Blimp-1 Attenuates Th1 Differentiation by Repression of ifng, , and Expression

This information is current as Luisa Cimmino, Gislaine A. Martins, Jerry Liao, Erna of September 24, 2021. Magnusdottir, Gabriele Grunig, Rocio K. Perez and Kathryn L. Calame J Immunol 2008; 181:2338-2347; ; doi: 10.4049/jimmunol.181.4.2338

http://www.jimmunol.org/content/181/4/2338 Downloaded from

References This article cites 68 articles, 29 of which you can access for free at: http://www.jimmunol.org/content/181/4/2338.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 24, 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 Copyright © 2008 by The American Association of Immunologists All rights reserved. Print ISSN: 0022-1767 Online ISSN: 1550-6606. The Journal of Immunology

Blimp-1 Attenuates Th1 Differentiation by Repression of ifng, tbx21, and bcl6

Luisa Cimmino,* Gislaine A. Martins,† Jerry Liao,† Erna Magnusdottir,‡ Gabriele Grunig,† Rocio K. Perez,‡ and Kathryn L. Calame1*§

T cell-specific deletion of Blimp-1 causes abnormal homeostasis and function, leading to spontaneous, fatal colitis in mice. Herein we explore the role of Blimp-1 in Th1/Th2 differentiation. Blimp-1 mRNA and are more highly expressed in Th2 cells compared with Th1 cells, and Blimp-1 attenuates IFN-␥ production in CD4 cells activated under nonpolarizing conditions. Although Blimp-1-deficient T cells differentiate normally to Th2 cytokines in vitro, Blimp-1 is required in vivo for normal Th2 humoral responses to NP-KLH (4-hydroxy-3-nitrophenylacetyl/keyhole lymphocyte hemocyanin) immunization. Lack of Blimp-1 in CD4 T cells causes increased IFN-␥, T-bet, and Bcl-6 mRNA. By chromatin immunoprecipitation we show that Blimp-1 binds

directly to a distal regulatory region in the ifng gene and at multiple sites in tbx21 and bcl6 . Our data provide evidence that Downloaded from Blimp-1 functions in Th2 cells to reinforce Th2 differentiation by repressing critical Th1 genes. The Journal of Immunology, 2008, 181: 2338–2347.

ifferentiation of the CD4 Th cell lineages is dictated IL-13 genes (7) and can autoactivate its own expression both by the strength of antigenic stimulation, cytokine en- independently and as a result of IL-4-mediated STAT-6 signal- D vironment, and the complement of transcription fac- ing (8). How T-bet and GATA-3 compete to drive Th1 or Th2 http://www.jimmunol.org/ tors activated by these processes. T-bet and GATA-3 transcrip- differentiation has been intensively studied and has laid the tion factors are master regulators of Th1 and Th2 lineage foundation for understanding how other transcription factors differentiation, respectively; Foxp3 is critical for regulatory T influence the way CD4 lineage decisions are made. cell development and function, and, more recently, ROR␥t (ret- The transcriptional repressor B lymphocyte-induced maturation ␥ 2 inoid-related orphan t) has been shown to act as a protein-1 (Blimp-1) has recently been identified as a regulator of master regulator of Th17 development (1–3). Additionally, T cell homeostasis and function (9, 10). Encoded by the complicated regulatory pathways and feedback loops cooperate gene, Blimp-1 is well established as a master regulator of plasma with the master regulators to establish and maintain CD4 dif- cell differentiation and maintenance (11, 12). Blimp-1 also plays a ferentiation states. role in multiple developmental checkpoints in nonlymphoid lin- by guest on September 24, 2021 Th polarization into Th1 and Th2 lineages can be divided into eages including germ cell formation during embryogenesis (13– hierarchical steps of initiation, reinforcement, and maintenance 15), myeloid differentiation (16), keratinocyte maturation (17), and (1). Initiation of lineage differentiation begins upon T cell recog- sebocyte differentiation (18). nition of Ag presented by APCs and response to the initial cyto- In the T cell lineage Blimp-1 is strongly induced upon TCR kine environment. Reinforcement of commitment is driven by the activation (10), with IL-2 playing an important role in the induc- up-regulation of cytokine production, positive feedback loops, and tion (19). In two independent studies, T cell-specific deletion of repression of genes necessary for the alternative fate. Maintenance Blimp-1 in mice caused spontaneous colitis (10) and a multiorgan of the differentiated state is conferred by heritable epigenetic inflammatory disease (9). These autoimmune pathologies were at- changes. Transcription factors activated by TCR-signaling such tributed in part to altered homeostasis within the CD4 T cell lin- ␬ as AP-1, NF- B, and NFAT can differentially induce both eage, including hyperproliferation in response to TCR stimulation, ␥ IFN- and IL-4 expression directly or via the induction of T-bet increased IL-2 and IFN-␥ production, and decreased IL-10 pro- and GATA-3, respectively (1, 4). In the case of Th1 differen- duction, possibly contributing to diminished regulatory T cell ␥ tiation, T-bet can be induced both by IFN- or IL-12 through function in vivo (9, 10). the transcription factors STAT-1 and STAT-4, respectively, to Blimp-1 is a transcriptional repressor. Interestingly, most of the ␥ directly reinforce expression of the IFN- gene while simulta- known direct targets of Blimp-1 are also transcription regulators, neously repressing GATA-3 function (2, 5, 6). GATA-3 acti- explaining in part its ability to control extensive developmental vates expression at the Th2 cytokine locus of IL-4, IL-5, and programs (11). In B cells Blimp-1 directly represses , spiB, CIITA, , and . In the epidermis it represses myc, fos, , *The Institute of Human Nutrition, †Department of Microbiology, ‡Department of and dusp16 (17). It has also been shown to repress in some Biological Sciences, and §Department of Biochemistry and Molecular Biophysicis, tumor cell lines (20). College of Physicians and Surgeons, Columbia University, New York, NY 10032 Received for publication March 18, 2008. Accepted for publication June 9, 2008. The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance 2 Abbreviations used in this paper: Blimp-1, B lymphocyte-induced maturation pro- with 18 U.S.C. Section 1734 solely to indicate this fact. tein-1; Bcl-6, B cell lymphoma-6; ChIP, chromatin immunoprecipitation; CNS, con- 1 Address correspondence and reprint requests to Dr. Kathryn Calame, Department of served noncoding sequence; IP, immunoprecipitation; MFI, mean fluorescence inten- Biochemistry and Molecular Biophysicis, College of Physicians and Surgeons, Co- sity; NP-KLH, 4-hydroxy-3-nitrophenylacetyl/keyhole lymphocyte hemocyanin. lumbia University, 701 168th Street, Hammer Health Science Center 1202, Manhat- tan, NY 10032. E-mail address: [email protected] Copyright © 2008 by The American Association of Immunologists, Inc. 0022-1767/08/$2.00

www.jimmunol.org The Journal of Immunology 2339

Direct targets for Blimp-1 in T cells have not yet been identified. In vitro CD4 Th cell differentiation However, it is known that B cell lymphoma-6 (Bcl-6) mRNA is For in vitro polarization, naive CD4 T cells were seeded at 1 ϫ 106 cells/ml elevated in Blimp-1-deficient CD4 T cell effectors (10). Mutual in 24-well plates coated with 5 ␮g/ml anti-CD3, 2.5 ␮g/ml anti-CD28 (BD repression by Bcl-6 and Blimp-1 in B cells is important for main- Pharmingen), and 25 U/ml of recombinant human IL-2 in RPMI 1640 with taining mutually exclusive programs of germinal center B cells and the addition of murine IL-12 (10 ng/ml) and anti-IL4 (10 ␮g/ml) or murine ␥ ␮ plasma cells (21). In T cells, Bcl-6 repression is involved in Th1 IL-4 (1000 U/ml) and anti-IFN- (10 g/ml). differentiation by repressing Th2 cytokine expression. It decreases GATA-3 protein levels (22), represses IL-5 transcription (23), and Western blot analysis competes for binding at STAT-6 recognition sequences in IL-4- CD4 T cells were cultured for 2 wk under polarizing conditions de- responsive genes (24, 25). scribed above, lysed in RIPA buffer (50 mM HEPES (pH 7.6), 1 mM EDTA, 0.7% sodium deoxycholate, 1% Nonidet P-40, and 0.5 M The goal of the present study was to learn how Bcl-6 and ␮ LiCl2), and 50 g of protein from the whole-cell extract was separated Blimp-1 interact during Th1/Th2 development and to determine by 8% SDS-PAGE and Western blotted using mouse monoclonal anti- the molecular mechanism(s) by which Blimp-1 alters expres- Blimp-1 (26), anti-GATA-3, and anti-␤-tubulin Abs (Santa Cruz sion of cytokines important for Th1 vs Th2 differentiation and Biotechnology). function. We show that Blimp-1 is most highly expressed in Th2 cells and that mice lacking Blimp-1 in CD4 T cells exhibit Quantitative RT-PCR impaired humoral Th2 responses, establishing a role for Total RNA was isolated according to the manufacturer’s instructions by Blimp-1 in Th2 differentiation. We also show that ifng, tbx21, using TRIzol reagent (Invitrogen) from in vitro-activated CD4 T cells and bcl6 are direct targets of Blimp-1-dependent repression in or sorted CD4 T cell subsets, and reverse transcription was performed Downloaded from CD4 T cells, providing a molecular basis for the idea that using SuperScript III (Invitrogen) following the manufacturer’s instruc- Blimp-1 helps to oppose Th1 differentiation during Th2 lineage tions. For quantitative PCR analysis, the cDNA was diluted 5-fold in water and 5 ␮l was amplified for 40 cycles using the ABI 7700 (Applied commitment. Biosystems) sequence detection system with the following primers: IFN-␥, forward 5Ј-GCTTTGCAGCTCTTCCTCAT and reverse 5Ј-GTC ACCATCCTTTTGCCAGT; T-bet, forward 5Ј-GGTGTCTGGGAAGCT Materials and Methods GAGAG and reverse 5Ј-CCACATCCACAAACATCCTG; Blimp-1, http://www.jimmunol.org/ Mice forward 5Ј-GACGGGGGTACTTCTGTTCA and reverse 5Ј-GGCATT CTTGGGAACTGTGT; Bcl-6, forward 5Ј-CTGCAGATGGAGCATGT ϩ/ϩ fl/fl Prdm1 or prdm1 mice (26) were crossed with mice expressing CD4- TGT and reverse 5Ј-CACCCGGGAGTATTTCTCAG; and 18S, for- ϩ ϩ ϩ ϩ/ϩ ϩ fl/fl Cre (27) or Lck-Cre (28) to generate Cre prdm1 or Cre prdm1 ward 5Ј-TCAAGAACGAAAGTCGGAGG and reverse 5Ј-GGA Ϫ/Ϫ mice. Bcl-6 mice were described previously (29). Mice were housed in CATCTAAGGGCATCACA. the barrier facility of Columbia University (New York, NY). All experi- ments were approved by the Institutional Animal Care and Use Committee of Columbia University. Chromatin immunoprecipitation (ChIP) and quantitative PCR CD4 T cells were grown for 6 days in culture and restimulated for 4 h CD4ϩ T cell purification and CFSE labeling with 20 ng/ml PMA and 1 ␮g/ml ionomycin before harvesting. ChIP assays were performed as previously described (30). Briefly, 30–40 ϫ by guest on September 24, 2021 ϩ Naive CD4 T cells were purified from the spleen and lymph nodes of 106 cells were used per ChIP after crosslinking with 1% formaldehyde ϩ ϩ ϩ 4-wk-old prdm1 / or prdm1fl/fl mice expressing CD4-Cre or Lck- at room temperature for 10 min, quenching with 125 mM glycine, nu- ϩ ϩ ϩ Ϫ Ϫ Cre and Bcl-6 / or Bcl-6 / littermates by flow cell-sorting for clear lysis and chromatin fragmentation by sonication at 4°C. Preim- ϩ CD4 CD62LhighCD44low cells. Staining with the division-sensitive dye mune serum bled from rabbits before immunization with Blimp-1 an- CFSE (Molecular Probes) was performed as previously described (10) tigenic peptide was used as a control for anti-Blimp-1 rabbit polyclonal before stimulation in vitro for 4 days with plate-bound anti-CD3 (5 serum generated in the laboratory (31). Abs were bound to protein A ␮g/ml), anti-CD28 (2.5 ␮g/ml), and recombinant human IL-2 Dynabeads (Dynal), immunoprecipitations were performed overnight at (25 U/ml). 4°C, and washed in RIPA buffer. Protein-DNA crosslinks were reversed at 65°C overnight, and precipitated DNA was treated with proteinase K FACS analysis and intracellular cytokine staining and RNase before phenol chloroform extraction and ethanol precipitation. FACS plots were gated on live, non-autofluorescent cells before anal- Analysis of and putative Blimp-1 binding sites ysis. Cells were stained with fluorochrome-conjugated Abs specific for was performed using the ECR browser (http://ecrbrowser.dcode.org/) CD3, CD4, CD8, CD25, CD44, CD62L, Thy-1.2, IFN-␥ (eBioscience), and rVista 2.0. Quantitative PCR analysis was performed with FastStart IL-2, IL-4, or IL-10 (BD Pharmingen). For intracellular staining, naive SYBR Green Master mix (Roche) using the Stratagene MX3000 real- CD4 T cells were stimulated for either 4–6 days in the presence of time PCR system. Relative fold-enrichment was determined by calcu- plate-bound anti-CD3 (5 ␮g/ml), anti-CD28 (2.5 ␮g/ml), and 25 U/ml lating the immunoprecipitation (IP) efficiency (ratio of the amount of of recombinant human IL-2 followed by restimulation for 6 h with PMA immunoprecipitated DNA to that of the input sample) of anti-Blimp-1 (20 ng/ml) and ionomycin (1 ␮g/ml) (Sigma-Aldrich) and monensin normalized to control antiserum. The primer sequences used are shown (GolgiStop; BD Pharmingen) to block cytokine secretion during the in Table I. final4hofstimulation. Cells were fixed with 4% paraformaldehyde and were made permeable with 0.5% saponin for intracellular staining with cytokine Abs. Results Blimp-1 decreases the number of IFN-␥-producing CD4ϩ cells ␥ 4-Hydroxy-3-nitrophenylacetyl/keyhole lymphocyte hemocyanin and the amount of IFN- made per cell (NP-KLH) immunizations Since our previous work showed that murine CD4ϩ T cells lacking ␥ ␥ Mice aged 4–6 wk were injected i.p. with 100 ␮g of alum-precipitated Blimp-1 had elevated IFN- production (10) and because IFN- is NP-KLH (Biosearch Technologies) in PBS. Imject Alum was purchased a key component in Th1 differentiation, we wanted to explore the from Pierce. Five to seven mice were included per genotype. potential role of Blimp-1 in the commitment and/or establishment of Th1 vs Th2 cells. First, we studied the role of Blimp-1 in IFN-␥ ELISA production in more detail. Naive CD4 T cells (CD44ϪCD62Lϩ) were purified from the spleen and lymph nodes of 4-wk-old Blimp NP-specific IgM, IgG1, and IgG2a in immunized mice for primary re- flox/flox sponses were detected by ELISA as previously described using NP-BSA CKO mice ( prdm1 with either Lck-Cre or CD4-Cre) and (Biosearch Technologies)-coated plates (68). analyzed for intracellular IFN-␥ following stimulation with 2340 Blimp-1 ATTENUATES Th1 DIFFERENTIATION

Table I. Sequences used

Site Location (bp) Forward Primer Reverse Primer

ifng 15ЈϪ31,884 CCTGGCCCTCTCCTGTAAG GGGGCTATAGGAATGGCATA 25ЈϪ27,497 GGCTAGACCAGCCAATAAGC TCTGGCCCTTCCTAGCTGTA 35ЈϪ21,572 CACCTGGGGTGAAAAGAAAT GTGAATCCCCAGAGAAGCAG 45Ј 5,592 GAACGGGTGGAGGAGGTAA GCCAAGGTGTGATTGCATTA 53Ј 7,480 TGGGTTTTACCTATCATCAT GGGCACAGACTAATCTCAAC 63Ј 14,895 GAAAGTTGGCGAGAAACAGG TCTGAATCCCTTGCTGACCT 73Ј 16,011 TCCACACACCCCTCCATT CCAAGGACTTCCTCTCCTCA tbx21 15ЈϪ8,673 GGTGGGTGAAGGAAGATCTA GACTGGGGACAAGAAGATGC 25ЈϪ627 CTGAACACTGACAGCCAAGC AAGAAGATTGCACCCAAAGC 3 Intron 1 1,637 CTCCACAGCCACCTACACTG GGGACTAGCGTAAGCCACTG 4 Intron 1 10,872 TATGGAGGTGGTGGTTGTCA CCAGGAAAAGCAGAAGTTGG 53Ј 3,846 TGCCAGCCATACCTGTTAAG GAGAGAGAGAGAGAGAGAGG 63Ј 4,378 CCTCATCTCGGCTCCTTTC CTATGGGCGATGTCAAGCTA bcl6 15Ј 4,778 CAGCCACCCTGAGTTTACAA CGTTCCAGCACTGTTTTGAA 2 Intron 1 2,117 CTGGCTTCCTCTGCAGTTC CTGTCTCTGGATGGATCACG 3 Intron 1 5,648 TGAGGCAGGACACGTAACAG GCGGGAATGACACAGAGATA Downloaded from 43Ј UTR838 TTGTCTAAAATGCCTCCGTGT CATTTCTCCTTCTGCAGCTTT 53Ј 4,993 ACAGCCAGGAGAGCTTTACCG TGTTTGCTTTCTGGCATTTG anti-CD3, anti-CD28, and IL-2 for 6 days and subsequent restimu- after 6 days in culture (Fig. 2A). Furthermore, Blimp-1 protein,

lation. As shown in Fig. 1, A and B, the fraction of cells producing analyzed by Western blotting of CD4 T cells cultured for 2 wk http://www.jimmunol.org/ IFN-␥ was 2-fold higher in Blimp-1 CKO CD4 T cells compared under Th1 and Th2 polarizing conditions, showed more with controls, and the IFN-␥-producing cells from the CKOs also Blimp-1 in Th2 cells compared with Th1 cells (Fig. 2B). showed a 2-fold higher mean fluorescence intensity (MFI), indi- Blimp-1 mRNA was also analyzed in CD4 T cells following cating elevated IFN-␥ production on a per cell basis. The fraction stimulation with different cytokines important for Th cell differ- of cells expressing IFN-␥ increases in frequency with successive entiation. Naive CD4 T cells were cultured for 6 days either with cell cycles (32); therefore, a previously observed proliferative ad- anti-CD3 and anti-CD28 stimulation alone or with added IL-2, vantage in Blimp-1 CKO CD4 T cells (10) might explain increased IL-4, IL-12, or IFN-␥. Stimulation with anti-CD3, anti-CD28, and ␥ IFN- production during activation in culture. To determine the exogenous IL-2 increased Blimp-1 mRNA modestly compared ␥ by guest on September 24, 2021 relationship between IFN- production and cell division in with TCR stimulation alone. IL-4 induced Blimp-1 mRNA - Blimp-1 CKO cells, naive CD4 T cells were stained with the di- imally under these culture conditions (Fig. 2C), while IFN-␥ ac- vision-sensitive dye CFSE before culture for 4 days under various tually inhibited induction of Blimp-1 mRNA (Fig. 2C). Thus, types of stimulation, comparing IFN-␥ production as a function Blimp-1 mRNA is induced by IL-4 and repressed by IFN-␥ sig- of proliferation. Consistent with our previous findings, Blimp-1 naling, consistent with highest expression in Th2 cells. These data CKO CD4 T cells are more proliferative than wild-type control are consistent with the idea that Blimp-1 is induced in Th2 cells cells under weak stimulatory conditions (10). Stimulation for 4 days with anti-CD3 alone gave rise to 6-fold more IFN-␥-pro- and plays a role in Th2 commitment or function. duction in the Blimp CKO cultures compared with a 2-fold increase with anti-CD3 and anti-CD28 (Fig. 1C). However, upon stimulation with anti-CD3, anti-CD28, and IL-2, despite Blimp-1 is required for a normal Th2 humoral response in vivo, equivalent rates of proliferation, the frequency of IFN-␥-pro- but not for Th2 differentiation in vitro ducing cells was 1.5-fold higher in the CKO in total culture To test if Blimp-1 deficiency alters differentiation of Th cells in (Fig. 1C) and, with one exception, at each round of division vitro, naive CD4 T cells were stimulated for 6 days with plate- (Fig. 1D). These results show a cell cycle-independent effect of bound anti-CD3, anti-CD28, and IL-2 with the addition of ei- Blimp-1 on the number of cells producing IFN-␥. Additionally, ther IL-12 and anti-IL-4 or IL-4 and anti-IFN-␥ to promote Th1 the increased IFN-␥ MFI per cell in the CKO CD4 T cells indicates that each CKO cell makes more IFN-␥. Thus, we con- or Th2 polarization, respectively. There was no significant dif- ␥ clude that Blimp-1 inhibits both the fraction of cells making ference in the fraction of IFN- -producing cells between the IFN-␥ and the amount of IFN-␥ made per cell. wild-type and CKO Th1 cultures, and, similarly, under Th2 culture conditions, equivalent fractions of IL-4-producing cells Blimp-1 is more highly expressed in Th2 cells were present (Fig. 3A). However, there was a 3–5-fold decrease Increased IFN-␥ production by Blimp-1-deficient CD4 T cells in the number of IL-10-producing cells upon Th2 differentiation stimulated under neutral conditions indicates that Blimp-1 may in Blimp CKO cultures (Fig. 3B), consistent with the decreased normally play a role in Th2 differentiation. The expression pattern IL-10 production observed in Blimp-1 CKO CD4 T cell effec- of Blimp-1 under Th1 and Th2 polarizing conditions was therefore tors ex vivo and under neutral conditions of stimulation in vitro analyzed in wild-type CD4 T cells to determine whether the ex- (10). Although the fraction of IFN-␥- or IL-4-producing cells pression pattern of Blimp-1 was consistent with a role in Th2 was not significantly different between the wild-type and CKO differentiation. Th cell cultures (Fig. 3C), the average MFI for IFN-␥ was 1.6- Blimp-1 steady-state mRNA, measured by quantitative RT- fold higher for Blimp-1 CKO cells under Th1 conditions, show- PCR, was 5-fold higher in Th2 cells compared with Th1 cells ing that these cells produce more IFN-␥ (supplementary Fig. The Journal of Immunology 2341

A WT CKO B 40 3 16% 30% 30 2

20

1

10

MFI CKO/WT MFI IFN- γ γ

producing cells producing IFN- % γ 0 0 IFN- γ FIGURE 1. Blimp-deficient CD4 T cells WT CKO WT CKO produce higher amounts of IFN-␥ on a per CD4 cell basis. A, Representative FACS plots of naive CD4 T cells cultured for 6 days with C Proliferation WT CKO anti-CD3, anti-CD28, and IL-2 and restim- 1% 6% ulated for 6 h with PMA and ionomycin fol- anti-CD3 lowed by intracellular cytokine staining for IFN-␥. B, Average percentage and MFI of IFN-␥-producing cells in wild-type (open bar) and CKO (gray bar) CD4 Th cell cul- tures (n ϭ 6). C, IFN-␥ production and pro- Downloaded from liferation measured by CFSE staining of 15% 27% wild-type and CKO CD4 T cells cultured for anti-CD3 MFI MFI 4 days with combinations of anti-CD3, anti- anti-CD28 280 350 CD28, and IL-2. Cells were restimulated with PMA and ionomycin for 6 h before har- vesting for intracellular cytokine staining of IFN-␥. Representative histograms for prolif- http://www.jimmunol.org/ eration between wild-type (gray) and CKO 29% 40% (open) T cells are overlaid and representa- anti-CD3 MFI MFI tive dot plots are shown with the total fre- anti-CD28 320 365 quency and MFI of IFN-␥-producing cells IL-2 per condition. D, Percentage of IFN-␥-pro- ducing cells per cell division in WT (open bar) and CKO (gray bar) CD4 T cells stim- IFN- γ ulated with anti-CD3, anti-CD28, and IL-2. Cell count CFSE CFSE

Data are representative of two experiments. by guest on September 24, 2021 D 12 10 WT 8 CKO

6

4

2

0

producing cells producing IFN- % γ 123456 Cell Division

1A).3 Thus, we conclude that Blimp-1 is not required for Th1 or elevated because the T cell cytokine milieu in the germinal center Th2 cell polarization in vitro, although in its absence, expres- is not normal, but this hypothesis requires further study. These data sion of IL-10 is abnormal. show that Blimp-1 is required for a normal Th2 humoral response Polarization in vitro provides saturating amounts of polarizing to NP-KLH in alum. cytokines and is unlikely to reflect the more subtle signals that determine Th1/Th2 polarization in vivo. To determine whether Blimp-1 represses IFN-␥ and T-bet mRNA expression in Blimp-1 is important for a Th2 response in vivo, we immunized activated CD4 T cells mice with NP-KLH in alum, which is known to cause a Th2- We wanted to explore the molecular mechanisms responsible for dependent response primarily of IgG1 (33). Sera from immunized the role of Blimp-1 in Th2 differentiation. The data showing ele- mice were measured over the course of 21 days for NP-specific vated IFN-␥ production in CKO CD4ϩ cells suggested that ϳ IgM, IgG1, and IgG2a by ELISA. IgG1 was decreased 2-fold in Blimp-1 might repress IFN-␥ or other genes expressed in Th1 CKO mice at each time point, indicating that the Th2 responses cells. In CD4ϩ cells, IFN-␥ transcription is activated by TCR sig- were impaired (Fig. 4B). Interestingly, CKO mice also exhibited naling (34, 35), and subsequent IFN-␥ secretion initiates more 2–3-fold higher titers of NP-specific IgM compared with control IFN-␥ production via a positive feedback loop involving the IFN- mice (Fig. 4A). The IgG2a response of the CKO mice was similar ␥R-STAT-1 pathway. Activated STAT-1 also induces the tran- to controls late in the response, but 1.5–2-fold higher at days 12 scription of tbx21, encoding T-bet, and T-bet further induces ifng and 18 (Fig. 4C). It may be that the IgM and IgG2a responses are transcription and interferes with GATA-3 function, thus reinforc- ing IFN-␥ production and Th1 polarization (5, 36, 37). To inves- 3 The online version of this article contains supplemental material. tigate the possible role of Blimp-1 in expression of IFN-␥ and 2342 Blimp-1 ATTENUATES Th1 DIFFERENTIATION

steady-state mRNA in wild-type and CKO CD4 T cells. IFN-␥ A 80 B mRNA was increased 2–3-fold in Blimp-1-deficient CD4 T cells Blimp-1 compared with wild-type cells, in naive cells, and in activated cells 60 at each time during the 6 day stimulation and upon restimulation with PMA and ionomycin at days 3 and 6 of culture (Fig. 5A). 40 GATA-3 T-bet mRNA was also elevated in the CKO, most noticeably at day 2, with 5–6-fold increased T-bet expression in CKO CD4 T cells 20 Tubulin compared with wild-type cells and upon restimulation with PMA

Blimp-1 mRNA/18S and ionomycin after 3 days in culture (Fig. 5B). Further evidence 0 Th1 Th2 Th1 Th2 for a requirement for Blimp-1 to repress IFN-␥ expression was ␥ C obtained by quantitative RT-PCR in which the levels of IFN- 150 mRNA from CKO Th1 and Th2 cells were observed to be 5-fold +IL-4 120 and 3-fold increased, respectively, as compared with wild-type cells (supplementary Fig. 1B). Collectively, these data show 90 that Blimp-1 normally represses expression of IFN-␥ and T-bet mRNA. Under neutral cytokine conditions, acute stimulation of Stim +IL-2 60 +IL-12 naive CD4 T cells for 24 h with PMA and ionomycin also +IFN-g induced 8-fold higher levels of IFN-␥ and 15-fold higher levels

Blimp-1 mRNA/18S 30 of T-bet mRNA in Blimp-1 CKO cells (Fig. 5C). Since stimu- Downloaded from

0 lation by PMA and ionomycin bypasses early TCR signaling anti-CD3/CD28 +++++ events, this provides evidence that Blimp-1 regulation occurs IL-2 -+--- downstream of PKC activation and Ca2ϩ mobilization and sug- IL-4 --+-- IL-12 ---+- gests that Blimp-1 might directly repress IFN-␥ and T-bet IFN-g ----+ expression.

FIGURE 2. Blimp-1 is more highly expressed in Th2 cells and http://www.jimmunol.org/ Blimp-1 and Bcl-6 repress one another in CD4 T cells Blimp-1 mRNA expression is differentially regulated by polarizing cy- tokines. A, Blimp-1 steady-state mRNA levels in CD4 T cells stimu- Bcl-6 is known to play a role in repressing Th2 differentiation (22, lated for 6 days under Th1 and Th2 culture conditions (n ϭ 5 with 29, 38). Furthermore, in B cells, Bcl-6 and Blimp-1 repress one SEM). B, Western blot analysis of Blimp-1 and GATA-3 protein levels another, resulting in mutually exclusive transcriptional programs ␤ normalized to -tubulin in CD4 T cells cultured for 2 wk under Th1 and that maintain cells either in a germinal center state, where Bcl-6 is Th2 conditions. C, Blimp-1 steady-state mRNA levels in CD4 T cells expressed, or in a plasma cell state, where Blimp-1 is present (21). stimulated for 6 days with plate-bound anti-CD3 and anti-CD28 in the In CD4ϩ T cells, Bcl-6 is induced by activated STAT-1 (39) and presence of IL-2, IL-4, IL-12, and IFN-␥ cytokines. Data represent averages and SEM of three experiments. Bcl-6 mRNA expression is elevated in Blimp-1 CKO CD4 T cell effectors purified ex vivo (10), suggesting that Blimp-1 also re- by guest on September 24, 2021 presses Bcl-6 in T cells, which, we hypothesized, might also be T-bet mRNAs, we employed quantitative RT-PCR. During acti- important for Th1 vs Th2 differentiation. vation in vitro under nonpolarizing conditions, total mRNA was We measured Bcl-6 mRNA by quantitative RT-PCR in cells collected at various times and analyzed for IFN-␥ and T-bet stimulated in nonpolarizing conditions for 6 days. Lack of Blimp-1

CKO A WT C WT WT 77% 87% CKO CKO 80 40 Th1 60 30

40 20

FIGURE 3. Blimp-1-deficient T cells are 1% 0.3% producing cells producing able to polarize in vitro but exhibit decreased γ 20 10

IL-10 production under Th2 conditions. Na- % IL-4 producing cells producing IL-4 % ive CD4 T cells were cultured for 6 days IFN- % 7% 8% 0 0 under Th1 or Th2 conditions and analyzed Th2 Th1 Th2 Th1 Th2 by FACS intracellular cytokine staining for (A) IFN-␥ and IL-4 production or (B) IL-10 production. Representative FACS plots are shown with percentage of IFN-␥-, IL-4-, or 24% 27% IFN- γ IL-10-producing cells (inset). Bar graphs represent (C) average percentage of IFN-␥- IL-4 or IL-4-producing cells in Th1 and Th2 cul- tures. Results are averages and SEM of four B WT CKO experiments. FACS data of IL-10-producing 21% 4% cells are representative of three experiments. Th2 IL-10

CD4 The Journal of Immunology 2343

A 30 A Control 40 500 anti-CD3 * RS WT CKO anti-CD28 400 * 30 CKO 20 Control Ave IL-2 300 CKO Ave mRNA/18S * γ 20 200 * 10 10 * 100

0 Relative IFN- 0 01236 3 6 Anti-NP IgM (arbitrary units) Anti-NP 0 Time (day) Time (day) WT CKO WT CKO WT CKO WT CKO 7 12 18 21 B 200 RS anti-CD3 WT 60 anti-CD28 Time (day) 160 CKO IL-2 80 B 120 40 ** 80 60 20 * 40 * 40 Relative Tbet mRNA/18S 0 0 01236 36 Time (day) Time (day) Downloaded from 20 C 1000 PMA+Iono WT

Anti-NP IgG1 (arbitrary units) Anti-NP 0 CKO 100 WT CKO WT CKO WT CKO 12 18 21 Time (day) 10 C 20 http://www.jimmunol.org/ Relative mRNA/18S

1 15 IFNg Tbet FIGURE 5. IFN-␥ and T-bet mRNA expression levels are elevated in 10 Blimp-1-deficient CD4 T cells. Quantitative PCR analysis of steady-state mRNA levels of (A) IFN-␥ and (B) T-bet from flow cell-sorted wild-type 5 (open bar) and CKO (gray bar) naive CD4 T cells stimulated for 6 days with plate-bound anti-CD3(5 ␮g/ml), anti-CD28 (2.5 ␮g/ml), and IL-2 (20

Anti-NP IgG2a (arbitrary units) Anti-NP 0 U/ml) and restimulated (RS) at days 3 and 6 with PMA (20 ng/ml) and ␮ WT CKO WT CKO WT CKO ionomycin (1 g/ml) for 4 h. Data represent averages and SEM of four by guest on September 24, 2021 12 18 21 experiments and were analyzed using a Students paired t test with p values as indicated. C, Steady-state mRNA levels of (ءء) orϽ0.01(ء) Time (day) Ͻ0.05 ϩ ␥ FIGURE 4. Prdm1fl/fl CD4/LCK-Cre (Blimp-1 CKO) mice exhibit IFN- and T-bet in CD4 T cells activated under nonpolarizing conditions increased IgM and decreased IgG1 production in response to immuni- for 24 h with PMA and ionomycin (Iono). Data are representative of two zation with alum-precipitated NP-KLH. Blimp-1 CKO and control mice experiments. were immunized with alum-precipitated NP-KLH, and serum at indi- cated days was analyzed by ELISA for NP-specific (A) IgM, (B) IgG1, and (C) IgG2a. For all panels, control mice are represented by open circles with averages (gray bar) and CKO mice by gray circles with differentiation. Naive CD4 T cells were cultured for 6 days either averages (black bar). with anti-CD3 and anti-CD28 alone or in addition to IL-2, IL-4, IL-12, or IFN-␥. Stimulation with anti-CD3 and anti-CD28 in the presence of IL-2, IL-4, or IL-12 caused no change in the level of had no effect on the early induction of Bcl-6 mRNA; however, by steady-state Bcl-6 mRNA (Fig. 6E). However, IFN-␥ was able to day 6, without restimulation, Bcl-6 mRNA was 2-fold higher in increase the level of Bcl-6 mRNA ϳ7-fold higher than anti-CD3 CKO cells and 2–3-fold increased in CKOs upon restimulation at and anti-CD28 stimulation alone (Fig. 6E). These data show that day 3 and day 6 compared with wild-type controls (Fig. 6A). This Blimp-1 and Bcl-6 mutually repress each other in CD4 T cells. provides further evidence that Blimp-1 represses Bcl-6 expression Bcl-6 mRNA is more highly expressed in Th1 cells compared with in CD4 T cells. Th2 cells and high levels are maintained upon stimulation in the We also wanted to determine whether Bcl-6 represses Blimp-1 presence of IFN-␥. mRNA in T cells as it does in B cells. Ex vivo naive, memory, and effector CD4 T cell populations were sort-purified from 4-wk-old Blimp-1 binds directly to the ifng, tbx21, and bcl6 genes bcl6Ϫ/Ϫ mice, and Blimp-1 mRNA was determined by quantitative To determine whether Blimp-1 directly represses ifng, tbx21,or RT-PCR. As previously shown by our laboratory, Bcl-6 mRNA is bcl6 transcription, potential Blimp-1 binding sites in these genes 3-fold higher in Blimp-1 CKO effector CD4 T cells (Fig. 6B). were identified using the Vista genomic analysis tool (40, 41) and Additionally, we found that Blimp-1 mRNA was ϳ2.5- fold higher tested for Blimp-1 binding in vivo by ChIP (Fig. 7). Candidate in Bcl-6 KO CD4 effectors (Fig. 6C). Blimp-1 binding sites were chosen based on their similarity to Under polarizing culture conditions, Blimp-1 levels are highest previously identified sites in other Blimp-1 targets and to the ex- in Th2 cells (Fig. 2, A and B). Under the same conditions we find perimentally determined consensus binding site (42). For the ChIP that Bcl-6 levels are 3-fold higher in Th1 compared with Th2 cells experiments, binding to an irrelevant site on the gene snail3 was (Fig. 6D). The levels of Bcl-6 mRNA were also analyzed in CD4 used as a negative control, as well as an irrelevant site within each T cells following stimulation with cytokines important for Th cell target gene. Known targets of Blimp-1 in B cells, myc and id3, 2344 Blimp-1 ATTENUATES Th1 DIFFERENTIATION

A A 8 160 300 +RS anti-CD3 WT anti-CD28 6 * CKO 120 IL-2 200

80 4

100

40 * Fold Enrichment 2

Relative Bcl-6 mRNA/18S 0 0 01236 36 0 Time (day) Time (day) 1 2 3 4 5 6 7 1 2 3 4 5 6 B myc id3 30 C 30 snail3 ifng tbx21 WT WT 25 25 B C Blimp CKO Bcl6 -/- 6 6 20 20 CD4 T cells P3X Plasmacytoma cells 15 15 4 4 10 10

Bcl-6 mRNA/18S 5 5 Blimp-1 mRNA/18S 2 2 Fold Enrichment Fold Enrichment 0 0 Downloaded from NME NME D 0 0 100 E 100 1 2 3 4 5 1 2 3 4 5 myc id3 myc id3 +IFN-g snail3 bcl6 snail3 bcl6 80 80 FIGURE 7. Blimp-1 binds directly to regions in the ifng, tbx21, and 60 60 bcl6 genes in CD4 T cells. Naive CD4 T cells for ChIP were stimulated for

6 days in vitro with anti-CD3, anti-CD28, and IL-2 followed by restimu- http://www.jimmunol.org/ 40 40 +IL-4 lation for 4 h with PMA/ionomycin before crosslinking. Quantitative PCR +IL-2 +IL-12 20 20 Stim analysis of immunoprecipitated DNA is displayed as relative fold-enrich- Bcl-6 mRNA/18S Bcl-6 mRNA/18S ment compared with control Abs for sites investigated within the (A) ifng 0 0 and tbx21 genes and bcl6 genomic loci in (B) CD4 T cells and (C) P3X Th1 Th2 anti-CD3/CD28 +++++ IL-2 -+- - - plasmacytoma cells. Irrelevant sites in the snail3 gene and within each IL-4 --+-- IL-12 ---+- target gene serve as negative controls and are indicated by the open bar. IFN-g ----+ Previously identified Blimp-1 binding sites within myc, as well as id3 genes, were used as a positive control. Data represent averages and SEM FIGURE 6. Blimp-1 and Bcl-6 mRNA expression is inversely regulated of four experiments. in CD4 T cells. Steady-state levels of Bcl-6 mRNA were measured by by guest on September 24, 2021 quantitative RT-PCR in (A) CD62LhighCD44low flow cell-sorted wild-type (open bar) and CKO (gray bar) naive CD4 T cells stimulated for 6 days with plate-bound anti-CD3, anti-CD28, and IL-2 and restimulated (RS) at days 3 and 6 with PMA and ionomycin for 4 h. Data represent averages and In tbx21, five potential Blimp-1 binding sites and one irrel- SEM of four experiments. B, Steady-state levels of Bcl-6 mRNA from flow evant site were tested by ChIP. Of those tested, sites 2 and 3 are cell-sorted peripheral naive (N), memory (M), and effector (E) CD4 T cells conserved between mice and humans, whereas sites 4–6 are not from Blimp-1 wild-type (open bar) and CKO (gray bar) mice and (C) Blimp-1 mRNA levels from Bcl-6ϩ/ϩ (open bar) and Bcl-6Ϫ/Ϫ (gray bar) (Fig. 8B). Blimp-1 binding on the tbx21 gene was most highly ϳ mice. Data represent averages and SEM of four experiments and were enriched for the conserved site 2, 600 bp upstream of the indicated. transcriptional start site, and sites 3 and 5, ϳ1.5 kb and 11 kb (ء) analyzed using a Student’s paired t test with p values Ͻ0.05 D, Bcl-6 steady-state mRNA levels in CD4 T cells stimulated for 6 days downstream of exon 1 within the first intron of the tbx21 gene under Th1 and Th2 culture conditions (n ϭ 5 with SEM) and with (E) (Fig. 7A). These data, in conjunction with the expression data plate-bound anti-CD3 and anti-CD28 in the presence of IL-2, IL-4, IL-12, above, demonstrate that ifng and tbx21 are direct targets of ␥ and IFN- cytokines. Data represent averages and SEM of three Blimp-1 repression in CD4 T cells. experiments. Four candidate sites in bcl6 were analyzed, including an ir- relevant site 5 kb downstream of the last bcl6 exon (site 5), used as a negative control (Fig. 8C), along with an irrelevant site were also tested but only the myc gene was bound by Blimp-1 in snail3 bcl6 CD4 T cells (Fig. 7). from . Blimp-1 bound reproducibly to the gene at Seven candidate sites were tested in the ifng gene: sites 3, 4, site 1, located 5 kb upstream of the first exon (Fig. 7B). Thus, 6, and 7 are conserved between mice and humans, whereas sites Blimp-1 directly represses bcl6 in T cells. 1 and 2 are not conserved and site 5 contains an irrelevant It has been previously shown that Blimp-1 can repress Bcl-6 sequence (Fig. 8A). None of the sites tested were found to lo- when ectopically expressed in activated B cells (21), but direct calize to previously described proximal regulatory regions of repression of bcl6 by Blimp-1 has not been demonstrated in B the ifng gene (43, 44). However, sites 3–6, excluding the irrel- cells. To determine whether bcl6 was also a direct target of evant site 5, fall within or near distal conserved noncoding se- Blimp-1 repression in B cells, ChIP was performed for the quence (CNS) regions 22 kb and 5 kb upstream of, and 17 kb Blimp-1 binding sites identified in T cells using P3X plasma- downstream of, the ifng gene, which have been shown to reg- cytoma cells. In these B cells, both myc and id3 were repro- ulate the expression of IFN-␥ in T cells (45). Blimp-1 binding ducibly bound by endogenous Blimp-1 at their known binding site 3 is within the region of the IFN-␥ locus known as CNS-22 sites, as was site 1 on the bcl6 gene (Fig. 7C). These data show that was recently shown to be essential for IFN-␥ expression in that bcl6 is a direct target of Blimp-1 repression in both CD4 T T cells (46) (Fig. 7A). cells and in B cells. The Journal of Immunology 2345

Key: Gene structure Blimp-1 Enrichment Coding Strong A Non-Coding Weak * Conserved None ifng 1 23* 4* 5 6* 7* FIGURE 8. Schematic representa- -31,884 -27,497 -21,572 -5,592 7,480 14,895 16,011 tion of sites investigated within the (A) ifng,(B) tbx21, and (C) bcl6 murine TTCACTTTCCT genomic loci. Blimp-1 recognition se- quences are shown for the site most B 2* 3* 4 5 6 greatly enriched by ChIP within each tbx21 1 gene. Conserved noncoding sequences -8,673 -627 1,637 10,872 3,846 4,378 (CNS) are indicated by an asterisk and locations are shown in bases. TGCTTTCACTT

C 4* 1* 2* 3* 5 bcl6 -4,778 2,117 5,648 838 4,993 Downloaded from

AAGTGAAAACC

Discussion Blimp-1-dependent repression of tbx21, encoding T-bet

We have shown that the transcriptional repressor Blimp-1 is most T-bet is the master initiating and maintaining http://www.jimmunol.org/ highly expressed in Th2 cells and directly represses transcription Th1 differentiation by driving high level of IFN-␥ expression. In of three genes that encode with critical roles in Th1 dif- addition to ifng, T-bet also induces additional genes in the Th1 ferentiation: IFN-␥, T-bet, and Bcl-6. Furthermore, the Th2 IgG1 program, including those encoding IL-12R␤2, HLX, and T-bet response to immunization with NP-KLH in alum is defective in itself. When ectopically expressed in Th2 cells, T-bet reprograms mice lacking Blimp-1 in their T cells. them to a Th1 phenotype (56) both by inducing a Th1 program and by interfering with the Th2 program. Interference with the Th2 Blimp-1 repression of ifng program occurs via a tyrosine phosphorylated form of T-bet that IFN-␥ is a critical cytokine for commitment to the Th1 lineage and interferes with the ability of the master Th2 transcription factor for establishment of feedback loops that reinforce Th1 lineage GATA-3 to bind its cognate sites (57). Thus, by repressing tbx21 by guest on September 24, 2021 commitment. Many activators but few repressors are known for and ifng, Blimp-1 attenuates expression of both the master Th1 ifng, which is induced by both TCR stimulation and cytokines transcription factor and the critical Th1 cytokine. IL-12 and IL-18. Th1 selective expression of IFN-␥ is regulated by Both STAT-1 and NF-␬B can induce tbx21 expression; how- T-bet binding in both the proximal promoter (43, 44) and in distal ever, studies in T cells have not yet confirmed their cis-activating regulatory elements of the ifng gene (46–48). Many activators of potential (6, 54, 58). Notch1 signaling activated by DCs can con- the ifng gene have been identified in the proximal promoter region, vert a basal repressor of T-bet, the recombination signal binding including STATs, NFAT, NF-␬B, AP-1, Oct-1, and T-bet (43, 44, protein Jk (RBJk), into an activator that will drive T-bet expression 49–51), but only Yin-Yang 1 (YY-1) has been shown to directly (59, 60). Additionally, the forkhead transcription factor Foxj1 can repress ifng gene expression (52). Later, however, this was found indirectly repress T-bet induction and Th1 differentiation by re- to result in transactivation of IFN-␥ expression in partnership with pressing NF-␬B activation (54). While GATA-3 can reverse Th1 nearby NFAT sites (49). Foxo3a and Foxj1 can indirectly repress differentiation and block T-bet activity, it also does so indirectly by IFN-␥ expression by directly repressing NF-␬B (53, 54). Thus, downregulating STAT-4 (61). We have shown that Blimp-1 binds Blimp-1 is one of the first direct repressors of ifng to be iden- directly to a region containing a conserved consensus binding site tified. Interestingly, the site recognized by Blimp-1, ϳ22 kb in the proximal promoter of the tbx21 gene and to regions con- upstream of the ifng transcription start site, is in a conserved taining conserved and nonconserved consensus sites in the first region that was shown to have histone 3 K4 methylation in Th1 intron. Binding of Blimp-1 to the tbx21 gene provides a mecha- cells but histone 3 K27 methylation in Th2 cells (55). Further- nism by which this repressor can directly regulate Th1/Th2 polar- more, this region has been shown to be required for IFN-␥ ization; however, future studies using retroviral transduction could expression in CD4 T cells, CD8 T cells, and NK cells (46), provide further insight into the role of Blimp-1 in this process. suggesting that Blimp-1 regulation of Th1 differentiation could extend beyond the setting of Th cells. Blimp-1 also increases IL-10 production, evidenced by lower Mutual repression by Blimp-1 and Bcl-6 IL-10 production in CKO CD4 T cells (Fig. 3) (10). However, this Data in this paper, in conjunction with previous reports (21, 31), probably involves regulatory as well as nonregulatory CD4 cells provide strong evidence that Blimp-1 and Bcl-6 mutually repress and is predicted to proceed by an indirect mechanism, since one another by direct repression in both T and B cells. In B cells, Blimp-1 is a transcriptional repressor in all situations studied thus this is important for mutually exclusive establishment of germinal far. Although we have not explored the mechanism of Blimp-1- center vs plasmacytic states of B cell differentiation (62, 63). Al- dependent IL-10 regulation in this study, it is likely that in addition though it was known that Bcl-6 directly repressed prdm1 in B cells to direct repression by Blimp-1, increased IL-10 is another mech- (31, 64), the data presented herein provide the first evidence that anism by which Blimp-1 attenuates IFN-␥ production. Blimp-1 directly represses bcl6 in B cells. 2346 Blimp-1 ATTENUATES Th1 DIFFERENTIATION

In CD4 T cells, our data support the notion that Blimp-1-de- Acknowledgments pendent repression of bcl6 is important for mutually exclusive Th1 We greatly appreciate the work of Kristie Gordon and members of the vs Th2 commitment. Bcl-6 is more highly expressed in Th1 cells Columbia FACS Facility for guidance and expertise in flow cytometry. We than in Th2 cells (65, 66) and plays a role in Th cells of attenuating thank members of the Calame laboratory for helpful advice. the expression or function of Th2 genes. Bcl-6-deficient mice have eosinophilic inflammation caused by elevated Th2 cytokines (29), Disclosures and Bcl-6 competes with STAT-6 for binding to Th2 genes (24, The authors have no financial conflicts of interest. 25). Bcl-6 also functions by other mechanisms in Th1 cells. It References binds to a silencer region in il5 and directly represses its transcrip- 1. Ansel, K. M., I. Djuretic, B. Tanasa, and A. Rao. 2006. Regulation of Th2 dif- tion (23). It induces degradation of GATA-3 by a poorly under- ferentiation and Il4 locus accessibility. Annu. Rev. Immunol. 24: 607–656. stood mechanism, leading to decreased expression of IL-10 (22). 2. Murphy, K. M., and S. L. Reiner. 2002. The lineage decisions of helper T cells. Finally, it directly represses prdm1, and we show in this work that Nat. Rev. Immunol. 2: 933–944. 3. Kim, J. M., and A. Rudensky. 2006. The role of the transcription factor Foxp3 in Blimp-1 represses Th1 genes ifng and tbx21. Thus, repression of the development of regulatory T cells. Immunol. Rev. 212: 86–98. bcl6 by Blimp-1 would block the ability of Bcl-6 to repress Th2 4. Mowen, K. A., and L. H. Glimcher. 2004. Signaling pathways in Th2 develop- ment. Immunol. Rev. 202: 203–222. gene expression. Reciprocally, repression of prdm1 by Bcl-6 5. Berenson, L. S., N. Ota, and K. M. Murphy. 2004. Issues in T-helper 1 devel- blocks the ability of Blimp-1 to repress Th1 gene expression. This opment: resolved and unresolved. Immunol. Rev. 202: 157–174. negative feedback loop between Blimp-1 and Bcl-6 in CD4 T cells 6. Mullen, A. C., F. A. High, A. S. Hutchins, H. W. Lee, A. V. Villarino, D. M. Livingston, A. L. Kung, N. Cereb, T. P. Yao, S. Y. Yang, and S. L. Reiner. is reminiscent in some ways of the feedback loop between T-bet 2001. Role of T-bet in commitment of TH1 cells before IL-12-dependent selec- and GATA-3, which also reinforces Th1 or Th2 differentiation tion. Science 292: 1907–1910. Downloaded from states. Further studies using retroviral transduction of Blimp-1 or 7. Avni, O., D. Lee, F. Macian, S. J. Szabo, L. H. Glimcher, and A. Rao. 2002. TH cell differentiation is accompanied by dynamic changes in histone acetylation of Bcl-6 in CD4 T cells will help to determine the extent to which Th cytokine genes. Nat. Immunol. 3: 643–651. lineage choice depends on direct mutual repression by these an- 8. Ouyang, W., M. Lohning, Z. Gao, M. Assenmacher, S. Ranganath, A. Radbruch, and K. M. Murphy. 2000. Stat6-independent GATA-3 autoactivation directs IL- tagonistic transcription factors. 4-independent Th2 development and commitment. Immunity 12: 27–37. 9. Kallies, A., E. D. Hawkins, G. T. Belz, D. Metcalf, M. Hommel, L. M. Corcoran, P. D. Hodgkin, and S. L. Nutt. 2006. Transcriptional repressor Blimp-1 is essen- http://www.jimmunol.org/ Importance of Blimp-1 in Th2 differentiation tial for T cell homeostasis and self-tolerance. Nat. Immunol. 7: 466–474. 10. Martins, G. A., L. Cimmino, M. Shapiro-Shelef, M. Szabolcs, A. Herron, We and others have shown that IL-4 signaling increases Blimp-1 E. Magnusdottir, and K. Calame. 2006. Transcriptional repressor Blimp-1 regu- mRNA expression (Fig. 2) (19). These data, along with the finding lates T cell homeostasis and function. Nat. Immunol. 7: 457–465. 11. Savitsky, D., L. Cimmino, T. Kuo, G. A. Martins, and K. Calame. 2007. Multiple that Blimp-1 mRNA is highest in Th2 cells (Fig. 2A), support a roles for Blimp-1 in B and T lymphocytes. Adv. Exp. Med. Biol. 596: 9–30. particular role for Blimp-1 in Th2 cells. 12. Calame, K. L. 2001. Plasma cells: finding new light at the end of B cell devel- opment. Nat. Immunol. 2: 1103–1108. Blimp-1-dependent repression of ifng, tbx21, and bcl6 provide a 13. Chang, D. H., G. Cattoretti, and K. L. Calame. 2002. The dynamic expression molecular mechanism for how Blimp-1 may function during Th2 pattern of B lymphocyte induced maturation protein-1 (Blimp-1) during mouse polarization to repress Th1 genes. Indeed, an important role for embryonic development. Mech. Dev. 117: 305–309.

14. Hayashi, K., S. M. de Sousa Lopes, and M. A. Surani. 2007. Germ cell specifi- by guest on September 24, 2021 Blimp-1 in Th2 responses is supported by our demonstration that cation in mice. Science 316: 394–396. the response of CKO mice to immunization with NP-KLH in alum, 15. Ohinata, Y., B. Payer, D. O’Carroll, K. Ancelin, Y. Ono, M. Sano, S. C. Barton, an adjuvant causing a predominantly IgG1 Th2 response (33), is T. Obukhanych, M. Nussenzweig, A. Tarakhovsky, et al. 2005. Blimp1 is a critical determinant of the germ cell lineage in mice. Nature 436: 207–213. defective, showing reduced IgG1 and increased IgM. It is also 16. Chang, D. H., C. Angelin-Duclos, and K. Calame. 2000. BLIMP-1: trigger for consistent with the acute colitis the CKO mice develop, because differentiation of myeloid lineage. Nat. Immunol. 1: 169–176. 17. Magnusdottir, E., S. Kalachikov, K. Mizukoshi, D. Savitsky, A. Ishida- colitis is usually an aberrant Th1 response. However, since CKO Yamamoto, A. A. Panteleyev, and K. Calame. 2007. Epidermal terminal differ- mice show abnormally activated T cells as early as 4–5 wk it is entiation depends on B lymphocyte-induced maturation protein-1. Proc. Natl. possible that they were already predisposed to a Th1-like response, Acad. Sci. USA 104: 14988–14993. 18. Horsley, V., D. O’Carroll, R. Tooze, Y. Ohinata, M. Saitou, T. Obukhanych, having an environment that disfavors Th2 responses. M. Nussenzweig, A. Tarakhovsky, and E. Fuchs. 2006. Blimp1 defines a pro- However, Blimp-1 is not strictly required for Th2 differentiation genitor population that governs cellular input to the sebaceous gland. Cell 126: because Blimp-1-deficient CD4 cells develop normally into IL-4- 597–609. 19. Gong, D., and T. R. Malek. 2007. Cytokine-dependent Blimp-1 expression in producing cells when stimulated with strongly polarizing condi- activated T cells inhibits IL-2 production. J. Immunol. 178: 242–252. tions (IL-4 and anti-IFN-␥) in vitro (Fig. 3A). Rather, we think that 20. Yan, J., J. Jiang, C. A. Lim, Q. Wu, H. H. Ng, and K. C. Chin. 2007. BLIMP1 regulates cell growth through repression of p53 transcription. Proc. Natl. Acad. the role of Blimp-1 is to attenuate Th1 responses, thereby enhanc- Sci. USA 104: 1841–1846. ing the commitment and establishment of Th2 responses in vivo. In 21. Shaffer, A. L., K. I. Lin, T. C. Kuo, X. Yu, E. M. Hurt, A. Rosenwald, this way its role is similar to that of Bcl-6. Bcl-6 KO CD4 T cells J. M. Giltnane, L. Yang, H. Zhao, K. Calame, and L. M. Staudt. 2002. Blimp-1 orchestrates plasma cell differentiation by extinguishing the mature B cell gene (67) polarize normally in vitro in contrast to their spontaneous expression program. Immunity 17: 51–62. cytokine skewing toward Th2 cell differentiation in vivo. In vitro 22. Kusam, S., L. M. Toney, H. Sato, and A. L. Dent. 2003. Inhibition of Th2 polarization bypasses the effect of imbalances in endogenous cy- differentiation and GATA-3 expression by BCL-6. J. Immunol. 170: 2435–2441. 23. Arima, M., H. Toyama, H. Ichii, S. Kojima, S. Okada, M. Hatano, G. Cheng, tokine production; therefore, unlike the obligate requirement for M. Kubo, T. Fukuda, and T. Tokuhisa. 2002. A putative silencer element in the STAT-1 and STAT-6 or T-bet and GATA-3 to potentiate cytokine IL-5 gene recognized by Bcl6. J. Immunol. 169: 829–836. 24. Harris, M. B., C. C. Chang, M. T. Berton, N. N. Danial, J. Zhang, D. Kuehner, signaling, Blimp-1 and Bcl-6 are repressors that attenuate Th dif- B. H. Ye, M. Kvatyuk, P. P. Pandolfi, G. Cattoretti, et al. 1999. Transcriptional ferentiation during differentiation in vivo when signals for Th1 vs repression of Stat6-dependent interleukin-4-induced genes by BCL-6: specific Th2 polarization are likely to be more subtle. regulation of I␧ transcription and immunoglobulin E switching. Mol. Cell. Biol. 19: 7264–7275. In summary, we have shown that Blimp-1 attenuates Th1 dif- 25. Harris, M. B., J. Mostecki, and P. B. Rothman. 2005. Repression of an interleu- ferentiation by directly repressing transcription of the signature kin-4-responsive promoter requires cooperative BCL-6 function. J. Biol. Chem. cytokine, IFN-␥, and the master regulator of Th1 cell development, 280: 13114–13121. 26. Shapiro-Shelef, M., K. I. Lin, L. J. McHeyzer-Williams, J. Liao, T-bet. Additionally, Blimp-1 also represses Bcl-6, thus expanding M. G. McHeyzer-Williams, and K. Calame. 2003. Blimp-1 is required for the from B cells to T cells the settings in which these two transcrip- formation of immunoglobulin secreting plasma cells and pre-plasma memory B cells. Immunity 19: 607–620. tional repressors can antagonize one another to promote alternative 27. Lee, P. P., D. R. Fitzpatrick, C. Beard, H. K. Jessup, S. Lehar, K. W. Makar, cell fates. M. Perez-Melgosa, M. T. Sweetser, M. S. Schlissel, S. Nguyen, S. R. Cherry, The Journal of Immunology 2347

et al. 2001. A critical role for Dnmt1 and DNA methylation in T cell develop- elements that positively regulate IFN-␥ expression in T cells. Proc. Natl. Acad. ment, function, and survival. Immunity 15: 763–774. Sci. USA 101: 12622–12627. 28. Takahama, Y., K. Ohishi, Y. Tokoro, T. Sugawara, Y. Yoshimura, M. Okabe, 49. Sweetser, M. T., T. Hoey, Y. L. Sun, W. M. Weaver, G. A. Price, and T. Kinoshita, and J. Takeda. 1998. Functional competence of T cells in the ab- C. B. Wilson. 1998. The roles of nuclear factor of activated T cells and ying-yang sence of glycosylphosphatidylinositol-anchored proteins caused by T cell-specific 1 in activation-induced expression of the interferon-␥ promoter in T cells. J. Biol. disruption of the Pig-a gene. Eur. J. Immunol. 28: 2159–2166. Chem. 273: 34775–34783. 29. Dent, A. L., A. L. Shaffer, X. Yu, D. Allman, and L. M. Staudt. 1997. Control of 50. Sica, A., L. Dorman, V. Viggiano, M. Cippitelli, P. Ghosh, N. Rice, and inflammation, cytokine expression, and germinal center formation by BCL-6. H. A. Young. 1997. Interaction of NF-␬B and NFAT with the interferon-␥ pro- Science 276: 589–592. moter. J. Biol. Chem. 272: 30412–30420. 30. Palomero, T., W. K. Lim, D. T. Odom, M. L. Sulis, P. J. Real, A. Margolin, 51. Xu, X., Y. L. Sun, and T. Hoey. 1996. Cooperative DNA binding and sequence- K. C. Barnes, J. O’Neil, D. Neuberg, A. P. Weng, et al. 2006. NOTCH1 directly selective recognition conferred by the STAT amino-terminal domain. Science regulates c-MYC and activates a feed-forward-loop transcriptional network pro- 273: 794–797. moting leukemic cell growth. Proc. Natl. Acad. Sci. USA 103: 18261–18266. 52. Ye, J., M. Cippitelli, L. Dorman, J. R. Ortaldo, and H. A. Young. 1996. The 31. Tunyaplin, C., A. L. Shaffer, C. D. Angelin-Duclos, X. Yu, L. M. Staudt, and nuclear factor YY1 suppresses the human gamma interferon promoter through K. L. Calame. 2004. Direct repression of prdm1 by Bcl-6 inhibits plasmacytic two mechanisms: inhibition of AP1 binding and activation of a silencer element. differentiation. J. Immunol. 173: 1158–1165. Mol. Cell. Biol. 16: 4744–4753. 32. Bird, J. J., D. R. Brown, A. C. Mullen, N. H. Moskowitz, M. A. Mahowald, 53. Lin, L., J. D. Hron, and S. L. Peng. 2004. Regulation of NF-␬B, Th activation, J. R. Sider, T. F. Gajewski, C. R. Wang, and S. L. Reiner. 1998. Helper T cell and autoinflammation by the forkhead transcription factor Foxo3a. Immunity 21: differentiation is controlled by the cell cycle. Immunity 9: 229–237. 203–213. 33. McKee, A. S., M. W. Munks, and P. Marrack. 2007. How do adjuvants work? 54. Lin, L., M. S. Spoor, A. J. Gerth, S. L. Brody, and S. L. Peng. 2004. Modulation Important considerations for new generation adjuvants. Immunity 27: 687–690. of Th1 activation and inflammation by the NF-␬B repressor Foxj1. Science 303: 34. Ise, W., M. Totsuka, Y. Sogawa, A. Ametani, S. Hachimura, T. Sato, 1017–1020. Y. Kumagai, S. Habu, and S. Kaminogawa. 2002. Naive CD4ϩ T cells exhibit 55. Schoenborn, J. R., M. O. Dorschner, M. Sekimata, D. M. Santer, M. Shnyreva, distinct expression patterns of cytokines and cell surface molecules on their pri- D. R. Fitzpatrick, J. A. Stamatoyannopoulos, and C. B. Wilson. 2007. Compre- mary responses to varying doses of antigen. J. Immunol. 168: 3242–3250. hensive epigenetic profiling identifies multiple distal regulatory elements direct- 35. Porter, C. M., and N. A. Clipstone. 2002. Sustained NFAT signaling promotes a ing transcription of the gene encoding interferon-␥. Nat. Immunol. 8: 732–742. Downloaded from Th1-like pattern of gene expression in primary murine CD4ϩ T cells. J. Immunol. 56. Szabo, S. J., S. T. Kim, G. L. Costa, X. Zhang, C. G. Fathman, and 168: 4936–4945. L. H. Glimcher. 2000. A novel transcription factor, T-bet, directs Th1 lineage 36. Glimcher, L. H., and K. M. Murphy. 2000. Lineage commitment in the immune commitment. Cell 100: 655–669. system: the T helper lymphocyte grows up. Genes Dev. 14: 1693–1711. 57. Hwang, E. S., S. J. Szabo, P. L. Schwartzberg, and L. H. Glimcher. 2005. T 37. Lee, G. R., S. T. Kim, C. G. Spilianakis, P. E. Fields, and R. A. Flavell. 2006. fate specified by kinase-mediated interaction of T-bet with GATA-3. helper cell differentiation: regulation by cis elements and epigenetics. Immunity Science 307: 430–433. 24: 369–379. 58. Murphy, K. M., W. Ouyang, S. J. Szabo, N. G. Jacobson, M. L. Guler,

38. Dent, A. L., J. Hu-Li, W. E. Paul, and L. M. Staudt. 1998. T helper type 2 J. D. Gorham, U. Gubler, and T. L. Murphy. 1999. T helper differentiation pro- http://www.jimmunol.org/ inflammatory disease in the absence of interleukin 4 and transcription factor ceeds through Stat1-dependent, Stat4-dependent and Stat4-independent phases. STAT6. Proc. Natl. Acad. Sci. USA 95: 13823–13828. Curr. Top. Microbiol. Immunol. 238: 13–26. 39. Zhou, G., and S. J. Ono. 2005. Induction of BCL-6 gene expression by interfer- 59. Minter, L. M., D. M. Turley, P. Das, H. M. Shin, I. Joshi, R. G. Lawlor, on-␥ and identification of an IRE in exon I. Exp. Mol. Pathol. 78: 25–35. O. H. Cho, T. Palaga, S. Gottipati, J. C. Telfer, et al. 2005. Inhibitors of ␥-secre- 40. Dubchak, I., and D. V. Ryaboy. 2006. VISTA family of computational tools for tase block in vivo and in vitro T helper type 1 polarization by preventing Notch comparative analysis of DNA sequences and whole genomes. Methods Mol. Biol. upregulation of Tbx21. Nat. Immunol. 6: 680–688. 338: 69–89. 60. Maekawa, Y., S. Tsukumo, S. Chiba, H. Hirai, Y. Hayashi, H. Okada, 41. Loots, G. G., and I. Ovcharenko. 2004. rVISTA 2.0: evolutionary analysis of K. Kishihara, and K. Yasutomo. 2003. Delta1-Notch3 interactions bias the func- transcription factor binding sites. Nucleic Acids Res. 32: W217–W221. tional differentiation of activated CD4ϩ T cells. Immunity 19: 549–559. 42. Kuo, T. C., and K. L. Calame. 2004. B lymphocyte-induced maturation protein 61. Usui, T., R. Nishikomori, A. Kitani, and W. Strober. 2003. GATA-3 suppresses (Blimp)-1, IFN regulatory factor (IRF)-1, and IRF-2 can bind to the same regu- Th1 development by downregulation of Stat4 and not through effects on IL-

latory sites. J. Immunol. 173: 5556–5563. 12R␤2 chain or T-bet. Immunity 18: 415–428. by guest on September 24, 2021 43. Teixeira, L. K., B. P. Fonseca, A. Vieira-de-Abreu, B. A. Barboza, B. K. Robbs, 62. Johnson, K., M. Shapiro-Shelef, C. Tunyaplin, and K. Calame. 2005. Regulatory P. T. Bozza, and J. P. Viola. 2005. IFN-␥ production by CD8ϩ T cells depends events in early and late B-cell differentiation. Mol. Immunol. 42: 749–761. on NFAT1 transcription factor and regulates Th differentiation. J. Immunol. 175: 63. Shapiro-Shelef, M., and K. Calame. 2005. Regulation of plasma-cell develop- 5931–5939. ment. Nat. Rev. Immunol. 5: 230–242. 44. Soutto, M., F. Zhang, B. Enerson, Y. Tong, M. Boothby, and T. M. Aune. 2002. 64. Dent, A. L., F. H. Vasanwala, and L. M. Toney. 2002. Regulation of gene ex- A minimal IFN-␥ promoter confers Th1 selective expression. J. Immunol. 169: pression by the proto-oncogene BCL-6. Crit. Rev. Oncol. Hematol. 41: 1–9. 4205–4212. 65. Lund, R. J., Z. Chen, J. Scheinin, and R. Lahesmaa. 2004. Early target genes of 45. Wilson, C. B., and J. Schoenborn. 2006. BACing up the interferon-␥ locus. Im- IL-12 and STAT4 signaling in Th cells. J. Immunol. 172: 6775–6782. munity 25: 691–693. 66. Lund, R. J., E. K. Ylikoski, T. Aittokallio, O. Nevalainen, and R. Lahesmaa. 46. Hatton, R. D., L. E. Harrington, R. J. Luther, T. Wakefield, K. M. Janowski, 2003. Kinetics and STAT4- or STAT6-mediated regulation of genes involved in J. R. Oliver, R. L. Lallone, K. M. Murphy, and C. T. Weaver. 2006. A distal lymphocyte polarization to Th1 and Th2 cells. Eur. J. Immunol. 33: 1105–1116. conserved sequence element controls Ifng gene expression by T cells and NK 67. Dent, A. L., T. M. Doherty, W. E. Paul, A. Sher, and L. M. Staudt. 1999. BCL- cells. Immunity 25: 717–729. 6-deficient mice reveal an IL-4-independent, STAT6-dependent pathway that 47. Lee, D. U., O. Avni, L. Chen, and A. Rao. 2004. A distal enhancer in the inter- controls susceptibility to infection by Leishmania major. J. Immunol. 163: feron-␥ (IFN-␥) locus revealed by genome sequence comparison. J. Biol. Chem. 2098–2103. 279: 4802–4810. 68. Angelin-Duclos, C., K. Johnson, J. Liao, and K. Calame. 2002. An interfering 48. Shnyreva, M., W. M. Weaver, M. Blanchette, S. L. Taylor, M. Tompa, form of Blimp-1 increases IgM secreting plasma cells and blocks maturation of D. R. Fitzpatrick, and C. B. Wilson. 2004. Evolutionarily conserved sequence peripheral B cells. Eur. J. Immunol. 32: 3765–3775.