f intergenic region, suggesting that STAT3 may be dis- there can be some redundancy, because previous work also dem- placed from the Il17a promoter by a STAT6-induced transcrip- onstrated that inhibition of IL-4R signaling in Th1 cells may be tional repressor that specifically binds this region. We screened mediated by SOCS1 (43, 44). In fact, there is disagreement on several candidates but were unable to identify the direct target of the underlying mechanisms of IL-4R desensitization in mature Downloaded from STAT6 that ultimately mediates IL-17 gene silencing. Th1 cells: Seki et al. (42) demonstrated selective upregulation of Although much attention has been given to the cytokines that SOCS5 in committed Th1 cells and SOCS5-dependent inhibition regulate Th cell differentiation from naive T cells, there is a paucity of IL-4R signaling, whereas Huang et al. (37) found no increase in of data on how cells are regulated beyond this early window. For expression of SOCS1, SOCS3, or SOCS5 and suggested that re- instance, current dogma states that TGF-b and IL-6 act on naive cruitment of STAT6 to IL-4R was impaired through an unknown

T cells to induce Th17 differentiation, whereas IL-23 acts on mechanism. In our in vitro-matured Th17 cells, there was upreg- http://www.jimmunol.org/ existing Th17 cells; however, our observations suggested that ulation of both SOCS1 and SOCS5; however, Th17 cells from TGF-b and IL-6 may be equally as important as IL-23 for aug- SOCS5-deficient mice (a kind gift from Dr. Sandra Nicholson, menting cytokine production by effector or memory Th17 cells. Walter and Eliza Hall Institute) showed no loss of IL-4R de- In addition, recent observations demonstrated a remarkable degree sensitization upon maturation (data not shown). Future experi- of fluidity within Th cell lineages in vivo, with cells converting ments are needed to address the role of SOCS1 in Th17 matu- from one lineage to another or stably expressing multiple cyto- ration. This line of inquiry is made difficult by the lethal pheno- kines characteristic of different lineages. Thus, much more work is type of SOCS12/2 mice, which has been attributed to overexpres- needed to address the role of cytokine-mediated regulation through- sion of IFN-g (45). out the full lifespan of each of the Th cell subsets, as well as the The simple observation that inflamed joints from arthritic mice by guest on September 30, 2021 ways in which our notions of divergent Th lineages break down coexpress large quantities of IL-17, IFN-g, and IL-4 (46) suggests and the lines between distinct subsets become blurred. that Th17 cells at the site of inflammation are resistant to sup- In our experiments, Th17 cells cultured under Th2 conditions did pression and that more work is needed to determine the role of not take on a Th2 phenotype and were unable to re-express IL-17 Th17 maturation in disease. Similarly, much work is needed to after the suppressive signals were removed. However, we also assess the sensitivity and resistance to suppression by IL-4 in found that three rounds of in vitro stimulation rendered Th17 cells human Th17 cells from patients with various immune-mediated resistant to suppression by IL-4 as a result of desensitization of IL- diseases. The concept of distinct stages of human Th17 maturation 4R, suggesting that there may be a specific window of opportunity raises many exciting new questions and ideas about the etiology of for Th17 suppression. Importantly, we observed a similar matu- Th17-mediated disease. Better understanding of the molecular ration process in Th17 cells generated in vivo, and maturation state mechanisms mediating stabilization of committed cytokine pro- correlated with disease progression. The in vitro maturation ki- duction may lead to new approaches for targeted therapies. netics closely mirrored what was demonstrated for Th1 cells, suggesting that this process is not Th17 specific, but rather may be Acknowledgments a universal property of chronically activated Th cells (to our We thank the following people for kind donation of mutant mouse tissues knowledge, there is no evidence regarding Th2 maturation). Al- and reagents: Arian Laurence, John O’Shea, Chia-Jui Ku, Tomonori though we found that Th17 cells generated both in vitro and in vivo Hosoya, Douglas Engel, and Sandra Nicholson. We also thank the staff gradually become more stable and resistant to suppression, data at the University of Michigan hybridoma core facility and the flow cytom- from other groups suggested that in vitro-derived Th17 cells etry core facility for technical assistance. We are grateful to Donna Cash quickly lose their IL-17 expression and can be converted to other for assistance with preparation of the manuscript. lineages, even after three rounds of stimulation. However, in vivo- derived Th17 cells maintained their phenotype and were resistant to Disclosures suppression (17–19). The conflicting observations of lineage sta- The authors have no financial conflicts of interest. bility following in vitro differentiation may be due to differences in culture conditions or APC populations. For example, other References investigators used peptide plus irradiated spleens cells, whereas 1. Langrish, C. L., Y. Chen, W. M. Blumenschein, J. Mattson, B. Basham, we used anti-CD3 and BM-DCs. Several groups showed that J. D. Sedgwick, T. McClanahan, R. A. Kastelein, and D. J. Cua. 2005. IL-23 in vitro-derived Th17 cells differ greatly from in vivo-derived drives a pathogenic T cell population that induces autoimmune inflammation. J. Th17 cells, and one of these reports suggested that in vivo- Exp. Med. 201: 233–240. 2. Bettelli, E., Y. Carrier, W. Gao, T. Korn, T. B. Strom, M. Oukka, H. L. Weiner, derived memory Th17 cells are more resistant to suppression and V. K. Kuchroo. 2006. Reciprocal developmental pathways for the generation and lineage conversion than in vitro-derived Th17 cells (19, 40, of pathogenic effector TH17 and regulatory T cells. Nature 441: 235–238. 4450 EFFECTS OF IL-4 ON Th17 CELLS

3. Veldhoen, M., R. J. Hocking, C. J. Atkins, R. M. Locksley, and B. Stockinger. receptor activates STAT5 by a mechanism that relies upon common gamma- 2006. TGFbeta in the context of an inflammatory cytokine milieu supports de chain. J. Biol. Chem. 273: 31222–31229. novo differentiation of IL-17-producing T cells. Immunity 24: 179–189. 26. Laurence, A., C. M. Tato, T. S. Davidson, Y. Kanno, Z. Chen, Z. Yao, 4. Korn, T., E. Bettelli, W. Gao, A. Awasthi, A. Ja¨ger, T. B. Strom, M. Oukka, and R. B. Blank, F. Meylan, R. Siegel, L. Hennighausen, et al. 2007. Interleukin-2 V. K. Kuchroo. 2007. IL-21 initiates an alternative pathway to induce proin- signaling via STAT5 constrains T helper 17 cell generation. Immunity 26: 371– flammatory T(H)17 cells. Nature 448: 484–487. 381. 5. Nurieva, R., X. O. Yang, G. Martinez, Y. Zhang, A. D. Panopoulos, L. Ma, 27. Fields, P. E., S. T. Kim, and R. A. Flavell. 2002. Cutting edge: changes in histone K. Schluns, Q. Tian, S. S. Watowich, A. M. Jetten, and C. Dong. 2007. Essential acetylation at the IL-4 and IFN-gamma loci accompany Th1/Th2 differentiation. autocrine regulation by IL-21 in the generation of inflammatory T cells. Nature J. Immunol. 169: 647–650. 448: 480–483. 28. Hwang, E. S., S. J. Szabo, P. L. Schwartzberg, and L. H. Glimcher. 2005. T 6. McGeachy, M. J., Y. Chen, C. M. Tato, A. Laurence, B. Joyce-Shaikh, helper cell fate specified by kinase-mediated interaction of T-bet with GATA-3. W. M. Blumenschein, T. K. McClanahan, J. J. O’Shea, and D. J. Cua. 2009. The Science 307: 430–433. interleukin 23 receptor is essential for the terminal differentiation of interleukin 29. Ouyang, W., S. H. Ranganath, K. Weindel, D. Bhattacharya, T. L. Murphy, 17-producing effector T helper cells in vivo. Nat. Immunol. 10: 314–324. W. C. Sha, and K. M. Murphy. 1998. Inhibition of Th1 development mediated by 7. Zhou, L., I. I. Ivanov, R. Spolski, R. Min, K. Shenderov, T. Egawa, D. E. Levy, GATA-3 through an IL-4-independent mechanism. Immunity 9: 745–755. W. J. Leonard, and D. R. Littman. 2007. IL-6 programs T(H)-17 cell differen- 30. Usui, T., J. C. Preiss, Y. Kanno, Z. J. Yao, J. H. Bream, J. J. O’Shea, and tiation by promoting sequential engagement of the IL-21 and IL-23 pathways. W. Strober. 2006. T-bet regulates Th1 responses through essential effects on Nat. Immunol. 8: 967–974. GATA-3 function rather than on IFNG gene acetylation and transcription. J. Exp. 8. Ivanov, I. I., B. S. McKenzie, L. Zhou, C. E. Tadokoro, A. Lepelley, J. J. Lafaille, Med. 203: 755–766. D. J. Cua, and D. R. Littman. 2006. The orphan nuclear receptor RORgammat 31. Hosoya, T., T. Kuroha, T. Moriguchi, D. Cummings, I. Maillard, K. C. Lim, and directs the differentiation program of proinflammatory IL-17+ T helper cells. J. D. Engel. 2009. GATA-3 is required for early T lineage progenitor de- Cell 126: 1121–1133. velopment. J. Exp. Med. 206: 2987–3000. 9. Yang, X. O., B. P. Pappu, R. Nurieva, A. Akimzhanov, H. S. Kang, Y. Chung, 32. Chen, Z., A. Laurence, Y. Kanno, M. Pacher-Zavisin, B.-M. Zhu, C. Tato, L. Ma, B. Shah, A. D. Panopoulos, K. S. Schluns, et al. 2008. T helper 17 lineage differentiation is programmed by orphan nuclear receptors ROR alpha and ROR A. Yoshimura, L. Hennighausen, and J. J. O’Shea. 2006. Selective regulatory gamma. Immunity 28: 29–39. function of Socs3 in the formation of IL-17-secreting T cells. Proc. Natl. Acad. Downloaded from 10. Yang, X. O., A. D. Panopoulos, R. Nurieva, S. H. Chang, D. Wang, Sci. USA 103: 8137–8142. S. S. Watowich, and C. Dong. 2007. STAT3 regulates cytokine-mediated gen- 33. Mathur, A. N., H.-C. Chang, D. G. Zisoulis, G. L. Stritesky, Q. Yu, J. T. O’Malley, eration of inflammatory helper T cells. J. Biol. Chem. 282: 9358–9363. R. Kapur, D. E. Levy, G. S. Kansas, and M. H. Kaplan. 2007. Stat3 and Stat4 11. Bru¨stle, A., S. Heink, M. Huber, C. Rosenpla¨nter, C. Stadelmann, P. Yu, direct development of IL-17-secreting Th cells. J. Immunol. 178: 4901–4907. E. Arpaia, T. W. Mak, T. Kamradt, and M. Lohoff. 2007. The development of 34. Nishihara, M., H. Ogura, N. Ueda, M. Tsuruoka, C. Kitabayashi, F. Tsuji, inflammatory T(H)-17 cells requires interferon-regulatory factor 4. Nat. Immu- H. Aono, K. Ishihara, E. Huseby, U. A. K. Betz, et al. 2007. IL-6-gp130-STAT3 nol. 8: 958–966. in T cells directs the development of IL-17+ Th with a minimum effect on that of Treg in the steady state. Int. Immunol. 19: 695–702. 12. Harrington, L. E., R. D. Hatton, P. R. Mangan, H. Turner, T. L. Murphy, http://www.jimmunol.org/ K. M. Murphy, and C. T. Weaver. 2005. Interleukin 17-producing CD4+ effector 35. Mukasa, R., A. Balasubramani, Y. K. Lee, S. K. Whitley, B. T. Weaver, T cells develop via a lineage distinct from the T helper type 1 and 2 lineages. Y. Shibata, G. E. Crawford, R. D. Hatton, and C. T. Weaver. 2010. Epigenetic Nat. Immunol. 6: 1123–1132. instability of cytokine and transcription factor gene loci underlies plasticity of 13. Park, H., Z. Li, X. O. Yang, S. H. Chang, R. Nurieva, Y.-H. Wang, Y. Wang, the T helper 17 cell lineage. Immunity 32: 616–627. L. Hood, Z. Zhu, Q. Tian, and C. Dong. 2005. A distinct lineage of CD4 T cells 36. Huang, H., and W. E. Paul. 1998. Impaired interleukin 4 signaling in T helper regulates tissue inflammation by producing interleukin 17. Nat. Immunol. 6: type 1 cells. J. Exp. Med. 187: 1305–1313. 1133–1141. 37. Huang, Z., J. Xin, J. Coleman, and H. Huang. 2005. IFN-gamma suppresses 14. Newcomb, D. C., W. Zhou, M. L. Moore, K. Goleniewska, G. K. K. Hershey, STAT6 phosphorylation by inhibiting its recruitment to the IL-4 receptor. J. J. K. Kolls, and R. S. Peebles, Jr. 2009. A functional IL-13 receptor is expressed Immunol. 174: 1332–1337. on polarized murine CD4+ Th17 cells and IL-13 signaling attenuates Th17 38. Sarkar, S., L. A. Tesmer, V. Hindnavis, J. L. Endres, and D. A. Fox. 2007. cytokine production. J. Immunol. 182: 5317–5321. Interleukin-17 as a molecular target in immune-mediated arthritis: immunoreg-

15. El-behi, M., B. Ciric, S. Yu, G.-X. Zhang, D. C. Fitzgerald, and A. Rostami. ulatory properties of genetically modified murine dendritic cells that secrete by guest on September 30, 2021 2009. Differential effect of IL-27 on developing versus committed Th17 cells. J. interleukin-4. Arthritis Rheum. 56: 89–100. Immunol. 183: 4957–4967. 39. Morita, Y., J. Yang, R. Gupta, K. Shimizu, E. A. Shelden, J. Endres, J. J. Mule´, 16. Ivanov, I. I., L. Zhou, and D. R. Littman. 2007. Transcriptional regulation of K. T. McDonagh, and D. A. Fox. 2001. Dendritic cells genetically engineered to Th17 cell differentiation. Semin. Immunol. 19: 409–417. express IL-4 inhibit murine collagen-induced arthritis. J. Clin. Invest. 107: 1275– 17. Lee, Y. K., H. Turner, C. L. Maynard, J. R. Oliver, D. Chen, C. O. Elson, and 1284. C. T. Weaver. 2009. Late developmental plasticity in the T helper 17 lineage. 40. Hirota, K., H. Yoshitomi, M. Hashimoto, S. Maeda, S. Teradaira, N. Sugimoto, Immunity 30: 92–107. T. Yamaguchi, T. Nomura, H. Ito, T. Nakamura, et al. 2007. Preferential re- 18. Nurieva, R., X. O. Yang, Y. Chung, and C. Dong. 2009. Cutting edge: in vitro cruitment of CCR6-expressing Th17 cells to inflamed joints via CCL20 in generated Th17 cells maintain their cytokine expression program in normal but rheumatoid arthritis and its animal model. J. Exp. Med. 204: 2803–2812. not lymphopenic hosts. J. Immunol. 182: 2565–2568. 41. Zhou, L., M. M. W. Chong, and D. R. Littman. 2009. Plasticity of CD4+ T cell 19. Lexberg, M. H., A. Taubner, A. Fo¨rster, I. Albrecht, A. Richter, T. Kamradt, lineage differentiation. Immunity 30: 646–655. A. Radbruch, and H.-D. Chang. 2008. Th memory for interleukin-17 expression 42. Seki, Y., K. Hayashi, A. Matsumoto, N. Seki, J. Tsukada, J. Ransom, T. Naka, is stable in vivo. Eur. J. Immunol. 38: 2654–2664. T. Kishimoto, A. Yoshimura, and M. Kubo. 2002. Expression of the suppressor 20. McKenzie, B. S., R. A. Kastelein, and D. J. Cua. 2006. Understanding the IL-23- of cytokine signaling-5 (SOCS5) negatively regulates IL-4-dependent STAT6 IL-17 immune pathway. Trends Immunol. 27: 17–23. activation and Th2 differentiation. Proc. Natl. Acad. Sci. USA 99: 13003–13008. 21. Annunziato, F., L. Cosmi, V. Santarlasci, L. Maggi, F. Liotta, B. Mazzinghi, 43. Yu, C.-R., R. M. Mahdi, S. Ebong, B. P. Vistica, J. Chen, Y. Guo, I. Gery, and E. Parente, L. Filı`, S. Ferri, F. Frosali, et al. 2007. Phenotypic and functional C. E. Egwuagu. 2004. Cell proliferation and STAT6 pathways are negatively features of human Th17 cells. J. Exp. Med. 204: 1849–1861. 22. Boniface, K., W. M. Blumenschein, K. Brovont-Porth, M. J. McGeachy, regulated in T cells by STAT1 and suppressors of cytokine signaling. J. Immunol. B. Basham, B. Desai, R. Pierce, T. K. McClanahan, S. Sadekova, and R. de Waal 173: 737–746. Malefyt. 2010. Human Th17 cells comprise heterogeneous subsets including 44. Cornish, A. L., M. M. Chong, G. M. Davey, R. Darwiche, N. A. Nicola, IFN-gamma-producing cells with distinct properties from the Th1 lineage. J. D. J. Hilton, T. W. Kay, R. Starr, and W. S. Alexander. 2003. Suppressor of Immunol. 185: 679–687. cytokine signaling-1 regulates signaling in response to interleukin-2 and other 23. Raymond, M., V. Q. Van, K. Wakahara, M. Rubio, and M. Sarfati. 2011. Lung gamma c-dependent cytokines in peripheral T cells. J. Biol. Chem. 278: 22755– dendritic cells induce T(H)17 cells that produce T(H)2 cytokines, express 22761. GATA-3, and promote airway inflammation. J. Allergy Clin. Immunol. 128: 192– 45. Alexander, W. S., R. Starr, J. E. Fenner, C. L. Scott, E. Handman, N. S. Sprigg, 201. J. E. Corbin, A. L. Cornish, R. Darwiche, C. M. Owczarek, et al. 1999. SOCS1 is 24. Mangan, P. R., L. E. Harrington, D. B. O’Quinn, W. S. Helms, D. C. Bullard, a critical inhibitor of interferon gamma signaling and prevents the potentially C. O. Elson, R. D. Hatton, S. M. Wahl, T. R. Schoeb, and C. T. Weaver. 2006. fatal neonatal actions of this cytokine. Cell 98: 597–608. Transforming growth factor-beta induces development of the T(H)17 lineage. 46. Sarkar, S., L. A. Cooney, P. White, D. B. Dunlop, J. Endres, J. M. Jorns, Nature 441: 231–234. M. J. Wasco, and D. A. Fox. 2009. Regulation of pathogenic IL-17 responses in 25. Lischke, A., R. Moriggl, S. Bra¨ndlein, S. Berchtold, W. Kammer, W. Sebald, collagen-induced arthritis: roles of endogenous interferon-gamma and IL-4. B. Groner, X. Liu, L. Hennighausen, and K. Friedrich. 1998. The interleukin-4 Arthritis Res. Ther. 11: R158.