STAT6-Dependent Regulation of Th9 Development Ritobrata Goswami, Rukhsana Jabeen, Ryoji Yagi, Duy Pham, Jinfang Zhu, Shreevrat Goenka and Mark H. Kaplan This information is current as of September 23, 2021. J Immunol 2012; 188:968-975; Prepublished online 16 December 2011; doi: 10.4049/jimmunol.1102840 http://www.jimmunol.org/content/188/3/968 Downloaded from

References This article cites 39 articles, 11 of which you can access for free at: http://www.jimmunol.org/content/188/3/968.full#ref-list-1 http://www.jimmunol.org/ Why The JI? Submit online.

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

• No Triage! Every submission reviewed by practicing scientists

• Fast Publication! 4 weeks from acceptance to publication

by guest on September 23, 2021 *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 All rights reserved. Print ISSN: 0022-1767 Online ISSN: 1550-6606. The Journal of Immunology

STAT6-Dependent Regulation of Th9 Development

Ritobrata Goswami,*,†,1 Rukhsana Jabeen,*,1 Ryoji Yagi,‡ Duy Pham,*,† Jinfang Zhu,‡ Shreevrat Goenka,*,† and Mark H. Kaplan*,†

Th cell effector subsets develop in response to specific environments. The development of a particular cytokine-secreting pattern requires an integration of signals that may promote the development of opposing pathways. A recent example of this par- adigm is the IL-9–secreting Th9 cell that develops in response to TGF-b and IL-4, that, in isolation, promote the development of inducible regulatory T cells and Th2 cells, respectively. To determine how the balance of these factors results in priming for IL-9 secretion, we examined the effects of each pathway on transcription factors that regulate Th cell differentiation. We demonstrated that TGF-b induces the PU.1-encoding Sfpi1 locus and that this is independent of IL-4–induced STAT6 activation. IL-4–activated STAT6 is required for repressing the expression of T-bet and Foxp3 in Th9 cells, transcription factors that inhibit IL-9 production, and STAT6 is required for the induction of IRF4, which promotes Th9 development. These data established a network that regulates IL-9 and demonstrated how combinations of cytokine signals generate Downloaded from cytokine-secreting potential by altering the expression of a panel of transcription factors. The Journal of Immunology, 2012, 188: 968–975.

D4 Th cells acquire specific patterns of cytokine secretion cells, but increased production of IL-9. However, it is still unclear in response to environmental cues. Although many fac- how the integration of each signal results in the unique Th9 http://www.jimmunol.org/ C tors, including Ag dose and costimulation, impact the phenotype. differentiation of Th subsets, cytokines remain among the most Although the development of Th9 cells has not been studied as potent agents for instructing the development of cytokine-secreting extensively as other Th subsets, several transcription factors are potential. The paradigm was supported by data showing that the known to be required for Th9 development, including PU.1 and cytokines IL-12 and IL-4 promote the development of Th1 and Th2 IRF4 (8, 9). It is not known whether induction of these factors is in cells, respectively (1). However, T cells, particularly in vivo, rarely response to specific cytokines. Moreover, as Th subsets differen- exist in an environment where they are exposed to only one cy- tiate, the expression of other lineage-associated factors may in- tokine. As the IL-17–secreting subset Th17 was described, it be- terfere with the production of cytokines characteristic of other came clearer that input from a variety of cytokines, including lineages (10, 11). These relationships have not been examined in by guest on September 23, 2021 TGF-b, IL-6, IL-21, IL-1, and IL-23, was required for optimal Th9 cells. In this article, we document the ability of transcription differentiation (2). For this subset, the cell fate decision of the factors regulated by TGF-b– and IL-4–induced signaling to con- responding T cells relies upon integration of a balance of signals. tribute to the development of IL-9–secreting T cells. The most recent example of a Th subset that requires multiple balanced signals to develop is the Th9 cell that secretes IL-9, Materials and Methods a cytokine important for inflammatory disease (3, 4). Th9 cells Mice develop following exposure to TGF-b and IL-4 (5–7). Although Wild-type (WT) female mice (C57BL/6 and BALB/c) were purchased from TGF-b alone promotes differentiation of T regulatory cells Harlan Biosciences (Indianapolis, IN). Mice with conditional Stat3 deletion, (Tregs), and IL-4 stimulates Th2 development, the integration of which were provided by Dr. David Levy (New York University, New York, both signals results in a Th subset that has lower Foxp3 expression NY) and described previously (12), were mated to Cd4-cre mice (13). fl/fl than Treg cultures and lower Th2 cytokine production than Th2 Runx3 mice were mated to dLck-Cre mice (14). Stat6-deficient mice and Parp14-deficient mice were described previously (15, 16) and are on a BALB/c and C57BL/6 background, respectively. Il4-deficient mice were *Department of Pediatrics, Herman B. Wells Center for Pediatric Research, Indiana purchased from Jackson Laboratories and were bred in the Indiana Uni- University School of Medicine, Indianapolis, IN 46202; †Department of Microbiol- versity animal facility. Mice were maintained in pathogen-free conditions, ogy and Immunology, Indiana University School of Medicine, Indianapolis, IN and studies were approved by the Indiana University Institutional Animal 46202; and ‡Laboratory of Immunology, National Institute of Allergy and Infectious Care and Use Committee. Diseases, National Institutes of Health, Bethesda, MD 20892 1R.G. and R.J. contributed equally to this work. Differentiation of murine T cells Received for publication September 30, 2011. Accepted for publication November Naive CD4+CD62L+ T cells were purified from spleens and lymph nodes 17, 2011. by magnetic selection (Miltenyi Biotec). Naive CD4+ T cells (1 3 106 This work was supported by Public Health Service Grants AI057459 (to M.H.K.) and cells/ml complete RPMI 1640 medium) were activated with plate-bound HL093105 (to S.G.). Y.R. and J.Z. were supported by the Division of Intramural anti-CD3 (2 mg/ml; 145-2C11; BioXcell) and soluble anti-CD28 (1 mg/ml; Research of the National Institute of Allergy and Infectious Diseases, the National 37.51; BD Biosciences) and cultured under Th9 conditions (IL-4 [20 ng/ Institutes of Health. ml; PeproTech], TGF-b [2 ng/ml; R&D Systems], and anti–IFN-g [10 mg/ Address correspondence and reprint requests to Dr. Mark H. Kaplan, Herman B. ml; XMG; BioXcell]); Th2 conditions (IL-4 and anti–IFN-g); Th1 con- Wells Center for Pediatric Research, Indiana University School of Medicine, 1044 ditions (IL-12 [5 ng/ml; R&D Systems], IL-2 [50 U/ml; PeproTech], and West Walnut Street, Room 202, Indianapolis, IN 46202. E-mail address: mkaplan2@ anti–IL-4 [10 mg/ml; 11B11; BioXcell]); Th17 conditions (TGF-b, IL-6 iupui.edu [100 ng/ml; R&D Systems], IL-1b [10 ng/ml; eBioscience], IL-23 [10 ng/ Abbreviations used in this article: ChIP, chromatin immunoprecipitation; CNS, con- ml; R&D Systems], anti–IFN-g, and anti–IL-4); and Treg conditions served noncoding sequence; qPCR, quantitative RT-PCR; Treg, regulatory ; (TGF-b and anti–IL-4). After 3 d, cultures were expanded with fresh WT, wild-type. complete RPMI 1640 medium with IL-4 and TGF-b added to the Th9 www.jimmunol.org/cgi/doi/10.4049/jimmunol.1102840 The Journal of Immunology 969 cells; half the dose of IL-1b, IL-23, and IL-6 added to the Th17 cells; and was precipitated with A-agarose beads at 4˚C overnight. The su- IL-2 added to Tregs. After 5 d of differentiation, cells were stimulated with pernatant was incubated in the presence of rabbit polyclonal Abs (anti- plate-bound anti-CD3 (4 mg/ml) for 1 d. Cell-free supernatant was col- PU.1 [Santa Cruz Biotechnology]) and rabbit IgG (Millipore). DNA cross- lected, and the amount of IL-9 was assessed by ELISA using anti–IL-9 links were reversed by incubating precipitates overnight at 65˚C, and DNA capture Ab (D8402E8; BD Biosciences) and biotin-labeled anti–IL-9 was purified by phenol-chloroform extraction and ethanol precipitation. (D9302C12; BioLegend) as the secondary Ab. DNA quantification was performed with SYBR Green Fast PCR Master Mix via an ABI 7500 Fast Real-time PCR system. A standard curve was Intracellular cytokine staining generated from serial dilutions of input DNA to quantify immunopreci- After 5 d of culture, cells were stimulated with PMA (phorbol 12-myristate pitated DNA. The amount of immunoprecipitated DNA from the IgG control was subtracted from the amount of immunoprecipitated DNA from 13-acetate) and ionomycin for 5 h. Monensin was added to the cells for the specific Ab ChIP and normalized against the amount of input DNA to the last 3 h of stimulation. Cells were fixed with paraformaldehyde after stimulation and permeabilized with saponin. Cells were then stained with calculate percentage input. fluorochrome-conjugated anti-mouse IL-9 (RM9A4; BioLegend), anti- Immunoblot mouse IL-4 (11B11; BioLegend), and anti–IL-17A (eBio17B7; eBio- science). Foxp3 staining was performed according to the manufacturer’s Total cell lysates were prepared from Th9, Th2, or Th1 cells after 5 d of instructions (FJK-16s; eBioscience). For intracellular phospho-STAT differentiation. The lysates were resuspended in SDS-PAGE loading buffer, staining, Th2 and Th9 cells were harvested each day during the 5-d dif- boiled at 100˚C for 5 min, and run on SDS-PAGE gels (Invitrogen Life ferentiation, fixed with 1.5% paraformaldehyde for 10 min at room tem- Technologies) before immunoblot. perature, and permeabilized for 10 min with 100% methanol at 4˚C. Cells were then stained for phospho-STAT3 and phospho-STAT6 (BD Phar- mingen) after 30 min at room temperature. Cells were analyzed by flow Results cytometry with FACSCalibur (Beckton Dickinson). Differential requirement for STAT in Th2 and Th9 cells Downloaded from Retroviral transduction In addition to the requirement for STAT6 in the development of Th2 and Th9 cells (5, 7, 16, 20, 21), we recently showed that Bicistronic retroviral vectors encoding mouse GATA3 and human CD4, T-bet, Foxp3, JunB, or c-Maf (kindly provided by I.C. Ho, Brigham and STAT3 is also required for the development of Th2 cells (13). To Women’s Hospital, Boston, MA) and enhanced GFP, or Runx3 and Thy1.1 compare the phosphorylation of STAT3 and STAT6 during the were used to generate virus, as described previously (17, 18). Naive CD4+ development of Th2 and Th9 cells, we cultured naive CD4+ T cells from WT and/or Il4-deficient and Stat6-deficient mice were cul-

T cells under Th2 or Th9 conditions for 5 d and analyzed STAT http://www.jimmunol.org/ tured under Th9 cell conditions; after 2 d of differentiation, Th cells were phosphorylation each day during culture. The percentages of cells transduced with retroviral supernatant containing polybrene. On day 5, + transduced cells (eGFP+, Thy1.1+, or hCD4+) were analyzed by flow that were phospho-STAT6 , as well as the intensity of phospho- cytometry for IL-9, IL-4, and IL-17A secretion. STAT6 staining, were identical between Th2 and Th9 cells (Fig. 1A). STAT3 was also phosphorylated during Th9 differentiation, Quantitative RT-PCR although with slightly lower percentages of cells and mean fluo- Total RNA was isolated from unstimulated or stimulated (plate-bound anti- rescence intensity than in Th2 cells (Fig. 1A). To define the role CD3; 2 mg/ml) cells using TRIzol and reverse transcribed, according to the for each STAT protein in the development of IL-9–secreting manufacturer’s instructions (Invitrogen Life Technologies). For quantita- tive RT-PCR (qPCR), TaqMan Fast Universal Master Mix and commer- T cells, we differentiated naive T cells from mice with conditional CD42/2 cially available primers were used for Gata-3, Maf, T-bet, Runx3, Foxp3, deletion of Stat3 in T cells (Stat3 ) and littermate controls by guest on September 23, 2021 2/2 and Il9, and RNA expression was normalized to the expression of b2- or from WT and Stat6 mice. Consistent with previous obser- microglobulin. The relative expression was calculated by the change-in- vations, IL-9 production was absent in cells lacking expression of threshold (2DDC ) method. T STAT6 (Fig. 1B,1C). In contrast, and despite the activation of Chromatin immunoprecipitation STAT3 throughout Th9 differentiation, STAT3 was not required Chromatin immunoprecipitation (ChIP) assay was performed, as previously for the development of IL-9–secreting cells (Fig. 1B,1C). described (19), with minor modifications. Cells were kept unstimulated We next examined the requirement for STAT6 cofactors in the before cross-linking of protein–chromatin complexes. Immunocomplex development of Th9 cells. PARP-14, previously identified as

FIGURE 1. Differential requirement for STAT proteins in Th2 and Th9 cells. A, Naive CD4+ T cells from WT mice were activated with anti-CD3 and anti-CD28 and cultured with IL-4 and anti–IFN-g (Th2 conditions) or with IL-4, TGF-b, and anti–IFN-g (Th9 conditions). Each day during differentiation, cells were stained for intracellular phospho-STAT3 and phospho-STAT6. The graphs represent percen- tages of cells (upper two panels) and mean fluores- cent intensity (MFI; lower two panels). Data are average 6 SEM of six mice from three experiments. B and C, Naive CD4+ T cells from WT and Stat62/2 (top row), WT and Stat3CD42/2 (middle row), and WT and Parp142/2 (bottom row) mice were cultured under Th9 conditions. After 5 d, cells were harvested and stimulated with PMA and ionomycin before in- tracellular staining for cytokines IL-9 and IL-4 was performed (B) or were stimulated with anti-CD3 and, 24 h later, cell-free supernatant was collected and IL-9 production was assessed by ELISA (C). Data are representative of three experiments with similar results. *p , 0.05, ***p , 0.001. 970 STAT6 TARGETS IN Th9 CELLS Downloaded from http://www.jimmunol.org/

FIGURE 2. Th2 transcription factors are expressed in Th9 cells. A, WT naive CD4+ T cells were activated with anti-CD3 and anti-CD28 and cultured under Th1, Th2, Th17, Th9, or Treg conditions or directly ex vivo (naive) for 5 d. Cell pellets were collected, RNA was isolated, and qPCR was performed + for the indicated transcription factors. B, Naive CD4 T cells from WT mice were differentiated under Th2 (left panel) or Th9 (right panel) cell conditions by guest on September 23, 2021 for 5 d. Cells were either kept unstimulated or stimulated for 6 h before intracellular staining for GATA3 was performed. Data are representative of three experiments with similar results. C, Naive CD4+ T cells from WT mice were differentiated under Th2 or Th9 cell conditions. RNA was isolated from cells on each day during differentiation and examined for the expression of the indicated genes. D, WT and Stat62/2 naive CD4+ T cells were cultured under Th9 cell conditions for 5 d, RNA was isolated, and qPCR was performed for the indicated genes. Data are representative of three independent experiments with two or three mice in each group. E, Naive CD4+ T cells from WT and Stat62/2 mice were differentiated under Th2 (left panel) or Th9 (right panel) cell conditions for 5 d. Cells were then stimulated for 6 h before intracellular staining for GATA3 was performed. a collaborator of STAT6 (22, 23), is a poly-ADP ribosyl poly- ment, we next compared the differentiation of WT and Parp142/2 merase. PARP-14 interacts with STAT6 target genes and increases Th9 cells. Consistent with the role of PARP-14 as a coactivator, STAT6-dependent by facilitating the clearance of we observed diminished production of IL-9 in the absence of repressive molecules from gene promoters and increasing binding PARP-14 (Fig. 1B,1C). Together, these data suggested that al- of STAT6 (24). Because STAT6 was required for Th9 develop- though STAT6 is required for both Th2 and Th9 development,

FIGURE 3. Transcription factor binding to the Il9 locus. A, Schematic diagram of the Il9 gene, with CNS used for ChIP analysis indicated. B, Naive CD4+ T cells from WT mice were activated with anti-CD3 and anti-CD28 and cultured under Th2 and Th9 cell conditions for 5 d. ChIP was performed for STAT6 before qPCR for two Il9 CNS sites, or for the Irf4 or Maf promoters, to determine the amount of Stat6 binding. Data are average 6 SD of four mice from two experiments. C, Naive CD4+ T cells from WT and Stat62/2 mice were cultured under Th9 conditions for 5 d. ChIP was performed for PU.1 (top panel) and IRF4 (bottom panel) before qPCR for two Il9 CNS sites. Data are average 6 SD of six to eight mice from three or four experiments. The Journal of Immunology 971

STAT3, although activated in both developing subsets, is only cultures (Fig. 3C). This suggested that IRF4 could be a relevant required for Th2 development. transcription factor downstream of IL-4/STAT6. However, trans- duction of IRF4 into Stat62/2 Th9 cultures did not rescue IL-9 Th2 transcription factors repress IL-9 in Th9 cultures production (data not shown), suggesting that it is not sufficient to Because STAT6 was required for the development of Th2 and Th9 induce Il9 in the absence of STAT6. cells, we examined the expression of transcription factors down- To determine whether either GATA3 or c-Maf contributed to stream of STAT6 in Th2 cells for a role in Th9 cells. GATA3 was IL-9 production, we transduced differentiating Th9 cells with shown to be necessary for the development of IL-9–secreting bicistronic retroviruses and examined the effects on IL-9 produc- T cells, but it also had low expression in Th9 cultures (5, 7). tion using flow cytometry. Transduction of GATA3 resulted in a Expression of other Th2 transcription factors has not been care- decrease in IL-9–producing cells compared with Th9 cultures fully examined. To explore the role of these factors further, we transduced with control retrovirus (Fig. 4A). We observed a simi- tested the expression of Gata3, Maf, and Junb in polarized Th cell lar result when we transduced Th9 cultures with a c-Maf– cultures by qPCR. Gata3 was expressed in the highest amount in Th2 cells, as has been well documented (25, 26), although it was also detected in Th9 cultures (Fig. 2A). Because the expression of Gata3 in Th9 cultures was previously described to be low, we confirmed our results by analyzing expression of GATA3 protein using intracellular staining for flow cytometry. We observed that

Th2 and Th9 cultures had similar levels of GATA3 protein before Downloaded from stimulation with anti-CD3 (Fig. 2B). However, GATA3 protein doubled following stimulation of Th2 cultures but was unaffected by stimulation of Th9 cultures (Fig. 2B). Maf, which is important for the production of IL-4 and IL-21 in Th2 and Th17 cells (27– 29), was expressed at the highest levels in Th9 cultures assessed

by qPCR (Fig. 2A). Junb, which also contributes to IL-4 pro- http://www.jimmunol.org/ duction (30, 31), did not show differential expression among Th subsets, although it was enriched in Treg cultures (data not shown). We examined the induction of expression of Gata3, Maf, and Irf4, the latter a factor previously documented to be expressed in Th9 cells and to regulate IL-9 (9), for differences in kinetics during the differentiation of Th2 and Th9 cells. We observed similar patterns of induction during differentiation, with greater induction of Gata3 and Maf in Th2 and Th9 cells, respectively, toward the end of differentiation (Fig. 2A,2C). Irf4 also showed by guest on September 23, 2021 a similar pattern of induction in Th2 and Th9 cells, although ex- pression was higher in Th9 cells throughout differentiation (Fig. 2C). To determine whether these factors might account for the IL-4/STAT6–dependent induction of IL-9, we examined the ex- pression of each of these genes, as well as Sfpi1, the gene en- coding PU.1, a transcription factor that we recently defined as promoting Th9 differentiation (8), in WT and Stat62/2 Th9 cul- tures. We observed that, although expression of Sfpi1 was not dependent upon STAT6, expression of Irf4, Gata3 and Maf,as well as GATA3 protein expression, was decreased in the absence of STAT6 (Fig. 2D,2E). We then determined whether STAT6 could be regulating IL-9 production by binding directly to the Il9 gene or whether it af- fected the binding of other factors to the Il9 gene. We performed ChIP experiments, testing the association of STAT6 to two con- served noncoding sequences (CNS) upstream of the Il9 locus (32). Although STAT6 binding could be detected to CNS1 (Il9 pro- moter) more than CNS0, the percent input values were very low (Fig. 3A,3B), and ,5% of the STAT6 binding detected to the Maf and Irf4 promoters in Th2 and Th9 cells (Fig. 3B). In agreement with this finding, STAT6 binding was not detected in the Il9 locus in Th2 cells analyzed using published ChIP-sequencing datasets (33). These data suggested that Il9 is not a major binding target of FIGURE 4. Th2 transcription factors repress IL-9 in Th9 cultures. WT 2/2 + STAT6. and Il4 naive CD4 T cells were activated with anti-CD3 and anti- CD28 and cultured under Th9 cell conditions for 48 h before being Because PU.1 and IRF4 were shown to regulate Il9 directly transduced with control or GATA3-expressing retroviruses (A), control or (8, 9), we tested whether STAT6 deficiency affected the binding c-Maf–expressing retroviruses (B), or control or JunB-expressing retro- of either factor to Il9 regulatory elements. We demonstrated that viruses (C). After 5 d in culture, cells were stimulated with PMA and binding of PU.1 to the Il9 promoter (CNS1) was indistinguishable ionomycin for intracellular staining of IL-9 and IL-4. Data are represen- 2/2 between WT and Stat6 Th9 cultures (Fig. 3C). In contrast, tative of two or three independent experiments with similar results. *p , 2 2 IRF4 binding to the Il9 gene was decreased in Stat6 / Th9 0.05, **p , 0.01. 972 STAT6 TARGETS IN Th9 CELLS expressing retrovirus (Fig. 4B). Transduction with a JunB- percentages of IL-9–secreting T cells compared with control expressing retrovirus did not affect IL-9 production (Fig. 4C). cultures (Fig. 5D). To determine whether STAT6 regulates the We speculated that because both GATA3 and c-Maf can induce the expression of either of these factors in Th9 cells, we tested mRNA production of IL-4, these factors might be inducing IL-4–secreting from WT and Stat62/2 Th9 cultures. Runx3 expression was de- cells in Th9 cultures at the expense of IL-9–secreting T cells. creased in the absence of STAT6; in contrast, T-bet expression was However, neither GATA3 nor c-Maf transduction increased the increased in the absence of STAT6 (Fig. 5E). Together, these data percentages of IL-4–secreting cells (Fig. 4A,4B). It was also suggested that both T-bet and Runx3 can limit IL-9 production and possible that increases in endogenous IL-4 during the culture period that the STAT6 signal represses T-bet expression in Th9 cells. were affecting the differentiation of IL-9–secreting T cells. To test A balance of IL-4 and TGF-b signals is required for the this, we performed transductions into differentiating Il42/2 Th9 development of IL-9–secreting T cells cultures. Differentiation of IL-9–secreting T cells, in the absence of endogenous IL-4, was normal (Fig. 4A,4B). Moreover, transduc- Although STAT6 was not required for expression of Sfpi1 in Th9 tion of GATA3 or c-Maf into Il42/2 Th9 cultures was as effective at cells, it is required to modify TGF-b–induced Treg development, reducing IL-9 production as was transduction into WT cells (Fig. because IL-9–secreting T cells develop from a balance between 4A,4B). Thus, although GATA3 and Maf are expressed in Th9 TGF-b– and IL-4–induced signals (5, 7). When the dose of IL-4 cells, ectopic expression actually decreased IL-9 production was held constant at 0 ng/ml (Treg condition) or 10 ng/ml (Th9 through effects that are independent from the induction of IL-4. condition), increasing doses of TGF-b resulted in increased pro- duction of Foxp3+ cells, attenuated by the presence of IL-4 in Th9 Repression of IL-9 by Th1-associated transcription factors conditions, and increasing amounts of IL-9 when IL-4 was present Downloaded from Because IL-4 promotes Th2 differentiation, it interferes with the (Fig. 6A). Although a small percentage of dim IL-9+ cells was expression and function of other transcription factors, such as T-bet observed in Treg conditions, there was no IL-9 detectable from and Runx3, which promote IFN-g expression and Th1 differen- Treg cultures using ELISA (Fig. 6A). There was a similar increase tiation (34, 35). Although T-bet is predominantly expressed in Th1 in the expression of Sfpi1 in response to TGF-b, with slightly cells, with expression in Th9 cells similar to other lineages, Runx3 higher expression under Treg conditions than under Th9 con-

is expressed in most Th lineages, with the highest expression in ditions, consistent with our previous work (Fig. 6B) (8). In con- http://www.jimmunol.org/ Th1 cells but detectable expression in Th9 cells (Fig. 5A,5B). To trast, with a constant dose of TGF-b, altering the dose of IL-4 determine whether T-bet and Runx3 limit Th9 development, we did not affect Sfpi1 expression (Fig. 6B), consistent with STAT6- transduced differentiating Th9 cells with retroviruses expressing independent expression (Fig. 2D). These results suggested that each factor. Ectopic expression of either T-bet or Runx3 decreased IL-9–secreting cells arise when there is a balance between the the percentages of IL-9–secreting T cells (Fig. 5C). Consistent TGF-b signal that induces Sfpi1 expression and the IL-4 signal with this observation, differentiation of Th9 cells using naive that limits Foxp3 expression. This is consistent with the require- CD4+ T cells deficient in T-bet or Runx3 resulted in increased ment for STAT6 in the ability of IL-4 to limit Foxp3 expression by guest on September 23, 2021

FIGURE 5. IL-9 is repressed by Th1-associated transcription factors. A, Naive CD4+ T cells from WT mice were analyzed directly ex vivo (naive) or cultured under Th1, Th2, Th17, Th9, and Treg conditions for 5 d. After RNA isolation, qPCR was performed for the indicated genes. Data are representative of two experiments with similar results. B, Total-cell lysates were prepared from Th1, Th2, and Th9 cells, and immunoblot was performed for Runx3; b-actin was used as loading control. C, WT naive CD4+ T cells were activated with anti-CD3 and anti-CD28 and cultured under Th9 cell conditions for 48 h before being transduced with control or Runx3-expressing retroviruses and control or T-bet–expressing retroviruses. Cells were harvested after 5 din culture and stimulated with PMA and ionomycin before intracellular IL-9 and IL-4 staining. Data are average 6 SD of two experiments with four mice. D, Naive CD4+ T cells from WT and Tbx212/2 (left two panels) or WT and Runx3fl/fl–dLck–Cre (right two panels) mice were cultured in Th9 cell conditions. After 5 d, cells were harvested and stimulated with PMA and ionomycin before intracellular IL-9 and IL-4 staining. Data are representative of two or three experiments with similar results. E, WT and Stat62/2 naive CD4+ T cells were cultured under Th9 cell conditions for 5 d. Cells were harvested, and qPCR was performed for Runx3 and Tbx21.*p , 0.05, **p , 0.01. The Journal of Immunology 973 Downloaded from

FIGURE 6. A balance of IL-4 and TGF-b signals is required for the development of IL-9–secreting T cells. A, WT naive CD4+ T cells were activated with anti-CD3 and anti-CD28 and were cultured with increasing doses of TGF-b in the presence (Th9) or absence (Treg) of IL-4 for 5 d. Cells were then stained for intracellular Foxp3 (top panel) and IL-9 (middle panel). Differentiated Th9 cells were stimulated with anti-CD3; 24 h later, IL-9 production was assessed by ELISA (bottom panel). B, Naive CD4+ T cells were cultured under Th9 and Treg conditions, as in A (left panel) or were cultured with 2 ng/ml TGF-b and increasing doses of IL-4 (Th9; right panel) for 5 d. After differentiation, RNA was isolated, and Sfpi1 expression was measured by qPCR. Data http://www.jimmunol.org/ are representative of two experiments with similar results. C, WT and Stat62/2 naive CD4+ T cells were cultured under Th9 cell conditions for 5 d, and Foxp3 expression was measured using qPCR. Data are average 6 SD of four mice from two experiments. D, Naive CD4+ T cells were activated with anti- CD3 and anti-CD28 and cultured under Th9 cell conditions for 48 h before being transduced with control or Foxp3-expressing retroviruses. After 5 d in culture, cells were stimulated with PMA and ionomycin and stained intracellularly for IL-9 and IL-17A. Data are representative of two experiments with similar results. E, Schematic diagram showing the summary of a transcription factor network in Th9 cells. *p , 0.05.

(Fig. 6C) (36, 37) and suggested that Foxp3 may be incompatible (35). This could occur through both direct and indirect mecha- with IL-9 production. To test this, we transduced developing Th9 nisms. In support of a direct effect, in previously generated ChIP- by guest on September 23, 2021 cells with Foxp3-expressing or control retroviruses. We observed sequencing datasets of Runx3/CBFb and T-bet binding in Th1 that ectopic expression of Foxp3 reduced IL-9 production without cells (14), both factors bound directly to the Il9 gene: T-bet at the induction of the Th17 cytokine IL-17A or IL-4 (Fig. 6D, data a major peak around 222.5 kb and smaller peaks between +7 and not shown). Thus, it is likely that an additional role for STAT6 +9 kb and Runx3 at a major peak around +5 kb and minor peaks at during the development of Th9 cells is the repression of Foxp3 225, 217, 27, and +11 kb. They might also have indirect effects expression that would otherwise reduce IL-9 production. by regulating other transcription factors required for IL-9 pro- duction. However, it seems unlikely that Runx3 contributes to the Discussion lack of IL-9 production in Stat62/2 Th9 cultures, because ex- Transcription factors direct the development of Th subsets with pression of Runx3 was decreased in the absence of STAT6 (Fig. distinct cytokine-secreting potentials. A multitude of factors has 5E). Increased expression of Tbx21 in the absence of STAT6 been identified that can promote one lineage while decreasing suggested that the IL-4/STAT6 signal is required to limit T-bet– differentiation of another. The developing Th subset integrates mediated inhibition in Th9 cells. multiple, sometimes conflicting, signals as it adopts a particular Similar to data in Th2 cells, STAT6 is required for the expression effector fate. This is particularly true in the development of IL-9– of Gata3 and Maf in Th9 cells. However, the role of these factors secreting Th9 cells, where the cell integrates the Treg-inducing in Th9 cells remains unclear. GATA3 was required for Th9 de- TGF-b signal and the Th2-inducing IL-4 signal to adopt a phe- velopment, although initial reports noted it was not expressed in notype discrete from either individual signal. Th9 cultures (5, 7). We found Gata3 expressed in Th9 cells, less In this study, we documented specific targets of both Th9- than in Th2 cells but in greater amounts than observed in other Th inducing signals (Fig. 6E). The TGF-b signal, which is well lineages (Fig. 2). Expression of GATA3 in Th9 cells was con- documented to induce Foxp3, also induces expression of Sfpi1 firmed by intracellular staining. However, ectopic expression of (Fig. 6). The IL-4 signal targets multiple genes, including the GATA3 reduced IL-9 production in WT cells and did not induce induction of Irf4, Gata3, and Maf, and is required for the re- IL-9 production when transduced into Stat62/2 Th9 cultures, pression of T-bet and Foxp3. Each of these factors likely plays suggesting that it was not directly regulating the Il9 gene. It is combinatorial roles in the regulation of Il9 gene expression. Foxp3 possible that GATA3 is an intermediate in the IL-4/STAT6–de- represses IL-9 production and would contribute to the lack of pendent reduction of Foxp3, as was demonstrated for GATA3 in appreciable IL-9 production from Tregs, despite their expression Tregs (38). The expression of Maf was higher in Th9 cultures than of Sfpi1. Thus, the IL-4 signal regulates the expression of both in either Th2 or Th17 cells, for which the function of c-Maf has positive and negative regulators of IL-9 production. been previously examined (27–29). c-Maf also repressed IL-9 T-bet and Runx3 repress IL-9 in Th9 cells and likely contribute to production, suggesting that it was not involved in regulation of diminished Il9 expression in Th1 cells, as they do at the Il4 locus the Il9 gene. c-Maf might be involved in the regulation of other 974 STAT6 TARGETS IN Th9 CELLS genes in Th9 cells. We showed that Th9 cells can secrete IL-21 essential for the developmental program of T helper 9 cells. Immunity 33: 192– 202. (39), and because this cytokine is a c-Maf target in other Th lin- 10. Zhou, L., M. M. Chong, and D. R. Littman. 2009. Plasticity of CD4+ T cell eages, there might be a role for c-Maf in promoting IL-21 pro- lineage differentiation. Immunity 30: 646–655. duction in Th9 cells. There could also be kinetic effects of the 11. Zhu, J., and W. E. Paul. 2010. Peripheral CD4+ T-cell differentiation regulated by networks of cytokines and transcription factors. Immunol. Rev. 238: 247–262. expression of each factor during Th cell differentiation that might 12. Chiarle, R., W. J. Simmons, H. Cai, G. Dhall, A. Zamo, R. Raz, J. G. Karras, be obscured by ectopic expression during in vitro culture. Al- D. E. Levy, and G. Inghirami. 2005. Stat3 is required for ALK-mediated lym- though we did not observe significant differences in expression of phomagenesis and provides a possible therapeutic target. Nat. Med. 11: 623–629. 13. Stritesky, G. L., R. Muthukrishnan, S. Sehra, R. Goswami, D. Pham, J. Travers, these factors until the final days of Th2 and Th9 differentiation, it E. T. Nguyen, D. E. Levy, and M. H. Kaplan. 2011. The transcription factor is still possible that a balance of these factors, at particular times STAT3 is required for T helper 2 cell development. Immunity 34: 39–49. during the differentiation process, is critical for appropriate IL-9 14. Yagi, R., I. S. Junttila, G. Wei, J. F. Urban, Jr., K. Zhao, W. E. Paul, and J. Zhu. 2010. The transcription factor GATA3 actively represses RUNX3 protein- production. regulated production of interferon-gamma. Immunity 32: 507–517. IRF4 is one of the factors downstream of IL-4 and STAT6 that 15. Cho, S. H., S. Goenka, T. Henttinen, P. Gudapati, A. Reinikainen, C. M. Eischen, Il9 R. Lahesmaa, and M. Boothby. 2009. PARP-14, a member of the B aggressive promotes IL-9 production. It was previously shown to bind the lymphoma family, transduces survival signals in primary B cells. Blood 113: promoter and is required for the expression of IL-9 (9). We con- 2416–2425. firmed those observations and demonstrated that, in the absence of 16. Kaplan, M. H., U. Schindler, S. T. Smiley, and M. J. Grusby. 1996. Stat6 is required for mediating responses to IL-4 and for development of Th2 cells. STAT6, there is decreased Irf4 expression, as well as decreased Immunity 4: 313–319. binding of IRF4 to the Il9 promoter. However, IRF4 is not the only 17. Chang, H. C., S. Zhang, V. T. Thieu, R. B. Slee, H. A. Bruns, R. N. Laribee, factor “missing” in Stat62/2 Th9 cells. Transduction of IRF4 into M. J. Klemsz, and M. H. Kaplan. 2005. PU.1 expression delineates heterogeneity 2 2 in primary Th2 cells. Immunity 22: 693–703.

/ Downloaded from Stat6 Th9 cultures did not rescue IL-9 production (data not 18. Thieu, V. T., Q. Yu, H. C. Chang, N. Yeh, E. T. Nguyen, S. Sehra, and shown). Moreover, IRF4 transduction did not decrease Foxp3 M. H. Kaplan. 2008. Signal transducer and activator of transcription 4 is required expression and only modestly affected T-bet expression, further for the transcription factor T-bet to promote T helper 1 cell-fate determination. Immunity 29: 679–690. suggesting that additional factors downstream of IL-4 and STAT6 19. Yu, Q., V. T. Thieu, and M. H. Kaplan. 2007. Stat4 limits DNA methyltransferase are required for the development of IL-9–secreting T cells. recruitment and DNA methylation of the IL-18Ralpha gene during Th1 differ- entiation. EMBO J. 26: 2052–2060. The Th9 subset of Th cells was only recently described, and the 20. Shimoda, K., J. van Deursen, M. Y. Sangster, S. R. Sarawar, R. T. Carson, factors that promote its development are still being defined. In R. A. Tripp, C. Chu, F. W. Quelle, T. Nosaka, D. A. A. Vignali, et al. 1996. Lack http://www.jimmunol.org/ this article, we defined a transcription factor network downstream of IL-4-induced Th2 response and IgE class switching in mice with disrupted Stat6 gene. Nature 380: 630–633. of the Th9-inducing cytokines TGF-b and IL-4 (Fig. 6E). TGF-b 21. Takeda, K., T. Tanaka, W. Shi, M. Matsumoto, M. Minami, S.-I. Kashiwamura, induction of Sfpi1 is complemented by the IL-4–dependent re- K. Nakanishi, N. Yoshida, T. Kishimoto, and S. Akira. 1996. Essential role of pression of Foxp3 and T-bet and the induction of IRF4, as well Stat6 in IL-4 signalling. Nature 380: 627–630. 22. Goenka, S., and M. Boothby. 2006. Selective potentiation of Stat-dependent gene as other likely factors that are not yet defined. Further work in expression by collaborator of Stat6 (CoaSt6), a transcriptional cofactor. Proc. defining these factors will add needed insight to the regulation Natl. Acad. Sci. USA 103: 4210–4215. of IL-9. 23. Goenka, S., S. H. Cho, and M. Boothby. 2007. Collaborator of Stat6 (CoaSt6)- associated poly(ADP-ribose) polymerase activity modulates Stat6-dependent gene transcription. J. Biol. Chem. 282: 18732–18739. Acknowledgments 24. Mehrotra, P., J. P. Riley, R. Patel, F. Li, L. Voss, and S. Goenka. 2011. PARP-14 by guest on September 23, 2021 functions as a transcriptional switch for Stat6-dependent gene activation. J. Biol. We thank Dr. David Levy for mice, Drs. Alexander Dent (Indiana Univer- Chem. 286: 1767–1776. sity, Indianapolis, IN) and I. Cheng Ho for providing retroviral vectors, and 25. Zheng, W.-P., and R. A. Flavell. 1997. The transcription factor GATA-3 is Dr. Gretta Stritesky (University of Minnesota, Minneapolis, MN) for re- necessary and sufficient for Th2 cytokine gene expression in CD4 T cells. Cell view of the manuscript. 89: 587–596. 26. Zhang, D. H., L. Cohn, P. Ray, K. Bottomly, and A. Ray. 1997. Transcription factor GATA-3 is differentially expressed in murine Th1 and Th2 cells and Disclosures controls Th2-specific expression of the interleukin-5 gene. J. Biol. Chem. 272: 21597–21603. The authors have no financial conflicts of interest. 27. Bauquet, A. T., H. Jin, A. M. Paterson, M. Mitsdoerffer, I. C. Ho, A. H. Sharpe, and V. K. Kuchroo. 2009. The costimulatory molecule ICOS regulates the ex- pression of c-Maf and IL-21 in the development of follicular T helper cells and References TH-17 cells. Nat. Immunol. 10: 167–175. 28. Ho, I. C., M. R. Hodge, J. W. Rooney, and L. H. Glimcher. 1996. The proto- 1. Zhu, J., H. Yamane, and W. E. Paul. 2010. Differentiation of effector CD4 T cell oncogene c- is responsible for tissue-specific expression of interleukin-4. populations (*). Annu. Rev. Immunol. 28: 445–489. Cell 85: 973–983. 2. Korn, T., E. Bettelli, M. Oukka, and V. K. Kuchroo. 2009. IL-17 and Th17 Cells. 29. Kim, J. I., I. C. Ho, M. J. Grusby, and L. H. Glimcher. 1999. The transcription Annu. Rev. Immunol. 27: 485–517. factor c-Maf controls the production of interleukin-4 but not other Th2 cyto- 3. Goswami, R., and M. H. Kaplan. 2011. A brief history of IL-9. J. Immunol. 186: kines. Immunity 10: 745–751. 3283–3288. 30. Hartenstein, B., S. Teurich, J. Hess, J. Schenkel, M. Schorpp-Kistner, and 4. Noelle, R. J., and E. C. Nowak. 2010. Cellular sources and immune functions of P. Angel. 2002. Th2 cell-specific cytokine expression and allergen-induced air- interleukin-9. Nat. Rev. Immunol. 10: 683–687. way inflammation depend on JunB. EMBO J. 21: 6321–6329. 5. Dardalhon, V., A. Awasthi, H. Kwon, G. Galileos, W. Gao, R. A. Sobel, 31. Li, B., C. Tournier, R. J. Davis, and R. A. Flavell. 1999. Regulation of IL-4 M. Mitsdoerffer, T. B. Strom, W. Elyaman, I. C. Ho, et al. 2008. IL-4 inhibits expression by the transcription factor JunB during differentiation. TGF-beta-induced Foxp3+ T cells and, together with TGF-beta, generates IL-9+ EMBO J. 18: 420–432. IL-10+ Foxp3(2) effector T cells. Nat. Immunol. 9: 1347–1355. 32. Perumal, N. B., and M. H. Kaplan. 2011. Regulating Il9 transcription in T helper 6. Schmitt, E., T. Germann, S. Goedert, P. Hoehn, C. Huels, S. Koelsch, R. Ku¨hn, cells. Trends Immunol. 32: 146–150. W. Mu¨ller, N. Palm, and E. Ru¨de. 1994. IL-9 production of naive CD4+ T cells 33. Wei, L., G. Vahedi, H. W. Sun, W. T. Watford, H. Takatori, H. L. Ramos, depends on IL-2, is synergistically enhanced by a combination of TGF-beta and H. Takahashi, J. Liang, G. Gutierrez-Cruz, C. Zang, et al. 2010. Discrete roles of IL-4, and is inhibited by IFN-gamma. J. Immunol. 153: 3989–3996. STAT4 and STAT6 transcription factors in tuning epigenetic modifications and 7. Veldhoen, M., C. Uyttenhove, J. van Snick, H. Helmby, A. Westendorf, J. Buer, transcription during T helper cell differentiation. Immunity 32: 840–851. 34. Afkarian, M., J. R. Sedy, J. Yang, N. G. Jacobson, N. Cereb, S. Y. Yang, B. Martin, C. Wilhelm, and B. Stockinger. 2008. Transforming - T. L. Murphy, and K. M. Murphy. 2002. T-bet is a STAT1-induced regulator of beta ‘reprograms’ the differentiation of T helper 2 cells and promotes an inter- IL-12R expression in naı¨ve CD4+ T cells. Nat. Immunol. 3: 549–557. leukin 9-producing subset. Nat. Immunol. 9: 1341–1346. 35. Djuretic, I. M., D. Levanon, V. Negreanu, Y. Groner, A. Rao, and K. M. Ansel. 8. Chang, H. C., S. Sehra, R. Goswami, W. Yao, Q. Yu, G. L. Stritesky, R. Jabeen, 2007. Transcription factors T-bet and Runx3 cooperate to activate Ifng and si- C. McKinley, A. N. Ahyi, L. Han, et al. 2010. The transcription factor PU.1 is lence Il4 in T helper type 1 cells. Nat. Immunol. 8: 145–153. required for the development of IL-9-producing T cells and allergic inflamma- 36. O’Malley, J. T., S. Sehra, V. T. Thieu, Q. Yu, H. C. Chang, G. L. Stritesky, tion. Nat. Immunol. 11: 527–534. E. T. Nguyen, A. N. Mathur, D. E. Levy, and M. H. Kaplan. 2009. Signal 9. Staudt, V., E. Bothur, M. Klein, K. Lingnau, S. Reuter, N. Grebe, B. Gerlitzki, transducer and activator of transcription 4 limits the development of adaptive M. Hoffmann, A. Ulges, C. Taube, et al. 2010. Interferon-regulatory factor 4 is regulatory T cells. Immunology 127: 587–595. The Journal of Immunology 975

37. Wei, J., O. Duramad, O. A. Perng, S. L. Reiner, Y. J. Liu, and F. X. Qin. 2007. GATA3-driven Th2 responses inhibit TGF-beta1-induced FOXP3 expression and Antagonistic nature of T helper 1/2 developmental programs in opposing pe- the formation of regulatory T cells. PLoS Biol. 5: e329. ripheral induction of Foxp3+ regulatory T cells. Proc. Natl. Acad. Sci. USA 104: 39. Kaplan, M. H., N. L. Glosson, G. L. Stritesky, N. Yeh, J. Kinzfogl, 18169–18174. S. L. Rohrabaugh, R. Goswami, D. Pham, D. E. Levy, R. R. Brutkiewicz, et al. 38. Mantel, P. Y., H. Kuipers, O. Boyman, C. Rhyner, N. Ouaked, B. Ru¨ckert, 2011. STAT3-dependent IL-21 production from T helper cells regulates hema- C. Karagiannidis, B. N. Lambrecht, R. W. Hendriks, R. Crameri, et al. 2007. topoietic progenitor cell homeostasis. Blood 117: 6198–6201. Downloaded from http://www.jimmunol.org/ by guest on September 23, 2021